Blop

* Add accessors for Snippet and HighlightSection

And add an example of custom highlighter

* Remove inline(always) and unnecessary empty lines

.gitignore

@@ -1,3 +1,4 @@
+tantivy.iml
 *.swp
 target
 target/debug

CHANGELOG.md

@@ -1,10 +1,24 @@
+Tantivy 0.8.1
+=====================
+*No change in the index format*
+- API Breaking change in the collector API. (@jwolfe, @fulmicoton)
+- Multithreaded search (@jwolfe, @fulmicoton) 
+
+
+Tantivy 0.7.1
+=====================
+*No change in the index format*
+- Bugfix: NGramTokenizer panics on non ascii chars
+- Added a space usage API
 
 Tantivy 0.7
 =====================
 - Skip data for doc ids and positions (@fulmicoton),
   greatly improving performance
 - Tantivy error now rely on the failure crate (@drusellers)
-
+- Added support for `AND`, `OR`, `NOT` syntax in addition to the `+`,`-` syntax
+- Added a snippet generator with highlight (@vigneshsarma, @fulmicoton)
+- Added a `TopFieldCollector` (@pentlander)
 
 Tantivy 0.6.1
 =========================

Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "tantivy"
-version = "0.7.0-dev"
+version = "0.8.0-dev"
 authors = ["Paul Masurel <paul.masurel@gmail.com>"]
 license = "MIT"
 categories = ["database-implementations", "data-structures"]
@@ -12,10 +12,9 @@ readme = "README.md"
 keywords = ["search", "information", "retrieval"]
 
 [dependencies]
-base64 = "0.9.1"
+base64 = "0.10.0"
 byteorder = "1.0"
 lazy_static = "1"
-tinysegmenter = "0.1.0"
 regex = "1.0"
 fst = {version="0.3", default-features=false}
 fst-regex = { version="0.2" }
@@ -33,9 +32,8 @@ num_cpus = "1.2"
 itertools = "0.7"
 levenshtein_automata = {version="0.1", features=["fst_automaton"]}
 bit-set = "0.5"
-uuid = { version = "0.6", features = ["v4", "serde"] }
-crossbeam = "0.4"
-crossbeam-channel = "0.2"
+uuid = { version = "0.7", features = ["v4", "serde"] }
+crossbeam = "0.5"
 futures = "0.1"
 futures-cpupool = "0.1"
 owning_ref = "0.4"
@@ -48,24 +46,36 @@ census = "0.1"
 fnv = "1.0.6"
 owned-read = "0.4"
 failure = "0.1"
+htmlescape = "0.3.1"
+fail = "0.2"
+scoped-pool = "1.0"
+aho-corasick = "0.6"
 
 [target.'cfg(windows)'.dependencies]
 winapi = "0.2"
 
 [dev-dependencies]
-rand = "0.5"
+rand = "0.6"
+maplit = "1"
 
 [profile.release]
 opt-level = 3
 debug = false
-lto = true
 debug-assertions = false
 
+[profile.test]
+debug-assertions = true
+overflow-checks = true
+
 [features]
-default = ["mmap"]
+# by default no-fail is disabled. We manually enable it when running test.
+default = ["mmap", "no_fail"]
 mmap = ["fst/mmap", "atomicwrites"]
 lz4-compression = ["lz4"]
+no_fail = ["fail/no_fail"]
+unstable = [] # useful for benches.
 
 [badges]
 travis-ci = { repository = "tantivy-search/tantivy" }
 
+

README.md

@@ -4,6 +4,7 @@
 [![Join the chat at https://gitter.im/tantivy-search/tantivy](https://badges.gitter.im/tantivy-search/tantivy.svg)](https://gitter.im/tantivy-search/tantivy?utm_source=badge&utm_medium=badge&utm_campaign=pr-badge&utm_content=badge)
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
 [![Build status](https://ci.appveyor.com/api/projects/status/r7nb13kj23u8m9pj/branch/master?svg=true)](https://ci.appveyor.com/project/fulmicoton/tantivy/branch/master)
+[![Say Thanks!](https://img.shields.io/badge/Say%20Thanks-!-1EAEDB.svg)](https://saythanks.io/to/fulmicoton)
 
 ![Tantivy](https://tantivy-search.github.io/logo/tantivy-logo.png)
 
@@ -20,7 +21,7 @@
 
 **Tantivy** is a **full text search engine library** written in rust.
 
-It is closer to Lucene than to Elastic Search and Solr in the sense it is not
+It is closer to [Apache Lucene](https://lucene.apache.org/) than to [Elastic Search](https://www.elastic.co/products/elasticsearch) and [Apache Solr](https://lucene.apache.org/solr/) in the sense it is not
 an off-the-shelf search engine server, but rather a crate that can be used
 to build such a search engine.
 
@@ -32,8 +33,8 @@ Tantivy is, in fact, strongly inspired by Lucene's design.
 - Fast (check out the :racehorse: :sparkles: [benchmark](https://tantivy-search.github.io/bench/) :sparkles: :racehorse:)
 - Tiny startup time (<10ms), perfect for command line tools
 - BM25 scoring (the same as lucene)
-- Basic query language (`+michael +jackson`)
-- Phrase queries search (\"michael jackson\"`)
+- Natural query language `(michael AND jackson) OR "king of pop"`
+- Phrase queries search (`"michael jackson"`)
 - Incremental indexing
 - Multithreaded indexing (indexing English Wikipedia takes < 3 minutes on my desktop)
 - Mmap directory
@@ -43,12 +44,14 @@ Tantivy is, in fact, strongly inspired by Lucene's design.
 - LZ4 compressed document store
 - Range queries
 - Faceted search
-- Configurable indexing (optional term frequency and position indexing
+- Configurable indexing (optional term frequency and position indexing)
 - Cheesy logo with a horse
 
 # Non-features
 
-- Distributed search and will not be in the scope of tantivy.
+- Distributed search is out of the scope of tantivy. That being said, tantivy is meant as a
+library upon which one could build a distributed search. Serializable/mergeable collector state for instance, 
+are within the scope of tantivy.
 
 
 # Supported OS and compiler
@@ -77,6 +80,10 @@ To check out and run tests, you can simply run :
     cd tantivy
     cargo build
 
+## Running tests
+
+Some tests will not run with just `cargo test` because of `fail-rs`.
+To run the tests exhaustively, run `./run-tests.sh`. 
 
 # Contribute
 

appveyor.yml

@@ -18,5 +18,5 @@ install:
 build: false
 
 test_script:
-  - REM SET RUST_LOG=tantivy,test & cargo test --verbose
-  - REM SET RUST_BACKTRACE=1 & cargo build --examples
+  - REM SET RUST_LOG=tantivy,test & cargo test --verbose --no-default-features --features mmap -- --test-threads 1
+  - REM SET RUST_BACKTRACE=1 & cargo build --examples

ci/script.sh

@@ -11,12 +11,11 @@ main() {
     else
         echo "Build"
         cross build --target $TARGET
-        cross build --target $TARGET --release
         if [ ! -z $DISABLE_TESTS ]; then
             return
         fi
         echo "Test"
-        cross test --target $TARGET
+        cross test --target $TARGET --no-default-features --features mmap -- --test-threads 1
     fi
     for example in $(ls examples/*.rs)
     do

doc/src/SUMMARY.md

@@ -9,6 +9,7 @@
 - [Facetting](./facetting.md)
 - [Innerworkings](./innerworkings.md)
   - [Inverted index](./inverted_index.md)
+- [Best practise](./inverted_index.md)
 
 [Frequently Asked Questions](./faq.md)
 [Examples](./examples.md)

doc/src/avant-propos.md

@@ -2,8 +2,8 @@
 
 > Tantivy is a **search** engine **library** for Rust.
 
-If you are familiar with Lucene, tantivy is heavily inspired by Lucene's design and
-they both have the same scope and targetted users.
+If you are familiar with Lucene, it's an excellent approximation to consider tantivy as Lucene for rust. tantivy is heavily inspired by Lucene's design and
+they both have the same scope and targetted use cases.
 
 If you are not familiar with Lucene, let's break down our little tagline.
 
@@ -17,15 +17,18 @@ relevancy, collapsing, highlighting, spatial search.
   experience. But keep in mind this is just a toolbox.
   Which bring us to the second keyword...
 
-- **Library** means that you will have to write code. tantivy is not an *all-in-one* server solution.
-  
-  Sometimes a functionality will not be available in tantivy because it is too specific to your use case. By design, tantivy should make it possible to extend
-  the available set of features using the existing rock-solid datastructures. 
+- **Library** means that you will have to write code. tantivy is not an *all-in-one* server solution like elastic search for instance.
 
-   Most frequently this will mean writing your own `Collector`, your own `Scorer` or your own 
-   `Tokenizer/TokenFilter`... But some of your requirement may also be related to 
-   architecture or operations. For instance, you may want to build a large corpus on Hadoop,
-   fine-tune the merge policy to keep your index sharded in a time-wise fashion, or you may want
-   to convert and existing index from a different format.
-   
-   Tantivy exposes its API to do all of these things.
+  Sometimes a functionality will not be available in tantivy because it is too
+  specific to your use case. By design, tantivy should make it possible to extend
+  the available set of features using the existing rock-solid datastructures.
+
+  Most frequently this will mean writing your own `Collector`, your own `Scorer` or your own
+  `TokenFilter`... Some of your requirements may also be related to
+  something closer to architecture or operations. For instance, you may
+  want to build a large corpus on Hadoop, fine-tune the merge policy to keep your
+  index sharded in a time-wise fashion, or you may want to convert and existing
+  index from a different format.
+
+  Tantivy exposes a lot of low level API to do all of these things.
+  

doc/src/basis.md

@@ -2,47 +2,76 @@
 
 ## Straight from disk
 
-By default, tantivy accesses its data using its `MMapDirectory`. 
-While this design has some downsides, this greatly simplifies the source code of tantivy, 
-and entirely delegates the caching to the OS.
+Tantivy accesses its data using an abstracting trait called `Directory`.
+In theory, one can come and override the data access logic. In practise, the
+trait somewhat assumes that your data can be mapped to memory, and tantivy
+seems deeply married to using `mmap` for its io [^1], and the only persisting
+directory shipped with tantivy is the `MmapDirectory`.
 
-`tantivy` works entirely (or almost) by directly reading the datastructures as they are layed on disk.
-As a result, the act of opening an indexing does not involve loading different datastructures 
-from the disk into random access memory : starting a process, opening an index, and performing a query 
-can typically be done in a matter of milliseconds. 
+While this design has some downsides, this greatly simplifies the source code of
+tantivy. Caching is also entirely delegated to the OS.
 
-This is an interesting property for a command line search engine, or for some multi-tenant log search engine.
-Spawning a new process for each new query can be a perfectly sensible solution in some use case.
+`tantivy` works entirely (or almost) by directly reading the datastructures as they are layed on disk. As a result, the act of opening an indexing does not involve loading different datastructures from the disk into random access memory : starting a process, opening an index, and performing your first query can typically be done in a matter of milliseconds.
+
+This is an interesting property for a command line search engine, or for some multi-tenant log search engine : spawning a new process for each new query can be a perfectly sensible solution in some use case.
 
 In later chapters, we will discuss tantivy's inverted index data layout.
-One key take away is that to achieve great performance, search indexes are extremely compact. 
+One key take away is that to achieve great performance, search indexes are extremely compact.
 Of course this is crucial to reduce IO, and ensure that as much of our index can sit in RAM.
 
-Also, whenever possible the data is accessed sequentially. Of course, this is an amazing property when tantivy needs to access
-the data from your spinning hard disk, but this is also a great property when working with `SSD` or `RAM`, 
-as it makes our read patterns very predictable for the CPU.
+Also, whenever possible its data is accessed sequentially. Of course, this is an amazing property when tantivy needs to access the data from your spinning hard disk, but this is also
+critical for performance, if your data is read from and an `SSD` or even already in your pagecache.
 
 
 ## Segments, and the log method
 
-That kind compact layout comes at one cost: it prevents our datastructures from being dynamic.
-In fact, a trait called `Directory` is in charge of abstracting all of tantivy's data access
-and its API does not even allow editing these file once they are written.
+That kind of compact layout comes at one cost: it prevents our datastructures from being dynamic.
+In fact, the `Directory` trait does not even allow you to modify part of a file.
 
 To allow the addition / deletion of documents, and create the illusion that
-your index is dynamic (i.e.: adding and deleting documents), tantivy uses a common database trick sometimes
-referred to as the *log method*.
+your index is dynamic (i.e.: adding and deleting documents), tantivy uses a common database trick sometimes referred to as the *log method*.
 
-Let's forget about deletes for a moment. As you add documents, these documents are processed and stored in 
-a dedicated datastructure, in a `RAM` buffer. This datastructure is designed to be dynamic but 
-cannot be accessed for search. As you add documents, this buffer will reach its capacity and tantivy will
-transparently stop adding document to it and start converting this datastructure to its final
-read-only format on disk. Once written, an brand empty buffer is available to resume adding documents. 
+Let's forget about deletes for a moment.
+
+As you add documents, these documents are processed and stored in a dedicated datastructure, in a `RAM` buffer. This datastructure is not ready for search, but it is useful to receive your data and rearrange it very rapidly.
+
+As you add documents, this buffer will reach its capacity and tantivy will transparently stop adding document to it and start converting this datastructure to its final read-only format on disk. Once written, an brand empty buffer is available to resume adding documents.
 
 The resulting chunk of index obtained after this serialization is called a `Segment`.
 
-> A segment is a self-contained atomic piece of index. It is identified with a UUID, and all of its files
-are identified using the naming scheme : `<UUID>.*`.
+> A segment is a self-contained atomic piece of index. It is identified with a UUID, and all of its files are identified using the naming scheme : `<UUID>.*`.
+
+Which brings us to the nature of a tantivy `Index`.
+
+> A tantivy `Index` is a collection of `Segments`.
+
+Physically, this really just means and index is a bunch of segment files in a given `Directory`,
+linked together by a `meta.json` file. This transparency can become extremely handy
+to get tantivy to fit your use case:
+
+*Example 1* You could for instance use hadoop to build a very large search index in a timely manner, copy all of the resulting segment files in the same directory and edit the `meta.json` to get a functional index.[^2]
+
+*Example 2* You could also disable your merge policy and enforce daily segments. Removing data after one week can then be done very efficiently by just editing the `meta.json` and deleting the files associated to segment `D-7`.
 
 
-> A tantivy `Index` is a collection of `Segments`.
+
+
+
+# Merging
+
+As you index more and more data, your index will accumulate more and more segments.
+Having a lot of small segments is not really optimal. There is a bit of redundancy in having
+all these term dictionary. Also when searching, we will need to do term lookups as many times as we have segments.  It can hurt search performance a bit.
+
+That's where merging or compacting comes into place. Tantivy will continuously consider merge
+opportunities and start merging segments in the background.
+
+
+# Indexing throughput, number of indexing threads
+
+
+
+
+[^1]: This may eventually change.
+
+[^2]: Be careful however. By default these files will not be considered as *managed* by tantivy. This means they will never be garbage collected by tantivy, regardless of whether they become obsolete or not.

doc/src/examples.md

@@ -1 +1,3 @@
 # Examples
+
+- [Basic search](/examples/basic_search.html)

examples/basic_search.rs

@@ -10,17 +10,17 @@
 // - search for the best document matchings "sea whale"
 // - retrieve the best document original content.
 
-
 extern crate tempdir;
 
 // ---
 // Importing tantivy...
 #[macro_use]
 extern crate tantivy;
-use tantivy::collector::TopCollector;
+use tantivy::collector::TopDocs;
 use tantivy::query::QueryParser;
 use tantivy::schema::*;
 use tantivy::Index;
+use tempdir::TempDir;
 
 fn main() -> tantivy::Result<()> {
     // Let's create a temporary directory for the
@@ -35,7 +35,7 @@ fn main() -> tantivy::Result<()> {
     // be indexed".
 
     // first we need to define a schema ...
-    let mut schema_builder = SchemaBuilder::default();
+    let mut schema_builder = Schema::builder();
 
     // Our first field is title.
     // We want full-text search for it, and we also want
@@ -213,15 +213,10 @@ fn main() -> tantivy::Result<()> {
     //
     // We are not interested in all of the documents but
     // only in the top 10. Keeping track of our top 10 best documents
-    // is the role of the TopCollector.
-    let mut top_collector = TopCollector::with_limit(10);
+    // is the role of the TopDocs.
 
     // We can now perform our query.
-    searcher.search(&*query, &mut top_collector)?;
-
-    // Our top collector now contains the 10
-    // most relevant doc ids...
-    let doc_addresses = top_collector.docs();
+    let top_docs = searcher.search(&query, &TopDocs::with_limit(10))?;
 
     // The actual documents still need to be
     // retrieved from Tantivy's store.
@@ -230,14 +225,10 @@ fn main() -> tantivy::Result<()> {
     // the document returned will only contain
     // a title.
 
-    for doc_address in doc_addresses {
-        let retrieved_doc = searcher.doc(&doc_address)?;
+    for (_score, doc_address) in top_docs {
+        let retrieved_doc = searcher.doc(doc_address)?;
         println!("{}", schema.to_json(&retrieved_doc));
     }
 
-
     Ok(())
 }
-
-
-use tempdir::TempDir;

examples/custom_collector.rs

@@ -0,0 +1,187 @@
+// # Custom collector example
+//
+// This example shows how you can implement your own
+// collector. As an example, we will compute a collector
+// that computes the standard deviation of a given fast field.
+//
+// Of course, you can have a look at the tantivy's built-in collectors
+// such as the `CountCollector` for more examples.
+
+extern crate tempdir;
+
+// ---
+// Importing tantivy...
+#[macro_use]
+extern crate tantivy;
+use tantivy::collector::{Collector, SegmentCollector};
+use tantivy::fastfield::FastFieldReader;
+use tantivy::query::QueryParser;
+use tantivy::schema::Field;
+use tantivy::schema::{Schema, FAST, INT_INDEXED, TEXT};
+use tantivy::Index;
+use tantivy::SegmentReader;
+
+#[derive(Default)]
+struct Stats {
+    count: usize,
+    sum: f64,
+    squared_sum: f64,
+}
+
+impl Stats {
+    pub fn count(&self) -> usize {
+        self.count
+    }
+
+    pub fn mean(&self) -> f64 {
+        self.sum / (self.count as f64)
+    }
+
+    fn square_mean(&self) -> f64 {
+        self.squared_sum / (self.count as f64)
+    }
+
+    pub fn standard_deviation(&self) -> f64 {
+        let mean = self.mean();
+        (self.square_mean() - mean * mean).sqrt()
+    }
+
+    fn non_zero_count(self) -> Option<Stats> {
+        if self.count == 0 {
+            None
+        } else {
+            Some(self)
+        }
+    }
+}
+
+struct StatsCollector {
+    field: Field,
+}
+
+impl StatsCollector {
+    fn with_field(field: Field) -> StatsCollector {
+        StatsCollector { field }
+    }
+}
+
+impl Collector for StatsCollector {
+    // That's the type of our result.
+    // Our standard deviation will be a float.
+    type Fruit = Option<Stats>;
+
+    type Child = StatsSegmentCollector;
+
+    fn for_segment(
+        &self,
+        _segment_local_id: u32,
+        segment: &SegmentReader,
+    ) -> tantivy::Result<StatsSegmentCollector> {
+        let fast_field_reader = segment.fast_field_reader(self.field)?;
+        Ok(StatsSegmentCollector {
+            fast_field_reader,
+            stats: Stats::default(),
+        })
+    }
+
+    fn requires_scoring(&self) -> bool {
+        // this collector does not care about score.
+        false
+    }
+
+    fn merge_fruits(&self, segment_stats: Vec<Option<Stats>>) -> tantivy::Result<Option<Stats>> {
+        let mut stats = Stats::default();
+        for segment_stats_opt in segment_stats {
+            if let Some(segment_stats) = segment_stats_opt {
+                stats.count += segment_stats.count;
+                stats.sum += segment_stats.sum;
+                stats.squared_sum += segment_stats.squared_sum;
+            }
+        }
+        Ok(stats.non_zero_count())
+    }
+}
+
+struct StatsSegmentCollector {
+    fast_field_reader: FastFieldReader<u64>,
+    stats: Stats,
+}
+
+impl SegmentCollector for StatsSegmentCollector {
+    type Fruit = Option<Stats>;
+
+    fn collect(&mut self, doc: u32, _score: f32) {
+        let value = self.fast_field_reader.get(doc) as f64;
+        self.stats.count += 1;
+        self.stats.sum += value;
+        self.stats.squared_sum += value * value;
+    }
+
+    fn harvest(self) -> <Self as SegmentCollector>::Fruit {
+        self.stats.non_zero_count()
+    }
+}
+
+fn main() -> tantivy::Result<()> {
+    // # Defining the schema
+    //
+    // The Tantivy index requires a very strict schema.
+    // The schema declares which fields are in the index,
+    // and for each field, its type and "the way it should
+    // be indexed".
+
+    // first we need to define a schema ...
+    let mut schema_builder = Schema::builder();
+
+    // We'll assume a fictional index containing
+    // products, and with a name, a description, and a price.
+    let product_name = schema_builder.add_text_field("name", TEXT);
+    let product_description = schema_builder.add_text_field("description", TEXT);
+    let price = schema_builder.add_u64_field("price", INT_INDEXED | FAST);
+    let schema = schema_builder.build();
+
+    // # Indexing documents
+    //
+    // Lets index a bunch of fake documents for the sake of
+    // this example.
+    let index = Index::create_in_ram(schema.clone());
+
+    let mut index_writer = index.writer(50_000_000)?;
+    index_writer.add_document(doc!(
+        product_name => "Super Broom 2000",
+        product_description => "While it is ok for short distance travel, this broom \
+        was designed quiditch. It will up your game.",
+        price => 30_200u64
+    ));
+    index_writer.add_document(doc!(
+        product_name => "Turbulobroom",
+        product_description => "You might have heard of this broom before : it is the sponsor of the Wales team.\
+            You'll enjoy its sharp turns, and rapid acceleration",
+        price => 29_240u64
+    ));
+    index_writer.add_document(doc!(
+        product_name => "Broomio",
+        product_description => "Great value for the price. This broom is a market favorite",
+        price => 21_240u64
+    ));
+    index_writer.add_document(doc!(
+        product_name => "Whack a Mole",
+        product_description => "Prime quality bat.",
+        price => 5_200u64
+    ));
+    index_writer.commit()?;
+    index.load_searchers()?;
+
+    let searcher = index.searcher();
+    let query_parser = QueryParser::for_index(&index, vec![product_name, product_description]);
+
+    // here we want to get a hit on the 'ken' in Frankenstein
+    let query = query_parser.parse_query("broom")?;
+    if let Some(stats) = searcher.search(&query, &StatsCollector::with_field(price))? {
+        println!("count: {}", stats.count());
+        println!("mean: {}", stats.mean());
+        println!("standard deviation: {}", stats.standard_deviation());
+    }
+
+    Ok(())
+}

examples/custom_tokenizer.rs

@@ -3,16 +3,14 @@
 // In this example, we'll see how to define a tokenizer pipeline
 // by aligning a bunch of `TokenFilter`.
 
-
 #[macro_use]
 extern crate tantivy;
-use tantivy::collector::TopCollector;
+use tantivy::collector::TopDocs;
 use tantivy::query::QueryParser;
 use tantivy::schema::*;
 use tantivy::tokenizer::NgramTokenizer;
 use tantivy::Index;
 
-
 fn main() -> tantivy::Result<()> {
     // # Defining the schema
     //
@@ -22,7 +20,7 @@ fn main() -> tantivy::Result<()> {
     // be indexed".
 
     // first we need to define a schema ...
-    let mut schema_builder = SchemaBuilder::default();
+    let mut schema_builder = Schema::builder();
 
     // Our first field is title.
     // In this example we want to use NGram searching
@@ -106,12 +104,10 @@ fn main() -> tantivy::Result<()> {
     // here we want to get a hit on the 'ken' in Frankenstein
     let query = query_parser.parse_query("ken")?;
 
-    let mut top_collector = TopCollector::with_limit(10);
-    searcher.search(&*query, &mut top_collector)?;
+    let top_docs = searcher.search(&query, &TopDocs::with_limit(10))?;
 
-    let doc_addresses = top_collector.docs();
-    for doc_address in doc_addresses {
-        let retrieved_doc = searcher.doc(&doc_address)?;
+    for (_, doc_address) in top_docs {
+        let retrieved_doc = searcher.doc(doc_address)?;
         println!("{}", schema.to_json(&retrieved_doc));
     }
 

examples/deleting_updating_documents.rs

@@ -10,11 +10,10 @@
 // Importing tantivy...
 #[macro_use]
 extern crate tantivy;
-use tantivy::collector::TopCollector;
+use tantivy::collector::TopDocs;
+use tantivy::query::TermQuery;
 use tantivy::schema::*;
 use tantivy::Index;
-use tantivy::query::TermQuery;
-
 
 // A simple helper function to fetch a single document
 // given its id from our index.
@@ -28,11 +27,10 @@ fn extract_doc_given_isbn(index: &Index, isbn_term: &Term) -> tantivy::Result<Op
     // The second argument is here to tell we don't care about decoding positions,
     // or term frequencies.
     let term_query = TermQuery::new(isbn_term.clone(), IndexRecordOption::Basic);
-    let mut top_collector = TopCollector::with_limit(1);
-    searcher.search(&term_query, &mut top_collector)?;
+    let top_docs = searcher.search(&term_query, &TopDocs::with_limit(1))?;
 
-    if let Some(doc_address) =  top_collector.docs().first() {
-        let doc = searcher.doc(doc_address)?;
+    if let Some((_score, doc_address)) = top_docs.first() {
+        let doc = searcher.doc(*doc_address)?;
         Ok(Some(doc))
     } else {
         // no doc matching this ID.
@@ -41,12 +39,11 @@ fn extract_doc_given_isbn(index: &Index, isbn_term: &Term) -> tantivy::Result<Op
 }
 
 fn main() -> tantivy::Result<()> {
-
     // # Defining the schema
     //
     // Check out the *basic_search* example if this makes
     // small sense to you.
-    let mut schema_builder = SchemaBuilder::default();
+    let mut schema_builder = Schema::builder();
 
     // Tantivy does not really have a notion of primary id.
     // This may change in the future.
@@ -126,7 +123,6 @@ fn main() -> tantivy::Result<()> {
        isbn => "978-9176370711",
     ));
 
-
     // You are guaranteed that your clients will only observe your index in
     // the state it was in after a commit.
     // In this example, your search engine will at no point be missing the *Frankenstein* document.
@@ -143,4 +139,4 @@ fn main() -> tantivy::Result<()> {
     );
 
     Ok(())
-}
+}

examples/faceted_search.rs

@@ -22,60 +22,59 @@ use tantivy::schema::*;
 use tantivy::Index;
 
 fn main() -> tantivy::Result<()> {
-  // Let's create a temporary directory for the
-  // sake of this example
-  let index_path = TempDir::new("tantivy_facet_example_dir")?;
-  let mut schema_builder = SchemaBuilder::default();
+    // Let's create a temporary directory for the
+    // sake of this example
+    let index_path = TempDir::new("tantivy_facet_example_dir")?;
+    let mut schema_builder = Schema::builder();
 
-  schema_builder.add_text_field("name", TEXT | STORED);
+    schema_builder.add_text_field("name", TEXT | STORED);
 
-  // this is our faceted field
-  schema_builder.add_facet_field("tags");
+    // this is our faceted field
+    schema_builder.add_facet_field("tags");
 
-  let schema = schema_builder.build();
+    let schema = schema_builder.build();
 
-  let index = Index::create_in_dir(&index_path, schema.clone())?;
+    let index = Index::create_in_dir(&index_path, schema.clone())?;
 
-  let mut index_writer = index.writer(50_000_000)?;
+    let mut index_writer = index.writer(50_000_000)?;
 
-  let name = schema.get_field("name").unwrap();
-  let tags = schema.get_field("tags").unwrap();
+    let name = schema.get_field("name").unwrap();
+    let tags = schema.get_field("tags").unwrap();
 
-  // For convenience, tantivy also comes with a macro to
-  // reduce the boilerplate above.
-  index_writer.add_document(doc!(
+    // For convenience, tantivy also comes with a macro to
+    // reduce the boilerplate above.
+    index_writer.add_document(doc!(
         name => "the ditch",
         tags => Facet::from("/pools/north")
     ));
 
-  index_writer.add_document(doc!(
+    index_writer.add_document(doc!(
         name => "little stacey",
         tags => Facet::from("/pools/south")
     ));
 
-  index_writer.commit()?;
+    index_writer.commit()?;
 
-  index.load_searchers()?;
+    index.load_searchers()?;
 
-  let searcher = index.searcher();
+    let searcher = index.searcher();
 
-  let mut facet_collector = FacetCollector::for_field(tags);
-  facet_collector.add_facet("/pools");
+    let mut facet_collector = FacetCollector::for_field(tags);
+    facet_collector.add_facet("/pools");
 
-  searcher.search(&AllQuery, &mut facet_collector).unwrap();
+    let facet_counts = searcher.search(&AllQuery, &facet_collector).unwrap();
 
-  let counts = facet_collector.harvest();
-  // This lists all of the facet counts
-  let facets: Vec<(&Facet, u64)> = counts.get("/pools").collect();
-  assert_eq!(
-    facets,
-    vec![
-      (&Facet::from("/pools/north"), 1),
-      (&Facet::from("/pools/south"), 1)
-    ]
-  );
+    // This lists all of the facet counts
+    let facets: Vec<(&Facet, u64)> = facet_counts.get("/pools").collect();
+    assert_eq!(
+        facets,
+        vec![
+            (&Facet::from("/pools/north"), 1),
+            (&Facet::from("/pools/south"), 1),
+        ]
+    );
 
-  Ok(())
+    Ok(())
 }
 
 use tempdir::TempDir;

examples/iterating_docs_and_positions.rs

@@ -7,21 +7,18 @@
 // the list of documents containing a term, getting
 // its term frequency, and accessing its positions.
 
-
 // ---
 // Importing tantivy...
 #[macro_use]
 extern crate tantivy;
 use tantivy::schema::*;
 use tantivy::Index;
-use tantivy::{DocSet, DocId, Postings};
+use tantivy::{DocId, DocSet, Postings};
 
 fn main() -> tantivy::Result<()> {
-
-
     // We first create a schema for the sake of the
     // example. Check the `basic_search` example for more information.
-    let mut schema_builder = SchemaBuilder::default();
+    let mut schema_builder = Schema::builder();
 
     // For this example, we need to make sure to index positions for our title
     // field. `TEXT` precisely does this.
@@ -47,7 +44,6 @@ fn main() -> tantivy::Result<()> {
     // there is actually only one segment here, but let's iterate through the list
     // anyway)
     for segment_reader in searcher.segment_readers() {
-
         // A segment contains different data structure.
         // Inverted index stands for the combination of
         // - the term dictionary
@@ -58,19 +54,18 @@ fn main() -> tantivy::Result<()> {
         // Let's go through all docs containing the term `title:the` and access their position
         let term_the = Term::from_field_text(title, "the");
 
-
         // This segment posting object is like a cursor over the documents matching the term.
         // The `IndexRecordOption` arguments tells tantivy we will be interested in both term frequencies
         // and positions.
         //
         // If you don't need all this information, you may get better performance by decompressing less
         // information.
-        if let Some(mut segment_postings) = inverted_index.read_postings(&term_the, IndexRecordOption::WithFreqsAndPositions) {
-
+        if let Some(mut segment_postings) =
+            inverted_index.read_postings(&term_the, IndexRecordOption::WithFreqsAndPositions)
+        {
             // this buffer will be used to request for positions
             let mut positions: Vec<u32> = Vec::with_capacity(100);
             while segment_postings.advance() {
-
                 // the number of time the term appears in the document.
                 let doc_id: DocId = segment_postings.doc(); //< do not try to access this before calling advance once.
 
@@ -98,7 +93,6 @@ fn main() -> tantivy::Result<()> {
         }
     }
 
-
     // A `Term` is a text token associated with a field.
     // Let's go through all docs containing the term `title:the` and access their position
     let term_the = Term::from_field_text(title, "the");
@@ -111,7 +105,6 @@ fn main() -> tantivy::Result<()> {
     // Also, for some VERY specific high performance use case like an OLAP analysis of logs,
     // you can get better performance by accessing directly the blocks of doc ids.
     for segment_reader in searcher.segment_readers() {
-
         // A segment contains different data structure.
         // Inverted index stands for the combination of
         // - the term dictionary
@@ -124,7 +117,9 @@ fn main() -> tantivy::Result<()> {
         //
         // If you don't need all this information, you may get better performance by decompressing less
         // information.
-        if let Some(mut block_segment_postings) = inverted_index.read_block_postings(&term_the, IndexRecordOption::Basic) {
+        if let Some(mut block_segment_postings) =
+            inverted_index.read_block_postings(&term_the, IndexRecordOption::Basic)
+        {
             while block_segment_postings.advance() {
                 // Once again these docs MAY contains deleted documents as well.
                 let docs = block_segment_postings.docs();
@@ -136,4 +131,3 @@ fn main() -> tantivy::Result<()> {
 
     Ok(())
 }
-

examples/snippet.rs

@@ -0,0 +1,87 @@
+// # Snippet example
+//
+// This example shows how to return a representative snippet of
+// your hit result.
+// Snippet are an extracted of a target document, and returned in HTML format.
+// The keyword searched by the user are highlighted with a `<b>` tag.
+extern crate tempdir;
+
+// ---
+// Importing tantivy...
+#[macro_use]
+extern crate tantivy;
+use tantivy::collector::TopDocs;
+use tantivy::query::QueryParser;
+use tantivy::schema::*;
+use tantivy::Index;
+use tantivy::{Snippet, SnippetGenerator};
+use tempdir::TempDir;
+
+fn main() -> tantivy::Result<()> {
+    // Let's create a temporary directory for the
+    // sake of this example
+    let index_path = TempDir::new("tantivy_example_dir")?;
+
+    // # Defining the schema
+    let mut schema_builder = Schema::builder();
+    let title = schema_builder.add_text_field("title", TEXT | STORED);
+    let body = schema_builder.add_text_field("body", TEXT | STORED);
+    let schema = schema_builder.build();
+
+    // # Indexing documents
+    let index = Index::create_in_dir(&index_path, schema.clone())?;
+
+    let mut index_writer = index.writer(50_000_000)?;
+
+    // we'll only need one doc for this example.
+    index_writer.add_document(doc!(
+        title => "Of Mice and Men",
+        body => "A few miles south of Soledad, the Salinas River drops in close to the hillside \
+                bank and runs deep and green. The water is warm too, for it has slipped twinkling \
+                over the yellow sands in the sunlight before reaching the narrow pool. On one \
+                side of the river the golden foothill slopes curve up to the strong and rocky \
+                Gabilan Mountains, but on the valley side the water is lined with trees—willows \
+                fresh and green with every spring, carrying in their lower leaf junctures the \
+                debris of the winter’s flooding; and sycamores with mottled, white, recumbent \
+                limbs and branches that arch over the pool"
+    ));
+    // ...
+    index_writer.commit()?;
+
+    index.load_searchers()?;
+
+    let searcher = index.searcher();
+    let query_parser = QueryParser::for_index(&index, vec![title, body]);
+    let query = query_parser.parse_query("sycamore spring")?;
+
+    let top_docs = searcher.search(&query, &TopDocs::with_limit(10))?;
+
+    let snippet_generator = SnippetGenerator::new(&searcher, &*query, body)?;
+
+    for (score, doc_address) in top_docs {
+        let doc = searcher.doc(doc_address)?;
+        let snippet = snippet_generator.snippet_from_doc(&doc);
+        println!("Document score {}:", score);
+        println!("title: {}", doc.get_first(title).unwrap().text().unwrap());
+        println!("snippet: {}", snippet.to_html());
+        println!("custom highlighting: {}", highlight(snippet));
+    }
+
+    Ok(())
+}
+
+fn highlight(snippet: Snippet) -> String {
+    let mut result = String::new();
+    let mut start_from = 0;
+
+    for (start, end) in snippet.highlighted().iter().map(|h| h.bounds()) {
+        result.push_str(&snippet.fragments()[start_from..start]);
+        result.push_str(" --> ");
+        result.push_str(&snippet.fragments()[start..end]);
+        result.push_str(" <-- ");
+        start_from = end;
+    }
+
+    result.push_str(&snippet.fragments()[start_from..]);
+    result
+}

examples/stop_words.rs

@@ -15,79 +15,78 @@ extern crate tempdir;
 // Importing tantivy...
 #[macro_use]
 extern crate tantivy;
-use tantivy::collector::TopCollector;
+use tantivy::collector::TopDocs;
 use tantivy::query::QueryParser;
 use tantivy::schema::*;
 use tantivy::tokenizer::*;
 use tantivy::Index;
 
 fn main() -> tantivy::Result<()> {
-  // this example assumes you understand the content in `basic_search`
-  let index_path = TempDir::new("tantivy_stopwords_example_dir")?;
-  let mut schema_builder = SchemaBuilder::default();
+    // this example assumes you understand the content in `basic_search`
+    let mut schema_builder = Schema::builder();
 
-  // This configures your custom options for how tantivy will
-  // store and process your content in the index; The key
-  // to note is that we are setting the tokenizer to `stoppy`
-  // which will be defined and registered below.
-  let text_field_indexing = TextFieldIndexing::default()
-    .set_tokenizer("stoppy")
-    .set_index_option(IndexRecordOption::WithFreqsAndPositions);
-  let text_options = TextOptions::default()
-    .set_indexing_options(text_field_indexing)
-    .set_stored();
+    // This configures your custom options for how tantivy will
+    // store and process your content in the index; The key
+    // to note is that we are setting the tokenizer to `stoppy`
+    // which will be defined and registered below.
+    let text_field_indexing = TextFieldIndexing::default()
+        .set_tokenizer("stoppy")
+        .set_index_option(IndexRecordOption::WithFreqsAndPositions);
+    let text_options = TextOptions::default()
+        .set_indexing_options(text_field_indexing)
+        .set_stored();
 
-  // Our first field is title.
-  schema_builder.add_text_field("title", text_options);
+    // Our first field is title.
+    schema_builder.add_text_field("title", text_options);
 
-  // Our second field is body.
-  let text_field_indexing = TextFieldIndexing::default()
-    .set_tokenizer("stoppy")
-    .set_index_option(IndexRecordOption::WithFreqsAndPositions);
-  let text_options = TextOptions::default()
-    .set_indexing_options(text_field_indexing)
-    .set_stored();
-  schema_builder.add_text_field("body", text_options);
+    // Our second field is body.
+    let text_field_indexing = TextFieldIndexing::default()
+        .set_tokenizer("stoppy")
+        .set_index_option(IndexRecordOption::WithFreqsAndPositions);
+    let text_options = TextOptions::default()
+        .set_indexing_options(text_field_indexing)
+        .set_stored();
+    schema_builder.add_text_field("body", text_options);
 
-  let schema = schema_builder.build();
+    let schema = schema_builder.build();
 
-  let index = Index::create_in_dir(&index_path, schema.clone())?;
+    let index = Index::create_in_ram(schema.clone());
 
-  // This tokenizer lowers all of the text (to help with stop word matching)
-  // then removes all instances of `the` and `and` from the corpus
-  let tokenizer = SimpleTokenizer
-    .filter(LowerCaser)
-    .filter(StopWordFilter::remove(vec![
-      "the".to_string(),
-      "and".to_string(),
-    ]));
+    // This tokenizer lowers all of the text (to help with stop word matching)
+    // then removes all instances of `the` and `and` from the corpus
+    let tokenizer = SimpleTokenizer
+        .filter(LowerCaser)
+        .filter(StopWordFilter::remove(vec![
+            "the".to_string(),
+            "and".to_string(),
+        ]));
 
-  index.tokenizers().register("stoppy", tokenizer);
+    index.tokenizers().register("stoppy", tokenizer);
 
-  let mut index_writer = index.writer(50_000_000)?;
+    let mut index_writer = index.writer(50_000_000)?;
 
-  let title = schema.get_field("title").unwrap();
-  let body = schema.get_field("body").unwrap();
+    let title = schema.get_field("title").unwrap();
+    let body = schema.get_field("body").unwrap();
 
-  index_writer.add_document(doc!(
+    index_writer.add_document(doc!(
     title => "The Old Man and the Sea",
     body => "He was an old man who fished alone in a skiff in the Gulf Stream and \
      he had gone eighty-four days now without taking a fish."
   ));
 
-  index_writer.add_document(doc!(
-        title => "Of Mice and Men",
-        body => "A few miles south of Soledad, the Salinas River drops in close to the hillside \
-                bank and runs deep and green. The water is warm too, for it has slipped twinkling \
-                over the yellow sands in the sunlight before reaching the narrow pool. On one \
-                side of the river the golden foothill slopes curve up to the strong and rocky \
-                Gabilan Mountains, but on the valley side the water is lined with trees—willows \
-                fresh and green with every spring, carrying in their lower leaf junctures the \
-                debris of the winter’s flooding; and sycamores with mottled, white, recumbent \
-                limbs and branches that arch over the pool"
-    ));
+    index_writer.add_document(doc!(
+      title => "Of Mice and Men",
+      body => "A few miles south of Soledad, the Salinas River drops in close to the hillside \
+              bank and runs deep and green. The water is warm too, for it has slipped twinkling \
+              over the yellow sands in the sunlight before reaching the narrow pool. On one \
+              side of the river the golden foothill slopes curve up to the strong and rocky \
+              Gabilan Mountains, but on the valley side the water is lined with trees—willows \
+              fresh and green with every spring, carrying in their lower leaf junctures the \
+              debris of the winter’s flooding; and sycamores with mottled, white, recumbent \
+              limbs and branches that arch over the pool"
+  ));
 
-  index_writer.add_document(doc!(
+    index_writer.add_document(doc!(
        title => "Frankenstein",
        body => "You will rejoice to hear that no disaster has accompanied the commencement of an \
                 enterprise which you have regarded with such evil forebodings.  I arrived here \
@@ -95,35 +94,24 @@ fn main() -> tantivy::Result<()> {
                 increasing confidence in the success of my undertaking."
     ));
 
-  index_writer.commit()?;
+    index_writer.commit()?;
 
-  index.load_searchers()?;
+    index.load_searchers()?;
 
-  let searcher = index.searcher();
+    let searcher = index.searcher();
 
-  let query_parser = QueryParser::for_index(&index, vec![title, body]);
+    let query_parser = QueryParser::for_index(&index, vec![title, body]);
 
-  // this will have NO hits because it was filtered out
-  // because the query is run through the analyzer you
-  // actually will get an error here because the query becomes
-  // empty
-  assert!(query_parser.parse_query("the").is_err());
+    // stop words are applied on the query as well.
+    // The following will be equivalent to `title:frankenstein`
+    let query = query_parser.parse_query("title:\"the Frankenstein\"")?;
+    let top_docs = searcher.search(&query, &TopDocs::with_limit(10))?;
 
-  // this will have hits
-  let query = query_parser.parse_query("is")?;
+    for (score, doc_address) in top_docs {
+        let retrieved_doc = searcher.doc(doc_address)?;
+        println!("\n==\nDocument score {}:", score);
+        println!("{}", schema.to_json(&retrieved_doc));
+    }
 
-  let mut top_collector = TopCollector::with_limit(10);
-
-  searcher.search(&*query, &mut top_collector)?;
-
-  let doc_addresses = top_collector.docs();
-
-  for doc_address in doc_addresses {
-    let retrieved_doc = searcher.doc(&doc_address)?;
-    println!("{}", schema.to_json(&retrieved_doc));
-  }
-
-  Ok(())
+    Ok(())
 }
-
-use tempdir::TempDir;

examples/working_with_json.rs

@@ -9,7 +9,7 @@ fn main() -> tantivy::Result<()> {
     // Check out the basic example if this is confusing to you.
     //
     // first we need to define a schema ...
-    let mut schema_builder = SchemaBuilder::default();
+    let mut schema_builder = Schema::builder();
     schema_builder.add_text_field("title", TEXT | STORED);
     schema_builder.add_text_field("body", TEXT);
     schema_builder.add_u64_field("year", INT_INDEXED);

run-tests.sh

@@ -0,0 +1,2 @@
+#!/bin/bash
+cargo test --no-default-features --features mmap -- --test-threads 1

src/collector/chained_collector.rs

@@ -1,142 +0,0 @@
-use collector::Collector;
-use DocId;
-use Result;
-use Score;
-use SegmentLocalId;
-use SegmentReader;
-
-/// Collector that does nothing.
-/// This is used in the chain Collector and will hopefully
-/// be optimized away by the compiler.
-pub struct DoNothingCollector;
-impl Collector for DoNothingCollector {
-    #[inline]
-    fn set_segment(&mut self, _: SegmentLocalId, _: &SegmentReader) -> Result<()> {
-        Ok(())
-    }
-    #[inline]
-    fn collect(&mut self, _doc: DocId, _score: Score) {}
-    #[inline]
-    fn requires_scoring(&self) -> bool {
-        false
-    }
-}
-
-/// Zero-cost abstraction used to collect on multiple collectors.
-/// This contraption is only usable if the type of your collectors
-/// are known at compile time.
-///
-/// ```rust
-/// #[macro_use]
-/// extern crate tantivy;
-/// use tantivy::schema::{SchemaBuilder, TEXT};
-/// use tantivy::{Index, Result};
-/// use tantivy::collector::{CountCollector, TopCollector, chain};
-/// use tantivy::query::QueryParser;
-///
-/// # fn main() { example().unwrap(); }
-/// fn example() -> Result<()> {
-///     let mut schema_builder = SchemaBuilder::new();
-///     let title = schema_builder.add_text_field("title", TEXT);
-///     let schema = schema_builder.build();
-///     let index = Index::create_in_ram(schema);
-///     {
-///         let mut index_writer = index.writer(3_000_000)?;
-///         index_writer.add_document(doc!(
-///             title => "The Name of the Wind",
-///         ));
-///         index_writer.add_document(doc!(
-///             title => "The Diary of Muadib",
-///         ));
-///         index_writer.add_document(doc!(
-///             title => "A Dairy Cow",
-///         ));
-///         index_writer.add_document(doc!(
-///             title => "The Diary of a Young Girl",
-///         ));
-///         index_writer.commit().unwrap();
-///     }
-///
-///     index.load_searchers()?;
-///     let searcher = index.searcher();
-///
-///     {
-///         let mut top_collector = TopCollector::with_limit(2);
-///         let mut count_collector = CountCollector::default();
-///         {
-///             let mut collectors = chain().push(&mut top_collector).push(&mut count_collector);
-///             let query_parser = QueryParser::for_index(&index, vec![title]);
-///             let query = query_parser.parse_query("diary")?;
-///             searcher.search(&*query, &mut collectors).unwrap();
-///         }
-///         assert_eq!(count_collector.count(), 2);
-///         assert!(top_collector.at_capacity());
-///     }
-///
-///     Ok(())
-/// }
-/// ```
-pub struct ChainedCollector<Left: Collector, Right: Collector> {
-    left: Left,
-    right: Right,
-}
-
-impl<Left: Collector, Right: Collector> ChainedCollector<Left, Right> {
-    /// Adds a collector
-    pub fn push<C: Collector>(self, new_collector: &mut C) -> ChainedCollector<Self, &mut C> {
-        ChainedCollector {
-            left: self,
-            right: new_collector,
-        }
-    }
-}
-
-impl<Left: Collector, Right: Collector> Collector for ChainedCollector<Left, Right> {
-    fn set_segment(
-        &mut self,
-        segment_local_id: SegmentLocalId,
-        segment: &SegmentReader,
-    ) -> Result<()> {
-        self.left.set_segment(segment_local_id, segment)?;
-        self.right.set_segment(segment_local_id, segment)?;
-        Ok(())
-    }
-
-    fn collect(&mut self, doc: DocId, score: Score) {
-        self.left.collect(doc, score);
-        self.right.collect(doc, score);
-    }
-
-    fn requires_scoring(&self) -> bool {
-        self.left.requires_scoring() || self.right.requires_scoring()
-    }
-}
-
-/// Creates a `ChainedCollector`
-pub fn chain() -> ChainedCollector<DoNothingCollector, DoNothingCollector> {
-    ChainedCollector {
-        left: DoNothingCollector,
-        right: DoNothingCollector,
-    }
-}
-
-#[cfg(test)]
-mod tests {
-
-    use super::*;
-    use collector::{Collector, CountCollector, TopCollector};
-
-    #[test]
-    fn test_chained_collector() {
-        let mut top_collector = TopCollector::with_limit(2);
-        let mut count_collector = CountCollector::default();
-        {
-            let mut collectors = chain().push(&mut top_collector).push(&mut count_collector);
-            collectors.collect(1, 0.2);
-            collectors.collect(2, 0.1);
-            collectors.collect(3, 0.5);
-        }
-        assert_eq!(count_collector.count(), 3);
-        assert!(top_collector.at_capacity());
-    }
-}

src/collector/count_collector.rs

@@ -1,4 +1,5 @@
 use super::Collector;
+use collector::SegmentCollector;
 use DocId;
 use Result;
 use Score;
@@ -11,14 +12,14 @@ use SegmentReader;
 /// ```rust
 /// #[macro_use]
 /// extern crate tantivy;
-/// use tantivy::schema::{SchemaBuilder, TEXT};
+/// use tantivy::schema::{Schema, TEXT};
 /// use tantivy::{Index, Result};
-/// use tantivy::collector::CountCollector;
+/// use tantivy::collector::Count;
 /// use tantivy::query::QueryParser;
 ///
 /// # fn main() { example().unwrap(); }
 /// fn example() -> Result<()> {
-///     let mut schema_builder = SchemaBuilder::new();
+///     let mut schema_builder = Schema::builder();
 ///     let title = schema_builder.add_text_field("title", TEXT);
 ///     let schema = schema_builder.build();
 ///     let index = Index::create_in_ram(schema);
@@ -43,59 +44,86 @@ use SegmentReader;
 ///     let searcher = index.searcher();
 ///
 ///     {
-///	        let mut count_collector = CountCollector::default();
 ///         let query_parser = QueryParser::for_index(&index, vec![title]);
 ///         let query = query_parser.parse_query("diary")?;
-///         searcher.search(&*query, &mut count_collector).unwrap();
+///         let count = searcher.search(&query, &Count).unwrap();
 ///
-///         assert_eq!(count_collector.count(), 2);
+///         assert_eq!(count, 2);
 ///     }
 ///
 ///     Ok(())
 /// }
 /// ```
-#[derive(Default)]
-pub struct CountCollector {
-    count: usize,
-}
+pub struct Count;
 
-impl CountCollector {
-    /// Returns the count of documents that were
-    /// collected.
-    pub fn count(&self) -> usize {
-        self.count
-    }
-}
+impl Collector for Count {
+    type Fruit = usize;
 
-impl Collector for CountCollector {
-    fn set_segment(&mut self, _: SegmentLocalId, _: &SegmentReader) -> Result<()> {
-        Ok(())
-    }
+    type Child = SegmentCountCollector;
 
-    fn collect(&mut self, _: DocId, _: Score) {
-        self.count += 1;
+    fn for_segment(&self, _: SegmentLocalId, _: &SegmentReader) -> Result<SegmentCountCollector> {
+        Ok(SegmentCountCollector::default())
     }
 
     fn requires_scoring(&self) -> bool {
         false
     }
+
+    fn merge_fruits(&self, segment_counts: Vec<usize>) -> Result<usize> {
+        Ok(segment_counts.into_iter().sum())
+    }
+}
+
+#[derive(Default)]
+pub struct SegmentCountCollector {
+    count: usize,
+}
+
+impl SegmentCollector for SegmentCountCollector {
+    type Fruit = usize;
+
+    fn collect(&mut self, _: DocId, _: Score) {
+        self.count += 1;
+    }
+
+    fn harvest(self) -> usize {
+        self.count
+    }
 }
 
 #[cfg(test)]
 mod tests {
-
-    use collector::{Collector, CountCollector};
+    use super::{Count, SegmentCountCollector};
+    use collector::Collector;
+    use collector::SegmentCollector;
 
     #[test]
-    fn test_count_collector() {
-        let mut count_collector = CountCollector::default();
-        assert_eq!(count_collector.count(), 0);
-        count_collector.collect(0u32, 1f32);
-        assert_eq!(count_collector.count(), 1);
-        assert_eq!(count_collector.count(), 1);
-        count_collector.collect(1u32, 1f32);
-        assert_eq!(count_collector.count(), 2);
-        assert!(!count_collector.requires_scoring());
+    fn test_count_collect_does_not_requires_scoring() {
+        assert!(!Count.requires_scoring());
+    }
+
+    #[test]
+    fn test_segment_count_collector() {
+        {
+            let count_collector = SegmentCountCollector::default();
+            assert_eq!(count_collector.harvest(), 0);
+        }
+        {
+            let mut count_collector = SegmentCountCollector::default();
+            count_collector.collect(0u32, 1f32);
+            assert_eq!(count_collector.harvest(), 1);
+        }
+        {
+            let mut count_collector = SegmentCountCollector::default();
+            count_collector.collect(0u32, 1f32);
+            assert_eq!(count_collector.harvest(), 1);
+        }
+        {
+            let mut count_collector = SegmentCountCollector::default();
+            count_collector.collect(0u32, 1f32);
+            count_collector.collect(1u32, 1f32);
+            assert_eq!(count_collector.harvest(), 2);
+        }
     }
 
 }

src/collector/facet_collector.rs

@@ -1,20 +1,17 @@
 use collector::Collector;
+use collector::SegmentCollector;
 use docset::SkipResult;
 use fastfield::FacetReader;
 use schema::Facet;
 use schema::Field;
-use std::cell::UnsafeCell;
+use std::cmp::Ordering;
 use std::collections::btree_map;
 use std::collections::BTreeMap;
 use std::collections::BTreeSet;
 use std::collections::BinaryHeap;
 use std::collections::Bound;
 use std::iter::Peekable;
-use std::mem;
 use std::{u64, usize};
-use termdict::TermMerger;
-
-use std::cmp::Ordering;
 use DocId;
 use Result;
 use Score;
@@ -46,12 +43,6 @@ impl<'a> Ord for Hit<'a> {
     }
 }
 
-struct SegmentFacetCounter {
-    pub facet_reader: FacetReader,
-    pub facet_ords: Vec<u64>,
-    pub facet_counts: Vec<u64>,
-}
-
 fn facet_depth(facet_bytes: &[u8]) -> usize {
     if facet_bytes.is_empty() {
         0
@@ -91,14 +82,14 @@ fn facet_depth(facet_bytes: &[u8]) -> usize {
 /// ```rust
 /// #[macro_use]
 /// extern crate tantivy;
-/// use tantivy::schema::{Facet, SchemaBuilder, TEXT};
+/// use tantivy::schema::{Facet, Schema, TEXT};
 /// use tantivy::{Index, Result};
 /// use tantivy::collector::FacetCollector;
 /// use tantivy::query::AllQuery;
 ///
 /// # fn main() { example().unwrap(); }
 /// fn example() -> Result<()> {
-///     let mut schema_builder = SchemaBuilder::new();
+///     let mut schema_builder = Schema::builder();
 ///
 ///     // Facet have their own specific type.
 ///     // It is not a bad practise to put all of your
@@ -141,13 +132,10 @@ fn facet_depth(facet_bytes: &[u8]) -> usize {
 ///			let mut facet_collector = FacetCollector::for_field(facet);
 ///         facet_collector.add_facet("/lang");
 ///         facet_collector.add_facet("/category");
-///         searcher.search(&AllQuery, &mut facet_collector).unwrap();
-///
-///         // this object contains count aggregate for all of the facets.
-///         let counts = facet_collector.harvest();
+///         let facet_counts = searcher.search(&AllQuery, &facet_collector).unwrap();
 ///
 ///         // This lists all of the facet counts
-///         let facets: Vec<(&Facet, u64)> = counts
+///         let facets: Vec<(&Facet, u64)> = facet_counts
 ///             .get("/category")
 ///             .collect();
 ///         assert_eq!(facets, vec![
@@ -159,13 +147,10 @@ fn facet_depth(facet_bytes: &[u8]) -> usize {
 ///     {
 ///			let mut facet_collector = FacetCollector::for_field(facet);
 ///         facet_collector.add_facet("/category/fiction");
-///         searcher.search(&AllQuery, &mut facet_collector).unwrap();
-///
-///         // this object contains count aggregate for all of the facets.
-///         let counts = facet_collector.harvest();
+///         let facet_counts = searcher.search(&AllQuery, &facet_collector).unwrap();
 ///
 ///         // This lists all of the facet counts
-///         let facets: Vec<(&Facet, u64)> = counts
+///         let facets: Vec<(&Facet, u64)> = facet_counts
 ///             .get("/category/fiction")
 ///             .collect();
 ///         assert_eq!(facets, vec![
@@ -178,13 +163,10 @@ fn facet_depth(facet_bytes: &[u8]) -> usize {
 ///    {
 ///			let mut facet_collector = FacetCollector::for_field(facet);
 ///         facet_collector.add_facet("/category/fiction");
-///         searcher.search(&AllQuery, &mut facet_collector).unwrap();
-///
-///         // this object contains count aggregate for all of the facets.
-///         let counts = facet_collector.harvest();
+///         let facet_counts = searcher.search(&AllQuery, &facet_collector).unwrap();
 ///
 ///         // This lists all of the facet counts
-///         let facets: Vec<(&Facet, u64)> = counts.top_k("/category/fiction", 1);
+///         let facets: Vec<(&Facet, u64)> = facet_counts.top_k("/category/fiction", 1);
 ///         assert_eq!(facets, vec![
 ///             (&Facet::from("/category/fiction/fantasy"), 2)
 ///         ]);
@@ -194,21 +176,21 @@ fn facet_depth(facet_bytes: &[u8]) -> usize {
 /// }
 /// ```
 pub struct FacetCollector {
-    facet_ords: Vec<u64>,
     field: Field,
-    ff_reader: Option<UnsafeCell<FacetReader>>,
-    segment_counters: Vec<SegmentFacetCounter>,
-
-    // facet_ord -> collapse facet_id
-    current_segment_collapse_mapping: Vec<usize>,
-    // collapse facet_id -> count
-    current_segment_counts: Vec<u64>,
-    // collapse facet_id -> facet_ord
-    current_collapse_facet_ords: Vec<u64>,
-
     facets: BTreeSet<Facet>,
 }
 
+pub struct FacetSegmentCollector {
+    reader: FacetReader,
+    facet_ords_buf: Vec<u64>,
+    // facet_ord -> collapse facet_id
+    collapse_mapping: Vec<usize>,
+    // collapse facet_id -> count
+    counts: Vec<u64>,
+    // collapse facet_id -> facet_ord
+    collapse_facet_ords: Vec<u64>,
+}
+
 fn skip<'a, I: Iterator<Item = &'a Facet>>(
     target: &[u8],
     collapse_it: &mut Peekable<I>,
@@ -240,15 +222,8 @@ impl FacetCollector {
     /// is of the proper type.
     pub fn for_field(field: Field) -> FacetCollector {
         FacetCollector {
-            facet_ords: Vec::with_capacity(255),
-            segment_counters: Vec::new(),
             field,
-            ff_reader: None,
-            facets: BTreeSet::new(),
-
-            current_segment_collapse_mapping: Vec::new(),
-            current_collapse_facet_ords: Vec::new(),
-            current_segment_counts: Vec::new(),
+            facets: BTreeSet::default(),
         }
     }
 
@@ -278,141 +253,100 @@ impl FacetCollector {
         }
         self.facets.insert(facet);
     }
-
-    fn set_collapse_mapping(&mut self, facet_reader: &FacetReader) {
-        self.current_segment_collapse_mapping.clear();
-        self.current_collapse_facet_ords.clear();
-        self.current_segment_counts.clear();
-        let mut collapse_facet_it = self.facets.iter().peekable();
-        self.current_collapse_facet_ords.push(0);
-        let mut facet_streamer = facet_reader.facet_dict().range().into_stream();
-        if !facet_streamer.advance() {
-            return;
-        }
-        'outer: loop {
-            // at the begining of this loop, facet_streamer
-            // is positionned on a term that has not been processed yet.
-            let skip_result = skip(facet_streamer.key(), &mut collapse_facet_it);
-            match skip_result {
-                SkipResult::Reached => {
-                    // we reach a facet we decided to collapse.
-                    let collapse_depth = facet_depth(facet_streamer.key());
-                    let mut collapsed_id = 0;
-                    self.current_segment_collapse_mapping.push(0);
-                    while facet_streamer.advance() {
-                        let depth = facet_depth(facet_streamer.key());
-                        if depth <= collapse_depth {
-                            continue 'outer;
-                        }
-                        if depth == collapse_depth + 1 {
-                            collapsed_id = self.current_collapse_facet_ords.len();
-                            self.current_collapse_facet_ords
-                                .push(facet_streamer.term_ord());
-                            self.current_segment_collapse_mapping.push(collapsed_id);
-                        } else {
-                            self.current_segment_collapse_mapping.push(collapsed_id);
-                        }
-                    }
-                    break;
-                }
-                SkipResult::End | SkipResult::OverStep => {
-                    self.current_segment_collapse_mapping.push(0);
-                    if !facet_streamer.advance() {
-                        break;
-                    }
-                }
-            }
-        }
-    }
-
-    fn finalize_segment(&mut self) {
-        if self.ff_reader.is_some() {
-            self.segment_counters.push(SegmentFacetCounter {
-                facet_reader: self.ff_reader.take().unwrap().into_inner(),
-                facet_ords: mem::replace(&mut self.current_collapse_facet_ords, Vec::new()),
-                facet_counts: mem::replace(&mut self.current_segment_counts, Vec::new()),
-            });
-        }
-    }
-
-    /// Returns the results of the collection.
-    ///
-    /// This method does not just return the counters,
-    /// it also translates the facet ordinals of the last segment.
-    pub fn harvest(mut self) -> FacetCounts {
-        self.finalize_segment();
-
-        let collapsed_facet_ords: Vec<&[u64]> = self.segment_counters
-            .iter()
-            .map(|segment_counter| &segment_counter.facet_ords[..])
-            .collect();
-        let collapsed_facet_counts: Vec<&[u64]> = self.segment_counters
-            .iter()
-            .map(|segment_counter| &segment_counter.facet_counts[..])
-            .collect();
-
-        let facet_streams = self.segment_counters
-            .iter()
-            .map(|seg_counts| seg_counts.facet_reader.facet_dict().range().into_stream())
-            .collect::<Vec<_>>();
-
-        let mut facet_merger = TermMerger::new(facet_streams);
-        let mut facet_counts = BTreeMap::new();
-
-        while facet_merger.advance() {
-            let count = facet_merger
-                .current_kvs()
-                .iter()
-                .map(|it| {
-                    let seg_ord = it.segment_ord;
-                    let term_ord = it.streamer.term_ord();
-                    collapsed_facet_ords[seg_ord]
-                        .binary_search(&term_ord)
-                        .map(|collapsed_term_id| {
-                            if collapsed_term_id == 0 {
-                                0
-                            } else {
-                                collapsed_facet_counts[seg_ord][collapsed_term_id]
-                            }
-                        })
-                        .unwrap_or(0)
-                })
-                .sum();
-            if count > 0u64 {
-                let bytes: Vec<u8> = facet_merger.key().to_owned();
-                // may create an corrupted facet if the term dicitonary is corrupted
-                let facet = unsafe { Facet::from_encoded(bytes) };
-                facet_counts.insert(facet, count);
-            }
-        }
-        FacetCounts { facet_counts }
-    }
 }
 
 impl Collector for FacetCollector {
-    fn set_segment(&mut self, _: SegmentLocalId, reader: &SegmentReader) -> Result<()> {
-        self.finalize_segment();
+    type Fruit = FacetCounts;
+
+    type Child = FacetSegmentCollector;
+
+    fn for_segment(
+        &self,
+        _: SegmentLocalId,
+        reader: &SegmentReader,
+    ) -> Result<FacetSegmentCollector> {
         let facet_reader = reader.facet_reader(self.field)?;
-        self.set_collapse_mapping(&facet_reader);
-        self.current_segment_counts
-            .resize(self.current_collapse_facet_ords.len(), 0);
-        self.ff_reader = Some(UnsafeCell::new(facet_reader));
-        Ok(())
+
+        let mut collapse_mapping = Vec::new();
+        let mut counts = Vec::new();
+        let mut collapse_facet_ords = Vec::new();
+
+        let mut collapse_facet_it = self.facets.iter().peekable();
+        collapse_facet_ords.push(0);
+        {
+            let mut facet_streamer = facet_reader.facet_dict().range().into_stream();
+            if facet_streamer.advance() {
+                'outer: loop {
+                    // at the begining of this loop, facet_streamer
+                    // is positionned on a term that has not been processed yet.
+                    let skip_result = skip(facet_streamer.key(), &mut collapse_facet_it);
+                    match skip_result {
+                        SkipResult::Reached => {
+                            // we reach a facet we decided to collapse.
+                            let collapse_depth = facet_depth(facet_streamer.key());
+                            let mut collapsed_id = 0;
+                            collapse_mapping.push(0);
+                            while facet_streamer.advance() {
+                                let depth = facet_depth(facet_streamer.key());
+                                if depth <= collapse_depth {
+                                    continue 'outer;
+                                }
+                                if depth == collapse_depth + 1 {
+                                    collapsed_id = collapse_facet_ords.len();
+                                    collapse_facet_ords.push(facet_streamer.term_ord());
+                                    collapse_mapping.push(collapsed_id);
+                                } else {
+                                    collapse_mapping.push(collapsed_id);
+                                }
+                            }
+                            break;
+                        }
+                        SkipResult::End | SkipResult::OverStep => {
+                            collapse_mapping.push(0);
+                            if !facet_streamer.advance() {
+                                break;
+                            }
+                        }
+                    }
+                }
+            }
+        }
+
+        counts.resize(collapse_facet_ords.len(), 0);
+
+        Ok(FacetSegmentCollector {
+            reader: facet_reader,
+            facet_ords_buf: Vec::with_capacity(255),
+            collapse_mapping,
+            counts,
+            collapse_facet_ords,
+        })
     }
 
+    fn requires_scoring(&self) -> bool {
+        false
+    }
+
+    fn merge_fruits(&self, segments_facet_counts: Vec<FacetCounts>) -> Result<FacetCounts> {
+        let mut facet_counts: BTreeMap<Facet, u64> = BTreeMap::new();
+        for segment_facet_counts in segments_facet_counts {
+            for (facet, count) in segment_facet_counts.facet_counts {
+                *(facet_counts.entry(facet).or_insert(0)) += count;
+            }
+        }
+        Ok(FacetCounts { facet_counts })
+    }
+}
+
+impl SegmentCollector for FacetSegmentCollector {
+    type Fruit = FacetCounts;
+
     fn collect(&mut self, doc: DocId, _: Score) {
-        let facet_reader: &mut FacetReader = unsafe {
-            &mut *self.ff_reader
-                .as_ref()
-                .expect("collect() was called before set_segment. This should never happen.")
-                .get()
-        };
-        facet_reader.facet_ords(doc, &mut self.facet_ords);
+        self.reader.facet_ords(doc, &mut self.facet_ords_buf);
         let mut previous_collapsed_ord: usize = usize::MAX;
-        for &facet_ord in &self.facet_ords {
-            let collapsed_ord = self.current_segment_collapse_mapping[facet_ord as usize];
-            self.current_segment_counts[collapsed_ord] += if collapsed_ord == previous_collapsed_ord
-            {
+        for &facet_ord in &self.facet_ords_buf {
+            let collapsed_ord = self.collapse_mapping[facet_ord as usize];
+            self.counts[collapsed_ord] += if collapsed_ord == previous_collapsed_ord {
                 0
             } else {
                 1
@@ -421,8 +355,23 @@ impl Collector for FacetCollector {
         }
     }
 
-    fn requires_scoring(&self) -> bool {
-        false
+    /// Returns the results of the collection.
+    ///
+    /// This method does not just return the counters,
+    /// it also translates the facet ordinals of the last segment.
+    fn harvest(self) -> FacetCounts {
+        let mut facet_counts = BTreeMap::new();
+        let facet_dict = self.reader.facet_dict();
+        for (collapsed_facet_ord, count) in self.counts.iter().cloned().enumerate() {
+            if count == 0 {
+                continue;
+            }
+            let mut facet = vec![];
+            let facet_ord = self.collapse_facet_ords[collapsed_facet_ord];
+            facet_dict.ord_to_term(facet_ord as u64, &mut facet);
+            facet_counts.insert(unsafe { Facet::from_encoded(facet) }, count);
+        }
+        FacetCounts { facet_counts }
     }
 }
 
@@ -476,9 +425,8 @@ impl FacetCounts {
             heap.push(Hit { count, facet });
         }
 
-        let mut lowest_count: u64 = heap.peek().map(|hit| hit.count)
-            .unwrap_or(u64::MIN); //< the `unwrap_or` case may be triggered but the value
-                                  // is never used in that case.
+        let mut lowest_count: u64 = heap.peek().map(|hit| hit.count).unwrap_or(u64::MIN); //< the `unwrap_or` case may be triggered but the value
+                                                                                          // is never used in that case.
 
         for (facet, count) in it {
             if count > lowest_count {
@@ -504,14 +452,14 @@ mod tests {
     use core::Index;
     use query::AllQuery;
     use rand::distributions::Uniform;
+    use rand::prelude::SliceRandom;
     use rand::{thread_rng, Rng};
-    use schema::Field;
-    use schema::{Document, Facet, SchemaBuilder};
+    use schema::{Document, Facet, Field, Schema};
     use std::iter;
 
     #[test]
     fn test_facet_collector_drilldown() {
-        let mut schema_builder = SchemaBuilder::new();
+        let mut schema_builder = Schema::builder();
         let facet_field = schema_builder.add_facet_field("facet");
         let schema = schema_builder.build();
         let index = Index::create_in_ram(schema);
@@ -526,8 +474,7 @@ mod tests {
                 n /= 4;
                 let leaf = n % 5;
                 Facet::from(&format!("/top{}/mid{}/leaf{}", top, mid, leaf))
-            })
-            .collect();
+            }).collect();
         for i in 0..num_facets * 10 {
             let mut doc = Document::new();
             doc.add_facet(facet_field, facets[i % num_facets].clone());
@@ -536,12 +483,10 @@ mod tests {
         index_writer.commit().unwrap();
         index.load_searchers().unwrap();
         let searcher = index.searcher();
-
         let mut facet_collector = FacetCollector::for_field(facet_field);
         facet_collector.add_facet(Facet::from("/top1"));
-        searcher.search(&AllQuery, &mut facet_collector).unwrap();
+        let counts = searcher.search(&AllQuery, &facet_collector).unwrap();
 
-        let counts: FacetCounts = facet_collector.harvest();
         {
             let facets: Vec<(String, u64)> = counts
                 .get("/top1")
@@ -554,7 +499,8 @@ mod tests {
                     ("/top1/mid1", 50),
                     ("/top1/mid2", 50),
                     ("/top1/mid3", 50),
-                ].iter()
+                ]
+                    .iter()
                     .map(|&(facet_str, count)| (String::from(facet_str), count))
                     .collect::<Vec<_>>()
             );
@@ -574,7 +520,7 @@ mod tests {
 
     #[test]
     fn test_doc_unsorted_multifacet() {
-        let mut schema_builder = SchemaBuilder::new();
+        let mut schema_builder = Schema::builder();
         let facet_field = schema_builder.add_facet_field("facets");
         let schema = schema_builder.build();
         let index = Index::create_in_ram(schema);
@@ -591,8 +537,7 @@ mod tests {
         assert_eq!(searcher.num_docs(), 1);
         let mut facet_collector = FacetCollector::for_field(facet_field);
         facet_collector.add_facet("/subjects");
-        searcher.search(&AllQuery, &mut facet_collector).unwrap();
-        let counts = facet_collector.harvest();
+        let counts = searcher.search(&AllQuery, &facet_collector).unwrap();
         let facets: Vec<(&Facet, u64)> = counts.get("/subjects").collect();
         assert_eq!(facets[0].1, 1);
     }
@@ -606,7 +551,7 @@ mod tests {
 
     #[test]
     fn test_facet_collector_topk() {
-        let mut schema_builder = SchemaBuilder::new();
+        let mut schema_builder = Schema::builder();
         let facet_field = schema_builder.add_facet_field("facet");
         let schema = schema_builder.build();
         let index = Index::create_in_ram(schema);
@@ -618,10 +563,14 @@ mod tests {
                 let facet = Facet::from(&format!("/facet/{}", c));
                 let doc = doc!(facet_field => facet);
                 iter::repeat(doc).take(count)
-            })
-            .map(|mut doc| { doc.add_facet(facet_field, &format!("/facet/{}", thread_rng().sample(&uniform) )); doc})
-            .collect();
-        thread_rng().shuffle(&mut docs[..]);
+            }).map(|mut doc| {
+                doc.add_facet(
+                    facet_field,
+                    &format!("/facet/{}", thread_rng().sample(&uniform)),
+                );
+                doc
+            }).collect();
+        docs[..].shuffle(&mut thread_rng());
 
         let mut index_writer = index.writer_with_num_threads(1, 3_000_000).unwrap();
         for doc in docs {
@@ -634,9 +583,8 @@ mod tests {
 
         let mut facet_collector = FacetCollector::for_field(facet_field);
         facet_collector.add_facet("/facet");
-        searcher.search(&AllQuery, &mut facet_collector).unwrap();
+        let counts: FacetCounts = searcher.search(&AllQuery, &facet_collector).unwrap();
 
-        let counts: FacetCounts = facet_collector.harvest();
         {
             let facets: Vec<(&Facet, u64)> = counts.top_k("/facet", 3);
             assert_eq!(
@@ -659,13 +607,13 @@ mod bench {
     use query::AllQuery;
     use rand::{thread_rng, Rng};
     use schema::Facet;
-    use schema::SchemaBuilder;
+    use schema::Schema;
     use test::Bencher;
     use Index;
 
     #[bench]
     fn bench_facet_collector(b: &mut Bencher) {
-        let mut schema_builder = SchemaBuilder::new();
+        let mut schema_builder = Schema::builder();
         let facet_field = schema_builder.add_facet_field("facet");
         let schema = schema_builder.build();
         let index = Index::create_in_ram(schema);
@@ -689,8 +637,8 @@ mod bench {
 
         b.iter(|| {
             let searcher = index.searcher();
-            let mut facet_collector = FacetCollector::for_field(facet_field);
-            searcher.search(&AllQuery, &mut facet_collector).unwrap();
+            let facet_collector = FacetCollector::for_field(facet_field);
+            searcher.search(&AllQuery, &facet_collector).unwrap();
         });
     }
 }

src/collector/int_facet_collector.rs

@@ -79,7 +79,7 @@ mod tests {
     // make sure we have facet counters correctly filled
     fn test_facet_collector_results() {
 
-        let mut schema_builder = schema::SchemaBuilder::new();
+        let mut schema_builder = schema::Schema::builder();
         let num_field_i64 = schema_builder.add_i64_field("num_i64", FAST);
         let num_field_u64 = schema_builder.add_u64_field("num_u64", FAST);
         let text_field = schema_builder.add_text_field("text", STRING);

src/collector/mod.rs

@@ -1,7 +1,91 @@
 /*!
-Defines how the documents matching a search query should be processed.
+
+# Collectors
+
+Collectors define the information you want to extract from the documents matching the queries.
+In tantivy jargon, we call this information your search "fruit".
+
+Your fruit could for instance be :
+- [the count of matching documents](./struct.Count.html)
+- [the top 10 documents, by relevancy or by a fast field](./struct.TopDocs.html)
+- [facet counts](./struct.FacetCollector.html)
+
+At one point in your code, you will trigger the actual search operation by calling
+[the `search(...)` method of your `Searcher` object](../struct.Searcher.html#method.search).
+This call will look like this.
+
+```verbatim
+let fruit = searcher.search(&query, &collector)?;
+```
+
+Here the type of fruit is actually determined as an associated type of the collector (`Collector::Fruit`).
+
+
+# Combining several collectors
+
+A rich search experience often requires to run several collectors on your search query.
+For instance,
+- selecting the top-K products matching your query
+- counting the matching documents
+- computing several facets
+- computing statistics about the matching product prices
+
+A simple and efficient way to do that is to pass your collectors as one tuple.
+The resulting `Fruit` will then be a typed tuple with each collector's original fruits
+in their respective position.
+
+```rust
+# extern crate tantivy;
+# use tantivy::schema::*;
+# use tantivy::*;
+# use tantivy::query::*;
+use tantivy::collector::{Count, TopDocs};
+#
+# fn main() -> tantivy::Result<()> {
+# let mut schema_builder = Schema::builder();
+#     let title = schema_builder.add_text_field("title", TEXT);
+#     let schema = schema_builder.build();
+#     let index = Index::create_in_ram(schema);
+#     let mut index_writer = index.writer(3_000_000)?;
+#       index_writer.add_document(doc!(
+#       title => "The Name of the Wind",
+#      ));
+#     index_writer.add_document(doc!(
+#        title => "The Diary of Muadib",
+#     ));
+#     index_writer.commit().unwrap();
+#     index.load_searchers()?;
+#     let searcher = index.searcher();
+#     let query_parser = QueryParser::for_index(&index, vec![title]);
+#     let query = query_parser.parse_query("diary")?;
+let (doc_count, top_docs): (usize, Vec<(Score, DocAddress)>) =
+    searcher.search(&query, &(Count, TopDocs::with_limit(2)))?;
+#     Ok(())
+# }
+```
+
+The `Collector` trait is implemented for up to 4 collectors.
+If you have more than 4 collectors, you can either group them into
+tuples of tuples `(a,(b,(c,d)))`, or rely on `MultiCollector`'s.
+
+# Combining several collectors dynamically
+
+Combining collectors into a tuple is a zero-cost abstraction: everything
+happens as if you had manually implemented a single collector
+combining all of our features.
+
+Unfortunately it requires you to know at compile time your collector types.
+If on the other hand, the collectors depend on some query parameter,
+you can rely on `MultiCollector`'s.
+
+
+# Implementing your own collectors.
+
+See the `custom_collector` example.
+
 */
 
+use downcast;
 use DocId;
 use Result;
 use Score;
@@ -9,238 +93,275 @@ use SegmentLocalId;
 use SegmentReader;
 
 mod count_collector;
-pub use self::count_collector::CountCollector;
+pub use self::count_collector::Count;
 
 mod multi_collector;
 pub use self::multi_collector::MultiCollector;
 
 mod top_collector;
-pub use self::top_collector::TopCollector;
+
+mod top_score_collector;
+pub use self::top_score_collector::TopDocs;
+
+mod top_field_collector;
+pub use self::top_field_collector::TopDocsByField;
 
 mod facet_collector;
 pub use self::facet_collector::FacetCollector;
 
-mod chained_collector;
-pub use self::chained_collector::{chain, ChainedCollector};
+/// `Fruit` is the type for the result of our collection.
+/// e.g. `usize` for the `Count` collector.
+pub trait Fruit: Send + downcast::Any {}
+
+impl<T> Fruit for T where T: Send + downcast::Any {}
 
 /// Collectors are in charge of collecting and retaining relevant
 /// information from the document found and scored by the query.
 ///
-///
 /// For instance,
 ///
 /// - keeping track of the top 10 best documents
 /// - computing a breakdown over a fast field
 /// - computing the number of documents matching the query
 ///
-/// Queries are in charge of pushing the `DocSet` to the collector.
+/// Our search index is in fact a collection of segments, so
+/// a `Collector` trait is actually more of a factory to instance
+/// `SegmentCollector`s for each segments.
 ///
-/// As they work on multiple segments, they first inform
-/// the collector of a change in a segment and then
-/// call the `collect` method to push the document to the collector.
-///
-/// Temporally, our collector will receive calls
-/// - `.set_segment(0, segment_reader_0)`
-/// - `.collect(doc0_of_segment_0)`
-/// - `.collect(...)`
-/// - `.collect(last_doc_of_segment_0)`
-/// - `.set_segment(1, segment_reader_1)`
-/// - `.collect(doc0_of_segment_1)`
-/// - `.collect(...)`
-/// - `.collect(last_doc_of_segment_1)`
-/// - `...`
-/// - `.collect(last_doc_of_last_segment)`
+/// The collection logic itself is in the `SegmentCollector`.
 ///
 /// Segments are not guaranteed to be visited in any specific order.
-pub trait Collector {
+pub trait Collector: Sync {
+    /// `Fruit` is the type for the result of our collection.
+    /// e.g. `usize` for the `Count` collector.
+    type Fruit: Fruit;
+
+    /// Type of the `SegmentCollector` associated to this collector.
+    type Child: SegmentCollector<Fruit = Self::Fruit>;
+
     /// `set_segment` is called before beginning to enumerate
     /// on this segment.
-    fn set_segment(
-        &mut self,
+    fn for_segment(
+        &self,
         segment_local_id: SegmentLocalId,
         segment: &SegmentReader,
-    ) -> Result<()>;
-    /// The query pushes the scored document to the collector via this method.
-    fn collect(&mut self, doc: DocId, score: Score);
+    ) -> Result<Self::Child>;
 
     /// Returns true iff the collector requires to compute scores for documents.
     fn requires_scoring(&self) -> bool;
+
+    /// Combines the fruit associated to the collection of each segments
+    /// into one fruit.
+    fn merge_fruits(&self, segment_fruits: Vec<Self::Fruit>) -> Result<Self::Fruit>;
 }
 
-impl<'a, C: Collector> Collector for &'a mut C {
-    fn set_segment(
-        &mut self,
-        segment_local_id: SegmentLocalId,
-        segment: &SegmentReader,
-    ) -> Result<()> {
-        (*self).set_segment(segment_local_id, segment)
-    }
+/// The `SegmentCollector` is the trait in charge of defining the
+/// collect operation at the scale of the segment.
+///
+/// `.collect(doc, score)` will be called for every documents
+/// matching the query.
+pub trait SegmentCollector: 'static {
+    /// `Fruit` is the type for the result of our collection.
+    /// e.g. `usize` for the `Count` collector.
+    type Fruit: Fruit;
+
     /// The query pushes the scored document to the collector via this method.
-    fn collect(&mut self, doc: DocId, score: Score) {
-        C::collect(self, doc, score)
+    fn collect(&mut self, doc: DocId, score: Score);
+
+    /// Extract the fruit of the collection from the `SegmentCollector`.
+    fn harvest(self) -> Self::Fruit;
+}
+
+// -----------------------------------------------
+// Tuple implementations.
+
+impl<Left, Right> Collector for (Left, Right)
+where
+    Left: Collector,
+    Right: Collector,
+{
+    type Fruit = (Left::Fruit, Right::Fruit);
+    type Child = (Left::Child, Right::Child);
+
+    fn for_segment(&self, segment_local_id: u32, segment: &SegmentReader) -> Result<Self::Child> {
+        let left = self.0.for_segment(segment_local_id, segment)?;
+        let right = self.1.for_segment(segment_local_id, segment)?;
+        Ok((left, right))
     }
 
     fn requires_scoring(&self) -> bool {
-        C::requires_scoring(self)
+        self.0.requires_scoring() || self.1.requires_scoring()
     }
+
+    fn merge_fruits(
+        &self,
+        children: Vec<(Left::Fruit, Right::Fruit)>,
+    ) -> Result<(Left::Fruit, Right::Fruit)> {
+        let mut left_fruits = vec![];
+        let mut right_fruits = vec![];
+        for (left_fruit, right_fruit) in children {
+            left_fruits.push(left_fruit);
+            right_fruits.push(right_fruit);
+        }
+        Ok((
+            self.0.merge_fruits(left_fruits)?,
+            self.1.merge_fruits(right_fruits)?,
+        ))
+    }
+}
+
+impl<Left, Right> SegmentCollector for (Left, Right)
+where
+    Left: SegmentCollector,
+    Right: SegmentCollector,
+{
+    type Fruit = (Left::Fruit, Right::Fruit);
+
+    fn collect(&mut self, doc: DocId, score: Score) {
+        self.0.collect(doc, score);
+        self.1.collect(doc, score);
+    }
+
+    fn harvest(self) -> <Self as SegmentCollector>::Fruit {
+        (self.0.harvest(), self.1.harvest())
+    }
+}
+
+// 3-Tuple
+
+impl<One, Two, Three> Collector for (One, Two, Three)
+where
+    One: Collector,
+    Two: Collector,
+    Three: Collector,
+{
+    type Fruit = (One::Fruit, Two::Fruit, Three::Fruit);
+    type Child = (One::Child, Two::Child, Three::Child);
+
+    fn for_segment(&self, segment_local_id: u32, segment: &SegmentReader) -> Result<Self::Child> {
+        let one = self.0.for_segment(segment_local_id, segment)?;
+        let two = self.1.for_segment(segment_local_id, segment)?;
+        let three = self.2.for_segment(segment_local_id, segment)?;
+        Ok((one, two, three))
+    }
+
+    fn requires_scoring(&self) -> bool {
+        self.0.requires_scoring() || self.1.requires_scoring() || self.2.requires_scoring()
+    }
+
+    fn merge_fruits(&self, children: Vec<Self::Fruit>) -> Result<Self::Fruit> {
+        let mut one_fruits = vec![];
+        let mut two_fruits = vec![];
+        let mut three_fruits = vec![];
+        for (one_fruit, two_fruit, three_fruit) in children {
+            one_fruits.push(one_fruit);
+            two_fruits.push(two_fruit);
+            three_fruits.push(three_fruit);
+        }
+        Ok((
+            self.0.merge_fruits(one_fruits)?,
+            self.1.merge_fruits(two_fruits)?,
+            self.2.merge_fruits(three_fruits)?,
+        ))
+    }
+}
+
+impl<One, Two, Three> SegmentCollector for (One, Two, Three)
+where
+    One: SegmentCollector,
+    Two: SegmentCollector,
+    Three: SegmentCollector,
+{
+    type Fruit = (One::Fruit, Two::Fruit, Three::Fruit);
+
+    fn collect(&mut self, doc: DocId, score: Score) {
+        self.0.collect(doc, score);
+        self.1.collect(doc, score);
+        self.2.collect(doc, score);
+    }
+
+    fn harvest(self) -> <Self as SegmentCollector>::Fruit {
+        (self.0.harvest(), self.1.harvest(), self.2.harvest())
+    }
+}
+
+// 4-Tuple
+
+impl<One, Two, Three, Four> Collector for (One, Two, Three, Four)
+where
+    One: Collector,
+    Two: Collector,
+    Three: Collector,
+    Four: Collector,
+{
+    type Fruit = (One::Fruit, Two::Fruit, Three::Fruit, Four::Fruit);
+    type Child = (One::Child, Two::Child, Three::Child, Four::Child);
+
+    fn for_segment(&self, segment_local_id: u32, segment: &SegmentReader) -> Result<Self::Child> {
+        let one = self.0.for_segment(segment_local_id, segment)?;
+        let two = self.1.for_segment(segment_local_id, segment)?;
+        let three = self.2.for_segment(segment_local_id, segment)?;
+        let four = self.3.for_segment(segment_local_id, segment)?;
+        Ok((one, two, three, four))
+    }
+
+    fn requires_scoring(&self) -> bool {
+        self.0.requires_scoring()
+            || self.1.requires_scoring()
+            || self.2.requires_scoring()
+            || self.3.requires_scoring()
+    }
+
+    fn merge_fruits(&self, children: Vec<Self::Fruit>) -> Result<Self::Fruit> {
+        let mut one_fruits = vec![];
+        let mut two_fruits = vec![];
+        let mut three_fruits = vec![];
+        let mut four_fruits = vec![];
+        for (one_fruit, two_fruit, three_fruit, four_fruit) in children {
+            one_fruits.push(one_fruit);
+            two_fruits.push(two_fruit);
+            three_fruits.push(three_fruit);
+            four_fruits.push(four_fruit);
+        }
+        Ok((
+            self.0.merge_fruits(one_fruits)?,
+            self.1.merge_fruits(two_fruits)?,
+            self.2.merge_fruits(three_fruits)?,
+            self.3.merge_fruits(four_fruits)?,
+        ))
+    }
+}
+
+impl<One, Two, Three, Four> SegmentCollector for (One, Two, Three, Four)
+where
+    One: SegmentCollector,
+    Two: SegmentCollector,
+    Three: SegmentCollector,
+    Four: SegmentCollector,
+{
+    type Fruit = (One::Fruit, Two::Fruit, Three::Fruit, Four::Fruit);
+
+    fn collect(&mut self, doc: DocId, score: Score) {
+        self.0.collect(doc, score);
+        self.1.collect(doc, score);
+        self.2.collect(doc, score);
+        self.3.collect(doc, score);
+    }
+
+    fn harvest(self) -> <Self as SegmentCollector>::Fruit {
+        (
+            self.0.harvest(),
+            self.1.harvest(),
+            self.2.harvest(),
+            self.3.harvest(),
+        )
+    }
+}
+
+#[allow(missing_docs)]
+mod downcast_impl {
+    downcast!(super::Fruit);
 }
 
 #[cfg(test)]
-pub mod tests {
-
-    use super::*;
-    use core::SegmentReader;
-    use fastfield::BytesFastFieldReader;
-    use fastfield::FastFieldReader;
-    use schema::Field;
-    use DocId;
-    use Score;
-    use SegmentLocalId;
-
-    /// Stores all of the doc ids.
-    /// This collector is only used for tests.
-    /// It is unusable in practise, as it does not store
-    /// the segment ordinals
-    pub struct TestCollector {
-        offset: DocId,
-        segment_max_doc: DocId,
-        docs: Vec<DocId>,
-        scores: Vec<Score>,
-    }
-
-    impl TestCollector {
-        /// Return the exhalist of documents.
-        pub fn docs(self) -> Vec<DocId> {
-            self.docs
-        }
-
-        pub fn scores(self) -> Vec<Score> {
-            self.scores
-        }
-    }
-
-    impl Default for TestCollector {
-        fn default() -> TestCollector {
-            TestCollector {
-                offset: 0,
-                segment_max_doc: 0,
-                docs: Vec::new(),
-                scores: Vec::new(),
-            }
-        }
-    }
-
-    impl Collector for TestCollector {
-        fn set_segment(&mut self, _: SegmentLocalId, reader: &SegmentReader) -> Result<()> {
-            self.offset += self.segment_max_doc;
-            self.segment_max_doc = reader.max_doc();
-            Ok(())
-        }
-
-        fn collect(&mut self, doc: DocId, score: Score) {
-            self.docs.push(doc + self.offset);
-            self.scores.push(score);
-        }
-
-        fn requires_scoring(&self) -> bool {
-            true
-        }
-    }
-
-    /// Collects in order all of the fast fields for all of the
-    /// doc in the `DocSet`
-    ///
-    /// This collector is mainly useful for tests.
-    pub struct FastFieldTestCollector {
-        vals: Vec<u64>,
-        field: Field,
-        ff_reader: Option<FastFieldReader<u64>>,
-    }
-
-    impl FastFieldTestCollector {
-        pub fn for_field(field: Field) -> FastFieldTestCollector {
-            FastFieldTestCollector {
-                vals: Vec::new(),
-                field,
-                ff_reader: None,
-            }
-        }
-
-        pub fn vals(self) -> Vec<u64> {
-            self.vals
-        }
-    }
-
-    impl Collector for FastFieldTestCollector {
-        fn set_segment(&mut self, _: SegmentLocalId, reader: &SegmentReader) -> Result<()> {
-            self.ff_reader = Some(reader.fast_field_reader(self.field)?);
-            Ok(())
-        }
-
-        fn collect(&mut self, doc: DocId, _score: Score) {
-            let val = self.ff_reader.as_ref().unwrap().get(doc);
-            self.vals.push(val);
-        }
-        fn requires_scoring(&self) -> bool {
-            false
-        }
-    }
-
-    /// Collects in order all of the fast field bytes for all of the
-    /// docs in the `DocSet`
-    ///
-    /// This collector is mainly useful for tests.
-    pub struct BytesFastFieldTestCollector {
-        vals: Vec<u8>,
-        field: Field,
-        ff_reader: Option<BytesFastFieldReader>,
-    }
-
-    impl BytesFastFieldTestCollector {
-        pub fn for_field(field: Field) -> BytesFastFieldTestCollector {
-            BytesFastFieldTestCollector {
-                vals: Vec::new(),
-                field,
-                ff_reader: None,
-            }
-        }
-
-        pub fn vals(self) -> Vec<u8> {
-            self.vals
-        }
-    }
-
-    impl Collector for BytesFastFieldTestCollector {
-        fn set_segment(&mut self, _segment_local_id: u32, segment: &SegmentReader) -> Result<()> {
-            self.ff_reader = Some(segment.bytes_fast_field_reader(self.field)?);
-            Ok(())
-        }
-
-        fn collect(&mut self, doc: u32, _score: f32) {
-            let val = self.ff_reader.as_ref().unwrap().get_val(doc);
-            self.vals.extend(val);
-        }
-
-        fn requires_scoring(&self) -> bool {
-            false
-        }
-    }
-}
-
-#[cfg(all(test, feature = "unstable"))]
-mod bench {
-    use collector::{Collector, CountCollector};
-    use test::Bencher;
-
-    #[bench]
-    fn build_collector(b: &mut Bencher) {
-        b.iter(|| {
-            let mut count_collector = CountCollector::default();
-            let docs: Vec<u32> = (0..1_000_000).collect();
-            for doc in docs {
-                count_collector.collect(doc, 1f32);
-            }
-            count_collector.count()
-        });
-    }
-}
+pub mod tests;

src/collector/multi_collector.rs

@@ -1,9 +1,96 @@
 use super::Collector;
+use super::SegmentCollector;
+use collector::Fruit;
+use downcast::Downcast;
+use std::marker::PhantomData;
 use DocId;
 use Result;
 use Score;
 use SegmentLocalId;
 use SegmentReader;
+use TantivyError;
+
+pub struct MultiFruit {
+    sub_fruits: Vec<Option<Box<Fruit>>>,
+}
+
+pub struct CollectorWrapper<TCollector: Collector>(TCollector);
+
+impl<TCollector: Collector> Collector for CollectorWrapper<TCollector> {
+    type Fruit = Box<Fruit>;
+    type Child = Box<BoxableSegmentCollector>;
+
+    fn for_segment(
+        &self,
+        segment_local_id: u32,
+        reader: &SegmentReader,
+    ) -> Result<Box<BoxableSegmentCollector>> {
+        let child = self.0.for_segment(segment_local_id, reader)?;
+        Ok(Box::new(SegmentCollectorWrapper(child)))
+    }
+
+    fn requires_scoring(&self) -> bool {
+        self.0.requires_scoring()
+    }
+
+    fn merge_fruits(&self, children: Vec<<Self as Collector>::Fruit>) -> Result<Box<Fruit>> {
+        let typed_fruit: Vec<TCollector::Fruit> = children
+            .into_iter()
+            .map(|untyped_fruit| {
+                Downcast::<TCollector::Fruit>::downcast(untyped_fruit)
+                    .map(|boxed_but_typed| *boxed_but_typed)
+                    .map_err(|e| {
+                        let err_msg = format!("Failed to cast child collector fruit. {:?}", e);
+                        TantivyError::InvalidArgument(err_msg)
+                    })
+            }).collect::<Result<_>>()?;
+        let merged_fruit = self.0.merge_fruits(typed_fruit)?;
+        Ok(Box::new(merged_fruit))
+    }
+}
+
+impl SegmentCollector for Box<BoxableSegmentCollector> {
+    type Fruit = Box<Fruit>;
+
+    fn collect(&mut self, doc: u32, score: f32) {
+        self.as_mut().collect(doc, score);
+    }
+
+    fn harvest(self) -> Box<Fruit> {
+        BoxableSegmentCollector::harvest_from_box(self)
+    }
+}
+
+pub trait BoxableSegmentCollector {
+    fn collect(&mut self, doc: u32, score: f32);
+    fn harvest_from_box(self: Box<Self>) -> Box<Fruit>;
+}
+
+pub struct SegmentCollectorWrapper<TSegmentCollector: SegmentCollector>(TSegmentCollector);
+
+impl<TSegmentCollector: SegmentCollector> BoxableSegmentCollector
+    for SegmentCollectorWrapper<TSegmentCollector>
+{
+    fn collect(&mut self, doc: u32, score: f32) {
+        self.0.collect(doc, score);
+    }
+
+    fn harvest_from_box(self: Box<Self>) -> Box<Fruit> {
+        Box::new(self.0.harvest())
+    }
+}
+
+pub struct FruitHandle<TFruit: Fruit> {
+    pos: usize,
+    _phantom: PhantomData<TFruit>,
+}
+
+impl<TFruit: Fruit> FruitHandle<TFruit> {
+    pub fn extract(self, fruits: &mut MultiFruit) -> TFruit {
+        let boxed_fruit = fruits.sub_fruits[self.pos].take().expect("");
+        *Downcast::<TFruit>::downcast(boxed_fruit).expect("Failed")
+    }
+}
 
 /// Multicollector makes it possible to collect on more than one collector.
 /// It should only be used for use cases where the Collector types is unknown
@@ -13,14 +100,14 @@ use SegmentReader;
 /// ```rust
 /// #[macro_use]
 /// extern crate tantivy;
-/// use tantivy::schema::{SchemaBuilder, TEXT};
+/// use tantivy::schema::{Schema, TEXT};
 /// use tantivy::{Index, Result};
-/// use tantivy::collector::{CountCollector, TopCollector, MultiCollector};
+/// use tantivy::collector::{Count, TopDocs, MultiCollector};
 /// use tantivy::query::QueryParser;
 ///
 /// # fn main() { example().unwrap(); }
 /// fn example() -> Result<()> {
-///     let mut schema_builder = SchemaBuilder::new();
+///     let mut schema_builder = Schema::builder();
 ///     let title = schema_builder.add_text_field("title", TEXT);
 ///     let schema = schema_builder.build();
 ///     let index = Index::create_in_ram(schema);
@@ -44,55 +131,114 @@ use SegmentReader;
 ///     index.load_searchers()?;
 ///     let searcher = index.searcher();
 ///
-///     {
-///         let mut top_collector = TopCollector::with_limit(2);
-///         let mut count_collector = CountCollector::default();
-///         {
-///             let mut collectors =
-///                 MultiCollector::from(vec![&mut top_collector, &mut count_collector]);
-///             let query_parser = QueryParser::for_index(&index, vec![title]);
-///             let query = query_parser.parse_query("diary")?;
-///             searcher.search(&*query, &mut collectors).unwrap();
-///         }
-///         assert_eq!(count_collector.count(), 2);
-///         assert!(top_collector.at_capacity());
-///     }
+///     let mut collectors = MultiCollector::new();
+///     let top_docs_handle = collectors.add_collector(TopDocs::with_limit(2));
+///     let count_handle = collectors.add_collector(Count);
+///     let query_parser = QueryParser::for_index(&index, vec![title]);
+///     let query = query_parser.parse_query("diary")?;
+///     let mut multi_fruit = searcher.search(&query, &collectors)?;
+///
+///     let count = count_handle.extract(&mut multi_fruit);
+///     let top_docs = top_docs_handle.extract(&mut multi_fruit);
+///
+///     # assert_eq!(count, 2);
+///     # assert_eq!(top_docs.len(), 2);
 ///
 ///     Ok(())
 /// }
 /// ```
 pub struct MultiCollector<'a> {
-    collectors: Vec<&'a mut Collector>,
+    collector_wrappers:
+        Vec<Box<Collector<Child = Box<BoxableSegmentCollector>, Fruit = Box<Fruit>> + 'a>>,
 }
 
 impl<'a> MultiCollector<'a> {
-    /// Constructor
-    pub fn from(collectors: Vec<&'a mut Collector>) -> MultiCollector {
-        MultiCollector { collectors }
+    /// Create a new `MultiCollector`
+    pub fn new() -> MultiCollector<'a> {
+        MultiCollector {
+            collector_wrappers: Vec::new(),
+        }
+    }
+
+    /// Add a new collector to our `MultiCollector`.
+    pub fn add_collector<'b: 'a, TCollector: Collector + 'b>(
+        &mut self,
+        collector: TCollector,
+    ) -> FruitHandle<TCollector::Fruit> {
+        let pos = self.collector_wrappers.len();
+        self.collector_wrappers
+            .push(Box::new(CollectorWrapper(collector)));
+        FruitHandle {
+            pos,
+            _phantom: PhantomData,
+        }
     }
 }
 
 impl<'a> Collector for MultiCollector<'a> {
-    fn set_segment(
-        &mut self,
+    type Fruit = MultiFruit;
+    type Child = MultiCollectorChild;
+
+    fn for_segment(
+        &self,
         segment_local_id: SegmentLocalId,
         segment: &SegmentReader,
-    ) -> Result<()> {
-        for collector in &mut self.collectors {
-            collector.set_segment(segment_local_id, segment)?;
-        }
-        Ok(())
+    ) -> Result<MultiCollectorChild> {
+        let children = self
+            .collector_wrappers
+            .iter()
+            .map(|collector_wrapper| collector_wrapper.for_segment(segment_local_id, segment))
+            .collect::<Result<Vec<_>>>()?;
+        Ok(MultiCollectorChild { children })
     }
 
+    fn requires_scoring(&self) -> bool {
+        self.collector_wrappers.iter().any(|c| c.requires_scoring())
+    }
+
+    fn merge_fruits(&self, segments_multifruits: Vec<MultiFruit>) -> Result<MultiFruit> {
+        let mut segment_fruits_list: Vec<Vec<Box<Fruit>>> = (0..self.collector_wrappers.len())
+            .map(|_| Vec::with_capacity(segments_multifruits.len()))
+            .collect::<Vec<_>>();
+        for segment_multifruit in segments_multifruits {
+            for (idx, segment_fruit_opt) in segment_multifruit.sub_fruits.into_iter().enumerate() {
+                if let Some(segment_fruit) = segment_fruit_opt {
+                    segment_fruits_list[idx].push(segment_fruit);
+                }
+            }
+        }
+        let sub_fruits = self
+            .collector_wrappers
+            .iter()
+            .zip(segment_fruits_list)
+            .map(|(child_collector, segment_fruits)| {
+                Ok(Some(child_collector.merge_fruits(segment_fruits)?))
+            }).collect::<Result<_>>()?;
+        Ok(MultiFruit { sub_fruits })
+    }
+}
+
+pub struct MultiCollectorChild {
+    children: Vec<Box<BoxableSegmentCollector>>,
+}
+
+impl SegmentCollector for MultiCollectorChild {
+    type Fruit = MultiFruit;
+
     fn collect(&mut self, doc: DocId, score: Score) {
-        for collector in &mut self.collectors {
-            collector.collect(doc, score);
+        for child in &mut self.children {
+            child.collect(doc, score);
         }
     }
-    fn requires_scoring(&self) -> bool {
-        self.collectors
-            .iter()
-            .any(|collector| collector.requires_scoring())
+
+    fn harvest(self) -> MultiFruit {
+        MultiFruit {
+            sub_fruits: self
+                .children
+                .into_iter()
+                .map(|child| Some(child.harvest()))
+                .collect(),
+        }
     }
 }
 
@@ -100,20 +246,42 @@ impl<'a> Collector for MultiCollector<'a> {
 mod tests {
 
     use super::*;
-    use collector::{Collector, CountCollector, TopCollector};
+    use collector::{Count, TopDocs};
+    use query::TermQuery;
+    use schema::IndexRecordOption;
+    use schema::{Schema, TEXT};
+    use Index;
+    use Term;
 
     #[test]
     fn test_multi_collector() {
-        let mut top_collector = TopCollector::with_limit(2);
-        let mut count_collector = CountCollector::default();
+        let mut schema_builder = Schema::builder();
+        let text = schema_builder.add_text_field("text", TEXT);
+        let schema = schema_builder.build();
+
+        let index = Index::create_in_ram(schema);
         {
-            let mut collectors =
-                MultiCollector::from(vec![&mut top_collector, &mut count_collector]);
-            collectors.collect(1, 0.2);
-            collectors.collect(2, 0.1);
-            collectors.collect(3, 0.5);
+            let mut index_writer = index.writer_with_num_threads(1, 3_000_000).unwrap();
+            index_writer.add_document(doc!(text=>"abc"));
+            index_writer.add_document(doc!(text=>"abc abc abc"));
+            index_writer.add_document(doc!(text=>"abc abc"));
+            index_writer.commit().unwrap();
+            index_writer.add_document(doc!(text=>""));
+            index_writer.add_document(doc!(text=>"abc abc abc abc"));
+            index_writer.add_document(doc!(text=>"abc"));
+            index_writer.commit().unwrap();
         }
-        assert_eq!(count_collector.count(), 3);
-        assert!(top_collector.at_capacity());
+        index.load_searchers().unwrap();
+        let searcher = index.searcher();
+        let term = Term::from_field_text(text, "abc");
+        let query = TermQuery::new(term, IndexRecordOption::Basic);
+
+        let mut collectors = MultiCollector::new();
+        let topdocs_handler = collectors.add_collector(TopDocs::with_limit(2));
+        let count_handler = collectors.add_collector(Count);
+        let mut multifruits = searcher.search(&query, &mut collectors).unwrap();
+
+        assert_eq!(count_handler.extract(&mut multifruits), 5);
+        assert_eq!(topdocs_handler.extract(&mut multifruits).len(), 2);
     }
 }

src/collector/tests.rs

@@ -0,0 +1,201 @@
+use super::*;
+use core::SegmentReader;
+use fastfield::BytesFastFieldReader;
+use fastfield::FastFieldReader;
+use schema::Field;
+use DocAddress;
+use DocId;
+use Score;
+use SegmentLocalId;
+
+/// Stores all of the doc ids.
+/// This collector is only used for tests.
+/// It is unusable in pr
+///
+/// actise, as it does not store
+/// the segment ordinals
+pub struct TestCollector;
+
+pub struct TestSegmentCollector {
+    segment_id: SegmentLocalId,
+    fruit: TestFruit,
+}
+
+#[derive(Default)]
+pub struct TestFruit {
+    docs: Vec<DocAddress>,
+    scores: Vec<Score>,
+}
+
+impl TestFruit {
+    /// Return the list of matching documents exhaustively.
+    pub fn docs(&self) -> &[DocAddress] {
+        &self.docs[..]
+    }
+
+    pub fn scores(&self) -> &[Score] {
+        &self.scores[..]
+    }
+}
+
+impl Collector for TestCollector {
+    type Fruit = TestFruit;
+    type Child = TestSegmentCollector;
+
+    fn for_segment(
+        &self,
+        segment_id: SegmentLocalId,
+        _reader: &SegmentReader,
+    ) -> Result<TestSegmentCollector> {
+        Ok(TestSegmentCollector {
+            segment_id,
+            fruit: TestFruit::default(),
+        })
+    }
+
+    fn requires_scoring(&self) -> bool {
+        true
+    }
+
+    fn merge_fruits(&self, mut children: Vec<TestFruit>) -> Result<TestFruit> {
+        children.sort_by_key(|fruit| {
+            if fruit.docs().is_empty() {
+                0
+            } else {
+                fruit.docs()[0].segment_ord()
+            }
+        });
+        let mut docs = vec![];
+        let mut scores = vec![];
+        for child in children {
+            docs.extend(child.docs());
+            scores.extend(child.scores);
+        }
+        Ok(TestFruit { docs, scores })
+    }
+}
+
+impl SegmentCollector for TestSegmentCollector {
+    type Fruit = TestFruit;
+
+    fn collect(&mut self, doc: DocId, score: Score) {
+        self.fruit.docs.push(DocAddress(self.segment_id, doc));
+        self.fruit.scores.push(score);
+    }
+
+    fn harvest(self) -> <Self as SegmentCollector>::Fruit {
+        self.fruit
+    }
+}
+
+/// Collects in order all of the fast fields for all of the
+/// doc in the `DocSet`
+///
+/// This collector is mainly useful for tests.
+pub struct FastFieldTestCollector {
+    field: Field,
+}
+
+pub struct FastFieldSegmentCollector {
+    vals: Vec<u64>,
+    reader: FastFieldReader<u64>,
+}
+
+impl FastFieldTestCollector {
+    pub fn for_field(field: Field) -> FastFieldTestCollector {
+        FastFieldTestCollector { field }
+    }
+}
+
+impl Collector for FastFieldTestCollector {
+    type Fruit = Vec<u64>;
+    type Child = FastFieldSegmentCollector;
+
+    fn for_segment(
+        &self,
+        _: SegmentLocalId,
+        reader: &SegmentReader,
+    ) -> Result<FastFieldSegmentCollector> {
+        Ok(FastFieldSegmentCollector {
+            vals: Vec::new(),
+            reader: reader.fast_field_reader(self.field)?,
+        })
+    }
+
+    fn requires_scoring(&self) -> bool {
+        false
+    }
+
+    fn merge_fruits(&self, children: Vec<Vec<u64>>) -> Result<Vec<u64>> {
+        Ok(children.into_iter().flat_map(|v| v.into_iter()).collect())
+    }
+}
+
+impl SegmentCollector for FastFieldSegmentCollector {
+    type Fruit = Vec<u64>;
+
+    fn collect(&mut self, doc: DocId, _score: Score) {
+        let val = self.reader.get(doc);
+        self.vals.push(val);
+    }
+
+    fn harvest(self) -> Vec<u64> {
+        self.vals
+    }
+}
+
+/// Collects in order all of the fast field bytes for all of the
+/// docs in the `DocSet`
+///
+/// This collector is mainly useful for tests.
+pub struct BytesFastFieldTestCollector {
+    field: Field,
+}
+
+pub struct BytesFastFieldSegmentCollector {
+    vals: Vec<u8>,
+    reader: BytesFastFieldReader,
+}
+
+impl BytesFastFieldTestCollector {
+    pub fn for_field(field: Field) -> BytesFastFieldTestCollector {
+        BytesFastFieldTestCollector { field }
+    }
+}
+
+impl Collector for BytesFastFieldTestCollector {
+    type Fruit = Vec<u8>;
+    type Child = BytesFastFieldSegmentCollector;
+
+    fn for_segment(
+        &self,
+        _segment_local_id: u32,
+        segment: &SegmentReader,
+    ) -> Result<BytesFastFieldSegmentCollector> {
+        Ok(BytesFastFieldSegmentCollector {
+            vals: Vec::new(),
+            reader: segment.bytes_fast_field_reader(self.field)?,
+        })
+    }
+
+    fn requires_scoring(&self) -> bool {
+        false
+    }
+
+    fn merge_fruits(&self, children: Vec<Vec<u8>>) -> Result<Vec<u8>> {
+        Ok(children.into_iter().flat_map(|c| c.into_iter()).collect())
+    }
+}
+
+impl SegmentCollector for BytesFastFieldSegmentCollector {
+    type Fruit = Vec<u8>;
+
+    fn collect(&mut self, doc: u32, _score: f32) {
+        let data = self.reader.get_val(doc);
+        self.vals.extend(data);
+    }
+
+    fn harvest(self) -> <Self as SegmentCollector>::Fruit {
+        self.vals
+    }
+}

src/collector/top_collector.rs

@@ -1,244 +1,225 @@
-use super::Collector;
+use serde::export::PhantomData;
 use std::cmp::Ordering;
 use std::collections::BinaryHeap;
 use DocAddress;
 use DocId;
 use Result;
-use Score;
 use SegmentLocalId;
 use SegmentReader;
 
-// Rust heap is a max-heap and we need a min heap.
-#[derive(Clone, Copy)]
-struct GlobalScoredDoc {
-    score: Score,
-    doc_address: DocAddress,
+/// Contains a feature (field, score, etc.) of a document along with the document address.
+///
+/// It has a custom implementation of `PartialOrd` that reverses the order. This is because the
+/// default Rust heap is a max heap, whereas a min heap is needed.
+///
+/// WARNING: equality is not what you would expect here.
+/// Two elements are equal if their feature is equal, and regardless of whether `doc`
+/// is equal. This should be perfectly fine for this usage, but let's make sure this
+/// struct is never public.
+struct ComparableDoc<T, D> {
+    feature: T,
+    doc: D,
 }
 
-impl PartialOrd for GlobalScoredDoc {
-    fn partial_cmp(&self, other: &GlobalScoredDoc) -> Option<Ordering> {
+impl<T: PartialOrd, D> PartialOrd for ComparableDoc<T, D> {
+    fn partial_cmp(&self, other: &Self) -> Option<Ordering> {
         Some(self.cmp(other))
     }
 }
 
-impl Ord for GlobalScoredDoc {
+impl<T: PartialOrd, D> Ord for ComparableDoc<T, D> {
     #[inline]
-    fn cmp(&self, other: &GlobalScoredDoc) -> Ordering {
+    fn cmp(&self, other: &Self) -> Ordering {
         other
-            .score
-            .partial_cmp(&self.score)
-            .unwrap_or_else(|| other.doc_address.cmp(&self.doc_address))
+            .feature
+            .partial_cmp(&self.feature)
+            .unwrap_or_else(|| Ordering::Equal)
     }
 }
 
-impl PartialEq for GlobalScoredDoc {
-    fn eq(&self, other: &GlobalScoredDoc) -> bool {
+impl<T: PartialOrd, D> PartialEq for ComparableDoc<T, D> {
+    fn eq(&self, other: &Self) -> bool {
         self.cmp(other) == Ordering::Equal
     }
 }
 
-impl Eq for GlobalScoredDoc {}
+impl<T: PartialOrd, D> Eq for ComparableDoc<T, D> {}
 
-/// The Top Collector keeps track of the K documents
-/// with the best scores.
-///
-/// The implementation is based on a `BinaryHeap`.
-/// The theorical complexity for collecting the top `K` out of `n` documents
-/// is `O(n log K)`.
-///
-/// ```rust
-/// #[macro_use]
-/// extern crate tantivy;
-/// use tantivy::schema::{SchemaBuilder, TEXT};
-/// use tantivy::{Index, Result, DocId, Score};
-/// use tantivy::collector::TopCollector;
-/// use tantivy::query::QueryParser;
-///
-/// # fn main() { example().unwrap(); }
-/// fn example() -> Result<()> {
-///     let mut schema_builder = SchemaBuilder::new();
-///     let title = schema_builder.add_text_field("title", TEXT);
-///     let schema = schema_builder.build();
-///     let index = Index::create_in_ram(schema);
-///     {
-///         let mut index_writer = index.writer_with_num_threads(1, 3_000_000)?;
-///         index_writer.add_document(doc!(
-///             title => "The Name of the Wind",
-///         ));
-///         index_writer.add_document(doc!(
-///             title => "The Diary of Muadib",
-///         ));
-///         index_writer.add_document(doc!(
-///             title => "A Dairy Cow",
-///         ));
-///         index_writer.add_document(doc!(
-///             title => "The Diary of a Young Girl",
-///         ));
-///         index_writer.commit().unwrap();
-///     }
-///
-///     index.load_searchers()?;
-///     let searcher = index.searcher();
-///
-///     {
-///	        let mut top_collector = TopCollector::with_limit(2);
-///         let query_parser = QueryParser::for_index(&index, vec![title]);
-///         let query = query_parser.parse_query("diary")?;
-///         searcher.search(&*query, &mut top_collector).unwrap();
-///
-///         let score_docs: Vec<(Score, DocId)> = top_collector
-///           .score_docs()
-///           .into_iter()
-///           .map(|(score, doc_address)| (score, doc_address.doc()))
-///           .collect();
-///
-///         assert_eq!(score_docs, vec![(0.7261542, 1), (0.6099695, 3)]);
-///     }
-///
-///     Ok(())
-/// }
-/// ```
-pub struct TopCollector {
+pub(crate) struct TopCollector<T> {
     limit: usize,
-    heap: BinaryHeap<GlobalScoredDoc>,
-    segment_id: u32,
+    _marker: PhantomData<T>,
 }
 
-impl TopCollector {
+impl<T> TopCollector<T>
+where
+    T: PartialOrd + Clone,
+{
     /// Creates a top collector, with a number of documents equal to "limit".
     ///
     /// # Panics
     /// The method panics if limit is 0
-    pub fn with_limit(limit: usize) -> TopCollector {
+    pub fn with_limit(limit: usize) -> TopCollector<T> {
         if limit < 1 {
             panic!("Limit must be strictly greater than 0.");
         }
         TopCollector {
             limit,
-            heap: BinaryHeap::with_capacity(limit),
-            segment_id: 0,
+            _marker: PhantomData,
         }
     }
 
-    /// Returns K best documents sorted in decreasing order.
-    ///
-    /// Calling this method triggers the sort.
-    /// The result of the sort is not cached.
-    pub fn docs(&self) -> Vec<DocAddress> {
-        self.score_docs()
-            .into_iter()
-            .map(|score_doc| score_doc.1)
-            .collect()
+    pub fn limit(&self) -> usize {
+        self.limit
     }
 
-    /// Returns K best ScoredDocument sorted in decreasing order.
-    ///
-    /// Calling this method triggers the sort.
-    /// The result of the sort is not cached.
-    pub fn score_docs(&self) -> Vec<(Score, DocAddress)> {
-        let mut scored_docs: Vec<GlobalScoredDoc> = self.heap.iter().cloned().collect();
-        scored_docs.sort();
-        scored_docs
+    pub fn merge_fruits(
+        &self,
+        children: Vec<Vec<(T, DocAddress)>>,
+    ) -> Result<Vec<(T, DocAddress)>> {
+        if self.limit == 0 {
+            return Ok(Vec::new());
+        }
+        let mut top_collector = BinaryHeap::new();
+        for child_fruit in children {
+            for (feature, doc) in child_fruit {
+                if top_collector.len() < self.limit {
+                    top_collector.push(ComparableDoc { feature, doc });
+                } else {
+                    if let Some(mut head) = top_collector.peek_mut() {
+                        if head.feature < feature {
+                            *head = ComparableDoc { feature, doc };
+                        }
+                    }
+                }
+            }
+        }
+        Ok(top_collector
+            .into_sorted_vec()
             .into_iter()
-            .map(|GlobalScoredDoc { score, doc_address }| (score, doc_address))
-            .collect()
+            .map(|cdoc| (cdoc.feature, cdoc.doc))
+            .collect())
+    }
+
+    pub(crate) fn for_segment(
+        &self,
+        segment_id: SegmentLocalId,
+        _: &SegmentReader,
+    ) -> Result<TopSegmentCollector<T>> {
+        Ok(TopSegmentCollector::new(segment_id, self.limit))
+    }
+}
+
+/// The Top Collector keeps track of the K documents
+/// sorted by type `T`.
+///
+/// The implementation is based on a `BinaryHeap`.
+/// The theorical complexity for collecting the top `K` out of `n` documents
+/// is `O(n log K)`.
+pub(crate) struct TopSegmentCollector<T> {
+    limit: usize,
+    heap: BinaryHeap<ComparableDoc<T, DocId>>,
+    segment_id: u32,
+}
+
+impl<T: PartialOrd> TopSegmentCollector<T> {
+    fn new(segment_id: SegmentLocalId, limit: usize) -> TopSegmentCollector<T> {
+        TopSegmentCollector {
+            limit,
+            heap: BinaryHeap::with_capacity(limit),
+            segment_id,
+        }
+    }
+}
+
+impl<T: PartialOrd + Clone> TopSegmentCollector<T> {
+    pub fn harvest(self) -> Vec<(T, DocAddress)> {
+        let segment_id = self.segment_id;
+        self.heap
+            .into_sorted_vec()
+            .into_iter()
+            .map(|comparable_doc| {
+                (
+                    comparable_doc.feature,
+                    DocAddress(segment_id, comparable_doc.doc),
+                )
+            }).collect()
     }
 
     /// Return true iff at least K documents have gone through
     /// the collector.
-    #[inline]
-    pub fn at_capacity(&self) -> bool {
+    #[inline(always)]
+    pub(crate) fn at_capacity(&self) -> bool {
         self.heap.len() >= self.limit
     }
-}
 
-impl Collector for TopCollector {
-    fn set_segment(&mut self, segment_id: SegmentLocalId, _: &SegmentReader) -> Result<()> {
-        self.segment_id = segment_id;
-        Ok(())
-    }
-
-    fn collect(&mut self, doc: DocId, score: Score) {
+    /// Collects a document scored by the given feature
+    ///
+    /// It collects documents until it has reached the max capacity. Once it reaches capacity, it
+    /// will compare the lowest scoring item with the given one and keep whichever is greater.
+    #[inline(always)]
+    pub fn collect(&mut self, doc: DocId, feature: T) {
         if self.at_capacity() {
             // It's ok to unwrap as long as a limit of 0 is forbidden.
-            let limit_doc: GlobalScoredDoc = *self.heap
-                .peek()
-                .expect("Top collector with size 0 is forbidden");
-            if limit_doc.score < score {
-                let mut mut_head = self.heap
-                    .peek_mut()
-                    .expect("Top collector with size 0 is forbidden");
-                mut_head.score = score;
-                mut_head.doc_address = DocAddress(self.segment_id, doc);
+            if let Some(limit_feature) = self.heap.peek().map(|head| head.feature.clone()) {
+                if limit_feature < feature {
+                    if let Some(mut head) = self.heap.peek_mut() {
+                        head.feature = feature;
+                        head.doc = doc;
+                    }
+                }
             }
         } else {
-            let wrapped_doc = GlobalScoredDoc {
-                score,
-                doc_address: DocAddress(self.segment_id, doc),
-            };
-            self.heap.push(wrapped_doc);
+            // we have not reached capacity yet, so we can just push the
+            // element.
+            self.heap.push(ComparableDoc { feature, doc });
         }
     }
-
-    fn requires_scoring(&self) -> bool {
-        true
-    }
 }
 
 #[cfg(test)]
 mod tests {
-
-    use super::*;
-    use collector::Collector;
-    use DocId;
+    use super::{TopCollector, TopSegmentCollector};
+    use DocAddress;
     use Score;
 
     #[test]
     fn test_top_collector_not_at_capacity() {
-        let mut top_collector = TopCollector::with_limit(4);
+        let mut top_collector = TopSegmentCollector::new(0, 4);
         top_collector.collect(1, 0.8);
         top_collector.collect(3, 0.2);
         top_collector.collect(5, 0.3);
-        assert!(!top_collector.at_capacity());
-        let score_docs: Vec<(Score, DocId)> = top_collector
-            .score_docs()
-            .into_iter()
-            .map(|(score, doc_address)| (score, doc_address.doc()))
-            .collect();
-        assert_eq!(score_docs, vec![(0.8, 1), (0.3, 5), (0.2, 3)]);
+        assert_eq!(
+            top_collector.harvest(),
+            vec![
+                (0.8, DocAddress(0, 1)),
+                (0.3, DocAddress(0, 5)),
+                (0.2, DocAddress(0, 3))
+            ]
+        );
     }
 
     #[test]
     fn test_top_collector_at_capacity() {
-        let mut top_collector = TopCollector::with_limit(4);
+        let mut top_collector = TopSegmentCollector::new(0, 4);
         top_collector.collect(1, 0.8);
         top_collector.collect(3, 0.2);
         top_collector.collect(5, 0.3);
         top_collector.collect(7, 0.9);
         top_collector.collect(9, -0.2);
-        assert!(top_collector.at_capacity());
-        {
-            let score_docs: Vec<(Score, DocId)> = top_collector
-                .score_docs()
-                .into_iter()
-                .map(|(score, doc_address)| (score, doc_address.doc()))
-                .collect();
-            assert_eq!(score_docs, vec![(0.9, 7), (0.8, 1), (0.3, 5), (0.2, 3)]);
-        }
-        {
-            let docs: Vec<DocId> = top_collector
-                .docs()
-                .into_iter()
-                .map(|doc_address| doc_address.doc())
-                .collect();
-            assert_eq!(docs, vec![7, 1, 5, 3]);
-        }
+        assert_eq!(
+            top_collector.harvest(),
+            vec![
+                (0.9, DocAddress(0, 7)),
+                (0.8, DocAddress(0, 1)),
+                (0.3, DocAddress(0, 5)),
+                (0.2, DocAddress(0, 3))
+            ]
+        );
     }
 
     #[test]
     #[should_panic]
     fn test_top_0() {
-        TopCollector::with_limit(0);
+        let _collector: TopCollector<Score> = TopCollector::with_limit(0);
     }
-
 }

src/collector/top_field_collector.rs

@@ -0,0 +1,250 @@
+use super::Collector;
+use collector::top_collector::TopCollector;
+use collector::top_collector::TopSegmentCollector;
+use collector::SegmentCollector;
+use fastfield::FastFieldReader;
+use fastfield::FastValue;
+use schema::Field;
+use DocAddress;
+use Result;
+use SegmentLocalId;
+use SegmentReader;
+
+/// The Top Field Collector keeps track of the K documents
+/// sorted by a fast field in the index
+///
+/// The implementation is based on a `BinaryHeap`.
+/// The theorical complexity for collecting the top `K` out of `n` documents
+/// is `O(n log K)`.
+///
+/// ```rust
+/// #[macro_use]
+/// extern crate tantivy;
+/// # use tantivy::schema::{Schema, Field, FAST, TEXT};
+/// # use tantivy::{Index, Result, DocAddress};
+/// # use tantivy::query::{Query, QueryParser};
+/// use tantivy::collector::TopDocs;
+///
+/// # fn main() {
+/// #   let mut schema_builder = Schema::builder();
+/// #   let title = schema_builder.add_text_field("title", TEXT);
+/// #   let rating = schema_builder.add_u64_field("rating", FAST);
+/// #   let schema = schema_builder.build();
+/// #   let index = Index::create_in_ram(schema);
+/// #   let mut index_writer = index.writer_with_num_threads(1, 3_000_000).unwrap();
+/// #   index_writer.add_document(doc!(
+/// #       title => "The Name of the Wind",
+/// #       rating => 92u64,
+/// #   ));
+/// #   index_writer.add_document(doc!(title => "The Diary of Muadib", rating => 97u64));
+/// #   index_writer.add_document(doc!(title => "A Dairy Cow", rating => 63u64));
+/// #   index_writer.add_document(doc!(title => "The Diary of a Young Girl", rating => 80u64));
+/// #   index_writer.commit().unwrap();
+/// #   index.load_searchers().unwrap();
+///	#   let query = QueryParser::for_index(&index, vec![title]).parse_query("diary").unwrap();
+/// #   let top_docs = docs_sorted_by_rating(&index, &query, rating).unwrap();
+/// #   assert_eq!(top_docs,
+/// #            vec![(97u64, DocAddress(0u32, 1)),
+/// #                 (80u64, DocAddress(0u32, 3))]);
+/// # }
+/// #
+/// /// Searches the document matching the given query, and
+/// /// collects the top 10 documents, order by the `field`
+/// /// given in argument.
+/// ///
+/// /// `field` is required to be a FAST field.
+/// fn docs_sorted_by_rating(index: &Index, query: &Query, sort_by_field: Field)
+///     -> Result<Vec<(u64, DocAddress)>> {
+///
+///     // This is where we build our collector!
+///     let top_docs_by_rating = TopDocs::with_limit(2).order_by_field(sort_by_field);
+///
+///     // ... and here is our documents. Not this is a simple vec.
+///     // The `u64` in the pair is the value of our fast field for each documents.
+///     index.searcher()
+///          .search(query, &top_docs_by_rating)
+/// }
+/// ```
+pub struct TopDocsByField<T> {
+    collector: TopCollector<T>,
+    field: Field,
+}
+
+impl<T: FastValue + PartialOrd + Clone> TopDocsByField<T> {
+    /// Creates a top field collector, with a number of documents equal to "limit".
+    ///
+    /// The given field name must be a fast field, otherwise the collector have an error while
+    /// collecting results.
+    ///
+    /// # Panics
+    /// The method panics if limit is 0
+    pub(crate) fn new(field: Field, limit: usize) -> TopDocsByField<T> {
+        TopDocsByField {
+            collector: TopCollector::with_limit(limit),
+            field,
+        }
+    }
+}
+
+impl<T: FastValue + PartialOrd + Send + Sync + 'static> Collector for TopDocsByField<T> {
+    type Fruit = Vec<(T, DocAddress)>;
+
+    type Child = TopFieldSegmentCollector<T>;
+
+    fn for_segment(
+        &self,
+        segment_local_id: SegmentLocalId,
+        reader: &SegmentReader,
+    ) -> Result<TopFieldSegmentCollector<T>> {
+        let collector = self.collector.for_segment(segment_local_id, reader)?;
+        let reader = reader.fast_field_reader(self.field)?;
+        Ok(TopFieldSegmentCollector { collector, reader })
+    }
+
+    fn requires_scoring(&self) -> bool {
+        false
+    }
+
+    fn merge_fruits(
+        &self,
+        segment_fruits: Vec<Vec<(T, DocAddress)>>,
+    ) -> Result<Vec<(T, DocAddress)>> {
+        self.collector.merge_fruits(segment_fruits)
+    }
+}
+
+pub struct TopFieldSegmentCollector<T: FastValue + PartialOrd> {
+    collector: TopSegmentCollector<T>,
+    reader: FastFieldReader<T>,
+}
+
+impl<T: FastValue + PartialOrd + Send + Sync + 'static> SegmentCollector
+    for TopFieldSegmentCollector<T>
+{
+    type Fruit = Vec<(T, DocAddress)>;
+
+    fn collect(&mut self, doc: u32, _score: f32) {
+        let field_value = self.reader.get(doc);
+        self.collector.collect(doc, field_value);
+    }
+
+    fn harvest(self) -> Vec<(T, DocAddress)> {
+        self.collector.harvest()
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    use super::TopDocsByField;
+    use collector::Collector;
+    use collector::TopDocs;
+    use query::Query;
+    use query::QueryParser;
+    use schema::Field;
+    use schema::IntOptions;
+    use schema::{Schema, FAST, TEXT};
+    use DocAddress;
+    use Index;
+    use IndexWriter;
+    use TantivyError;
+
+    const TITLE: &str = "title";
+    const SIZE: &str = "size";
+
+    #[test]
+    fn test_top_collector_not_at_capacity() {
+        let mut schema_builder = Schema::builder();
+        let title = schema_builder.add_text_field(TITLE, TEXT);
+        let size = schema_builder.add_u64_field(SIZE, FAST);
+        let schema = schema_builder.build();
+        let (index, query) = index("beer", title, schema, |index_writer| {
+            index_writer.add_document(doc!(
+                title => "bottle of beer",
+                size => 12u64,
+            ));
+            index_writer.add_document(doc!(
+                title => "growler of beer",
+                size => 64u64,
+            ));
+            index_writer.add_document(doc!(
+                title => "pint of beer",
+                size => 16u64,
+            ));
+        });
+        let searcher = index.searcher();
+
+        let top_collector = TopDocs::with_limit(4).order_by_field(size);
+        let top_docs: Vec<(u64, DocAddress)> = searcher.search(&query, &top_collector).unwrap();
+        assert_eq!(
+            top_docs,
+            vec![
+                (64, DocAddress(0, 1)),
+                (16, DocAddress(0, 2)),
+                (12, DocAddress(0, 0))
+            ]
+        );
+    }
+
+    #[test]
+    #[should_panic]
+    fn test_field_does_not_exist() {
+        let mut schema_builder = Schema::builder();
+        let title = schema_builder.add_text_field(TITLE, TEXT);
+        let size = schema_builder.add_u64_field(SIZE, FAST);
+        let schema = schema_builder.build();
+        let (index, _) = index("beer", title, schema, |index_writer| {
+            index_writer.add_document(doc!(
+                title => "bottle of beer",
+                size => 12u64,
+            ));
+        });
+        let searcher = index.searcher();
+        let top_collector: TopDocsByField<u64> = TopDocs::with_limit(4).order_by_field(Field(2));
+        let segment_reader = searcher.segment_reader(0u32);
+        top_collector
+            .for_segment(0, segment_reader)
+            .expect("should panic");
+    }
+
+    #[test]
+    fn test_field_not_fast_field() {
+        let mut schema_builder = Schema::builder();
+        let title = schema_builder.add_text_field(TITLE, TEXT);
+        let size = schema_builder.add_u64_field(SIZE, IntOptions::default());
+        let schema = schema_builder.build();
+        let (index, _) = index("beer", title, schema, |index_writer| {
+            index_writer.add_document(doc!(
+                title => "bottle of beer",
+                size => 12u64,
+            ));
+        });
+        let searcher = index.searcher();
+        let segment = searcher.segment_reader(0);
+        let top_collector: TopDocsByField<u64> = TopDocs::with_limit(4).order_by_field(size);
+        assert_matches!(
+            top_collector
+                .for_segment(0, segment)
+                .map(|_| ())
+                .unwrap_err(),
+            TantivyError::FastFieldError(_)
+        );
+    }
+
+    fn index(
+        query: &str,
+        query_field: Field,
+        schema: Schema,
+        mut doc_adder: impl FnMut(&mut IndexWriter) -> (),
+    ) -> (Index, Box<Query>) {
+        let index = Index::create_in_ram(schema);
+
+        let mut index_writer = index.writer_with_num_threads(1, 3_000_000).unwrap();
+        doc_adder(&mut index_writer);
+        index_writer.commit().unwrap();
+        index.load_searchers().unwrap();
+
+        let query_parser = QueryParser::for_index(&index, vec![query_field]);
+        let query = query_parser.parse_query(query).unwrap();
+        (index, query)
+    }
+}

src/collector/top_score_collector.rs

@@ -0,0 +1,200 @@
+use super::Collector;
+use collector::top_collector::TopCollector;
+use collector::top_collector::TopSegmentCollector;
+use collector::SegmentCollector;
+use collector::TopDocsByField;
+use fastfield::FastValue;
+use schema::Field;
+use DocAddress;
+use DocId;
+use Result;
+use Score;
+use SegmentLocalId;
+use SegmentReader;
+
+/// The Top Score Collector keeps track of the K documents
+/// sorted by their score.
+///
+/// The implementation is based on a `BinaryHeap`.
+/// The theorical complexity for collecting the top `K` out of `n` documents
+/// is `O(n log K)`.
+///
+/// ```rust
+/// #[macro_use]
+/// extern crate tantivy;
+/// use tantivy::DocAddress;
+/// use tantivy::schema::{Schema, TEXT};
+/// use tantivy::{Index, Result};
+/// use tantivy::collector::TopDocs;
+/// use tantivy::query::QueryParser;
+///
+/// # fn main() { example().unwrap(); }
+/// fn example() -> Result<()> {
+///     let mut schema_builder = Schema::builder();
+///     let title = schema_builder.add_text_field("title", TEXT);
+///     let schema = schema_builder.build();
+///     let index = Index::create_in_ram(schema);
+///     {
+///         let mut index_writer = index.writer_with_num_threads(1, 3_000_000)?;
+///         index_writer.add_document(doc!(
+///             title => "The Name of the Wind",
+///         ));
+///         index_writer.add_document(doc!(
+///             title => "The Diary of Muadib",
+///         ));
+///         index_writer.add_document(doc!(
+///             title => "A Dairy Cow",
+///         ));
+///         index_writer.add_document(doc!(
+///             title => "The Diary of a Young Girl",
+///         ));
+///         index_writer.commit().unwrap();
+///     }
+///
+///     index.load_searchers()?;
+///     let searcher = index.searcher();
+///
+///     let query_parser = QueryParser::for_index(&index, vec![title]);
+///     let query = query_parser.parse_query("diary")?;
+///     let top_docs = searcher.search(&query, &TopDocs::with_limit(2))?;
+///
+///     assert_eq!(&top_docs[0], &(0.7261542, DocAddress(0, 1)));
+///     assert_eq!(&top_docs[1], &(0.6099695, DocAddress(0, 3)));
+///
+///     Ok(())
+/// }
+/// ```
+pub struct TopDocs(TopCollector<Score>);
+
+impl TopDocs {
+    /// Creates a top score collector, with a number of documents equal to "limit".
+    ///
+    /// # Panics
+    /// The method panics if limit is 0
+    pub fn with_limit(limit: usize) -> TopDocs {
+        TopDocs(TopCollector::with_limit(limit))
+    }
+
+    /// Set top-K to rank documents by a given fast field.
+    ///
+    /// (By default, `TopDocs` collects the top-K documents sorted by
+    /// the similarity score.)
+    pub fn order_by_field<T: PartialOrd + FastValue + Clone>(
+        self,
+        field: Field,
+    ) -> TopDocsByField<T> {
+        TopDocsByField::new(field, self.0.limit())
+    }
+}
+
+impl Collector for TopDocs {
+    type Fruit = Vec<(Score, DocAddress)>;
+
+    type Child = TopScoreSegmentCollector;
+
+    fn for_segment(
+        &self,
+        segment_local_id: SegmentLocalId,
+        reader: &SegmentReader,
+    ) -> Result<Self::Child> {
+        let collector = self.0.for_segment(segment_local_id, reader)?;
+        Ok(TopScoreSegmentCollector(collector))
+    }
+
+    fn requires_scoring(&self) -> bool {
+        true
+    }
+
+    fn merge_fruits(&self, child_fruits: Vec<Vec<(Score, DocAddress)>>) -> Result<Self::Fruit> {
+        self.0.merge_fruits(child_fruits)
+    }
+}
+
+/// Segment Collector associated to `TopDocs`.
+pub struct TopScoreSegmentCollector(TopSegmentCollector<Score>);
+
+impl SegmentCollector for TopScoreSegmentCollector {
+    type Fruit = Vec<(Score, DocAddress)>;
+
+    fn collect(&mut self, doc: DocId, score: Score) {
+        self.0.collect(doc, score)
+    }
+
+    fn harvest(self) -> Vec<(Score, DocAddress)> {
+        self.0.harvest()
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    use super::TopDocs;
+    use query::QueryParser;
+    use schema::Schema;
+    use schema::TEXT;
+    use DocAddress;
+    use Index;
+    use Score;
+
+    fn make_index() -> Index {
+        let mut schema_builder = Schema::builder();
+        let text_field = schema_builder.add_text_field("text", TEXT);
+        let schema = schema_builder.build();
+        let index = Index::create_in_ram(schema);
+        {
+            // writing the segment
+            let mut index_writer = index.writer_with_num_threads(1, 40_000_000).unwrap();
+            index_writer.add_document(doc!(text_field=>"Hello happy tax payer."));
+            index_writer.add_document(doc!(text_field=>"Droopy says hello happy tax payer"));
+            index_writer.add_document(doc!(text_field=>"I like Droopy"));
+            assert!(index_writer.commit().is_ok());
+        }
+        index.load_searchers().unwrap();
+        index
+    }
+
+    #[test]
+    fn test_top_collector_not_at_capacity() {
+        let index = make_index();
+        let field = index.schema().get_field("text").unwrap();
+        let query_parser = QueryParser::for_index(&index, vec![field]);
+        let text_query = query_parser.parse_query("droopy tax").unwrap();
+        let score_docs: Vec<(Score, DocAddress)> = index
+            .searcher()
+            .search(&text_query, &TopDocs::with_limit(4))
+            .unwrap();
+        assert_eq!(
+            score_docs,
+            vec![
+                (0.81221175, DocAddress(0u32, 1)),
+                (0.5376842, DocAddress(0u32, 2)),
+                (0.48527452, DocAddress(0, 0))
+            ]
+        );
+    }
+
+    #[test]
+    fn test_top_collector_at_capacity() {
+        let index = make_index();
+        let field = index.schema().get_field("text").unwrap();
+        let query_parser = QueryParser::for_index(&index, vec![field]);
+        let text_query = query_parser.parse_query("droopy tax").unwrap();
+        let score_docs: Vec<(Score, DocAddress)> = index
+            .searcher()
+            .search(&text_query, &TopDocs::with_limit(2))
+            .unwrap();
+        assert_eq!(
+            score_docs,
+            vec![
+                (0.81221175, DocAddress(0u32, 1)),
+                (0.5376842, DocAddress(0u32, 2)),
+            ]
+        );
+    }
+
+    #[test]
+    #[should_panic]
+    fn test_top_0() {
+        TopDocs::with_limit(0);
+    }
+
+}

src/common/bitpacker.rs

@@ -102,6 +102,7 @@ where
             addr + 8 <= data.len(),
             "The fast field field should have been padded with 7 bytes."
         );
+        #[cfg_attr(feature = "cargo-clippy", allow(clippy::cast_ptr_alignment))]
         let val_unshifted_unmasked: u64 =
             u64::from_le(unsafe { ptr::read_unaligned(data[addr..].as_ptr() as *const u64) });
         let val_shifted = (val_unshifted_unmasked >> bit_shift) as u64;
@@ -125,6 +126,7 @@ where
             for output_val in output.iter_mut() {
                 let addr = addr_in_bits >> 3;
                 let bit_shift = addr_in_bits & 7;
+                #[cfg_attr(feature = "cargo-clippy", allow(clippy::cast_ptr_alignment))]
                 let val_unshifted_unmasked: u64 =
                     unsafe { ptr::read_unaligned(data[addr..].as_ptr() as *const u64) };
                 let val_shifted = (val_unshifted_unmasked >> bit_shift) as u64;

src/common/bitset.rs

@@ -34,17 +34,17 @@ impl TinySet {
     }
 
     /// Returns the complement of the set in `[0, 64[`.
-    fn complement(&self) -> TinySet {
+    fn complement(self) -> TinySet {
         TinySet(!self.0)
     }
 
     /// Returns true iff the `TinySet` contains the element `el`.
-    pub fn contains(&self, el: u32) -> bool {
+    pub fn contains(self, el: u32) -> bool {
         !self.intersect(TinySet::singleton(el)).is_empty()
     }
 
     /// Returns the intersection of `self` and `other`
-    pub fn intersect(&self, other: TinySet) -> TinySet {
+    pub fn intersect(self, other: TinySet) -> TinySet {
         TinySet(self.0 & other.0)
     }
 
@@ -77,7 +77,7 @@ impl TinySet {
 
     /// Returns true iff the `TinySet` is empty.
     #[inline(always)]
-    pub fn is_empty(&self) -> bool {
+    pub fn is_empty(self) -> bool {
         self.0 == 0u64
     }
 
@@ -114,7 +114,7 @@ impl TinySet {
         self.0 = 0u64;
     }
 
-    pub fn len(&self) -> u32 {
+    pub fn len(self) -> u32 {
         self.0.count_ones()
     }
 }
@@ -266,14 +266,14 @@ mod tests {
 
     #[test]
     fn test_bitset_large() {
-        let arr = generate_nonunique_unsorted(1_000_000, 50_000);
+        let arr = generate_nonunique_unsorted(100_000, 5_000);
         let mut btreeset: BTreeSet<u32> = BTreeSet::new();
-        let mut bitset = BitSet::with_max_value(1_000_000);
+        let mut bitset = BitSet::with_max_value(100_000);
         for el in arr {
             btreeset.insert(el);
             bitset.insert(el);
         }
-        for i in 0..1_000_000 {
+        for i in 0..100_000 {
             assert_eq!(btreeset.contains(&i), bitset.contains(i));
         }
         assert_eq!(btreeset.len(), bitset.len());

src/common/composite_file.rs

@@ -4,6 +4,8 @@ use common::VInt;
 use directory::ReadOnlySource;
 use directory::WritePtr;
 use schema::Field;
+use space_usage::FieldUsage;
+use space_usage::PerFieldSpaceUsage;
 use std::collections::HashMap;
 use std::io::Write;
 use std::io::{self, Read};
@@ -72,7 +74,8 @@ impl<W: Write> CompositeWrite<W> {
         let footer_offset = self.write.written_bytes();
         VInt(self.offsets.len() as u64).serialize(&mut self.write)?;
 
-        let mut offset_fields: Vec<_> = self.offsets
+        let mut offset_fields: Vec<_> = self
+            .offsets
             .iter()
             .map(|(file_addr, offset)| (*offset, *file_addr))
             .collect();
@@ -165,6 +168,17 @@ impl CompositeFile {
             .get(&FileAddr { field, idx })
             .map(|&(from, to)| self.data.slice(from, to))
     }
+
+    pub fn space_usage(&self) -> PerFieldSpaceUsage {
+        let mut fields = HashMap::new();
+        for (&field_addr, &(start, end)) in self.offsets_index.iter() {
+            fields
+                .entry(field_addr.field)
+                .or_insert_with(|| FieldUsage::empty(field_addr.field))
+                .add_field_idx(field_addr.idx, end - start);
+        }
+        PerFieldSpaceUsage::new(fields)
+    }
 }
 
 #[cfg(test)]

src/common/vint.rs

@@ -10,8 +10,6 @@ pub struct VInt(pub u64);
 const STOP_BIT: u8 = 128;
 
 impl VInt {
-
-
     pub fn val(&self) -> u64 {
         self.0
     }
@@ -20,14 +18,13 @@ impl VInt {
         VInt::deserialize(reader).map(|vint| vint.0)
     }
 
-    pub fn serialize_into_vec(&self, output: &mut Vec<u8>){
+    pub fn serialize_into_vec(&self, output: &mut Vec<u8>) {
         let mut buffer = [0u8; 10];
         let num_bytes = self.serialize_into(&mut buffer);
         output.extend(&buffer[0..num_bytes]);
     }
 
     fn serialize_into(&self, buffer: &mut [u8; 10]) -> usize {
-
         let mut remaining = self.0;
         for (i, b) in buffer.iter_mut().enumerate() {
             let next_byte: u8 = (remaining % 128u64) as u8;
@@ -74,7 +71,6 @@ impl BinarySerializable for VInt {
     }
 }
 
-
 #[cfg(test)]
 mod tests {
 
@@ -89,10 +85,10 @@ mod tests {
         }
         assert!(num_bytes > 0);
         if num_bytes < 10 {
-            assert!(1u64 << (7*num_bytes) > val);
+            assert!(1u64 << (7 * num_bytes) > val);
         }
         if num_bytes > 1 {
-            assert!(1u64 << (7*(num_bytes-1)) <= val);
+            assert!(1u64 << (7 * (num_bytes - 1)) <= val);
         }
         let serdeser_val = VInt::deserialize(&mut &v[..]).unwrap();
         assert_eq!(val, serdeser_val.0);
@@ -105,11 +101,11 @@ mod tests {
         aux_test_vint(5);
         aux_test_vint(u64::max_value());
         for i in 1..9 {
-            let power_of_128 = 1u64 << (7*i);
+            let power_of_128 = 1u64 << (7 * i);
             aux_test_vint(power_of_128 - 1u64);
-            aux_test_vint(power_of_128 );
+            aux_test_vint(power_of_128);
             aux_test_vint(power_of_128 + 1u64);
         }
         aux_test_vint(10);
     }
-}
+}

src/core/executor.rs

@@ -0,0 +1,134 @@
+use crossbeam::channel;
+use scoped_pool::{Pool, ThreadConfig};
+use Result;
+
+/// Search executor whether search request are single thread or multithread.
+///
+/// We don't expose Rayon thread pool directly here for several reasons.
+///
+/// First dependency hell. It is not a good idea to expose the
+/// API of a dependency, knowing it might conflict with a different version
+/// used by the client. Second, we may stop using rayon in the future.
+pub enum Executor {
+    SingleThread,
+    ThreadPool(Pool),
+}
+
+impl Executor {
+    /// Creates an Executor that performs all task in the caller thread.
+    pub fn single_thread() -> Executor {
+        Executor::SingleThread
+    }
+
+    // Creates an Executor that dispatches the tasks in a thread pool.
+    pub fn multi_thread(num_threads: usize, prefix: &'static str) -> Executor {
+        let thread_config = ThreadConfig::new().prefix(prefix);
+        let pool = Pool::with_thread_config(num_threads, thread_config);
+        Executor::ThreadPool(pool)
+    }
+
+    // Perform a map in the thread pool.
+    //
+    // Regardless of the executor (`SingleThread` or `ThreadPool`), panics in the task
+    // will propagate to the caller.
+    pub fn map<
+        A: Send,
+        R: Send,
+        AIterator: Iterator<Item = A>,
+        F: Sized + Sync + Fn(A) -> Result<R>,
+    >(
+        &self,
+        f: F,
+        args: AIterator,
+    ) -> Result<Vec<R>> {
+        match self {
+            Executor::SingleThread => args.map(f).collect::<Result<_>>(),
+            Executor::ThreadPool(pool) => {
+                let args_with_indices: Vec<(usize, A)> = args.enumerate().collect();
+                let num_fruits = args_with_indices.len();
+                let fruit_receiver = {
+                    let (fruit_sender, fruit_receiver) = channel::unbounded();
+                    pool.scoped(|scope| {
+                        for arg_with_idx in args_with_indices {
+                            scope.execute(|| {
+                                let (idx, arg) = arg_with_idx;
+                                let fruit = f(arg);
+                                if let Err(err) = fruit_sender.send((idx, fruit)) {
+                                    error!("Failed to send search task. It probably means all search threads have panicked. {:?}", err);
+                                }
+                            });
+                        }
+                    });
+                    fruit_receiver
+                    // This ends the scope of fruit_sender.
+                    // This is important as it makes it possible for the fruit_receiver iteration to
+                    // terminate.
+                };
+                let mut results = Vec::with_capacity(num_fruits);
+                unsafe { results.set_len(num_fruits) };
+                let mut num_items = 0;
+                for (pos, fruit_res) in fruit_receiver {
+                    results[pos] = fruit_res?;
+                    num_items += 1;
+                }
+                // this checks ensures that we filled of this
+                // uninitialized memory.
+                assert_eq!(num_items, results.len());
+                Ok(results)
+            }
+        }
+    }
+}
+
+#[cfg(test)]
+mod tests {
+
+    use super::Executor;
+
+    #[test]
+    #[should_panic(expected = "panic should propagate")]
+    fn test_panic_propagates_single_thread() {
+        let _result: Vec<usize> = Executor::single_thread()
+            .map(
+                |_| {
+                    panic!("panic should propagate");
+                },
+                vec![0].into_iter(),
+            ).unwrap();
+    }
+
+    #[test]
+    #[should_panic] //< unfortunately the panic message is not propagated
+    fn test_panic_propagates_multi_thread() {
+        let _result: Vec<usize> = Executor::multi_thread(1, "search-test")
+            .map(
+                |_| {
+                    panic!("panic should propagate");
+                },
+                vec![0].into_iter(),
+            ).unwrap();
+    }
+
+    #[test]
+    fn test_map_singlethread() {
+        let result: Vec<usize> = Executor::single_thread()
+            .map(|i| Ok(i * 2), 0..1_000)
+            .unwrap();
+        assert_eq!(result.len(), 1_000);
+        for i in 0..1_000 {
+            assert_eq!(result[i], i * 2);
+        }
+    }
+
+}
+
+#[test]
+fn test_map_multithread() {
+    let result: Vec<usize> = Executor::multi_thread(3, "search-test")
+        .map(|i| Ok(i * 2), 0..10)
+        .unwrap();
+    assert_eq!(result.len(), 10);
+    for i in 0..10 {
+        assert_eq!(result[i], i * 2);
+    }
+}

src/core/index.rs

@@ -1,34 +1,37 @@
-use core::SegmentId;
-use error::TantivyError;
-use schema::Schema;
-use serde_json;
-use std::borrow::BorrowMut;
-use std::fmt;
-use std::sync::atomic::{AtomicUsize, Ordering};
-use std::sync::Arc;
-use Result;
-
 use super::pool::LeasedItem;
 use super::pool::Pool;
 use super::segment::create_segment;
 use super::segment::Segment;
 use core::searcher::Searcher;
+use core::Executor;
 use core::IndexMeta;
+use core::SegmentId;
 use core::SegmentMeta;
 use core::SegmentReader;
 use core::META_FILEPATH;
+use directory::ManagedDirectory;
 #[cfg(feature = "mmap")]
 use directory::MmapDirectory;
 use directory::{Directory, RAMDirectory};
-use directory::{DirectoryClone, ManagedDirectory};
+use error::TantivyError;
 use indexer::index_writer::open_index_writer;
 use indexer::index_writer::HEAP_SIZE_MIN;
 use indexer::segment_updater::save_new_metas;
-use indexer::DirectoryLock;
+use indexer::LockType;
 use num_cpus;
+use schema::Field;
+use schema::FieldType;
+use schema::Schema;
+use serde_json;
+use std::borrow::BorrowMut;
+use std::fmt;
 use std::path::Path;
+use std::sync::atomic::{AtomicUsize, Ordering};
+use std::sync::Arc;
+use tokenizer::BoxedTokenizer;
 use tokenizer::TokenizerManager;
 use IndexWriter;
+use Result;
 
 fn load_metas(directory: &Directory) -> Result<IndexMeta> {
     let meta_data = directory.atomic_read(&META_FILEPATH)?;
@@ -43,10 +46,39 @@ pub struct Index {
     schema: Schema,
     num_searchers: Arc<AtomicUsize>,
     searcher_pool: Arc<Pool<Searcher>>,
+    executor: Arc<Executor>,
     tokenizers: TokenizerManager,
 }
 
 impl Index {
+    /// Examines the director to see if it contains an index
+    pub fn exists<Dir: Directory>(dir: &Dir) -> bool {
+        dir.exists(&META_FILEPATH)
+    }
+
+    /// Accessor to the search executor.
+    ///
+    /// This pool is used by default when calling `searcher.search(...)`
+    /// to perform search on the individual segments.
+    ///
+    /// By default the executor is single thread, and simply runs in the calling thread.
+    pub fn search_executor(&self) -> &Executor {
+        self.executor.as_ref()
+    }
+
+    /// Replace the default single thread search executor pool
+    /// by a thread pool with a given number of threads.
+    pub fn set_multithread_executor(&mut self, num_threads: usize) {
+        self.executor = Arc::new(Executor::multi_thread(num_threads, "thrd-tantivy-search-"));
+    }
+
+    /// Replace the default single thread search executor pool
+    /// by a thread pool with a given number of threads.
+    pub fn set_default_multithread_executor(&mut self) {
+        let default_num_threads = num_cpus::get();
+        self.set_multithread_executor(default_num_threads);
+    }
+
     /// Creates a new index using the `RAMDirectory`.
     ///
     /// The index will be allocated in anonymous memory.
@@ -63,9 +95,30 @@ impl Index {
     #[cfg(feature = "mmap")]
     pub fn create_in_dir<P: AsRef<Path>>(directory_path: P, schema: Schema) -> Result<Index> {
         let mmap_directory = MmapDirectory::open(directory_path)?;
+        if Index::exists(&mmap_directory) {
+            return Err(TantivyError::IndexAlreadyExists);
+        }
+
         Index::create(mmap_directory, schema)
     }
 
+    /// Opens or creates a new index in the provided directory
+    #[cfg(feature = "mmap")]
+    pub fn open_or_create<Dir: Directory>(dir: Dir, schema: Schema) -> Result<Index> {
+        if Index::exists(&dir) {
+            let index = Index::open(dir)?;
+            if index.schema() == schema {
+                Ok(index)
+            } else {
+                Err(TantivyError::SchemaError(
+                    "An index exists but the schema does not match.".to_string(),
+                ))
+            }
+        } else {
+            Index::create(dir, schema)
+        }
+    }
+
     /// Creates a new index in a temp directory.
     ///
     /// The index will use the `MMapDirectory` in a newly created directory.
@@ -87,6 +140,8 @@ impl Index {
     }
 
     /// Create a new index from a directory.
+    ///
+    /// This will overwrite existing meta.json
     fn from_directory(mut directory: ManagedDirectory, schema: Schema) -> Result<Index> {
         save_new_metas(schema.clone(), 0, directory.borrow_mut())?;
         let metas = IndexMeta::with_schema(schema);
@@ -103,6 +158,7 @@ impl Index {
             num_searchers: Arc::new(AtomicUsize::new(n_cpus)),
             searcher_pool: Arc::new(Pool::new()),
             tokenizers: TokenizerManager::default(),
+            executor: Arc::new(Executor::single_thread()),
         };
         index.load_searchers()?;
         Ok(index)
@@ -113,6 +169,27 @@ impl Index {
         &self.tokenizers
     }
 
+    /// Helper to access the tokenizer associated to a specific field.
+    pub fn tokenizer_for_field(&self, field: Field) -> Result<Box<BoxedTokenizer>> {
+        let field_entry = self.schema.get_field_entry(field);
+        let field_type = field_entry.field_type();
+        let tokenizer_manager: &TokenizerManager = self.tokenizers();
+        let tokenizer_name_opt: Option<Box<BoxedTokenizer>> = match field_type {
+            FieldType::Str(text_options) => text_options
+                .get_indexing_options()
+                .map(|text_indexing_options| text_indexing_options.tokenizer().to_string())
+                .and_then(|tokenizer_name| tokenizer_manager.get(&tokenizer_name)),
+            _ => None,
+        };
+        match tokenizer_name_opt {
+            Some(tokenizer) => Ok(tokenizer),
+            None => Err(TantivyError::SchemaError(format!(
+                "{:?} is not a text field.",
+                field_entry.name()
+            ))),
+        }
+    }
+
     /// Opens a new directory from an index path.
     #[cfg(feature = "mmap")]
     pub fn open_in_dir<P: AsRef<Path>>(directory_path: P) -> Result<Index> {
@@ -156,7 +233,7 @@ impl Index {
         num_threads: usize,
         overall_heap_size_in_bytes: usize,
     ) -> Result<IndexWriter> {
-        let directory_lock = DirectoryLock::lock(self.directory().box_clone())?;
+        let directory_lock = LockType::IndexWriterLock.acquire_lock(&self.directory)?;
         let heap_size_in_bytes_per_thread = overall_heap_size_in_bytes / num_threads;
         open_index_writer(
             self,
@@ -194,7 +271,8 @@ impl Index {
 
     /// Returns the list of segments that are searchable
     pub fn searchable_segments(&self) -> Result<Vec<Segment>> {
-        Ok(self.searchable_segment_metas()?
+        Ok(self
+            .searchable_segment_metas()?
             .into_iter()
             .map(|segment_meta| self.segment(segment_meta))
             .collect())
@@ -229,7 +307,8 @@ impl Index {
 
     /// Returns the list of segment ids that are searchable.
     pub fn searchable_segment_ids(&self) -> Result<Vec<SegmentId>> {
-        Ok(self.searchable_segment_metas()?
+        Ok(self
+            .searchable_segment_metas()?
             .iter()
             .map(|segment_meta| segment_meta.id())
             .collect())
@@ -242,13 +321,18 @@ impl Index {
         self.num_searchers.store(num_searchers, Ordering::Release);
     }
 
-    /// Creates a new generation of searchers after
-
-    /// a change of the set of searchable indexes.
+    /// Update searchers so that they reflect the state of the last
+    /// `.commit()`.
     ///
-    /// This needs to be called when a new segment has been
-    /// published or after a merge.
+    /// If indexing happens in the same process as searching,
+    /// you most likely want to call `.load_searchers()` right after each
+    /// successful call to `.commit()`.
+    ///
+    /// If indexing and searching happen in different processes, the way to
+    /// get the freshest `index` at all time, is to watch `meta.json` and
+    /// call `load_searchers` whenever a changes happen.
     pub fn load_searchers(&self) -> Result<()> {
+        let _meta_lock = LockType::MetaLock.acquire_lock(self.directory())?;
         let searchable_segments = self.searchable_segments()?;
         let segment_readers: Vec<SegmentReader> = searchable_segments
             .iter()
@@ -257,7 +341,7 @@ impl Index {
         let schema = self.schema();
         let num_searchers: usize = self.num_searchers.load(Ordering::Acquire);
         let searchers = (0..num_searchers)
-            .map(|_| Searcher::new(schema.clone(), segment_readers.clone()))
+            .map(|_| Searcher::new(schema.clone(), self.clone(), segment_readers.clone()))
             .collect();
         self.searcher_pool.publish_new_generation(searchers);
         Ok(())
@@ -292,6 +376,79 @@ impl Clone for Index {
             num_searchers: Arc::clone(&self.num_searchers),
             searcher_pool: Arc::clone(&self.searcher_pool),
             tokenizers: self.tokenizers.clone(),
+            executor: self.executor.clone(),
         }
     }
 }
+
+#[cfg(test)]
+mod tests {
+    use directory::RAMDirectory;
+    use schema::{Schema, INT_INDEXED, TEXT};
+    use Index;
+
+    #[test]
+    fn test_indexer_for_field() {
+        let mut schema_builder = Schema::builder();
+        let num_likes_field = schema_builder.add_u64_field("num_likes", INT_INDEXED);
+        let body_field = schema_builder.add_text_field("body", TEXT);
+        let schema = schema_builder.build();
+        let index = Index::create_in_ram(schema);
+        assert!(index.tokenizer_for_field(body_field).is_ok());
+        assert_eq!(
+            format!("{:?}", index.tokenizer_for_field(num_likes_field).err()),
+            "Some(SchemaError(\"\\\"num_likes\\\" is not a text field.\"))"
+        );
+    }
+
+    #[test]
+    fn test_index_exists() {
+        let directory = RAMDirectory::create();
+        assert!(!Index::exists(&directory));
+        assert!(Index::create(directory.clone(), throw_away_schema()).is_ok());
+        assert!(Index::exists(&directory));
+    }
+
+    #[test]
+    fn open_or_create_should_create() {
+        let directory = RAMDirectory::create();
+        assert!(!Index::exists(&directory));
+        assert!(Index::open_or_create(directory.clone(), throw_away_schema()).is_ok());
+        assert!(Index::exists(&directory));
+    }
+
+    #[test]
+    fn open_or_create_should_open() {
+        let directory = RAMDirectory::create();
+        assert!(Index::create(directory.clone(), throw_away_schema()).is_ok());
+        assert!(Index::exists(&directory));
+        assert!(Index::open_or_create(directory, throw_away_schema()).is_ok());
+    }
+
+    #[test]
+    fn create_should_wipeoff_existing() {
+        let directory = RAMDirectory::create();
+        assert!(Index::create(directory.clone(), throw_away_schema()).is_ok());
+        assert!(Index::exists(&directory));
+        assert!(Index::create(directory.clone(), Schema::builder().build()).is_ok());
+    }
+
+    #[test]
+    fn open_or_create_exists_but_schema_does_not_match() {
+        let directory = RAMDirectory::create();
+        assert!(Index::create(directory.clone(), throw_away_schema()).is_ok());
+        assert!(Index::exists(&directory));
+        assert!(Index::open_or_create(directory.clone(), throw_away_schema()).is_ok());
+        let err = Index::open_or_create(directory, Schema::builder().build());
+        assert_eq!(
+            format!("{:?}", err.unwrap_err()),
+            "SchemaError(\"An index exists but the schema does not match.\")"
+        );
+    }
+
+    fn throw_away_schema() -> Schema {
+        let mut schema_builder = Schema::builder();
+        let _ = schema_builder.add_u64_field("num_likes", INT_INDEXED);
+        schema_builder.build()
+    }
+}

src/core/index_meta.rs

@@ -46,19 +46,19 @@ impl fmt::Debug for IndexMeta {
 mod tests {
 
     use super::IndexMeta;
-    use schema::{SchemaBuilder, TEXT};
+    use schema::{Schema, TEXT};
     use serde_json;
 
     #[test]
     fn test_serialize_metas() {
         let schema = {
-            let mut schema_builder = SchemaBuilder::new();
+            let mut schema_builder = Schema::builder();
             schema_builder.add_text_field("text", TEXT);
             schema_builder.build()
         };
         let index_metas = IndexMeta {
             segments: Vec::new(),
-            schema: schema,
+            schema,
             opstamp: 0u64,
             payload: None,
         };

src/core/inverted_index_reader.rs

@@ -1,13 +1,13 @@
 use common::BinarySerializable;
 use directory::ReadOnlySource;
+use owned_read::OwnedRead;
+use positions::PositionReader;
 use postings::TermInfo;
 use postings::{BlockSegmentPostings, SegmentPostings};
 use schema::FieldType;
 use schema::IndexRecordOption;
 use schema::Term;
 use termdict::TermDictionary;
-use owned_read::OwnedRead;
-use positions::PositionReader;
 
 /// The inverted index reader is in charge of accessing
 /// the inverted index associated to a specific field.
@@ -32,6 +32,10 @@ pub struct InvertedIndexReader {
 }
 
 impl InvertedIndexReader {
+    #[cfg_attr(
+        feature = "cargo-clippy",
+        allow(clippy::needless_pass_by_value)
+    )] // for symetry
     pub(crate) fn new(
         termdict: TermDictionary,
         postings_source: ReadOnlySource,
@@ -54,12 +58,12 @@ impl InvertedIndexReader {
 
     /// Creates an empty `InvertedIndexReader` object, which
     /// contains no terms at all.
-    pub fn empty(field_type: FieldType) -> InvertedIndexReader {
+    pub fn empty(field_type: &FieldType) -> InvertedIndexReader {
         let record_option = field_type
             .get_index_record_option()
             .unwrap_or(IndexRecordOption::Basic);
         InvertedIndexReader {
-            termdict: TermDictionary::empty(field_type),
+            termdict: TermDictionary::empty(&field_type),
             postings_source: ReadOnlySource::empty(),
             positions_source: ReadOnlySource::empty(),
             positions_idx_source: ReadOnlySource::empty(),
@@ -100,7 +104,6 @@ impl InvertedIndexReader {
         block_postings.reset(term_info.doc_freq, postings_reader);
     }
 
-
     /// Returns a block postings given a `Term`.
     /// This method is for an advanced usage only.
     ///
@@ -111,7 +114,7 @@ impl InvertedIndexReader {
         option: IndexRecordOption,
     ) -> Option<BlockSegmentPostings> {
         self.get_term_info(term)
-            .map(move|term_info| self.read_block_postings_from_terminfo(&term_info, option))
+            .map(move |term_info| self.read_block_postings_from_terminfo(&term_info, option))
     }
 
     /// Returns a block postings given a `term_info`.
@@ -147,7 +150,8 @@ impl InvertedIndexReader {
             if option.has_positions() {
                 let position_reader = self.positions_source.clone();
                 let skip_reader = self.positions_idx_source.clone();
-                let position_reader = PositionReader::new(position_reader, skip_reader, term_info.positions_idx);
+                let position_reader =
+                    PositionReader::new(position_reader, skip_reader, term_info.positions_idx);
                 Some(position_reader)
             } else {
                 None

src/core/mod.rs

@@ -1,3 +1,4 @@
+mod executor;
 pub mod index;
 mod index_meta;
 mod inverted_index_reader;
@@ -9,6 +10,7 @@ mod segment_id;
 mod segment_meta;
 mod segment_reader;
 
+pub use self::executor::Executor;
 pub use self::index::Index;
 pub use self::index_meta::IndexMeta;
 pub use self::inverted_index_reader::InvertedIndexReader;
@@ -33,10 +35,4 @@ lazy_static! {
     /// Removing this file is safe, but will prevent the garbage collection of all of the file that
     /// are currently in the directory
     pub static ref MANAGED_FILEPATH: PathBuf = PathBuf::from(".managed.json");
-
-    /// Only one process should be able to write tantivy's index at a time.
-    /// This file, when present, is in charge of preventing other processes to open an IndexWriter.
-    ///
-    /// If the process is killed and this file remains, it is safe to remove it manually.
-    pub static ref LOCKFILE_FILEPATH: PathBuf = PathBuf::from(".tantivy-indexer.lock");
 }

src/core/pool.rs

@@ -87,7 +87,8 @@ impl<T> Deref for LeasedItem<T> {
     type Target = T;
 
     fn deref(&self) -> &T {
-        &self.gen_item
+        &self
+            .gen_item
             .as_ref()
             .expect("Unwrapping a leased item should never fail")
             .item // unwrap is safe here
@@ -96,7 +97,8 @@ impl<T> Deref for LeasedItem<T> {
 
 impl<T> DerefMut for LeasedItem<T> {
     fn deref_mut(&mut self) -> &mut T {
-        &mut self.gen_item
+        &mut self
+            .gen_item
             .as_mut()
             .expect("Unwrapping a mut leased item should never fail")
             .item // unwrap is safe here

src/core/searcher.rs

@@ -1,16 +1,43 @@
 use collector::Collector;
+use collector::SegmentCollector;
+use core::Executor;
 use core::InvertedIndexReader;
 use core::SegmentReader;
 use query::Query;
+use query::Scorer;
+use query::Weight;
 use schema::Document;
 use schema::Schema;
 use schema::{Field, Term};
+use space_usage::SearcherSpaceUsage;
 use std::fmt;
 use std::sync::Arc;
+use store::StoreReader;
 use termdict::TermMerger;
 use DocAddress;
+use Index;
 use Result;
 
+fn collect_segment<C: Collector>(
+    collector: &C,
+    weight: &Weight,
+    segment_ord: u32,
+    segment_reader: &SegmentReader,
+) -> Result<C::Fruit> {
+    let mut scorer = weight.scorer(segment_reader)?;
+    let mut segment_collector = collector.for_segment(segment_ord as u32, segment_reader)?;
+    if let Some(delete_bitset) = segment_reader.delete_bitset() {
+        scorer.for_each(&mut |doc, score| {
+            if !delete_bitset.is_deleted(doc) {
+                segment_collector.collect(doc, score);
+            }
+        });
+    } else {
+        scorer.for_each(&mut |doc, score| segment_collector.collect(doc, score));
+    }
+    Ok(segment_collector.harvest())
+}
+
 /// Holds a list of `SegmentReader`s ready for search.
 ///
 /// It guarantees that the `Segment` will not be removed before
@@ -18,25 +45,43 @@ use Result;
 ///
 pub struct Searcher {
     schema: Schema,
+    index: Index,
     segment_readers: Vec<SegmentReader>,
+    store_readers: Vec<StoreReader>,
 }
 
 impl Searcher {
     /// Creates a new `Searcher`
-    pub(crate) fn new(schema: Schema, segment_readers: Vec<SegmentReader>) -> Searcher {
+    pub(crate) fn new(
+        schema: Schema,
+        index: Index,
+        segment_readers: Vec<SegmentReader>,
+    ) -> Searcher {
+        let store_readers = segment_readers
+            .iter()
+            .map(|segment_reader| segment_reader.get_store_reader())
+            .collect();
         Searcher {
             schema,
+            index,
             segment_readers,
+            store_readers,
         }
     }
+
+    /// Returns the `Index` associated to the `Searcher`
+    pub fn index(&self) -> &Index {
+        &self.index
+    }
+
     /// Fetches a document from tantivy's store given a `DocAddress`.
     ///
     /// The searcher uses the segment ordinal to route the
     /// the request to the right `Segment`.
-    pub fn doc(&self, doc_address: &DocAddress) -> Result<Document> {
-        let DocAddress(segment_local_id, doc_id) = *doc_address;
-        let segment_reader = &self.segment_readers[segment_local_id as usize];
-        segment_reader.doc(doc_id)
+    pub fn doc(&self, doc_address: DocAddress) -> Result<Document> {
+        let DocAddress(segment_local_id, doc_id) = doc_address;
+        let store_reader = &self.store_readers[segment_local_id as usize];
+        store_reader.get(doc_id)
     }
 
     /// Access the schema associated to the index of this searcher.
@@ -48,7 +93,7 @@ impl Searcher {
     pub fn num_docs(&self) -> u64 {
         self.segment_readers
             .iter()
-            .map(|segment_reader| segment_reader.num_docs() as u64)
+            .map(|segment_reader| u64::from(segment_reader.num_docs()))
             .sum::<u64>()
     }
 
@@ -57,8 +102,9 @@ impl Searcher {
     pub fn doc_freq(&self, term: &Term) -> u64 {
         self.segment_readers
             .iter()
-            .map(|segment_reader| segment_reader.inverted_index(term.field()).doc_freq(term) as u64)
-            .sum::<u64>()
+            .map(|segment_reader| {
+                u64::from(segment_reader.inverted_index(term.field()).doc_freq(term))
+            }).sum::<u64>()
     }
 
     /// Return the list of segment readers
@@ -71,19 +117,78 @@ impl Searcher {
         &self.segment_readers[segment_ord as usize]
     }
 
-    /// Runs a query on the segment readers wrapped by the searcher
-    pub fn search<C: Collector>(&self, query: &Query, collector: &mut C) -> Result<()> {
-        query.search(self, collector)
+    /// Runs a query on the segment readers wrapped by the searcher.
+    ///
+    /// Search works as follows :
+    ///
+    ///  First the weight object associated to the query is created.
+    ///
+    ///  Then, the query loops over the segments and for each segment :
+    ///  - setup the collector and informs it that the segment being processed has changed.
+    ///  - creates a SegmentCollector for collecting documents associated to the segment
+    ///  - creates a `Scorer` object associated for this segment
+    ///  - iterate through the matched documents and push them to the segment collector.
+    ///
+    ///  Finally, the Collector merges each of the child collectors into itself for result usability
+    ///  by the caller.
+    pub fn search<C: Collector>(&self, query: &Query, collector: &C) -> Result<C::Fruit> {
+        let executor = self.index.search_executor();
+        self.search_with_executor(query, collector, executor)
+    }
+
+    /// Same as [`search(...)`](#method.search) but multithreaded.
+    ///
+    /// The current implementation is rather naive :
+    /// multithreading is by splitting search into as many task
+    /// as there are segments.
+    ///
+    /// It is powerless at making search faster if your index consists in
+    /// one large segment.
+    ///
+    /// Also, keep in my multithreading a single query on several
+    /// threads will not improve your throughput. It can actually
+    /// hurt it. It will however, decrease the average response time.
+    pub fn search_with_executor<C: Collector>(
+        &self,
+        query: &Query,
+        collector: &C,
+        executor: &Executor,
+    ) -> Result<C::Fruit> {
+        let scoring_enabled = collector.requires_scoring();
+        let weight = query.weight(self, scoring_enabled)?;
+        let segment_readers = self.segment_readers();
+        let fruits = executor.map(
+            |(segment_ord, segment_reader)| {
+                collect_segment(
+                    collector,
+                    weight.as_ref(),
+                    segment_ord as u32,
+                    segment_reader,
+                )
+            },
+            segment_readers.iter().enumerate(),
+        )?;
+        collector.merge_fruits(fruits)
     }
 
     /// Return the field searcher associated to a `Field`.
     pub fn field(&self, field: Field) -> FieldSearcher {
-        let inv_index_readers = self.segment_readers
+        let inv_index_readers = self
+            .segment_readers
             .iter()
             .map(|segment_reader| segment_reader.inverted_index(field))
             .collect::<Vec<_>>();
         FieldSearcher::new(inv_index_readers)
     }
+
+    /// Summarize total space usage of this searcher.
+    pub fn space_usage(&self) -> SearcherSpaceUsage {
+        let mut space_usage = SearcherSpaceUsage::new();
+        for segment_reader in self.segment_readers.iter() {
+            space_usage.add_segment(segment_reader.space_usage());
+        }
+        space_usage
+    }
 }
 
 pub struct FieldSearcher {
@@ -98,7 +203,8 @@ impl FieldSearcher {
     /// Returns a Stream over all of the sorted unique terms of
     /// for the given field.
     pub fn terms(&self) -> TermMerger {
-        let term_streamers: Vec<_> = self.inv_index_readers
+        let term_streamers: Vec<_> = self
+            .inv_index_readers
             .iter()
             .map(|inverted_index| inverted_index.terms().stream())
             .collect();
@@ -108,7 +214,8 @@ impl FieldSearcher {
 
 impl fmt::Debug for Searcher {
     fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
-        let segment_ids = self.segment_readers
+        let segment_ids = self
+            .segment_readers
             .iter()
             .map(|segment_reader| segment_reader.segment_id())
             .collect::<Vec<_>>();

src/core/segment_id.rs

@@ -52,12 +52,12 @@ impl SegmentId {
     /// Picking the first 8 chars is ok to identify
     /// segments in a display message.
     pub fn short_uuid_string(&self) -> String {
-        (&self.0.simple().to_string()[..8]).to_string()
+        (&self.0.to_simple_ref().to_string()[..8]).to_string()
     }
 
     /// Returns a segment uuid string.
     pub fn uuid_string(&self) -> String {
-        self.0.simple().to_string()
+        self.0.to_simple_ref().to_string()
     }
 }
 

src/core/segment_meta.rs

@@ -50,7 +50,7 @@ impl<'a> serde::Deserialize<'a> for SegmentMeta {
     {
         let inner = InnerSegmentMeta::deserialize(deserializer)?;
         let tracked = INVENTORY.track(inner);
-        Ok(SegmentMeta { tracked: tracked })
+        Ok(SegmentMeta { tracked })
     }
 }
 

src/core/segment_reader.rs

@@ -4,7 +4,7 @@ use core::InvertedIndexReader;
 use core::Segment;
 use core::SegmentComponent;
 use core::SegmentId;
-use core::SegmentMeta;
+use directory::ReadOnlySource;
 use error::TantivyError;
 use fastfield::DeleteBitSet;
 use fastfield::FacetReader;
@@ -13,10 +13,10 @@ use fastfield::{self, FastFieldNotAvailableError};
 use fastfield::{BytesFastFieldReader, FastValue, MultiValueIntFastFieldReader};
 use fieldnorm::FieldNormReader;
 use schema::Cardinality;
-use schema::Document;
 use schema::Field;
 use schema::FieldType;
 use schema::Schema;
+use space_usage::SegmentSpaceUsage;
 use std::collections::HashMap;
 use std::fmt;
 use std::sync::Arc;
@@ -44,7 +44,8 @@ pub struct SegmentReader {
     inv_idx_reader_cache: Arc<RwLock<HashMap<Field, Arc<InvertedIndexReader>>>>,
 
     segment_id: SegmentId,
-    segment_meta: SegmentMeta,
+    max_doc: DocId,
+    num_docs: DocId,
 
     termdict_composite: CompositeFile,
     postings_composite: CompositeFile,
@@ -53,7 +54,7 @@ pub struct SegmentReader {
     fast_fields_composite: CompositeFile,
     fieldnorms_composite: CompositeFile,
 
-    store_reader: StoreReader,
+    store_source: ReadOnlySource,
     delete_bitset_opt: Option<DeleteBitSet>,
     schema: Schema,
 }
@@ -64,7 +65,7 @@ impl SegmentReader {
     /// Today, `tantivy` does not handle deletes, so it happens
     /// to also be the number of documents in the index.
     pub fn max_doc(&self) -> DocId {
-        self.segment_meta.max_doc()
+        self.max_doc
     }
 
     /// Returns the number of documents.
@@ -73,7 +74,7 @@ impl SegmentReader {
     /// Today, `tantivy` does not handle deletes so max doc and
     /// num_docs are the same.
     pub fn num_docs(&self) -> DocId {
-        self.segment_meta.num_docs()
+        self.num_docs
     }
 
     /// Returns the schema of the index this segment belongs to.
@@ -153,15 +154,17 @@ impl SegmentReader {
     /// Accessor to the `BytesFastFieldReader` associated to a given `Field`.
     pub fn bytes_fast_field_reader(&self, field: Field) -> fastfield::Result<BytesFastFieldReader> {
         let field_entry = self.schema.get_field_entry(field);
-        match field_entry.field_type() {
-            &FieldType::Bytes => {}
+        match *field_entry.field_type() {
+            FieldType::Bytes => {}
             _ => return Err(FastFieldNotAvailableError::new(field_entry)),
         }
-        let idx_reader = self.fast_fields_composite
+        let idx_reader = self
+            .fast_fields_composite
             .open_read_with_idx(field, 0)
             .ok_or_else(|| FastFieldNotAvailableError::new(field_entry))
             .map(FastFieldReader::open)?;
-        let values = self.fast_fields_composite
+        let values = self
+            .fast_fields_composite
             .open_read_with_idx(field, 1)
             .ok_or_else(|| FastFieldNotAvailableError::new(field_entry))?;
         Ok(BytesFastFieldReader::open(idx_reader, values))
@@ -175,7 +178,7 @@ impl SegmentReader {
                 "The field {:?} is not a \
                  hierarchical facet.",
                 field_entry
-            )).into());
+            )));
         }
         let term_ords_reader = self.multi_fast_field_reader(field)?;
         let termdict_source = self.termdict_composite.open_read(field).ok_or_else(|| {
@@ -186,7 +189,7 @@ impl SegmentReader {
                 field_entry.name()
             ))
         })?;
-        let termdict = TermDictionary::from_source(termdict_source);
+        let termdict = TermDictionary::from_source(&termdict_source);
         let facet_reader = FacetReader::new(term_ords_reader, termdict);
         Ok(facet_reader)
     }
@@ -194,8 +197,7 @@ impl SegmentReader {
     /// Accessor to the segment's `Field norms`'s reader.
     ///
     /// Field norms are the length (in tokens) of the fields.
-    /// It is used in the computation of the [TfIdf]
-    /// (https://fulmicoton.gitbooks.io/tantivy-doc/content/tfidf.html).
+    /// It is used in the computation of the [TfIdf](https://fulmicoton.gitbooks.io/tantivy-doc/content/tfidf.html).
     ///
     /// They are simply stored as a fast field, serialized in
     /// the `.fieldnorm` file of the segment.
@@ -213,8 +215,8 @@ impl SegmentReader {
     }
 
     /// Accessor to the segment's `StoreReader`.
-    pub fn get_store_reader(&self) -> &StoreReader {
-        &self.store_reader
+    pub fn get_store_reader(&self) -> StoreReader {
+        StoreReader::from_source(self.store_source.clone())
     }
 
     /// Open a new segment for reading.
@@ -223,7 +225,8 @@ impl SegmentReader {
         let termdict_composite = CompositeFile::open(&termdict_source)?;
 
         let store_source = segment.open_read(SegmentComponent::STORE)?;
-        let store_reader = StoreReader::from_source(store_source);
+
+        fail_point!("SegmentReader::open#middle");
 
         let postings_source = segment.open_read(SegmentComponent::POSTINGS)?;
         let postings_composite = CompositeFile::open(&postings_source)?;
@@ -260,13 +263,14 @@ impl SegmentReader {
         let schema = segment.schema();
         Ok(SegmentReader {
             inv_idx_reader_cache: Arc::new(RwLock::new(HashMap::new())),
-            segment_meta: segment.meta().clone(),
+            max_doc: segment.meta().max_doc(),
+            num_docs: segment.meta().num_docs(),
             termdict_composite,
             postings_composite,
             fast_fields_composite,
             fieldnorms_composite,
             segment_id: segment.id(),
-            store_reader,
+            store_source,
             delete_bitset_opt,
             positions_composite,
             positions_idx_composite,
@@ -282,7 +286,8 @@ impl SegmentReader {
     /// term dictionary associated to a specific field,
     /// and opening the posting list associated to any term.
     pub fn inverted_index(&self, field: Field) -> Arc<InvertedIndexReader> {
-        if let Some(inv_idx_reader) = self.inv_idx_reader_cache
+        if let Some(inv_idx_reader) = self
+            .inv_idx_reader_cache
             .read()
             .expect("Lock poisoned. This should never happen")
             .get(&field)
@@ -306,25 +311,28 @@ impl SegmentReader {
             // As a result, no data is associated to the inverted index.
             //
             // Returns an empty inverted index.
-            return Arc::new(InvertedIndexReader::empty(field_type.clone()));
+            return Arc::new(InvertedIndexReader::empty(field_type));
         }
 
         let postings_source = postings_source_opt.unwrap();
 
-        let termdict_source = self.termdict_composite
+        let termdict_source = self
+            .termdict_composite
             .open_read(field)
             .expect("Failed to open field term dictionary in composite file. Is the field indexed");
 
-        let positions_source = self.positions_composite
+        let positions_source = self
+            .positions_composite
             .open_read(field)
             .expect("Index corrupted. Failed to open field positions in composite file.");
 
-        let positions_idx_source = self.positions_idx_composite
+        let positions_idx_source = self
+            .positions_idx_composite
             .open_read(field)
             .expect("Index corrupted. Failed to open field positions in composite file.");
 
         let inv_idx_reader = Arc::new(InvertedIndexReader::new(
-            TermDictionary::from_source(termdict_source),
+            TermDictionary::from_source(&termdict_source),
             postings_source,
             positions_source,
             positions_idx_source,
@@ -341,14 +349,6 @@ impl SegmentReader {
         inv_idx_reader
     }
 
-    /// Returns the document (or to be accurate, its stored field)
-    /// bearing the given doc id.
-    /// This method is slow and should seldom be called from
-    /// within a collector.
-    pub fn doc(&self, doc_id: DocId) -> Result<Document> {
-        self.store_reader.get(doc_id)
-    }
-
     /// Returns the segment id
     pub fn segment_id(&self) -> SegmentId {
         self.segment_id
@@ -372,6 +372,24 @@ impl SegmentReader {
     pub fn doc_ids_alive(&self) -> SegmentReaderAliveDocsIterator {
         SegmentReaderAliveDocsIterator::new(&self)
     }
+
+    /// Summarize total space usage of this segment.
+    pub fn space_usage(&self) -> SegmentSpaceUsage {
+        SegmentSpaceUsage::new(
+            self.num_docs(),
+            self.termdict_composite.space_usage(),
+            self.postings_composite.space_usage(),
+            self.positions_composite.space_usage(),
+            self.positions_idx_composite.space_usage(),
+            self.fast_fields_composite.space_usage(),
+            self.fieldnorms_composite.space_usage(),
+            self.get_store_reader().space_usage(),
+            self.delete_bitset_opt
+                .as_ref()
+                .map(|x| x.space_usage())
+                .unwrap_or(0),
+        )
+    }
 }
 
 impl fmt::Debug for SegmentReader {
@@ -391,7 +409,7 @@ pub struct SegmentReaderAliveDocsIterator<'a> {
 impl<'a> SegmentReaderAliveDocsIterator<'a> {
     pub fn new(reader: &'a SegmentReader) -> SegmentReaderAliveDocsIterator<'a> {
         SegmentReaderAliveDocsIterator {
-            reader: reader,
+            reader,
             max_doc: reader.max_doc(),
             current: 0,
         }
@@ -429,12 +447,12 @@ impl<'a> Iterator for SegmentReaderAliveDocsIterator<'a> {
 #[cfg(test)]
 mod test {
     use core::Index;
-    use schema::{SchemaBuilder, Term, STORED, TEXT};
+    use schema::{Schema, Term, STORED, TEXT};
     use DocId;
 
     #[test]
     fn test_alive_docs_iterator() {
-        let mut schema_builder = SchemaBuilder::new();
+        let mut schema_builder = Schema::builder();
         schema_builder.add_text_field("name", TEXT | STORED);
         let schema = schema_builder.build();
         let index = Index::create_in_ram(schema.clone());

src/directory/directory.rs

@@ -77,15 +77,15 @@ pub trait Directory: DirectoryClone + fmt::Debug + Send + Sync + 'static {
 
 /// DirectoryClone
 pub trait DirectoryClone {
-  /// Clones the directory and boxes the clone
-  fn box_clone(&self) -> Box<Directory>;
+    /// Clones the directory and boxes the clone
+    fn box_clone(&self) -> Box<Directory>;
 }
 
 impl<T> DirectoryClone for T
 where
-  T: 'static + Directory + Clone,
+    T: 'static + Directory + Clone,
 {
-  fn box_clone(&self) -> Box<Directory> {
-    Box::new(self.clone())
-  }
+    fn box_clone(&self) -> Box<Directory> {
+        Box::new(self.clone())
+    }
 }

src/directory/managed_directory.rs

@@ -2,6 +2,7 @@ use core::MANAGED_FILEPATH;
 use directory::error::{DeleteError, IOError, OpenReadError, OpenWriteError};
 use directory::{ReadOnlySource, WritePtr};
 use error::TantivyError;
+use indexer::LockType;
 use serde_json;
 use std::collections::HashSet;
 use std::io;
@@ -13,6 +14,17 @@ use std::sync::{Arc, RwLock};
 use Directory;
 use Result;
 
+/// Returns true iff the file is "managed".
+/// Non-managed file are not subject to garbage collection.
+///
+/// Filenames that starts by a "." -typically locks-
+/// are not managed.
+fn is_managed(path: &Path) -> bool {
+    path.to_str()
+        .map(|p_str| !p_str.starts_with('.'))
+        .unwrap_or(true)
+}
+
 /// Wrapper of directories that keeps track of files created by Tantivy.
 ///
 /// A managed directory is just a wrapper of a directory
@@ -40,7 +52,7 @@ fn save_managed_paths(
     wlock: &RwLockWriteGuard<MetaInformation>,
 ) -> io::Result<()> {
     let mut w = serde_json::to_vec(&wlock.managed_paths)?;
-    write!(&mut w, "\n")?;
+    writeln!(&mut w)?;
     directory.atomic_write(&MANAGED_FILEPATH, &w[..])?;
     Ok(())
 }
@@ -82,25 +94,35 @@ impl ManagedDirectory {
     pub fn garbage_collect<L: FnOnce() -> HashSet<PathBuf>>(&mut self, get_living_files: L) {
         info!("Garbage collect");
         let mut files_to_delete = vec![];
+
+        // It is crucial to get the living files after acquiring the
+        // read lock of meta informations. That way, we
+        // avoid the following scenario.
+        //
+        // 1) we get the list of living files.
+        // 2) someone creates a new file.
+        // 3) we start garbage collection and remove this file
+        // even though it is a living file.
+        //
+        // releasing the lock as .delete() will use it too.
         {
-            // releasing the lock as .delete() will use it too.
-            let meta_informations_rlock = self.meta_informations
+            let meta_informations_rlock = self
+                .meta_informations
                 .read()
                 .expect("Managed directory rlock poisoned in garbage collect.");
 
-            // It is crucial to get the living files after acquiring the
-            // read lock of meta informations. That way, we
-            // avoid the following scenario.
-            //
-            // 1) we get the list of living files.
-            // 2) someone creates a new file.
-            // 3) we start garbage collection and remove this file
-            // even though it is a living file.
-            let living_files = get_living_files();
-
-            for managed_path in &meta_informations_rlock.managed_paths {
-                if !living_files.contains(managed_path) {
-                    files_to_delete.push(managed_path.clone());
+            // The point of this second "file" lock is to enforce the following scenario
+            // 1) process B tries to load a new set of searcher.
+            // The list of segments is loaded
+            // 2) writer change meta.json (for instance after a merge or a commit)
+            // 3) gc kicks in.
+            // 4) gc removes a file that was useful for process B, before process B opened it.
+            if let Ok(_meta_lock) = LockType::MetaLock.acquire_lock(self) {
+                let living_files = get_living_files();
+                for managed_path in &meta_informations_rlock.managed_paths {
+                    if !living_files.contains(managed_path) {
+                        files_to_delete.push(managed_path.clone());
+                    }
                 }
             }
         }
@@ -134,7 +156,8 @@ impl ManagedDirectory {
         if !deleted_files.is_empty() {
             // update the list of managed files by removing
             // the file that were removed.
-            let mut meta_informations_wlock = self.meta_informations
+            let mut meta_informations_wlock = self
+                .meta_informations
                 .write()
                 .expect("Managed directory wlock poisoned (2).");
             {
@@ -156,8 +179,17 @@ impl ManagedDirectory {
     /// registering the filepath and creating the file
     /// will not lead to garbage files that will
     /// never get removed.
+    ///
+    /// File starting by "." are reserved to locks.
+    /// They are not managed and cannot be subjected
+    /// to garbage collection.
     fn register_file_as_managed(&mut self, filepath: &Path) -> io::Result<()> {
-        let mut meta_wlock = self.meta_informations
+        // Files starting by "." (e.g. lock files) are not managed.
+        if !is_managed(filepath) {
+            return Ok(());
+        }
+        let mut meta_wlock = self
+            .meta_informations
             .write()
             .expect("Managed file lock poisoned");
         let has_changed = meta_wlock.managed_paths.insert(filepath.to_owned());

src/directory/mmap_directory.rs

@@ -32,7 +32,8 @@ fn open_mmap(full_path: &Path) -> result::Result<Option<MmapReadOnly>, OpenReadE
         }
     })?;
 
-    let meta_data = file.metadata()
+    let meta_data = file
+        .metadata()
         .map_err(|e| IOError::with_path(full_path.to_owned(), e))?;
     if meta_data.len() == 0 {
         // if the file size is 0, it will not be possible
@@ -309,7 +310,8 @@ impl Directory for MmapDirectory {
         // when the last reference is gone.
         mmap_cache.cache.remove(&full_path);
         match fs::remove_file(&full_path) {
-            Ok(_) => self.sync_directory()
+            Ok(_) => self
+                .sync_directory()
                 .map_err(|e| IOError::with_path(path.to_owned(), e).into()),
             Err(e) => {
                 if e.kind() == io::ErrorKind::NotFound {
@@ -362,6 +364,11 @@ mod tests {
 
     use super::*;
 
+    #[test]
+    fn test_open_non_existant_path() {
+        assert!(MmapDirectory::open(PathBuf::from("./nowhere")).is_err());
+    }
+
     #[test]
     fn test_open_empty() {
         // empty file is actually an edge case because those

src/directory/ram_directory.rs

@@ -100,8 +100,7 @@ impl InnerDirectory {
                 );
                 let io_err = make_io_err(msg);
                 OpenReadError::IOError(IOError::with_path(path.to_owned(), io_err))
-            })
-            .and_then(|readable_map| {
+            }).and_then(|readable_map| {
                 readable_map
                     .get(path)
                     .ok_or_else(|| OpenReadError::FileDoesNotExist(PathBuf::from(path)))
@@ -121,8 +120,7 @@ impl InnerDirectory {
                 );
                 let io_err = make_io_err(msg);
                 DeleteError::IOError(IOError::with_path(path.to_owned(), io_err))
-            })
-            .and_then(|mut writable_map| match writable_map.remove(path) {
+            }).and_then(|mut writable_map| match writable_map.remove(path) {
                 Some(_) => Ok(()),
                 None => Err(DeleteError::FileDoesNotExist(PathBuf::from(path))),
             })
@@ -170,10 +168,10 @@ impl Directory for RAMDirectory {
         let path_buf = PathBuf::from(path);
         let vec_writer = VecWriter::new(path_buf.clone(), self.fs.clone());
 
-        let exists = self.fs
+        let exists = self
+            .fs
             .write(path_buf.clone(), &Vec::new())
             .map_err(|err| IOError::with_path(path.to_owned(), err))?;
-
         // force the creation of the file to mimic the MMap directory.
         if exists {
             Err(OpenWriteError::FileAlreadyExists(path_buf))
@@ -196,6 +194,10 @@ impl Directory for RAMDirectory {
     }
 
     fn atomic_write(&mut self, path: &Path, data: &[u8]) -> io::Result<()> {
+        fail_point!("RAMDirectory::atomic_write", |msg| Err(io::Error::new(
+            io::ErrorKind::Other,
+            msg.unwrap_or("Undefined".to_string())
+        )));
         let path_buf = PathBuf::from(path);
         let mut vec_writer = VecWriter::new(path_buf.clone(), self.fs.clone());
         self.fs.write(path_buf, &Vec::new())?;

src/directory/read_only_source.rs

@@ -5,7 +5,6 @@ use fst::raw::MmapReadOnly;
 use stable_deref_trait::{CloneStableDeref, StableDeref};
 use std::ops::Deref;
 
-
 /// Read object that represents files in tantivy.
 ///
 /// These read objects are only in charge to deliver

src/error.rs

@@ -4,6 +4,7 @@ use std::io;
 
 use directory::error::{IOError, OpenDirectoryError, OpenReadError, OpenWriteError};
 use fastfield::FastFieldNotAvailableError;
+use indexer::LockType;
 use query;
 use schema;
 use serde_json;
@@ -14,80 +15,88 @@ use std::sync::PoisonError;
 #[derive(Debug, Fail)]
 pub enum TantivyError {
     /// Path does not exist.
-    #[fail(display = "path does not exist: '{:?}'", _0)]
+    #[fail(display = "Path does not exist: '{:?}'", _0)]
     PathDoesNotExist(PathBuf),
     /// File already exists, this is a problem when we try to write into a new file.
-    #[fail(display = "file already exists: '{:?}'", _0)]
+    #[fail(display = "File already exists: '{:?}'", _0)]
     FileAlreadyExists(PathBuf),
+    /// Index already exists in this directory
+    #[fail(display = "Index already exists")]
+    IndexAlreadyExists,
+    /// Failed to acquire file lock
+    #[fail(
+        display = "Failed to acquire Lockfile: {:?}. Possible causes: another IndexWriter instance or panic during previous lock drop.",
+        _0
+    )]
+    LockFailure(LockType),
     /// IO Error.
-    #[fail(display = "an IO error occurred: '{}'", _0)]
+    #[fail(display = "An IO error occurred: '{}'", _0)]
     IOError(#[cause] IOError),
-    /// The data within is corrupted.
-    ///
-    /// For instance, it contains invalid JSON.
-    #[fail(display = "file contains corrupted data: '{:?}'", _0)]
+    /// Data corruption.
+    #[fail(display = "File contains corrupted data: '{:?}'", _0)]
     CorruptedFile(PathBuf),
     /// A thread holding the locked panicked and poisoned the lock.
-    #[fail(display = "a thread holding the locked panicked and poisoned the lock")]
+    #[fail(display = "A thread holding the locked panicked and poisoned the lock")]
     Poisoned,
     /// Invalid argument was passed by the user.
-    #[fail(display = "an invalid argument was passed: '{}'", _0)]
+    #[fail(display = "An invalid argument was passed: '{}'", _0)]
     InvalidArgument(String),
     /// An Error happened in one of the thread.
-    #[fail(display = "an error occurred in a thread: '{}'", _0)]
+    #[fail(display = "An error occurred in a thread: '{}'", _0)]
     ErrorInThread(String),
     /// An Error appeared related to the schema.
     #[fail(display = "Schema error: '{}'", _0)]
     SchemaError(String),
     /// Tried to access a fastfield reader for a field not configured accordingly.
-    #[fail(display = "fast field not available: '{:?}'", _0)]
+    #[fail(display = "Fast field not available: '{:?}'", _0)]
     FastFieldError(#[cause] FastFieldNotAvailableError),
+    /// System error. (e.g.: We failed spawning a new thread)
+    #[fail(display = "System error.'{}'", _0)]
+    SystemError(String),
 }
 
 impl From<FastFieldNotAvailableError> for TantivyError {
     fn from(fastfield_error: FastFieldNotAvailableError) -> TantivyError {
-        TantivyError::FastFieldError(fastfield_error).into()
+        TantivyError::FastFieldError(fastfield_error)
     }
 }
 
 impl From<IOError> for TantivyError {
     fn from(io_error: IOError) -> TantivyError {
-        TantivyError::IOError(io_error).into()
+        TantivyError::IOError(io_error)
     }
 }
 
 impl From<io::Error> for TantivyError {
     fn from(io_error: io::Error) -> TantivyError {
-        TantivyError::IOError(io_error.into()).into()
+        TantivyError::IOError(io_error.into())
     }
 }
 
 impl From<query::QueryParserError> for TantivyError {
     fn from(parsing_error: query::QueryParserError) -> TantivyError {
-        TantivyError::InvalidArgument(format!("Query is invalid. {:?}", parsing_error)).into()
+        TantivyError::InvalidArgument(format!("Query is invalid. {:?}", parsing_error))
     }
 }
 
 impl<Guard> From<PoisonError<Guard>> for TantivyError {
     fn from(_: PoisonError<Guard>) -> TantivyError {
-        TantivyError::Poisoned.into()
+        TantivyError::Poisoned
     }
 }
 
 impl From<OpenReadError> for TantivyError {
     fn from(error: OpenReadError) -> TantivyError {
         match error {
-            OpenReadError::FileDoesNotExist(filepath) => {
-                TantivyError::PathDoesNotExist(filepath).into()
-            }
-            OpenReadError::IOError(io_error) => TantivyError::IOError(io_error).into(),
+            OpenReadError::FileDoesNotExist(filepath) => TantivyError::PathDoesNotExist(filepath),
+            OpenReadError::IOError(io_error) => TantivyError::IOError(io_error),
         }
     }
 }
 
 impl From<schema::DocParsingError> for TantivyError {
     fn from(error: schema::DocParsingError) -> TantivyError {
-        TantivyError::InvalidArgument(format!("Failed to parse document {:?}", error)).into()
+        TantivyError::InvalidArgument(format!("Failed to parse document {:?}", error))
     }
 }
 
@@ -98,7 +107,7 @@ impl From<OpenWriteError> for TantivyError {
                 TantivyError::FileAlreadyExists(filepath)
             }
             OpenWriteError::IOError(io_error) => TantivyError::IOError(io_error),
-        }.into()
+        }
     }
 }
 
@@ -106,11 +115,11 @@ impl From<OpenDirectoryError> for TantivyError {
     fn from(error: OpenDirectoryError) -> TantivyError {
         match error {
             OpenDirectoryError::DoesNotExist(directory_path) => {
-                TantivyError::PathDoesNotExist(directory_path).into()
+                TantivyError::PathDoesNotExist(directory_path)
+            }
+            OpenDirectoryError::NotADirectory(directory_path) => {
+                TantivyError::InvalidArgument(format!("{:?} is not a directory", directory_path))
             }
-            OpenDirectoryError::NotADirectory(directory_path) => TantivyError::InvalidArgument(
-                format!("{:?} is not a directory", directory_path),
-            ).into(),
         }
     }
 }
@@ -118,6 +127,6 @@ impl From<OpenDirectoryError> for TantivyError {
 impl From<serde_json::Error> for TantivyError {
     fn from(error: serde_json::Error) -> TantivyError {
         let io_err = io::Error::from(error);
-        TantivyError::IOError(io_err.into()).into()
+        TantivyError::IOError(io_err.into())
     }
 }

src/fastfield/bytes/mod.rs

@@ -6,12 +6,12 @@ pub use self::writer::BytesFastFieldWriter;
 
 #[cfg(test)]
 mod tests {
-    use schema::SchemaBuilder;
+    use schema::Schema;
     use Index;
 
     #[test]
     fn test_bytes() {
-        let mut schema_builder = SchemaBuilder::default();
+        let mut schema_builder = Schema::builder();
         let field = schema_builder.add_bytes_field("bytesfield");
         let schema = schema_builder.build();
         let index = Index::create_in_ram(schema);

src/fastfield/bytes/writer.rs

@@ -51,7 +51,7 @@ impl BytesFastFieldWriter {
         self.next_doc();
         for field_value in doc.field_values() {
             if field_value.field() == self.field {
-                if let &Value::Bytes(ref bytes) = field_value.value() {
+                if let Value::Bytes(ref bytes) = *field_value.value() {
                     self.vals.extend_from_slice(bytes);
                 } else {
                     panic!(

src/fastfield/delete.rs

@@ -2,6 +2,7 @@ use bit_set::BitSet;
 use common::HasLen;
 use directory::ReadOnlySource;
 use directory::WritePtr;
+use space_usage::ByteCount;
 use std::io;
 use std::io::Write;
 use DocId;
@@ -41,7 +42,8 @@ pub struct DeleteBitSet {
 impl DeleteBitSet {
     /// Opens a delete bitset given its data source.
     pub fn open(data: ReadOnlySource) -> DeleteBitSet {
-        let num_deleted: usize = data.as_slice()
+        let num_deleted: usize = data
+            .as_slice()
             .iter()
             .map(|b| b.count_ones() as usize)
             .sum();
@@ -62,6 +64,11 @@ impl DeleteBitSet {
             b & (1u8 << shift) != 0
         }
     }
+
+    /// Summarize total space usage of this bitset.
+    pub fn space_usage(&self) -> ByteCount {
+        self.data.len()
+    }
 }
 
 impl HasLen for DeleteBitSet {

src/fastfield/facet_reader.rs

@@ -56,7 +56,8 @@ impl FacetReader {
 
     /// Given a term ordinal returns the term associated to it.
     pub fn facet_from_ord(&self, facet_ord: TermOrdinal, output: &mut Facet) {
-        let found_term = self.term_dict
+        let found_term = self
+            .term_dict
             .ord_to_term(facet_ord as u64, output.inner_buffer_mut());
         assert!(found_term, "Term ordinal {} no found.", facet_ord);
     }

src/fastfield/mod.rs

@@ -127,19 +127,19 @@ mod tests {
     use common::CompositeFile;
     use directory::{Directory, RAMDirectory, WritePtr};
     use fastfield::FastFieldReader;
-    use rand::Rng;
+    use rand::prelude::SliceRandom;
+    use rand::rngs::StdRng;
     use rand::SeedableRng;
-    use rand::XorShiftRng;
     use schema::Document;
     use schema::Field;
+    use schema::Schema;
     use schema::FAST;
-    use schema::{Schema, SchemaBuilder};
     use std::collections::HashMap;
     use std::path::Path;
 
     lazy_static! {
         pub static ref SCHEMA: Schema = {
-            let mut schema_builder = SchemaBuilder::default();
+            let mut schema_builder = Schema::builder();
             schema_builder.add_u64_field("field", FAST);
             schema_builder.build()
         };
@@ -298,7 +298,7 @@ mod tests {
     fn test_signed_intfastfield() {
         let path = Path::new("test");
         let mut directory: RAMDirectory = RAMDirectory::create();
-        let mut schema_builder = SchemaBuilder::new();
+        let mut schema_builder = Schema::builder();
 
         let i64_field = schema_builder.add_i64_field("field", FAST);
         let schema = schema_builder.build();
@@ -342,7 +342,7 @@ mod tests {
     fn test_signed_intfastfield_default_val() {
         let path = Path::new("test");
         let mut directory: RAMDirectory = RAMDirectory::create();
-        let mut schema_builder = SchemaBuilder::new();
+        let mut schema_builder = Schema::builder();
         let i64_field = schema_builder.add_i64_field("field", FAST);
         let schema = schema_builder.build();
 
@@ -367,11 +367,10 @@ mod tests {
         }
     }
 
+    // Warning: this generates the same permutation at each call
     pub fn generate_permutation() -> Vec<u64> {
-        let seed: [u8; 16] = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16];
-        let mut rng = XorShiftRng::from_seed(seed);
-        let mut permutation: Vec<u64> = (0u64..1_000_000u64).collect();
-        rng.shuffle(&mut permutation);
+        let mut permutation: Vec<u64> = (0u64..100_000u64).collect();
+        permutation.shuffle(&mut StdRng::from_seed([1u8; 32]));
         permutation
     }
 

src/fastfield/multivalued/mod.rs

@@ -9,12 +9,12 @@ mod tests {
 
     use schema::Cardinality;
     use schema::IntOptions;
-    use schema::SchemaBuilder;
+    use schema::Schema;
     use Index;
 
     #[test]
     fn test_multivalued_u64() {
-        let mut schema_builder = SchemaBuilder::default();
+        let mut schema_builder = Schema::builder();
         let field = schema_builder.add_u64_field(
             "multifield",
             IntOptions::default().set_fast(Cardinality::MultiValues),
@@ -49,7 +49,7 @@ mod tests {
 
     #[test]
     fn test_multivalued_i64() {
-        let mut schema_builder = SchemaBuilder::default();
+        let mut schema_builder = Schema::builder();
         let field = schema_builder.add_i64_field(
             "multifield",
             IntOptions::default().set_fast(Cardinality::MultiValues),

src/fastfield/multivalued/reader.rs

@@ -47,11 +47,11 @@ impl<Item: FastValue> MultiValueIntFastFieldReader<Item> {
 mod tests {
 
     use core::Index;
-    use schema::{Document, Facet, SchemaBuilder};
+    use schema::{Document, Facet, Schema};
 
     #[test]
     fn test_multifastfield_reader() {
-        let mut schema_builder = SchemaBuilder::new();
+        let mut schema_builder = Schema::builder();
         let facet_field = schema_builder.add_facet_field("facets");
         let schema = schema_builder.build();
         let index = Index::create_in_ram(schema);

src/fastfield/multivalued/writer.rs

@@ -132,7 +132,8 @@ impl MultiValueIntFastFieldWriter {
                     );
 
                     let mut doc_vals: Vec<u64> = Vec::with_capacity(100);
-                    for (start, stop) in self.doc_index
+                    for (start, stop) in self
+                        .doc_index
                         .windows(2)
                         .map(|interval| (interval[0], interval[1]))
                         .chain(Some(last_interval).into_iter())
@@ -148,7 +149,6 @@ impl MultiValueIntFastFieldWriter {
                             value_serializer.add_val(val)?;
                         }
                     }
-
                 }
                 None => {
                     let val_min_max = self.vals.iter().cloned().minmax();

src/fastfield/reader.rs

@@ -7,11 +7,10 @@ use directory::ReadOnlySource;
 use directory::{Directory, RAMDirectory, WritePtr};
 use fastfield::{FastFieldSerializer, FastFieldsWriter};
 use owning_ref::OwningRef;
-use schema::SchemaBuilder;
+use schema::Schema;
 use schema::FAST;
 use std::collections::HashMap;
 use std::marker::PhantomData;
-use std::mem;
 use std::path::Path;
 use DocId;
 
@@ -80,7 +79,8 @@ impl<Item: FastValue> FastFieldReader<Item> {
     // TODO change start to `u64`.
     // For multifastfield, start is an index in a second fastfield, not a `DocId`
     pub fn get_range(&self, start: u32, output: &mut [Item]) {
-        let output_u64: &mut [u64] = unsafe { mem::transmute(output) }; // ok: Item is either `u64` or `i64`
+        // ok: Item is either `u64` or `i64`
+        let output_u64: &mut [u64] = unsafe { &mut *(output as *mut [Item] as *mut [u64]) };
         self.bit_unpacker.get_range(start, output_u64);
         for out in output_u64.iter_mut() {
             *out = Item::from_u64(*out + self.min_value_u64).as_u64();
@@ -108,7 +108,7 @@ impl<Item: FastValue> FastFieldReader<Item> {
 
 impl<Item: FastValue> From<Vec<Item>> for FastFieldReader<Item> {
     fn from(vals: Vec<Item>) -> FastFieldReader<Item> {
-        let mut schema_builder = SchemaBuilder::default();
+        let mut schema_builder = Schema::builder();
         let field = schema_builder.add_u64_field("field", FAST);
         let schema = schema_builder.build();
         let path = Path::new("__dummy__");

src/fieldnorm/code.rs

@@ -10,27 +10,28 @@ pub fn fieldnorm_to_id(fieldnorm: u32) -> u8 {
         .unwrap_or_else(|idx| idx - 1) as u8
 }
 
+#[cfg_attr(feature = "cargo-clippy", allow(clippy::unreadable_literal))]
 pub const FIELD_NORMS_TABLE: [u32; 256] = [
     0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25,
     26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 42, 44, 46, 48, 50, 52, 54, 56, 60,
     64, 68, 72, 76, 80, 84, 88, 96, 104, 112, 120, 128, 136, 144, 152, 168, 184, 200, 216, 232,
-    248, 264, 280, 312, 344, 376, 408, 440, 472, 504, 536, 600, 664, 728, 792, 856, 920, 984, 1048,
-    1176, 1304, 1432, 1560, 1688, 1816, 1944, 2072, 2328, 2584, 2840, 3096, 3352, 3608, 3864, 4120,
-    4632, 5144, 5656, 6168, 6680, 7192, 7704, 8216, 9240, 10264, 11288, 12312, 13336, 14360, 15384,
-    16408, 18456, 20504, 22552, 24600, 26648, 28696, 30744, 32792, 36888, 40984, 45080, 49176,
-    53272, 57368, 61464, 65560, 73752, 81944, 90136, 98328, 106520, 114712, 122904, 131096, 147480,
-    163864, 180248, 196632, 213016, 229400, 245784, 262168, 294936, 327704, 360472, 393240, 426008,
-    458776, 491544, 524312, 589848, 655384, 720920, 786456, 851992, 917528, 983064, 1048600,
-    1179672, 1310744, 1441816, 1572888, 1703960, 1835032, 1966104, 2097176, 2359320, 2621464,
-    2883608, 3145752, 3407896, 3670040, 3932184, 4194328, 4718616, 5242904, 5767192, 6291480,
-    6815768, 7340056, 7864344, 8388632, 9437208, 10485784, 11534360, 12582936, 13631512, 14680088,
-    15728664, 16777240, 18874392, 20971544, 23068696, 25165848, 27263000, 29360152, 31457304,
-    33554456, 37748760, 41943064, 46137368, 50331672, 54525976, 58720280, 62914584, 67108888,
-    75497496, 83886104, 92274712, 100663320, 109051928, 117440536, 125829144, 134217752, 150994968,
-    167772184, 184549400, 201326616, 218103832, 234881048, 251658264, 268435480, 301989912,
-    335544344, 369098776, 402653208, 436207640, 469762072, 503316504, 536870936, 603979800,
-    671088664, 738197528, 805306392, 872415256, 939524120, 1006632984, 1073741848, 1207959576,
-    1342177304, 1476395032, 1610612760, 1744830488, 1879048216, 2013265944,
+    248, 264, 280, 312, 344, 376, 408, 440, 472, 504, 536, 600, 664, 728, 792, 856, 920, 984,
+    1_048, 1176, 1304, 1432, 1560, 1688, 1816, 1944, 2072, 2328, 2584, 2840, 3096, 3352, 3608,
+    3864, 4120, 4632, 5144, 5656, 6168, 6680, 7192, 7704, 8216, 9240, 10264, 11288, 12312, 13336,
+    14360, 15384, 16408, 18456, 20504, 22552, 24600, 26648, 28696, 30744, 32792, 36888, 40984,
+    45080, 49176, 53272, 57368, 61464, 65560, 73752, 81944, 90136, 98328, 106520, 114712, 122904,
+    131096, 147480, 163864, 180248, 196632, 213016, 229400, 245784, 262168, 294936, 327704, 360472,
+    393240, 426008, 458776, 491544, 524312, 589848, 655384, 720920, 786456, 851992, 917528, 983064,
+    1048600, 1179672, 1310744, 1441816, 1572888, 1703960, 1835032, 1966104, 2097176, 2359320,
+    2621464, 2883608, 3145752, 3407896, 3670040, 3932184, 4194328, 4718616, 5242904, 5767192,
+    6291480, 6815768, 7340056, 7864344, 8388632, 9437208, 10485784, 11534360, 12582936, 13631512,
+    14680088, 15728664, 16777240, 18874392, 20971544, 23068696, 25165848, 27263000, 29360152,
+    31457304, 33554456, 37748760, 41943064, 46137368, 50331672, 54525976, 58720280, 62914584,
+    67108888, 75497496, 83886104, 92274712, 100663320, 109051928, 117440536, 125829144, 134217752,
+    150994968, 167772184, 184549400, 201326616, 218103832, 234881048, 251658264, 268435480,
+    301989912, 335544344, 369098776, 402653208, 436207640, 469762072, 503316504, 536870936,
+    603979800, 671088664, 738197528, 805306392, 872415256, 939524120, 1006632984, 1073741848,
+    1207959576, 1342177304, 1476395032, 1610612760, 1744830488, 1879048216, 2013265944,
 ];
 
 #[cfg(test)]

src/fieldnorm/mod.rs

@@ -15,7 +15,7 @@
 //! precompute computationally expensive functions of the fieldnorm
 //! in a very short array.
 //!
-//! This trick is used by the [BM25 similarity]().
+//! This trick is used by the BM25 similarity.
 mod code;
 mod reader;
 mod serializer;

src/functional_test.rs

@@ -2,7 +2,6 @@ use rand::thread_rng;
 use std::collections::HashSet;
 
 use rand::Rng;
-use rand::distributions::Range;
 use schema::*;
 use Index;
 use Searcher;
@@ -16,7 +15,7 @@ fn check_index_content(searcher: &Searcher, vals: &HashSet<u64>) {
 #[ignore]
 #[cfg(feature = "mmap")]
 fn test_indexing() {
-    let mut schema_builder = SchemaBuilder::default();
+    let mut schema_builder = Schema::builder();
 
     let id_field = schema_builder.add_u64_field("id", INT_INDEXED);
     let multiples_field = schema_builder.add_u64_field("multiples", INT_INDEXED);
@@ -24,7 +23,6 @@ fn test_indexing() {
 
     let index = Index::create_from_tempdir(schema).unwrap();
 
-    let universe = Range::new(0u64, 20u64);
     let mut rng = thread_rng();
 
     let mut index_writer = index.writer_with_num_threads(3, 120_000_000).unwrap();
@@ -33,7 +31,7 @@ fn test_indexing() {
     let mut uncommitted_docs: HashSet<u64> = HashSet::new();
 
     for _ in 0..200 {
-        let random_val = rng.sample(&universe);
+        let random_val = rng.gen_range(0, 20);
         if random_val == 0 {
             index_writer.commit().expect("Commit failed");
             committed_docs.extend(&uncommitted_docs);

src/indexer/delete_queue.rs

@@ -52,7 +52,8 @@ impl DeleteQueue {
     //
     // Past delete operations are not accessible.
     pub fn cursor(&self) -> DeleteCursor {
-        let last_block = self.inner
+        let last_block = self
+            .inner
             .read()
             .expect("Read lock poisoned when opening delete queue cursor")
             .last_block
@@ -92,7 +93,8 @@ impl DeleteQueue {
     // be some unflushed operations.
     //
     fn flush(&self) -> Option<Arc<Block>> {
-        let mut self_wlock = self.inner
+        let mut self_wlock = self
+            .inner
             .write()
             .expect("Failed to acquire write lock on delete queue writer");
 
@@ -132,7 +134,8 @@ impl From<DeleteQueue> for NextBlock {
 impl NextBlock {
     fn next_block(&self) -> Option<Arc<Block>> {
         {
-            let next_read_lock = self.0
+            let next_read_lock = self
+                .0
                 .read()
                 .expect("Failed to acquire write lock in delete queue");
             if let InnerNextBlock::Closed(ref block) = *next_read_lock {
@@ -141,7 +144,8 @@ impl NextBlock {
         }
         let next_block;
         {
-            let mut next_write_lock = self.0
+            let mut next_write_lock = self
+                .0
                 .write()
                 .expect("Failed to acquire write lock in delete queue");
             match *next_write_lock {
@@ -182,19 +186,21 @@ impl DeleteCursor {
     /// `opstamp >= target_opstamp`.
     pub fn skip_to(&mut self, target_opstamp: u64) {
         // TODO Can be optimize as we work with block.
-        #[cfg_attr(feature = "cargo-clippy", allow(while_let_loop))]
-        loop {
-            if let Some(operation) = self.get() {
-                if operation.opstamp >= target_opstamp {
-                    break;
-                }
-            } else {
-                break;
-            }
+        while self.is_behind_opstamp(target_opstamp) {
             self.advance();
         }
     }
 
+    #[cfg_attr(
+        feature = "cargo-clippy",
+        allow(clippy::wrong_self_convention)
+    )]
+    fn is_behind_opstamp(&mut self, target_opstamp: u64) -> bool {
+        self.get()
+            .map(|operation| operation.opstamp < target_opstamp)
+            .unwrap_or(false)
+    }
+
     /// If the current block has been entirely
     /// consumed, try to load the next one.
     ///

src/indexer/directory_lock.rs

@@ -1,26 +1,130 @@
-use core::LOCKFILE_FILEPATH;
 use directory::error::OpenWriteError;
+use std::io::Write;
+use std::path::{Path, PathBuf};
+use std::thread;
+use std::time::Duration;
 use Directory;
+use TantivyError;
 
-/// The directory lock is a mechanism used to
-/// prevent the creation of two [`IndexWriter`](struct.IndexWriter.html)
-///
-/// Only one lock can exist at a time for a given directory.
-/// The lock is release automatically on `Drop`.
-pub struct DirectoryLock {
-    directory: Box<Directory>,
+#[derive(Debug, Clone, Copy)]
+pub enum LockType {
+    /// Only one process should be able to write tantivy's index at a time.
+    /// This lock file, when present, is in charge of preventing other processes to open an IndexWriter.
+    ///
+    /// If the process is killed and this file remains, it is safe to remove it manually.
+    ///
+    /// Failing to acquire this lock usually means a misuse of tantivy's API,
+    /// (creating more than one instance of the `IndexWriter`), are a spurious
+    /// lock file remaining after a crash. In the latter case, removing the file after
+    /// checking no process running tantivy is running is safe.
+    IndexWriterLock,
+    /// The meta lock file is here to protect the segment files being opened by
+    /// `.load_searchers()` from being garbage collected.
+    /// It makes it possible for another process to safely consume
+    /// our index in-writing. Ideally, we may have prefered `RWLock` semantics
+    /// here, but it is difficult to achieve on Windows.
+    ///
+    /// Opening segment readers is a very fast process.
+    /// Right now if the lock cannot be acquire on the first attempt, the logic
+    /// is very simplistic. We retry after `100ms` until we effectively
+    /// acquire the lock.
+    /// This lock should not have much contention in normal usage.
+    MetaLock,
 }
 
-impl DirectoryLock {
-    pub fn lock(mut directory: Box<Directory>) -> Result<DirectoryLock, OpenWriteError> {
-        directory.open_write(&*LOCKFILE_FILEPATH)?;
-        Ok(DirectoryLock { directory })
+/// Retry the logic of acquiring locks is pretty simple.
+/// We just retry `n` times after a given `duratio`, both
+/// depending on the type of lock.
+struct RetryPolicy {
+    num_retries: usize,
+    wait_in_ms: u64,
+}
+
+impl RetryPolicy {
+    fn no_retry() -> RetryPolicy {
+        RetryPolicy {
+            num_retries: 0,
+            wait_in_ms: 0,
+        }
     }
+
+    fn wait_and_retry(&mut self) -> bool {
+        if self.num_retries == 0 {
+            false
+        } else {
+            self.num_retries -= 1;
+            let wait_duration = Duration::from_millis(self.wait_in_ms);
+            thread::sleep(wait_duration);
+            true
+        }
+    }
+}
+
+impl LockType {
+    fn retry_policy(self) -> RetryPolicy {
+        match self {
+            LockType::IndexWriterLock => RetryPolicy::no_retry(),
+            LockType::MetaLock => RetryPolicy {
+                num_retries: 100,
+                wait_in_ms: 100,
+            },
+        }
+    }
+
+    fn try_acquire_lock(self, directory: &mut Directory) -> Result<DirectoryLock, TantivyError> {
+        let path = self.filename();
+        let mut write = directory.open_write(path).map_err(|e| match e {
+            OpenWriteError::FileAlreadyExists(_) => TantivyError::LockFailure(self),
+            OpenWriteError::IOError(io_error) => TantivyError::IOError(io_error),
+        })?;
+        write.flush()?;
+        Ok(DirectoryLock {
+            directory: directory.box_clone(),
+            path: path.to_owned(),
+        })
+    }
+
+    /// Acquire a lock in the given directory.
+    pub fn acquire_lock(self, directory: &Directory) -> Result<DirectoryLock, TantivyError> {
+        let mut box_directory = directory.box_clone();
+        let mut retry_policy = self.retry_policy();
+        loop {
+            let lock_result = self.try_acquire_lock(&mut *box_directory);
+            match lock_result {
+                Ok(result) => {
+                    return Ok(result);
+                }
+                Err(TantivyError::LockFailure(ref filepath)) => {
+                    if !retry_policy.wait_and_retry() {
+                        return Err(TantivyError::LockFailure(filepath.to_owned()));
+                    }
+                }
+                Err(_) => {}
+            }
+        }
+    }
+
+    fn filename(&self) -> &Path {
+        match *self {
+            LockType::MetaLock => Path::new(".tantivy-meta.lock"),
+            LockType::IndexWriterLock => Path::new(".tantivy-indexer.lock"),
+        }
+    }
+}
+
+/// The `DirectoryLock` is an object that represents a file lock.
+/// See  [`LockType`](struct.LockType.html)
+///
+/// It is transparently associated to a lock file, that gets deleted
+/// on `Drop.` The lock is release automatically on `Drop`.
+pub struct DirectoryLock {
+    directory: Box<Directory>,
+    path: PathBuf,
 }
 
 impl Drop for DirectoryLock {
     fn drop(&mut self) {
-        if let Err(e) = self.directory.delete(&*LOCKFILE_FILEPATH) {
+        if let Err(e) = self.directory.delete(&*self.path) {
             error!("Failed to remove the lock file. {:?}", e);
         }
     }

src/indexer/index_writer.rs

@@ -8,7 +8,7 @@ use core::SegmentComponent;
 use core::SegmentId;
 use core::SegmentMeta;
 use core::SegmentReader;
-use crossbeam_channel as channel;
+use crossbeam::channel;
 use docset::DocSet;
 use error::TantivyError;
 use fastfield::write_delete_bitset;
@@ -54,14 +54,14 @@ type DocumentReceiver = channel::Receiver<AddOperation>;
 fn initial_table_size(per_thread_memory_budget: usize) -> usize {
     let table_size_limit: usize = per_thread_memory_budget / 3;
     (1..)
-        .into_iter()
         .take_while(|num_bits: &usize| compute_table_size(*num_bits) < table_size_limit)
         .last()
-        .expect(&format!(
-            "Per thread memory is too small: {}",
-            per_thread_memory_budget
-        ))
-        .min(19) // we cap it at 512K
+        .unwrap_or_else(|| {
+            panic!(
+                "Per thread memory is too small: {}",
+                per_thread_memory_budget
+            )
+        }).min(19) // we cap it at 512K
 }
 
 /// `IndexWriter` is the user entry-point to add document to an index.
@@ -177,7 +177,7 @@ pub fn compute_deleted_bitset(
 ) -> Result<bool> {
     let mut might_have_changed = false;
 
-    #[cfg_attr(feature = "cargo-clippy", allow(while_let_loop))]
+    #[cfg_attr(feature = "cargo-clippy", allow(clippy::while_let_loop))]
     loop {
         if let Some(delete_op) = delete_cursor.get() {
             if delete_op.opstamp > target_opstamp {
@@ -301,25 +301,29 @@ fn index_documents(
 
     let last_docstamp: u64 = *(doc_opstamps.last().unwrap());
 
-    let doc_to_opstamps = DocToOpstampMapping::from(doc_opstamps);
-    let segment_reader = SegmentReader::open(segment)?;
-    let mut deleted_bitset = BitSet::with_capacity(num_docs as usize);
-    let may_have_deletes = compute_deleted_bitset(
-        &mut deleted_bitset,
-        &segment_reader,
-        &mut delete_cursor,
-        &doc_to_opstamps,
-        last_docstamp,
-    )?;
-
-    let segment_entry = SegmentEntry::new(segment_meta, delete_cursor, {
-        if may_have_deletes {
-            Some(deleted_bitset)
-        } else {
-            None
-        }
-    });
-
+    let segment_entry: SegmentEntry = if delete_cursor.get().is_some() {
+        let doc_to_opstamps = DocToOpstampMapping::from(doc_opstamps);
+        let segment_reader = SegmentReader::open(segment)?;
+        let mut deleted_bitset = BitSet::with_capacity(num_docs as usize);
+        let may_have_deletes = compute_deleted_bitset(
+            &mut deleted_bitset,
+            &segment_reader,
+            &mut delete_cursor,
+            &doc_to_opstamps,
+            last_docstamp,
+        )?;
+        SegmentEntry::new(segment_meta, delete_cursor, {
+            if may_have_deletes {
+                Some(deleted_bitset)
+            } else {
+                None
+            }
+        })
+    } else {
+        // if there are no delete operation in the queue, no need
+        // to even open the segment.
+        SegmentEntry::new(segment_meta, delete_cursor, None)
+    };
     Ok(segment_updater.add_segment(generation, segment_entry))
 }
 
@@ -341,7 +345,8 @@ impl IndexWriter {
         }
         drop(self.workers_join_handle);
 
-        let result = self.segment_updater
+        let result = self
+            .segment_updater
             .wait_merging_thread()
             .map_err(|_| TantivyError::ErrorInThread("Failed to join merging thread.".into()));
 
@@ -383,10 +388,9 @@ impl IndexWriter {
         let mem_budget = self.heap_size_in_bytes_per_thread;
         let join_handle: JoinHandle<Result<()>> = thread::Builder::new()
             .name(format!(
-                "indexing thread {} for gen {}",
+                "thrd-tantivy-index{}-gen{}",
                 self.worker_id, generation
-            ))
-            .spawn(move || {
+            )).spawn(move || {
                 loop {
                     let mut document_iterator =
                         document_receiver_clone.clone().into_iter().peekable();
@@ -488,7 +492,8 @@ impl IndexWriter {
         let document_receiver = self.document_receiver.clone();
 
         // take the directory lock to create a new index_writer.
-        let directory_lock = self._directory_lock
+        let directory_lock = self
+            ._directory_lock
             .take()
             .expect("The IndexWriter does not have any lock. This is a bug, please report.");
 
@@ -636,7 +641,10 @@ impl IndexWriter {
     pub fn add_document(&mut self, document: Document) -> u64 {
         let opstamp = self.stamper.stamp();
         let add_operation = AddOperation { opstamp, document };
-        self.document_sender.send(add_operation);
+        let send_result = self.document_sender.send(add_operation);
+        if let Err(e) = send_result {
+            panic!("Failed to index document. Sending to indexing channel failed. This probably means all of the indexing threads have panicked. {:?}", e);
+        }
         opstamp
     }
 }
@@ -653,18 +661,33 @@ mod tests {
 
     #[test]
     fn test_lockfile_stops_duplicates() {
-        let schema_builder = schema::SchemaBuilder::default();
+        let schema_builder = schema::Schema::builder();
         let index = Index::create_in_ram(schema_builder.build());
         let _index_writer = index.writer(40_000_000).unwrap();
         match index.writer(40_000_000) {
-            Err(TantivyError::FileAlreadyExists(_)) => {}
+            Err(TantivyError::LockFailure(_)) => {}
             _ => panic!("Expected FileAlreadyExists error"),
         }
     }
 
+    #[test]
+    fn test_lockfile_already_exists_error_msg() {
+        let schema_builder = schema::Schema::builder();
+        let index = Index::create_in_ram(schema_builder.build());
+        let _index_writer = index.writer_with_num_threads(1, 3_000_000).unwrap();
+        match index.writer_with_num_threads(1, 3_000_000) {
+            Err(err) => {
+                let err_msg = err.to_string();
+                assert!(err_msg.contains("Lockfile"));
+                assert!(err_msg.contains("Possible causes:"))
+            }
+            _ => panic!("Expected LockfileAlreadyExists error"),
+        }
+    }
+
     #[test]
     fn test_set_merge_policy() {
-        let schema_builder = schema::SchemaBuilder::default();
+        let schema_builder = schema::Schema::builder();
         let index = Index::create_in_ram(schema_builder.build());
         let index_writer = index.writer(40_000_000).unwrap();
         assert_eq!(
@@ -682,7 +705,7 @@ mod tests {
 
     #[test]
     fn test_lockfile_released_on_drop() {
-        let schema_builder = schema::SchemaBuilder::default();
+        let schema_builder = schema::Schema::builder();
         let index = Index::create_in_ram(schema_builder.build());
         {
             let _index_writer = index.writer(40_000_000).unwrap();
@@ -694,7 +717,7 @@ mod tests {
 
     #[test]
     fn test_commit_and_rollback() {
-        let mut schema_builder = schema::SchemaBuilder::default();
+        let mut schema_builder = schema::Schema::builder();
         let text_field = schema_builder.add_text_field("text", schema::TEXT);
         let index = Index::create_in_ram(schema_builder.build());
 
@@ -728,7 +751,7 @@ mod tests {
 
     #[test]
     fn test_with_merges() {
-        let mut schema_builder = schema::SchemaBuilder::default();
+        let mut schema_builder = schema::Schema::builder();
         let text_field = schema_builder.add_text_field("text", schema::TEXT);
         let index = Index::create_in_ram(schema_builder.build());
         let num_docs_containing = |s: &str| {
@@ -765,7 +788,7 @@ mod tests {
 
     #[test]
     fn test_prepare_with_commit_message() {
-        let mut schema_builder = schema::SchemaBuilder::default();
+        let mut schema_builder = schema::Schema::builder();
         let text_field = schema_builder.add_text_field("text", schema::TEXT);
         let index = Index::create_in_ram(schema_builder.build());
 
@@ -799,7 +822,7 @@ mod tests {
 
     #[test]
     fn test_prepare_but_rollback() {
-        let mut schema_builder = schema::SchemaBuilder::default();
+        let mut schema_builder = schema::Schema::builder();
         let text_field = schema_builder.add_text_field("text", schema::TEXT);
         let index = Index::create_in_ram(schema_builder.build());
 
@@ -843,4 +866,32 @@ mod tests {
         assert_eq!(initial_table_size(1_000_000_000), 19);
     }
 
+    #[cfg(not(feature = "no_fail"))]
+    #[test]
+    fn test_write_commit_fails() {
+        use fail;
+        let mut schema_builder = schema::Schema::builder();
+        let text_field = schema_builder.add_text_field("text", schema::TEXT);
+        let index = Index::create_in_ram(schema_builder.build());
+
+        let mut index_writer = index.writer_with_num_threads(1, 3_000_000).unwrap();
+        for _ in 0..100 {
+            index_writer.add_document(doc!(text_field => "a"));
+        }
+        index_writer.commit().unwrap();
+        fail::cfg("RAMDirectory::atomic_write", "return(error_write_failed)").unwrap();
+        for _ in 0..100 {
+            index_writer.add_document(doc!(text_field => "b"));
+        }
+        assert!(index_writer.commit().is_err());
+        index.load_searchers().unwrap();
+        let num_docs_containing = |s: &str| {
+            let searcher = index.searcher();
+            let term_a = Term::from_field_text(text_field, s);
+            searcher.doc_freq(&term_a)
+        };
+        assert_eq!(num_docs_containing("a"), 100);
+        assert_eq!(num_docs_containing("b"), 0);
+        fail::cfg("RAMDirectory::atomic_write", "off").unwrap();
+    }
 }

src/indexer/merge_policy.rs

@@ -21,17 +21,17 @@ pub trait MergePolicy: MergePolicyClone + marker::Send + marker::Sync + Debug {
 
 /// MergePolicyClone
 pub trait MergePolicyClone {
-  /// Returns a boxed clone of the MergePolicy.
-  fn box_clone(&self) -> Box<MergePolicy>;
+    /// Returns a boxed clone of the MergePolicy.
+    fn box_clone(&self) -> Box<MergePolicy>;
 }
 
 impl<T> MergePolicyClone for T
 where
-  T: 'static + MergePolicy + Clone,
+    T: 'static + MergePolicy + Clone,
 {
-  fn box_clone(&self) -> Box<MergePolicy> {
-    Box::new(self.clone())
-  }
+    fn box_clone(&self) -> Box<MergePolicy> {
+        Box::new(self.clone())
+    }
 }
 
 /// Never merge segments.

src/indexer/merger.rs

@@ -40,15 +40,13 @@ fn compute_total_num_tokens(readers: &[SegmentReader], field: Field) -> u64 {
             total_tokens += reader.inverted_index(field).total_num_tokens();
         }
     }
-    total_tokens
-        + count
-            .iter()
-            .cloned()
-            .enumerate()
-            .map(|(fieldnorm_ord, count)| {
-                count as u64 * FieldNormReader::id_to_fieldnorm(fieldnorm_ord as u8) as u64
-            })
-            .sum::<u64>()
+    total_tokens + count
+        .iter()
+        .cloned()
+        .enumerate()
+        .map(|(fieldnorm_ord, count)| {
+            count as u64 * u64::from(FieldNormReader::id_to_fieldnorm(fieldnorm_ord as u8))
+        }).sum::<u64>()
 }
 
 pub struct IndexMerger {
@@ -111,7 +109,7 @@ impl TermOrdinalMapping {
             .iter()
             .flat_map(|term_ordinals| term_ordinals.iter().cloned().max())
             .max()
-            .unwrap_or(TermOrdinal::default())
+            .unwrap_or_else(TermOrdinal::default)
     }
 }
 
@@ -190,7 +188,7 @@ impl IndexMerger {
                         `term_ordinal_mapping`.");
                     self.write_hierarchical_facet_field(
                         field,
-                        term_ordinal_mapping,
+                        &term_ordinal_mapping,
                         fast_field_serializer,
                     )?;
                 }
@@ -314,7 +312,7 @@ impl IndexMerger {
     fn write_hierarchical_facet_field(
         &self,
         field: Field,
-        term_ordinal_mappings: TermOrdinalMapping,
+        term_ordinal_mappings: &TermOrdinalMapping,
         fast_field_serializer: &mut FastFieldSerializer,
     ) -> Result<()> {
         // Multifastfield consists in 2 fastfields.
@@ -393,8 +391,8 @@ impl IndexMerger {
 
         // We can now initialize our serializer, and push it the different values
         {
-            let mut serialize_vals =
-                fast_field_serializer.new_u64_fast_field_with_idx(field, min_value, max_value, 1)?;
+            let mut serialize_vals = fast_field_serializer
+                .new_u64_fast_field_with_idx(field, min_value, max_value, 1)?;
             for reader in &self.readers {
                 let ff_reader: MultiValueIntFastFieldReader<u64> =
                     reader.multi_fast_field_reader(field)?;
@@ -440,7 +438,8 @@ impl IndexMerger {
     ) -> Result<Option<TermOrdinalMapping>> {
         let mut positions_buffer: Vec<u32> = Vec::with_capacity(1_000);
         let mut delta_computer = DeltaComputer::new();
-        let field_readers = self.readers
+        let field_readers = self
+            .readers
             .iter()
             .map(|reader| reader.inverted_index(indexed_field))
             .collect::<Vec<_>>();
@@ -524,8 +523,7 @@ impl IndexMerger {
                         }
                     }
                     None
-                })
-                .collect();
+                }).collect();
 
             // At this point, `segment_postings` contains the posting list
             // of all of the segments containing the given term.
@@ -616,7 +614,7 @@ impl IndexMerger {
                     store_writer.store(&doc)?;
                 }
             } else {
-                store_writer.stack(store_reader)?;
+                store_writer.stack(&store_reader)?;
             }
         }
         Ok(())
@@ -637,10 +635,9 @@ impl SerializableSegment for IndexMerger {
 #[cfg(test)]
 mod tests {
     use byteorder::{BigEndian, ReadBytesExt, WriteBytesExt};
-    use collector::chain;
     use collector::tests::TestCollector;
     use collector::tests::{BytesFastFieldTestCollector, FastFieldTestCollector};
-    use collector::FacetCollector;
+    use collector::{Count, FacetCollector};
     use core::Index;
     use futures::Future;
     use query::AllQuery;
@@ -649,6 +646,7 @@ mod tests {
     use schema;
     use schema::Cardinality;
     use schema::Document;
+    use schema::Facet;
     use schema::IndexRecordOption;
     use schema::IntOptions;
     use schema::Term;
@@ -660,14 +658,13 @@ mod tests {
 
     #[test]
     fn test_index_merger_no_deletes() {
-        let mut schema_builder = schema::SchemaBuilder::default();
+        let mut schema_builder = schema::Schema::builder();
         let text_fieldtype = schema::TextOptions::default()
             .set_indexing_options(
                 TextFieldIndexing::default()
                     .set_tokenizer("default")
                     .set_index_option(IndexRecordOption::WithFreqs),
-            )
-            .set_stored();
+            ).set_stored();
         let text_field = schema_builder.add_text_field("text", text_fieldtype);
         let score_fieldtype = schema::IntOptions::default().set_fast(Cardinality::SingleValue);
         let score_field = schema_builder.add_u64_field("score", score_fieldtype);
@@ -745,63 +742,68 @@ mod tests {
             index.load_searchers().unwrap();
             let searcher = index.searcher();
             let get_doc_ids = |terms: Vec<Term>| {
-                let mut collector = TestCollector::default();
                 let query = BooleanQuery::new_multiterms_query(terms);
-                assert!(searcher.search(&query, &mut collector).is_ok());
-                collector.docs()
+                let top_docs = searcher.search(&query, &TestCollector).unwrap();
+                top_docs.docs().to_vec()
             };
             {
                 assert_eq!(
                     get_doc_ids(vec![Term::from_field_text(text_field, "a")]),
-                    vec![1, 2, 4]
+                    vec![DocAddress(0, 1), DocAddress(0, 2), DocAddress(0, 4)]
                 );
                 assert_eq!(
                     get_doc_ids(vec![Term::from_field_text(text_field, "af")]),
-                    vec![0, 3]
+                    vec![DocAddress(0, 0), DocAddress(0, 3)]
                 );
                 assert_eq!(
                     get_doc_ids(vec![Term::from_field_text(text_field, "g")]),
-                    vec![4]
+                    vec![DocAddress(0, 4)]
                 );
                 assert_eq!(
                     get_doc_ids(vec![Term::from_field_text(text_field, "b")]),
-                    vec![0, 1, 2, 3, 4]
+                    vec![
+                        DocAddress(0, 0),
+                        DocAddress(0, 1),
+                        DocAddress(0, 2),
+                        DocAddress(0, 3),
+                        DocAddress(0, 4)
+                    ]
                 );
             }
             {
-                let doc = searcher.doc(&DocAddress(0, 0)).unwrap();
-                assert_eq!(doc.get_first(text_field).unwrap().text(), "af b");
+                let doc = searcher.doc(DocAddress(0, 0)).unwrap();
+                assert_eq!(doc.get_first(text_field).unwrap().text(), Some("af b"));
             }
             {
-                let doc = searcher.doc(&DocAddress(0, 1)).unwrap();
-                assert_eq!(doc.get_first(text_field).unwrap().text(), "a b c");
+                let doc = searcher.doc(DocAddress(0, 1)).unwrap();
+                assert_eq!(doc.get_first(text_field).unwrap().text(), Some("a b c"));
             }
             {
-                let doc = searcher.doc(&DocAddress(0, 2)).unwrap();
-                assert_eq!(doc.get_first(text_field).unwrap().text(), "a b c d");
+                let doc = searcher.doc(DocAddress(0, 2)).unwrap();
+                assert_eq!(doc.get_first(text_field).unwrap().text(), Some("a b c d"));
             }
             {
-                let doc = searcher.doc(&DocAddress(0, 3)).unwrap();
-                assert_eq!(doc.get_first(text_field).unwrap().text(), "af b");
+                let doc = searcher.doc(DocAddress(0, 3)).unwrap();
+                assert_eq!(doc.get_first(text_field).unwrap().text(), Some("af b"));
             }
             {
-                let doc = searcher.doc(&DocAddress(0, 4)).unwrap();
-                assert_eq!(doc.get_first(text_field).unwrap().text(), "a b c g");
+                let doc = searcher.doc(DocAddress(0, 4)).unwrap();
+                assert_eq!(doc.get_first(text_field).unwrap().text(), Some("a b c g"));
             }
             {
                 let get_fast_vals = |terms: Vec<Term>| {
                     let query = BooleanQuery::new_multiterms_query(terms);
-                    let mut collector = FastFieldTestCollector::for_field(score_field);
-                    assert!(searcher.search(&query, &mut collector).is_ok());
-                    collector.vals()
+                    searcher
+                        .search(&query, &FastFieldTestCollector::for_field(score_field))
+                        .unwrap()
                 };
                 let get_fast_vals_bytes = |terms: Vec<Term>| {
                     let query = BooleanQuery::new_multiterms_query(terms);
-                    let mut collector = BytesFastFieldTestCollector::for_field(bytes_score_field);
                     searcher
-                        .search(&query, &mut collector)
-                        .expect("failed to search");
-                    collector.vals()
+                        .search(
+                            &query,
+                            &BytesFastFieldTestCollector::for_field(bytes_score_field),
+                        ).expect("failed to search")
                 };
                 assert_eq!(
                     get_fast_vals(vec![Term::from_field_text(text_field, "a")]),
@@ -817,12 +819,11 @@ mod tests {
 
     #[test]
     fn test_index_merger_with_deletes() {
-        let mut schema_builder = schema::SchemaBuilder::default();
+        let mut schema_builder = schema::Schema::builder();
         let text_fieldtype = schema::TextOptions::default()
             .set_indexing_options(
                 TextFieldIndexing::default().set_index_option(IndexRecordOption::WithFreqs),
-            )
-            .set_stored();
+            ).set_stored();
         let text_field = schema_builder.add_text_field("text", text_fieldtype);
         let score_fieldtype = schema::IntOptions::default().set_fast(Cardinality::SingleValue);
         let score_field = schema_builder.add_u64_field("score", score_fieldtype);
@@ -831,21 +832,13 @@ mod tests {
         let mut index_writer = index.writer_with_num_threads(1, 40_000_000).unwrap();
 
         let search_term = |searcher: &Searcher, term: Term| {
-            let mut collector = FastFieldTestCollector::for_field(score_field);
-            let mut bytes_collector = BytesFastFieldTestCollector::for_field(bytes_score_field);
+            let collector = FastFieldTestCollector::for_field(score_field);
+            let bytes_collector = BytesFastFieldTestCollector::for_field(bytes_score_field);
             let term_query = TermQuery::new(term, IndexRecordOption::Basic);
-
-            {
-                let mut combined_collector =
-                    chain().push(&mut collector).push(&mut bytes_collector);
-                searcher
-                    .search(&term_query, &mut combined_collector)
-                    .unwrap();
-            }
-
-            let scores = collector.vals();
-
-            let mut score_bytes = Cursor::new(bytes_collector.vals());
+            let (scores, bytes) = searcher
+                .search(&term_query, &(collector, bytes_collector))
+                .unwrap();
+            let mut score_bytes = Cursor::new(bytes);
             for &score in &scores {
                 assert_eq!(score as u32, score_bytes.read_u32::<BigEndian>().unwrap());
             }
@@ -926,10 +919,10 @@ mod tests {
 
             assert_eq!(searcher.segment_readers().len(), 2);
             assert_eq!(searcher.num_docs(), 3);
-            assert_eq!(searcher.segment_readers()[0].num_docs(), 1);
-            assert_eq!(searcher.segment_readers()[0].max_doc(), 3);
-            assert_eq!(searcher.segment_readers()[1].num_docs(), 2);
-            assert_eq!(searcher.segment_readers()[1].max_doc(), 4);
+            assert_eq!(searcher.segment_readers()[0].num_docs(), 2);
+            assert_eq!(searcher.segment_readers()[0].max_doc(), 4);
+            assert_eq!(searcher.segment_readers()[1].num_docs(), 1);
+            assert_eq!(searcher.segment_readers()[1].max_doc(), 3);
             assert_eq!(
                 search_term(&searcher, Term::from_field_text(text_field, "a")),
                 empty_vec
@@ -963,15 +956,15 @@ mod tests {
                 .segment_reader(0)
                 .fast_field_reader::<u64>(score_field)
                 .unwrap();
-            assert_eq!(score_field_reader.min_value(), 1);
-            assert_eq!(score_field_reader.max_value(), 3);
+            assert_eq!(score_field_reader.min_value(), 4000);
+            assert_eq!(score_field_reader.max_value(), 7000);
 
             let score_field_reader = searcher
                 .segment_reader(1)
                 .fast_field_reader::<u64>(score_field)
                 .unwrap();
-            assert_eq!(score_field_reader.min_value(), 4000);
-            assert_eq!(score_field_reader.max_value(), 7000);
+            assert_eq!(score_field_reader.min_value(), 1);
+            assert_eq!(score_field_reader.max_value(), 3);
         }
         {
             // merging the segments
@@ -1144,10 +1137,9 @@ mod tests {
 
     #[test]
     fn test_merge_facets() {
-        let mut schema_builder = schema::SchemaBuilder::default();
+        let mut schema_builder = schema::Schema::builder();
         let facet_field = schema_builder.add_facet_field("facet");
         let index = Index::create_in_ram(schema_builder.build());
-        use schema::Facet;
         {
             let mut index_writer = index.writer_with_num_threads(1, 40_000_000).unwrap();
             let index_doc = |index_writer: &mut IndexWriter, doc_facets: &[&str]| {
@@ -1176,20 +1168,16 @@ mod tests {
             index_doc(&mut index_writer, &["/top/e", "/top/f"]);
             index_writer.commit().expect("committed");
         }
+
         index.load_searchers().unwrap();
         let test_searcher = |expected_num_docs: usize, expected: &[(&str, u64)]| {
             let searcher = index.searcher();
             let mut facet_collector = FacetCollector::for_field(facet_field);
             facet_collector.add_facet(Facet::from("/top"));
-            use collector::{CountCollector, MultiCollector};
-            let mut count_collector = CountCollector::default();
-            {
-                let mut multi_collectors =
-                    MultiCollector::from(vec![&mut count_collector, &mut facet_collector]);
-                searcher.search(&AllQuery, &mut multi_collectors).unwrap();
-            }
-            assert_eq!(count_collector.count(), expected_num_docs);
-            let facet_counts = facet_collector.harvest();
+            let (count, facet_counts) = searcher
+                .search(&AllQuery, &(Count, facet_collector))
+                .unwrap();
+            assert_eq!(count, expected_num_docs);
             let facets: Vec<(String, u64)> = facet_counts
                 .get("/top")
                 .map(|(facet, count)| (facet.to_string(), count))
@@ -1213,7 +1201,6 @@ mod tests {
                 ("/top/f", 1),
             ],
         );
-
         // Merging the segments
         {
             let segment_ids = index
@@ -1226,7 +1213,6 @@ mod tests {
                 .wait()
                 .expect("Merging failed");
             index_writer.wait_merging_threads().unwrap();
-
             index.load_searchers().unwrap();
             test_searcher(
                 11,
@@ -1265,7 +1251,7 @@ mod tests {
 
     #[test]
     fn test_merge_multivalued_int_fields_all_deleted() {
-        let mut schema_builder = schema::SchemaBuilder::default();
+        let mut schema_builder = schema::Schema::builder();
         let int_options = IntOptions::default()
             .set_fast(Cardinality::MultiValues)
             .set_indexed();
@@ -1306,7 +1292,7 @@ mod tests {
 
     #[test]
     fn test_merge_multivalued_int_fields() {
-        let mut schema_builder = schema::SchemaBuilder::default();
+        let mut schema_builder = schema::Schema::builder();
         let int_options = IntOptions::default()
             .set_fast(Cardinality::MultiValues)
             .set_indexed();
@@ -1372,15 +1358,17 @@ mod tests {
             assert_eq!(&vals, &[17]);
         }
 
-        {
-            let segment = searcher.segment_reader(1u32);
-            let ff_reader = segment.multi_fast_field_reader(int_field).unwrap();
-            ff_reader.get_vals(0, &mut vals);
-            assert_eq!(&vals, &[20]);
-        }
+        println!(
+            "{:?}",
+            searcher
+                .segment_readers()
+                .iter()
+                .map(|reader| reader.max_doc())
+                .collect::<Vec<_>>()
+        );
 
         {
-            let segment = searcher.segment_reader(2u32);
+            let segment = searcher.segment_reader(1u32);
             let ff_reader = segment.multi_fast_field_reader(int_field).unwrap();
             ff_reader.get_vals(0, &mut vals);
             assert_eq!(&vals, &[28, 27]);
@@ -1389,6 +1377,13 @@ mod tests {
             assert_eq!(&vals, &[1_000]);
         }
 
+        {
+            let segment = searcher.segment_reader(2u32);
+            let ff_reader = segment.multi_fast_field_reader(int_field).unwrap();
+            ff_reader.get_vals(0, &mut vals);
+            assert_eq!(&vals, &[20]);
+        }
+
         // Merging the segments
         {
             let segment_ids = index
@@ -1407,6 +1402,14 @@ mod tests {
 
         {
             let searcher = index.searcher();
+            println!(
+                "{:?}",
+                searcher
+                    .segment_readers()
+                    .iter()
+                    .map(|reader| reader.max_doc())
+                    .collect::<Vec<_>>()
+            );
             let segment = searcher.segment_reader(0u32);
             let ff_reader = segment.multi_fast_field_reader(int_field).unwrap();
 
@@ -1432,13 +1435,13 @@ mod tests {
             assert_eq!(&vals, &[17]);
 
             ff_reader.get_vals(7, &mut vals);
-            assert_eq!(&vals, &[20]);
-
-            ff_reader.get_vals(8, &mut vals);
             assert_eq!(&vals, &[28, 27]);
 
-            ff_reader.get_vals(9, &mut vals);
+            ff_reader.get_vals(8, &mut vals);
             assert_eq!(&vals, &[1_000]);
+
+            ff_reader.get_vals(9, &mut vals);
+            assert_eq!(&vals, &[20]);
         }
     }
 }

src/indexer/mod.rs

@@ -16,6 +16,8 @@ mod segment_writer;
 mod stamper;
 
 pub(crate) use self::directory_lock::DirectoryLock;
+pub use self::directory_lock::LockType;
+
 pub use self::index_writer::IndexWriter;
 pub use self::log_merge_policy::LogMergePolicy;
 pub use self::merge_policy::{MergeCandidate, MergePolicy, NoMergePolicy};

src/indexer/segment_entry.rs

@@ -11,8 +11,8 @@ pub enum SegmentState {
 }
 
 impl SegmentState {
-    pub fn letter_code(&self) -> char {
-        match *self {
+    pub fn letter_code(self) -> char {
+        match self {
             SegmentState::InMerge => 'M',
             SegmentState::Ready => 'R',
         }

src/indexer/segment_manager.rs

@@ -1,7 +1,7 @@
 use super::segment_register::SegmentRegister;
 use core::SegmentId;
 use core::SegmentMeta;
-use core::{LOCKFILE_FILEPATH, META_FILEPATH};
+use core::META_FILEPATH;
 use error::TantivyError;
 use indexer::delete_queue::DeleteCursor;
 use indexer::SegmentEntry;
@@ -78,10 +78,13 @@ impl SegmentManager {
         registers_lock.committed.len() + registers_lock.uncommitted.len()
     }
 
+    /// List the files that are useful to the index.
+    ///
+    /// This does not include lock files, or files that are obsolete
+    /// but have not yet been deleted by the garbage collector.
     pub fn list_files(&self) -> HashSet<PathBuf> {
         let mut files = HashSet::new();
         files.insert(META_FILEPATH.clone());
-        files.insert(LOCKFILE_FILEPATH.clone());
         for segment_meta in SegmentMeta::all() {
             files.extend(segment_meta.list_files());
         }

src/indexer/segment_register.rs

@@ -51,7 +51,8 @@ impl SegmentRegister {
     }
 
     pub fn segment_metas(&self) -> Vec<SegmentMeta> {
-        let mut segment_ids: Vec<SegmentMeta> = self.segment_states
+        let mut segment_ids: Vec<SegmentMeta> = self
+            .segment_states
             .values()
             .map(|segment_entry| segment_entry.meta().clone())
             .collect();

src/indexer/segment_updater.rs

@@ -72,7 +72,7 @@ pub fn save_metas(
         payload,
     };
     let mut buffer = serde_json::to_vec_pretty(&metas)?;
-    write!(&mut buffer, "\n")?;
+    writeln!(&mut buffer)?;
     directory.atomic_write(&META_FILEPATH, &buffer[..])?;
     debug!("Saved metas {:?}", serde_json::to_string_pretty(&metas));
     Ok(())
@@ -227,8 +227,24 @@ impl SegmentUpdater {
         if self.is_alive() {
             let index = &self.0.index;
             let directory = index.directory();
+            let mut commited_segment_metas = self.0.segment_manager.committed_segment_metas();
+
+            // We sort segment_readers by number of documents.
+            // This is an heuristic to make multithreading more efficient.
+            //
+            // This is not done at the searcher level because I had a strange
+            // use case in which I was dealing with a large static index,
+            // dispatched over 5 SSD drives.
+            //
+            // A `UnionDirectory` makes it possible to read from these
+            // 5 different drives and creates a meta.json on the fly.
+            // In order to optimize the throughput, it creates a lasagna of segments
+            // from the different drives.
+            //
+            // Segment 1 from disk 1, Segment 1 from disk 2, etc.
+            commited_segment_metas.sort_by_key(|segment_meta| -(segment_meta.max_doc() as i32));
             save_metas(
-                self.0.segment_manager.committed_segment_metas(),
+                commited_segment_metas,
                 index.schema(),
                 opstamp,
                 commit_message,
@@ -336,8 +352,7 @@ impl SegmentUpdater {
                     .unwrap()
                     .remove(&merging_thread_id);
                 Ok(())
-            })
-            .expect("Failed to spawn a thread.");
+            }).expect("Failed to spawn a thread.");
         self.0
             .merging_threads
             .write()
@@ -485,7 +500,7 @@ mod tests {
 
     #[test]
     fn test_delete_during_merge() {
-        let mut schema_builder = SchemaBuilder::default();
+        let mut schema_builder = Schema::builder();
         let text_field = schema_builder.add_text_field("text", TEXT);
         let schema = schema_builder.build();
 

src/indexer/segment_writer.rs

@@ -49,20 +49,20 @@ impl SegmentWriter {
     ) -> Result<SegmentWriter> {
         let segment_serializer = SegmentSerializer::for_segment(&mut segment)?;
         let multifield_postings = MultiFieldPostingsWriter::new(schema, table_bits);
-        let tokenizers = schema
-            .fields()
-            .iter()
-            .map(|field_entry| field_entry.field_type())
-            .map(|field_type| match *field_type {
-                FieldType::Str(ref text_options) => text_options.get_indexing_options().and_then(
-                    |text_index_option| {
-                        let tokenizer_name = &text_index_option.tokenizer();
-                        segment.index().tokenizers().get(tokenizer_name)
-                    },
-                ),
-                _ => None,
-            })
-            .collect();
+        let tokenizers =
+            schema
+                .fields()
+                .iter()
+                .map(|field_entry| field_entry.field_type())
+                .map(|field_type| match *field_type {
+                    FieldType::Str(ref text_options) => text_options
+                        .get_indexing_options()
+                        .and_then(|text_index_option| {
+                            let tokenizer_name = &text_index_option.tokenizer();
+                            segment.index().tokenizers().get(tokenizer_name)
+                        }),
+                    _ => None,
+                }).collect();
         Ok(SegmentWriter {
             max_doc: 0,
             multifield_postings,
@@ -117,8 +117,7 @@ impl SegmentWriter {
                             _ => {
                                 panic!("Expected hierarchical facet");
                             }
-                        })
-                        .collect();
+                        }).collect();
                     let mut term = Term::for_field(field); // we set the Term
                     for facet_bytes in facets {
                         let mut unordered_term_id_opt = None;
@@ -146,8 +145,7 @@ impl SegmentWriter {
                             .flat_map(|field_value| match *field_value.value() {
                                 Value::Str(ref text) => Some(text.as_str()),
                                 _ => None,
-                            })
-                            .collect();
+                            }).collect();
                         if texts.is_empty() {
                             0
                         } else {

src/lib.rs

@@ -1,11 +1,8 @@
 #![doc(html_logo_url = "http://fulmicoton.com/tantivy-logo/tantivy-logo.png")]
-#![cfg_attr(feature = "cargo-clippy", allow(module_inception))]
-#![cfg_attr(feature = "cargo-clippy", allow(inline_always))]
 #![cfg_attr(all(feature = "unstable", test), feature(test))]
+#![cfg_attr(feature = "cargo-clippy", feature(tool_lints))]
+#![cfg_attr(feature = "cargo-clippy", allow(clippy::module_inception))]
 #![doc(test(attr(allow(unused_variables), deny(warnings))))]
-#![allow(unknown_lints)]
-#![allow(new_without_default)]
-#![allow(decimal_literal_representation)]
 #![warn(missing_docs)]
 #![recursion_limit = "80"]
 
@@ -27,7 +24,8 @@
 //! # use tempdir::TempDir;
 //! # use tantivy::Index;
 //! # use tantivy::schema::*;
-//! # use tantivy::collector::TopCollector;
+//! # use tantivy::{Score, DocAddress};
+//! # use tantivy::collector::TopDocs;
 //! # use tantivy::query::QueryParser;
 //! #
 //! # fn main() {
@@ -49,7 +47,7 @@
 //! // in a compressed, row-oriented key-value store.
 //! // This store is useful to reconstruct the
 //! // documents that were selected during the search phase.
-//! let mut schema_builder = SchemaBuilder::default();
+//! let mut schema_builder = Schema::builder();
 //! let title = schema_builder.add_text_field("title", TEXT | STORED);
 //! let body = schema_builder.add_text_field("body", TEXT);
 //! let schema = schema_builder.build();
@@ -89,14 +87,14 @@
 //! // A ticket has been opened regarding this problem.
 //! let query = query_parser.parse_query("sea whale")?;
 //!
-//! let mut top_collector = TopCollector::with_limit(10);
-//! searcher.search(&*query, &mut top_collector)?;
+//! // Perform search.
+//! // `topdocs` contains the 10 most relevant doc ids, sorted by decreasing scores...
+//! let top_docs: Vec<(Score, DocAddress)> =
+//!     searcher.search(&query, &TopDocs::with_limit(10))?;
 //!
-//! // Our top collector now contains the 10
-//! // most relevant doc ids...
-//! let doc_addresses = top_collector.docs();
-//! for doc_address in doc_addresses {
-//!     let retrieved_doc = searcher.doc(&doc_address)?;
+//! for (_score, doc_address) in top_docs {
+//!     // Retrieve the actual content of documents given its `doc_address`.
+//!     let retrieved_doc = searcher.doc(doc_address)?;
 //!     println!("{}", schema.to_json(&retrieved_doc));
 //! }
 //!
@@ -132,17 +130,17 @@ extern crate base64;
 extern crate bit_set;
 extern crate bitpacking;
 extern crate byteorder;
+extern crate scoped_pool;
 
-#[macro_use]
 extern crate combine;
 
 extern crate crossbeam;
-extern crate crossbeam_channel;
 extern crate fnv;
 extern crate fst;
 extern crate fst_regex;
 extern crate futures;
 extern crate futures_cpupool;
+extern crate htmlescape;
 extern crate itertools;
 extern crate levenshtein_automata;
 extern crate num_cpus;
@@ -165,14 +163,19 @@ extern crate winapi;
 #[cfg(test)]
 extern crate rand;
 
+#[cfg(test)]
+#[macro_use]
+extern crate maplit;
+
 #[cfg(all(test, feature = "unstable"))]
 extern crate test;
 
-extern crate tinysegmenter;
-
 #[macro_use]
 extern crate downcast;
 
+#[macro_use]
+extern crate fail;
+
 #[cfg(test)]
 mod functional_test;
 
@@ -181,7 +184,10 @@ mod macros;
 
 pub use error::TantivyError;
 
-#[deprecated(since="0.7.0", note="please use `tantivy::TantivyError` instead")]
+#[deprecated(
+    since = "0.7.0",
+    note = "please use `tantivy::TantivyError` instead"
+)]
 pub use error::TantivyError as Error;
 
 extern crate census;
@@ -206,9 +212,13 @@ pub(crate) mod positions;
 pub mod postings;
 pub mod query;
 pub mod schema;
+pub mod space_usage;
 pub mod store;
 pub mod termdict;
 
+mod snippet;
+pub use self::snippet::{SnippetGenerator, Snippet};
+
 mod docset;
 pub use self::docset::{DocSet, SkipResult};
 
@@ -261,12 +271,12 @@ impl DocAddress {
     /// The segment ordinal is an id identifying the segment
     /// hosting the document. It is only meaningful, in the context
     /// of a searcher.
-    pub fn segment_ord(&self) -> SegmentLocalId {
+    pub fn segment_ord(self) -> SegmentLocalId {
         self.0
     }
 
     /// Return the segment local `DocId`
-    pub fn doc(&self) -> DocId {
+    pub fn doc(self) -> DocId {
         self.1
     }
 }
@@ -290,9 +300,11 @@ mod tests {
     use docset::DocSet;
     use query::BooleanQuery;
     use rand::distributions::Bernoulli;
-    use rand::distributions::Range;
-    use rand::{Rng, SeedableRng, XorShiftRng};
+    use rand::distributions::Uniform;
+    use rand::rngs::StdRng;
+    use rand::{Rng, SeedableRng};
     use schema::*;
+    use DocAddress;
     use Index;
     use IndexWriter;
     use Postings;
@@ -311,16 +323,15 @@ mod tests {
     }
 
     pub fn generate_nonunique_unsorted(max_value: u32, n_elems: usize) -> Vec<u32> {
-        let seed: [u8; 16] = [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15];
-        XorShiftRng::from_seed(seed)
-            .sample_iter(&Range::new(0u32, max_value))
+        let seed: [u8; 32] = [1; 32];
+        StdRng::from_seed(seed)
+            .sample_iter(&Uniform::new(0u32, max_value))
             .take(n_elems)
             .collect::<Vec<u32>>()
     }
 
     pub fn sample_with_seed(n: u32, ratio: f64, seed_val: u8) -> Vec<u32> {
-        let seed: [u8; 16] = [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, seed_val];
-        XorShiftRng::from_seed(seed)
+        StdRng::from_seed([seed_val; 32])
             .sample_iter(&Bernoulli::new(ratio))
             .take(n as usize)
             .enumerate()
@@ -335,7 +346,7 @@ mod tests {
     #[test]
     #[cfg(feature = "mmap")]
     fn test_indexing() {
-        let mut schema_builder = SchemaBuilder::default();
+        let mut schema_builder = Schema::builder();
         let text_field = schema_builder.add_text_field("text", TEXT);
         let schema = schema_builder.build();
         let index = Index::create_from_tempdir(schema).unwrap();
@@ -360,7 +371,7 @@ mod tests {
 
     #[test]
     fn test_docfreq1() {
-        let mut schema_builder = SchemaBuilder::default();
+        let mut schema_builder = Schema::builder();
         let text_field = schema_builder.add_text_field("text", TEXT);
         let index = Index::create_in_ram(schema_builder.build());
         let mut index_writer = index.writer_with_num_threads(1, 40_000_000).unwrap();
@@ -400,7 +411,7 @@ mod tests {
 
     #[test]
     fn test_fieldnorm_no_docs_with_field() {
-        let mut schema_builder = SchemaBuilder::default();
+        let mut schema_builder = Schema::builder();
         let title_field = schema_builder.add_text_field("title", TEXT);
         let text_field = schema_builder.add_text_field("text", TEXT);
         let index = Index::create_in_ram(schema_builder.build());
@@ -429,7 +440,7 @@ mod tests {
 
     #[test]
     fn test_fieldnorm() {
-        let mut schema_builder = SchemaBuilder::default();
+        let mut schema_builder = Schema::builder();
         let text_field = schema_builder.add_text_field("text", TEXT);
         let index = Index::create_in_ram(schema_builder.build());
         {
@@ -470,7 +481,7 @@ mod tests {
 
     #[test]
     fn test_delete_postings1() {
-        let mut schema_builder = SchemaBuilder::default();
+        let mut schema_builder = Schema::builder();
         let text_field = schema_builder.add_text_field("text", TEXT);
         let term_abcd = Term::from_field_text(text_field, "abcd");
         let term_a = Term::from_field_text(text_field, "a");
@@ -481,42 +492,21 @@ mod tests {
         {
             // writing the segment
             let mut index_writer = index.writer_with_num_threads(1, 40_000_000).unwrap();
-            {
-                // 0
-                let doc = doc!(text_field=>"a b");
-                index_writer.add_document(doc);
-            }
-            {
-                // 1
-                let doc = doc!(text_field=>" a c");
-                index_writer.add_document(doc);
-            }
-            {
-                // 2
-                let doc = doc!(text_field=>" b c");
-                index_writer.add_document(doc);
-            }
-            {
-                // 3
-                let doc = doc!(text_field=>" b d");
-                index_writer.add_document(doc);
-            }
-            {
-                index_writer.delete_term(Term::from_field_text(text_field, "c"));
-            }
-            {
-                index_writer.delete_term(Term::from_field_text(text_field, "a"));
-            }
-            {
-                // 4
-                let doc = doc!(text_field=>" b c");
-                index_writer.add_document(doc);
-            }
-            {
-                // 5
-                let doc = doc!(text_field=>" a");
-                index_writer.add_document(doc);
-            }
+            // 0
+            index_writer.add_document(doc!(text_field=>"a b"));
+            // 1
+            index_writer.add_document(doc!(text_field=>" a c"));
+            // 2
+            index_writer.add_document(doc!(text_field=>" b c"));
+            // 3
+            index_writer.add_document(doc!(text_field=>" b d"));
+
+            index_writer.delete_term(Term::from_field_text(text_field, "c"));
+            index_writer.delete_term(Term::from_field_text(text_field, "a"));
+            // 4
+            index_writer.add_document(doc!(text_field=>" b c"));
+            // 5
+            index_writer.add_document(doc!(text_field=>" a"));
             index_writer.commit().unwrap();
         }
         {
@@ -551,15 +541,10 @@ mod tests {
         {
             // writing the segment
             let mut index_writer = index.writer_with_num_threads(1, 40_000_000).unwrap();
-            {
-                // 0
-                let doc = doc!(text_field=>"a b");
-                index_writer.add_document(doc);
-            }
-            {
-                // 1
-                index_writer.delete_term(Term::from_field_text(text_field, "c"));
-            }
+            // 0
+            index_writer.add_document(doc!(text_field=>"a b"));
+            // 1
+            index_writer.delete_term(Term::from_field_text(text_field, "c"));
             index_writer.rollback().unwrap();
         }
         {
@@ -595,13 +580,8 @@ mod tests {
         {
             // writing the segment
             let mut index_writer = index.writer_with_num_threads(1, 40_000_000).unwrap();
-            {
-                let doc = doc!(text_field=>"a b");
-                index_writer.add_document(doc);
-            }
-            {
-                index_writer.delete_term(Term::from_field_text(text_field, "c"));
-            }
+            index_writer.add_document(doc!(text_field=>"a b"));
+            index_writer.delete_term(Term::from_field_text(text_field, "c"));
             index_writer.rollback().unwrap();
             index_writer.delete_term(Term::from_field_text(text_field, "a"));
             index_writer.commit().unwrap();
@@ -645,7 +625,7 @@ mod tests {
 
     #[test]
     fn test_indexed_u64() {
-        let mut schema_builder = SchemaBuilder::default();
+        let mut schema_builder = Schema::builder();
         let field = schema_builder.add_u64_field("value", INT_INDEXED);
         let schema = schema_builder.build();
 
@@ -668,7 +648,7 @@ mod tests {
 
     #[test]
     fn test_indexed_i64() {
-        let mut schema_builder = SchemaBuilder::default();
+        let mut schema_builder = Schema::builder();
         let value_field = schema_builder.add_i64_field("value", INT_INDEXED);
         let schema = schema_builder.build();
 
@@ -692,7 +672,7 @@ mod tests {
 
     #[test]
     fn test_indexedfield_not_in_documents() {
-        let mut schema_builder = SchemaBuilder::default();
+        let mut schema_builder = Schema::builder();
         let text_field = schema_builder.add_text_field("text", TEXT);
         let absent_field = schema_builder.add_text_field("text", TEXT);
         let schema = schema_builder.build();
@@ -708,7 +688,7 @@ mod tests {
 
     #[test]
     fn test_delete_postings2() {
-        let mut schema_builder = SchemaBuilder::default();
+        let mut schema_builder = Schema::builder();
         let text_field = schema_builder.add_text_field("text", TEXT);
         let schema = schema_builder.build();
         let index = Index::create_in_ram(schema);
@@ -744,7 +724,7 @@ mod tests {
 
     #[test]
     fn test_termfreq() {
-        let mut schema_builder = SchemaBuilder::default();
+        let mut schema_builder = Schema::builder();
         let text_field = schema_builder.add_text_field("text", TEXT);
         let schema = schema_builder.build();
         let index = Index::create_in_ram(schema);
@@ -781,7 +761,7 @@ mod tests {
 
     #[test]
     fn test_searcher_1() {
-        let mut schema_builder = SchemaBuilder::default();
+        let mut schema_builder = Schema::builder();
         let text_field = schema_builder.add_text_field("text", TEXT);
         let schema = schema_builder.build();
         let index = Index::create_in_ram(schema);
@@ -789,18 +769,9 @@ mod tests {
         {
             // writing the segment
             let mut index_writer = index.writer_with_num_threads(1, 40_000_000).unwrap();
-            {
-                let doc = doc!(text_field=>"af af af b");
-                index_writer.add_document(doc);
-            }
-            {
-                let doc = doc!(text_field=>"a b c");
-                index_writer.add_document(doc);
-            }
-            {
-                let doc = doc!(text_field=>"a b c d");
-                index_writer.add_document(doc);
-            }
+            index_writer.add_document(doc!(text_field=>"af af af b"));
+            index_writer.add_document(doc!(text_field=>"a b c"));
+            index_writer.add_document(doc!(text_field=>"a b c d"));
             index_writer.commit().unwrap();
         }
         {
@@ -808,55 +779,42 @@ mod tests {
             let searcher = index.searcher();
             let get_doc_ids = |terms: Vec<Term>| {
                 let query = BooleanQuery::new_multiterms_query(terms);
-                let mut collector = TestCollector::default();
-                assert!(searcher.search(&query, &mut collector).is_ok());
-                collector.docs()
+                let topdocs = searcher.search(&query, &TestCollector).unwrap();
+                topdocs.docs().to_vec()
             };
-            {
-                assert_eq!(
-                    get_doc_ids(vec![Term::from_field_text(text_field, "a")]),
-                    vec![1, 2]
-                );
-            }
-            {
-                assert_eq!(
-                    get_doc_ids(vec![Term::from_field_text(text_field, "af")]),
-                    vec![0]
-                );
-            }
-            {
-                assert_eq!(
-                    get_doc_ids(vec![Term::from_field_text(text_field, "b")]),
-                    vec![0, 1, 2]
-                );
-            }
-            {
-                assert_eq!(
-                    get_doc_ids(vec![Term::from_field_text(text_field, "c")]),
-                    vec![1, 2]
-                );
-            }
-            {
-                assert_eq!(
-                    get_doc_ids(vec![Term::from_field_text(text_field, "d")]),
-                    vec![2]
-                );
-            }
-            {
-                assert_eq!(
-                    get_doc_ids(vec![
-                        Term::from_field_text(text_field, "b"),
-                        Term::from_field_text(text_field, "a"),
-                    ]),
-                    vec![0, 1, 2]
-                );
-            }
+            assert_eq!(
+                get_doc_ids(vec![Term::from_field_text(text_field, "a")]),
+                vec![DocAddress(0, 1), DocAddress(0, 2)]
+            );
+            assert_eq!(
+                get_doc_ids(vec![Term::from_field_text(text_field, "af")]),
+                vec![DocAddress(0, 0)]
+            );
+            assert_eq!(
+                get_doc_ids(vec![Term::from_field_text(text_field, "b")]),
+                vec![DocAddress(0, 0), DocAddress(0, 1), DocAddress(0, 2)]
+            );
+            assert_eq!(
+                get_doc_ids(vec![Term::from_field_text(text_field, "c")]),
+                vec![DocAddress(0, 1), DocAddress(0, 2)]
+            );
+            assert_eq!(
+                get_doc_ids(vec![Term::from_field_text(text_field, "d")]),
+                vec![DocAddress(0, 2)]
+            );
+            assert_eq!(
+                get_doc_ids(vec![
+                    Term::from_field_text(text_field, "b"),
+                    Term::from_field_text(text_field, "a"),
+                ]),
+                vec![DocAddress(0, 0), DocAddress(0, 1), DocAddress(0, 2)]
+            );
         }
     }
 
     #[test]
     fn test_searcher_2() {
-        let mut schema_builder = SchemaBuilder::default();
+        let mut schema_builder = Schema::builder();
         let text_field = schema_builder.add_text_field("text", TEXT);
         let schema = schema_builder.build();
         let index = Index::create_in_ram(schema);
@@ -883,7 +841,7 @@ mod tests {
 
     #[test]
     fn test_doc_macro() {
-        let mut schema_builder = SchemaBuilder::default();
+        let mut schema_builder = Schema::builder();
         let text_field = schema_builder.add_text_field("text", TEXT);
         let other_text_field = schema_builder.add_text_field("text2", TEXT);
         let document = doc!(text_field => "tantivy",
@@ -892,16 +850,16 @@ mod tests {
         assert_eq!(document.len(), 3);
         let values = document.get_all(text_field);
         assert_eq!(values.len(), 2);
-        assert_eq!(values[0].text(), "tantivy");
-        assert_eq!(values[1].text(), "some other value");
+        assert_eq!(values[0].text(), Some("tantivy"));
+        assert_eq!(values[1].text(), Some("some other value"));
         let values = document.get_all(other_text_field);
         assert_eq!(values.len(), 1);
-        assert_eq!(values[0].text(), "short");
+        assert_eq!(values[0].text(), Some("short"));
     }
 
     #[test]
     fn test_wrong_fast_field_type() {
-        let mut schema_builder = SchemaBuilder::default();
+        let mut schema_builder = Schema::builder();
         let fast_field_unsigned = schema_builder.add_u64_field("unsigned", FAST);
         let fast_field_signed = schema_builder.add_i64_field("signed", FAST);
         let text_field = schema_builder.add_text_field("text", TEXT);

src/macros.rs

@@ -26,12 +26,12 @@
 /// #[macro_use]
 /// extern crate tantivy;
 ///
-/// use tantivy::schema::{SchemaBuilder, TEXT, FAST};
+/// use tantivy::schema::{Schema, TEXT, FAST};
 ///
 /// //...
 ///
 /// # fn main() {
-/// let mut schema_builder = SchemaBuilder::new();
+/// let mut schema_builder = Schema::builder();
 /// let title = schema_builder.add_text_field("title", TEXT);
 /// let author = schema_builder.add_text_field("text", TEXT);
 /// let likes = schema_builder.add_u64_field("num_u64", FAST);
@@ -67,11 +67,11 @@ macro_rules! doc(
 
 #[cfg(test)]
 mod test {
-    use schema::{SchemaBuilder, FAST, TEXT};
+    use schema::{Schema, FAST, TEXT};
 
     #[test]
     fn test_doc_basic() {
-        let mut schema_builder = SchemaBuilder::new();
+        let mut schema_builder = Schema::builder();
         let title = schema_builder.add_text_field("title", TEXT);
         let author = schema_builder.add_text_field("text", TEXT);
         let likes = schema_builder.add_u64_field("num_u64", FAST);
@@ -85,7 +85,7 @@ mod test {
 
     #[test]
     fn test_doc_trailing_comma() {
-        let mut schema_builder = SchemaBuilder::new();
+        let mut schema_builder = Schema::builder();
         let title = schema_builder.add_text_field("title", TEXT);
         let author = schema_builder.add_text_field("text", TEXT);
         let likes = schema_builder.add_u64_field("num_u64", FAST);

src/positions/mod.rs

@@ -1,4 +1,3 @@
-
 /// Positions are stored in three parts and over two files.
 //
 /// The `SegmentComponent::POSITIONS` file contains all of the bitpacked positions delta,
@@ -24,13 +23,12 @@
 /// The long skip structure makes it possible to skip rapidly to the a checkpoint close to this
 /// value, and then skip normally.
 ///
-
 mod reader;
 mod serializer;
 
 pub use self::reader::PositionReader;
 pub use self::serializer::PositionSerializer;
-use bitpacking::{BitPacker4x, BitPacker};
+use bitpacking::{BitPacker, BitPacker4x};
 
 const COMPRESSION_BLOCK_SIZE: usize = BitPacker4x::BLOCK_LEN;
 const LONG_SKIP_IN_BLOCKS: usize = 1_024;
@@ -43,10 +41,10 @@ lazy_static! {
 #[cfg(test)]
 pub mod tests {
 
-    use std::iter;
-    use super::{PositionSerializer, PositionReader};
+    use super::{PositionReader, PositionSerializer};
     use directory::ReadOnlySource;
     use positions::COMPRESSION_BLOCK_SIZE;
+    use std::iter;
 
     fn create_stream_buffer(vals: &[u32]) -> (ReadOnlySource, ReadOnlySource) {
         let mut skip_buffer = vec![];
@@ -59,7 +57,10 @@ pub mod tests {
             }
             serializer.close().unwrap();
         }
-        (ReadOnlySource::from(stream_buffer), ReadOnlySource::from(skip_buffer))
+        (
+            ReadOnlySource::from(stream_buffer),
+            ReadOnlySource::from(skip_buffer),
+        )
     }
 
     #[test]
@@ -103,7 +104,7 @@ pub mod tests {
         assert_eq!(skip.len(), 12);
         assert_eq!(stream.len(), 1168);
 
-        let mut position_reader = PositionReader::new(stream,skip, 0u64);
+        let mut position_reader = PositionReader::new(stream, skip, 0u64);
         let mut buf = [0u32; 7];
         let mut c = 0;
         for _ in 0..100 {
@@ -125,7 +126,7 @@ pub mod tests {
         let (stream, skip) = create_stream_buffer(&v[..]);
         assert_eq!(skip.len(), 15_749);
         assert_eq!(stream.len(), 1_000_000);
-        let mut position_reader = PositionReader::new(stream,skip, 128 * 1024);
+        let mut position_reader = PositionReader::new(stream, skip, 128 * 1024);
         let mut buf = [0u32; 1];
         position_reader.read(&mut buf);
         assert_eq!(buf[0], CONST_VAL);
@@ -137,12 +138,17 @@ pub mod tests {
         let (stream, skip) = create_stream_buffer(&v[..]);
         assert_eq!(skip.len(), 15_749);
         assert_eq!(stream.len(), 4_987_872);
-        for &offset in &[10, 128 * 1024, 128 * 1024 - 1, 128 * 1024 + 7, 128 * 10 * 1024 + 10] {
-            let mut position_reader = PositionReader::new(stream.clone(),skip.clone(), offset);
+        for &offset in &[
+            10,
+            128 * 1024,
+            128 * 1024 - 1,
+            128 * 1024 + 7,
+            128 * 10 * 1024 + 10,
+        ] {
+            let mut position_reader = PositionReader::new(stream.clone(), skip.clone(), offset);
             let mut buf = [0u32; 1];
             position_reader.read(&mut buf);
             assert_eq!(buf[0], offset as u32);
         }
     }
 }
-

src/positions/reader.rs

@@ -1,12 +1,12 @@
-use bitpacking::{BitPacker4x, BitPacker};
-use owned_read::OwnedRead;
-use common::{BinarySerializable, FixedSize};
-use postings::compression::compressed_block_size;
-use directory::ReadOnlySource;
-use positions::COMPRESSION_BLOCK_SIZE;
-use positions::LONG_SKIP_IN_BLOCKS;
-use positions::LONG_SKIP_INTERVAL;
 use super::BIT_PACKER;
+use bitpacking::{BitPacker, BitPacker4x};
+use common::{BinarySerializable, FixedSize};
+use directory::ReadOnlySource;
+use owned_read::OwnedRead;
+use positions::COMPRESSION_BLOCK_SIZE;
+use positions::LONG_SKIP_INTERVAL;
+use positions::LONG_SKIP_IN_BLOCKS;
+use postings::compression::compressed_block_size;
 
 pub struct PositionReader {
     skip_read: OwnedRead,
@@ -18,7 +18,6 @@ pub struct PositionReader {
                           // of the block of the next int to read.
 }
 
-
 // `ahead` represents the offset of the block currently loaded
 // compared to the cursor of the actual stream.
 //
@@ -32,7 +31,8 @@ fn read_impl(
     buffer: &mut [u32; 128],
     mut inner_offset: usize,
     num_bits: &[u8],
-    output: &mut [u32]) -> usize {
+    output: &mut [u32],
+) -> usize {
     let mut output_start = 0;
     let mut output_len = output.len();
     let mut ahead = 0;
@@ -47,8 +47,7 @@ fn read_impl(
             output_start += available_len;
             inner_offset = 0;
             let num_bits = num_bits[ahead];
-            BitPacker4x::new()
-                .decompress(position, &mut buffer[..], num_bits);
+            BitPacker4x::new().decompress(position, &mut buffer[..], num_bits);
             let block_len = compressed_block_size(num_bits);
             position = &position[block_len..];
             ahead += 1;
@@ -56,11 +55,12 @@ fn read_impl(
     }
 }
 
-
 impl PositionReader {
-    pub fn new(position_source: ReadOnlySource,
-               skip_source: ReadOnlySource,
-               offset: u64) -> PositionReader {
+    pub fn new(
+        position_source: ReadOnlySource,
+        skip_source: ReadOnlySource,
+        offset: u64,
+    ) -> PositionReader {
         let skip_len = skip_source.len();
         let (body, footer) = skip_source.split(skip_len - u32::SIZE_IN_BYTES);
         let num_long_skips = u32::deserialize(&mut footer.as_slice()).expect("Index corrupted");
@@ -70,7 +70,8 @@ impl PositionReader {
         let small_skip = (offset - (long_skip_id as u64) * (LONG_SKIP_INTERVAL as u64)) as usize;
         let offset_num_bytes: u64 = {
             if long_skip_id > 0 {
-                let mut long_skip_blocks: &[u8] = &long_skips.as_slice()[(long_skip_id - 1) * 8..][..8];
+                let mut long_skip_blocks: &[u8] =
+                    &long_skips.as_slice()[(long_skip_id - 1) * 8..][..8];
                 u64::deserialize(&mut long_skip_blocks).expect("Index corrupted") * 16
             } else {
                 0
@@ -79,13 +80,13 @@ impl PositionReader {
         let mut position_read = OwnedRead::new(position_source);
         position_read.advance(offset_num_bytes as usize);
         let mut skip_read = OwnedRead::new(skip_body);
-        skip_read.advance(long_skip_id  * LONG_SKIP_IN_BLOCKS);
+        skip_read.advance(long_skip_id * LONG_SKIP_IN_BLOCKS);
         let mut position_reader = PositionReader {
             skip_read,
             position_read,
             inner_offset: 0,
             buffer: Box::new([0u32; 128]),
-            ahead: None
+            ahead: None,
         };
         position_reader.skip(small_skip);
         position_reader
@@ -108,7 +109,8 @@ impl PositionReader {
             self.buffer.as_mut(),
             self.inner_offset,
             &skip_data[1..],
-            output));
+            output,
+        ));
     }
 
     /// Skip the next `skip_len` integer.
@@ -118,27 +120,25 @@ impl PositionReader {
     ///
     /// May panic if the end of the stream is reached.
     pub fn skip(&mut self, skip_len: usize) {
-
         let skip_len_plus_inner_offset = skip_len + self.inner_offset;
 
         let num_blocks_to_advance = skip_len_plus_inner_offset / COMPRESSION_BLOCK_SIZE;
         self.inner_offset = skip_len_plus_inner_offset % COMPRESSION_BLOCK_SIZE;
 
-        self.ahead = self.ahead
-            .and_then(|num_blocks| {
-                if num_blocks >= num_blocks_to_advance {
-                    Some(num_blocks_to_advance - num_blocks_to_advance)
-                } else {
-                    None
-                }
-            });
+        self.ahead = self.ahead.and_then(|num_blocks| {
+            if num_blocks >= num_blocks_to_advance {
+                Some(num_blocks - num_blocks_to_advance)
+            } else {
+                None
+            }
+        });
 
-        let skip_len = self.skip_read
-            .as_ref()[..num_blocks_to_advance]
+        let skip_len = self.skip_read.as_ref()[..num_blocks_to_advance]
             .iter()
             .cloned()
             .map(|num_bit| num_bit as usize)
-            .sum::<usize>() * (COMPRESSION_BLOCK_SIZE / 8);
+            .sum::<usize>()
+            * (COMPRESSION_BLOCK_SIZE / 8);
 
         self.skip_read.advance(num_blocks_to_advance);
         self.position_read.advance(skip_len);

src/positions/serializer.rs

@@ -1,8 +1,8 @@
-use std::io;
-use bitpacking::BitPacker;
-use positions::{COMPRESSION_BLOCK_SIZE, LONG_SKIP_INTERVAL};
-use common::BinarySerializable;
 use super::BIT_PACKER;
+use bitpacking::BitPacker;
+use common::BinarySerializable;
+use positions::{COMPRESSION_BLOCK_SIZE, LONG_SKIP_INTERVAL};
+use std::io;
 
 pub struct PositionSerializer<W: io::Write> {
     write_stream: W,
@@ -23,7 +23,7 @@ impl<W: io::Write> PositionSerializer<W> {
             buffer: vec![0u8; 128 * 4],
             num_ints: 0u64,
             long_skips: Vec::new(),
-            cumulated_num_bits: 0u64
+            cumulated_num_bits: 0u64,
         }
     }
 
@@ -31,7 +31,6 @@ impl<W: io::Write> PositionSerializer<W> {
         self.num_ints
     }
 
-
     fn remaining_block_len(&self) -> usize {
         COMPRESSION_BLOCK_SIZE - self.block.len()
     }
@@ -52,8 +51,8 @@ impl<W: io::Write> PositionSerializer<W> {
 
     fn flush_block(&mut self) -> io::Result<()> {
         let num_bits = BIT_PACKER.num_bits(&self.block[..]);
-        self.cumulated_num_bits += num_bits as u64;
-        self.write_skiplist.write(&[num_bits])?;
+        self.cumulated_num_bits += u64::from(num_bits);
+        self.write_skiplist.write_all(&[num_bits])?;
         let written_len = BIT_PACKER.compress(&self.block[..], &mut self.buffer, num_bits);
         self.write_stream.write_all(&self.buffer[..written_len])?;
         self.block.clear();

src/postings/compression/mod.rs

@@ -28,14 +28,16 @@ impl BlockEncoder {
 
     pub fn compress_block_sorted(&mut self, block: &[u32], offset: u32) -> (u8, &[u8]) {
         let num_bits = self.bitpacker.num_bits_sorted(offset, block);
-        let written_size = self.bitpacker
-            .compress_sorted(offset, block, &mut self.output[..], num_bits);
+        let written_size =
+            self.bitpacker
+                .compress_sorted(offset, block, &mut self.output[..], num_bits);
         (num_bits, &self.output[..written_size])
     }
 
     pub fn compress_block_unsorted(&mut self, block: &[u32]) -> (u8, &[u8]) {
         let num_bits = self.bitpacker.num_bits(block);
-        let written_size = self.bitpacker
+        let written_size = self
+            .bitpacker
             .compress(block, &mut self.output[..], num_bits);
         (num_bits, &self.output[..written_size])
     }
@@ -62,19 +64,21 @@ impl BlockDecoder {
         }
     }
 
-    pub fn uncompress_block_sorted(&mut self, compressed_data: &[u8], offset: u32, num_bits: u8) -> usize {
+    pub fn uncompress_block_sorted(
+        &mut self,
+        compressed_data: &[u8],
+        offset: u32,
+        num_bits: u8,
+    ) -> usize {
         self.output_len = COMPRESSION_BLOCK_SIZE;
-        self.bitpacker.decompress_sorted(
-            offset,
-            &compressed_data,
-            &mut self.output,
-            num_bits,
-        )
+        self.bitpacker
+            .decompress_sorted(offset, &compressed_data, &mut self.output, num_bits)
     }
 
     pub fn uncompress_block_unsorted(&mut self, compressed_data: &[u8], num_bits: u8) -> usize {
         self.output_len = COMPRESSION_BLOCK_SIZE;
-        self.bitpacker.decompress(&compressed_data, &mut self.output, num_bits)
+        self.bitpacker
+            .decompress(&compressed_data, &mut self.output, num_bits)
     }
 
     #[inline]
@@ -88,7 +92,6 @@ impl BlockDecoder {
     }
 }
 
-
 pub trait VIntEncoder {
     /// Compresses an array of `u32` integers,
     /// using [delta-encoding](https://en.wikipedia.org/wiki/Delta_ encoding)
@@ -263,21 +266,17 @@ pub mod tests {
 mod bench {
 
     use super::*;
-    use rand::Rng;
     use rand::SeedableRng;
-    use rand::XorShiftRng;
+    use rand::{Rng, XorShiftRng};
     use test::Bencher;
 
-    fn generate_array_with_seed(n: usize, ratio: f32, seed_val: u32) -> Vec<u32> {
-        let seed: &[u32; 4] = &[1, 2, 3, seed_val];
+    fn generate_array_with_seed(n: usize, ratio: f64, seed_val: u8) -> Vec<u32> {
+        let seed: &[u8; 16] = &[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, seed_val];
         let mut rng: XorShiftRng = XorShiftRng::from_seed(*seed);
-        (0..u32::max_value())
-            .filter(|_| rng.next_f32() < ratio)
-            .take(n)
-            .collect()
+        (0u32..).filter(|_| rng.gen_bool(ratio)).take(n).collect()
     }
 
-    pub fn generate_array(n: usize, ratio: f32) -> Vec<u32> {
+    pub fn generate_array(n: usize, ratio: f64) -> Vec<u32> {
         generate_array_with_seed(n, ratio, 4)
     }
 
@@ -294,24 +293,23 @@ mod bench {
     fn bench_uncompress(b: &mut Bencher) {
         let mut encoder = BlockEncoder::new();
         let data = generate_array(COMPRESSION_BLOCK_SIZE, 0.1);
-        let (_, compressed) = encoder.compress_block_sorted(&data, 0u32);
+        let (num_bits, compressed) = encoder.compress_block_sorted(&data, 0u32);
         let mut decoder = BlockDecoder::new();
         b.iter(|| {
-            decoder.uncompress_block_sorted(compressed, 0u32);
+            decoder.uncompress_block_sorted(compressed, 0u32, num_bits);
         });
     }
 
     #[test]
     fn test_all_docs_compression_numbits() {
-        for num_bits in 0..33 {
+        for expected_num_bits in 0u8.. {
             let mut data = [0u32; 128];
-            if num_bits > 0 {
-                data[0] = 1 << (num_bits - 1);
+            if expected_num_bits > 0 {
+                data[0] = (1u64 << (expected_num_bits as usize) - 1) as u32;
             }
             let mut encoder = BlockEncoder::new();
-            let compressed = encoder.compress_block_unsorted(&data);
-            assert_eq!(compressed[0] as usize, num_bits);
-            assert_eq!(compressed.len(), compressed_block_size(compressed[0]));
+            let (num_bits, compressed) = encoder.compress_block_unsorted(&data);
+            assert_eq!(compressed.len(), compressed_block_size(num_bits));
         }
     }
 

src/postings/compression/vint.rs

@@ -1,9 +1,5 @@
 #[inline(always)]
-pub fn compress_sorted<'a>(
-    input: &[u32],
-    output: &'a mut [u8],
-    mut offset: u32,
-) -> &'a [u8] {
+pub fn compress_sorted<'a>(input: &[u32], output: &'a mut [u8], mut offset: u32) -> &'a [u8] {
     let mut byte_written = 0;
     for &v in input {
         let mut to_encode: u32 = v - offset;
@@ -46,47 +42,41 @@ pub(crate) fn compress_unsorted<'a>(input: &[u32], output: &'a mut [u8]) -> &'a
 }
 
 #[inline(always)]
-pub fn uncompress_sorted<'a>(
-    compressed_data: &'a [u8],
-    output: &mut [u32],
-    offset: u32,
-) -> usize {
+pub fn uncompress_sorted<'a>(compressed_data: &'a [u8], output: &mut [u32], offset: u32) -> usize {
     let mut read_byte = 0;
     let mut result = offset;
-    let num_els = output.len();
-    for i in 0..num_els {
+    for output_mut in output.iter_mut() {
         let mut shift = 0u32;
         loop {
             let cur_byte = compressed_data[read_byte];
             read_byte += 1;
-            result += ((cur_byte % 128u8) as u32) << shift;
+            result += u32::from(cur_byte % 128u8) << shift;
             if cur_byte & 128u8 != 0u8 {
                 break;
             }
             shift += 7;
         }
-        output[i] = result;
+        *output_mut = result;
     }
     read_byte
 }
 
 #[inline(always)]
-pub(crate) fn uncompress_unsorted<'a>(compressed_data: &'a [u8], output: &mut [u32]) -> usize {
+pub(crate) fn uncompress_unsorted(compressed_data: &[u8], output_arr: &mut [u32]) -> usize {
     let mut read_byte = 0;
-    let num_els = output.len();
-    for i in 0..num_els {
+    for output_mut in output_arr.iter_mut() {
         let mut result = 0u32;
         let mut shift = 0u32;
         loop {
             let cur_byte = compressed_data[read_byte];
             read_byte += 1;
-            result += ((cur_byte % 128u8) as u32) << shift;
+            result += u32::from(cur_byte % 128u8) << shift;
             if cur_byte & 128u8 != 0u8 {
                 break;
             }
             shift += 7;
         }
-        output[i] = result;
+        *output_mut = result;
     }
     read_byte
 }

src/postings/mod.rs

@@ -2,6 +2,7 @@
 Postings module (also called inverted index)
 */
 
+pub(crate) mod compression;
 /// Postings module
 ///
 /// Postings, also called inverted lists, is the key datastructure
@@ -11,18 +12,17 @@ mod postings_writer;
 mod recorder;
 mod segment_postings;
 mod serializer;
-pub(crate) mod compression;
+mod skip;
 mod stacker;
 mod term_info;
-mod skip;
 
 pub(crate) use self::postings_writer::MultiFieldPostingsWriter;
 pub use self::serializer::{FieldSerializer, InvertedIndexSerializer};
 
+use self::compression::COMPRESSION_BLOCK_SIZE;
 pub use self::postings::Postings;
-pub use self::term_info::TermInfo;
 pub(crate) use self::skip::SkipReader;
-use self::compression::{COMPRESSION_BLOCK_SIZE};
+pub use self::term_info::TermInfo;
 
 pub use self::segment_postings::{BlockSegmentPostings, SegmentPostings};
 
@@ -34,7 +34,7 @@ pub(crate) const USE_SKIP_INFO_LIMIT: u32 = COMPRESSION_BLOCK_SIZE as u32;
 
 pub(crate) type UnorderedTermId = u64;
 
-#[allow(enum_variant_names)]
+#[cfg_attr(feature = "cargo-clippy", allow(clippy::enum_variant_names))]
 #[derive(Debug, PartialEq, Clone, Copy, Eq)]
 pub(crate) enum FreqReadingOption {
     NoFreq,
@@ -54,25 +54,25 @@ pub mod tests {
     use indexer::operation::AddOperation;
     use indexer::SegmentWriter;
     use query::Scorer;
-    use rand::{Rng, SeedableRng, XorShiftRng};
+    use rand::rngs::StdRng;
+    use rand::{Rng, SeedableRng};
     use schema::Field;
     use schema::IndexRecordOption;
-    use schema::{Document, SchemaBuilder, Term, INT_INDEXED, STRING, TEXT};
+    use schema::{Document, Schema, Term, INT_INDEXED, STRING, TEXT};
     use std::iter;
     use DocId;
     use Score;
 
     #[test]
     pub fn test_position_write() {
-        let mut schema_builder = SchemaBuilder::default();
+        let mut schema_builder = Schema::builder();
         let text_field = schema_builder.add_text_field("text", TEXT);
         let schema = schema_builder.build();
         let index = Index::create_in_ram(schema);
         let mut segment = index.new_segment();
         let mut posting_serializer = InvertedIndexSerializer::open(&mut segment).unwrap();
         {
-            let mut field_serializer = posting_serializer
-                .new_field(text_field, 120 * 4).unwrap();
+            let mut field_serializer = posting_serializer.new_field(text_field, 120 * 4).unwrap();
             field_serializer.new_term("abc".as_bytes()).unwrap();
             for doc_id in 0u32..120u32 {
                 let delta_positions = vec![1, 2, 3, 2];
@@ -89,7 +89,7 @@ pub mod tests {
 
     #[test]
     pub fn test_skip_positions() {
-        let mut schema_builder = SchemaBuilder::new();
+        let mut schema_builder = Schema::builder();
         let title = schema_builder.add_text_field("title", TEXT);
         let schema = schema_builder.build();
         let index = Index::create_in_ram(schema);
@@ -164,7 +164,7 @@ pub mod tests {
     #[test]
     pub fn test_position_and_fieldnorm1() {
         let mut positions = Vec::new();
-        let mut schema_builder = SchemaBuilder::default();
+        let mut schema_builder = Schema::builder();
         let text_field = schema_builder.add_text_field("text", TEXT);
         let schema = schema_builder.build();
         let index = Index::create_in_ram(schema.clone());
@@ -277,7 +277,7 @@ pub mod tests {
     #[test]
     pub fn test_position_and_fieldnorm2() {
         let mut positions: Vec<u32> = Vec::new();
-        let mut schema_builder = SchemaBuilder::default();
+        let mut schema_builder = Schema::builder();
         let text_field = schema_builder.add_text_field("text", TEXT);
         let schema = schema_builder.build();
         let index = Index::create_in_ram(schema);
@@ -318,7 +318,7 @@ pub mod tests {
         let num_docs = 300u32;
 
         let index = {
-            let mut schema_builder = SchemaBuilder::default();
+            let mut schema_builder = Schema::builder();
             let value_field = schema_builder.add_u64_field("value", INT_INDEXED);
             let schema = schema_builder.build();
 
@@ -499,12 +499,11 @@ pub mod tests {
             Term::from_field_text(field, "d")
         };
         pub static ref INDEX: Index = {
-            let mut schema_builder = SchemaBuilder::default();
+            let mut schema_builder = Schema::builder();
             let text_field = schema_builder.add_text_field("text", STRING);
             let schema = schema_builder.build();
 
-            let seed: [u8; 16] = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16];
-            let mut rng: XorShiftRng = XorShiftRng::from_seed(seed);
+            let mut rng: StdRng = StdRng::from_seed([1u8; 32]);
 
             let index = Index::create_in_ram(schema);
             let posting_list_size = 1_000_000;
@@ -512,13 +511,13 @@ pub mod tests {
                 let mut index_writer = index.writer_with_num_threads(1, 40_000_000).unwrap();
                 for _ in 0..posting_list_size {
                     let mut doc = Document::default();
-                    if rng.gen_bool(1f64/ 15f64) {
+                    if rng.gen_bool(1f64 / 15f64) {
                         doc.add_text(text_field, "a");
                     }
-                    if rng.gen_bool(1f64/ 10f64) {
+                    if rng.gen_bool(1f64 / 10f64) {
                         doc.add_text(text_field, "b");
                     }
-                    if rng.gen_bool(1f64/ 5f64) {
+                    if rng.gen_bool(1f64 / 5f64) {
                         doc.add_text(text_field, "c");
                     }
                     doc.add_text(text_field, "d");
@@ -655,7 +654,7 @@ mod bench {
         });
     }
 
-    fn bench_skip_next(p: f32, b: &mut Bencher) {
+    fn bench_skip_next(p: f64, b: &mut Bencher) {
         let searcher = INDEX.searcher();
         let segment_reader = searcher.segment_reader(0);
         let docs = tests::sample(segment_reader.num_docs(), p);

src/postings/postings_writer.rs

@@ -15,7 +15,7 @@ use tokenizer::TokenStream;
 use DocId;
 use Result;
 
-fn posting_from_field_entry<'a>(field_entry: &FieldEntry) -> Box<PostingsWriter> {
+fn posting_from_field_entry(field_entry: &FieldEntry) -> Box<PostingsWriter> {
     match *field_entry.field_type() {
         FieldType::Str(ref text_options) => text_options
             .get_indexing_options()
@@ -29,8 +29,7 @@ fn posting_from_field_entry<'a>(field_entry: &FieldEntry) -> Box<PostingsWriter>
                 IndexRecordOption::WithFreqsAndPositions => {
                     SpecializedPostingsWriter::<TFAndPositionRecorder>::new_boxed()
                 }
-            })
-            .unwrap_or_else(|| SpecializedPostingsWriter::<NothingRecorder>::new_boxed()),
+            }).unwrap_or_else(|| SpecializedPostingsWriter::<NothingRecorder>::new_boxed()),
         FieldType::U64(_) | FieldType::I64(_) | FieldType::HierarchicalFacet => {
             SpecializedPostingsWriter::<NothingRecorder>::new_boxed()
         }
@@ -94,11 +93,12 @@ impl MultiFieldPostingsWriter {
         &self,
         serializer: &mut InvertedIndexSerializer,
     ) -> Result<HashMap<Field, HashMap<UnorderedTermId, TermOrdinal>>> {
-        let mut term_offsets: Vec<(&[u8], Addr, UnorderedTermId)> = self.term_index
+        let mut term_offsets: Vec<(&[u8], Addr, UnorderedTermId)> = self
+            .term_index
             .iter()
             .map(|(term_bytes, addr, bucket_id)| (term_bytes, addr, bucket_id as UnorderedTermId))
             .collect();
-        term_offsets.sort_by_key(|&(k, _, _)| k);
+        term_offsets.sort_unstable_by_key(|&(k, _, _)| k);
 
         let mut offsets: Vec<(Field, usize)> = vec![];
         let term_offsets_it = term_offsets
@@ -127,8 +127,8 @@ impl MultiFieldPostingsWriter {
 
             let field_entry = self.schema.get_field_entry(field);
 
-            match field_entry.field_type() {
-                &FieldType::Str(_) | &FieldType::HierarchicalFacet => {
+            match *field_entry.field_type() {
+                FieldType::Str(_) | FieldType::HierarchicalFacet => {
                     // populating the (unordered term ord) -> (ordered term ord) mapping
                     // for the field.
                     let mut unordered_term_ids = term_offsets[start..stop]
@@ -138,12 +138,11 @@ impl MultiFieldPostingsWriter {
                         .enumerate()
                         .map(|(term_ord, unord_term_id)| {
                             (unord_term_id as UnorderedTermId, term_ord as TermOrdinal)
-                        })
-                        .collect();
+                        }).collect();
                     unordered_term_mappings.insert(field, mapping);
                 }
-                &FieldType::U64(_) | &FieldType::I64(_) => {}
-                &FieldType::Bytes => {}
+                FieldType::U64(_) | FieldType::I64(_) => {}
+                FieldType::Bytes => {}
             }
 
             let postings_writer = &self.per_field_postings_writers[field.0 as usize];
@@ -202,14 +201,11 @@ pub trait PostingsWriter {
         heap: &mut MemoryArena,
     ) -> u32 {
         let mut term = Term::for_field(field);
-        let num_tokens = {
-            let mut sink = |token: &Token| {
-                term.set_text(token.text.as_str());
-                self.subscribe(term_index, doc_id, token.position as u32, &term, heap);
-            };
-            token_stream.process(&mut sink)
+        let mut sink = |token: &Token| {
+            term.set_text(token.text.as_str());
+            self.subscribe(term_index, doc_id, token.position as u32, &term, heap);
         };
-        num_tokens
+        token_stream.process(&mut sink)
     }
 
     fn total_num_tokens(&self) -> u64;

src/postings/recorder.rs

@@ -107,7 +107,8 @@ impl Recorder for TermFrequencyRecorder {
     fn serialize(&self, serializer: &mut FieldSerializer, heap: &MemoryArena) -> io::Result<()> {
         // the last document has not been closed...
         // its term freq is self.current_tf.
-        let mut doc_iter = self.stack
+        let mut doc_iter = self
+            .stack
             .iter(heap)
             .chain(Some(self.current_tf).into_iter());
 

src/postings/segment_postings.rs

@@ -1,20 +1,20 @@
-use postings::compression::{BlockDecoder, VIntDecoder, COMPRESSION_BLOCK_SIZE};
-use DocId;
 use common::BitSet;
 use common::HasLen;
-use postings::compression::compressed_block_size;
+use common::{BinarySerializable, VInt};
 use docset::{DocSet, SkipResult};
 use fst::Streamer;
+use owned_read::OwnedRead;
+use positions::PositionReader;
+use postings::compression::compressed_block_size;
+use postings::compression::{BlockDecoder, VIntDecoder, COMPRESSION_BLOCK_SIZE};
 use postings::serializer::PostingsSerializer;
 use postings::FreqReadingOption;
 use postings::Postings;
-use owned_read::OwnedRead;
-use common::{VInt, BinarySerializable};
-use postings::USE_SKIP_INFO_LIMIT;
 use postings::SkipReader;
+use postings::USE_SKIP_INFO_LIMIT;
 use schema::IndexRecordOption;
-use positions::PositionReader;
 use std::cmp::Ordering;
+use DocId;
 
 const EMPTY_ARR: [u8; 0] = [];
 
@@ -98,7 +98,7 @@ impl SegmentPostings {
             docs.len() as u32,
             OwnedRead::new(buffer),
             IndexRecordOption::Basic,
-            IndexRecordOption::Basic
+            IndexRecordOption::Basic,
         );
         SegmentPostings::from_block_postings(block_segment_postings, None)
     }
@@ -151,7 +151,11 @@ fn exponential_search(target: u32, arr: &[u32]) -> (usize, usize) {
 /// The target is assumed smaller or equal to the last element.
 fn search_within_block(block_docs: &[u32], target: u32) -> usize {
     let (start, end) = exponential_search(target, block_docs);
-    start.wrapping_add(block_docs[start..end].binary_search(&target).unwrap_or_else(|e| e))
+    start.wrapping_add(
+        block_docs[start..end]
+            .binary_search(&target)
+            .unwrap_or_else(|e| e),
+    )
 }
 
 impl DocSet for SegmentPostings {
@@ -179,21 +183,20 @@ impl DocSet for SegmentPostings {
         // check if we need to go to the next block
         let need_positions = self.position_computer.is_some();
         let mut sum_freqs_skipped: u32 = 0;
-        if !self.block_cursor
-               .docs()
-               .last()
-               .map(|doc| *doc >= target)
-               .unwrap_or(false)  // there should always be at least a document in the block
-                                       // since advance returned.
+        if !self
+            .block_cursor
+            .docs()
+            .last()
+            .map(|doc| *doc >= target)
+            .unwrap_or(false)
+        // there should always be at least a document in the block
+        // since advance returned.
         {
             // we are not in the right block.
             //
             // First compute all of the freqs skipped from the current block.
             if need_positions {
-                sum_freqs_skipped = self.block_cursor
-                    .freqs()[self.cur..]
-                    .iter()
-                    .sum();
+                sum_freqs_skipped = self.block_cursor.freqs()[self.cur..].iter().sum();
                 match self.block_cursor.skip_to(target) {
                     BlockSegmentPostingsSkipResult::Success(block_skip_freqs) => {
                         sum_freqs_skipped += block_skip_freqs;
@@ -202,11 +205,11 @@ impl DocSet for SegmentPostings {
                         return SkipResult::End;
                     }
                 }
-            } else {
+            } else if self.block_cursor.skip_to(target)
+                == BlockSegmentPostingsSkipResult::Terminated
+            {
                 // no positions needed. no need to sum freqs.
-                if self.block_cursor.skip_to(target) == BlockSegmentPostingsSkipResult::Terminated {
-                    return SkipResult::End;
-                }
+                return SkipResult::End;
             }
             self.cur = 0;
         }
@@ -215,9 +218,13 @@ impl DocSet for SegmentPostings {
         let block_docs = self.block_cursor.docs();
 
         debug_assert!(target >= self.doc());
-        let new_cur = self.cur.wrapping_add(search_within_block(&block_docs[self.cur..], target));
+        let new_cur = self
+            .cur
+            .wrapping_add(search_within_block(&block_docs[self.cur..], target));
         if need_positions {
-            sum_freqs_skipped += self.block_cursor.freqs()[self.cur..new_cur].iter().sum::<u32>();
+            sum_freqs_skipped += self.block_cursor.freqs()[self.cur..new_cur]
+                .iter()
+                .sum::<u32>();
             self.position_computer
                 .as_mut()
                 .unwrap()
@@ -229,9 +236,9 @@ impl DocSet for SegmentPostings {
         let doc = block_docs[new_cur];
         debug_assert!(doc >= target);
         if doc == target {
-            return SkipResult::Reached;
+            SkipResult::Reached
         } else {
-            return SkipResult::OverStep;
+            SkipResult::OverStep
         }
     }
 
@@ -330,7 +337,10 @@ pub struct BlockSegmentPostings {
     skip_reader: SkipReader,
 }
 
-fn split_into_skips_and_postings(doc_freq: u32, mut data: OwnedRead) -> (Option<OwnedRead>, OwnedRead) {
+fn split_into_skips_and_postings(
+    doc_freq: u32,
+    mut data: OwnedRead,
+) -> (Option<OwnedRead>, OwnedRead) {
     if doc_freq >= USE_SKIP_INFO_LIMIT {
         let skip_len = VInt::deserialize(&mut data).expect("Data corrupted").0 as usize;
         let mut postings_data = data.clone();
@@ -345,7 +355,7 @@ fn split_into_skips_and_postings(doc_freq: u32, mut data: OwnedRead) -> (Option<
 #[derive(Debug, Eq, PartialEq)]
 pub enum BlockSegmentPostingsSkipResult {
     Terminated,
-    Success(u32) //< number of term freqs to skip
+    Success(u32), //< number of term freqs to skip
 }
 
 impl BlockSegmentPostings {
@@ -353,7 +363,7 @@ impl BlockSegmentPostings {
         doc_freq: u32,
         data: OwnedRead,
         record_option: IndexRecordOption,
-        requested_option: IndexRecordOption
+        requested_option: IndexRecordOption,
     ) -> BlockSegmentPostings {
         let freq_reading_option = match (record_option, requested_option) {
             (IndexRecordOption::Basic, _) => FreqReadingOption::NoFreq,
@@ -362,11 +372,10 @@ impl BlockSegmentPostings {
         };
 
         let (skip_data_opt, postings_data) = split_into_skips_and_postings(doc_freq, data);
-        let skip_reader =
-            match skip_data_opt {
-                Some(skip_data) => SkipReader::new(skip_data, record_option),
-                None => SkipReader::new(OwnedRead::new(&EMPTY_ARR[..]), record_option)
-            };
+        let skip_reader = match skip_data_opt {
+            Some(skip_data) => SkipReader::new(skip_data, record_option),
+            None => SkipReader::new(OwnedRead::new(&EMPTY_ARR[..]), record_option),
+        };
         let doc_freq = doc_freq as usize;
         let num_vint_docs = doc_freq % COMPRESSION_BLOCK_SIZE;
         BlockSegmentPostings {
@@ -450,7 +459,6 @@ impl BlockSegmentPostings {
         self.doc_decoder.output_len
     }
 
-
     /// position on a block that may contains `doc_id`.
     /// Always advance the current block.
     ///
@@ -461,9 +469,7 @@ impl BlockSegmentPostings {
     /// Returns false iff all of the document remaining are smaller than
     /// `doc_id`. In that case, all of these document are consumed.
     ///
-    pub fn skip_to(&mut self,
-                   target_doc: DocId) -> BlockSegmentPostingsSkipResult {
-
+    pub fn skip_to(&mut self, target_doc: DocId) -> BlockSegmentPostingsSkipResult {
         let mut skip_freqs = 0u32;
         while self.skip_reader.advance() {
             if self.skip_reader.doc() >= target_doc {
@@ -472,11 +478,11 @@ impl BlockSegmentPostings {
                 //
                 // We found our block!
                 let num_bits = self.skip_reader.doc_num_bits();
-                let num_consumed_bytes = self.doc_decoder
-                    .uncompress_block_sorted(
-                        self.remaining_data.as_ref(),
-                        self.doc_offset,
-                        num_bits);
+                let num_consumed_bytes = self.doc_decoder.uncompress_block_sorted(
+                    self.remaining_data.as_ref(),
+                    self.doc_offset,
+                    num_bits,
+                );
                 self.remaining_data.advance(num_consumed_bytes);
                 let tf_num_bits = self.skip_reader.tf_num_bits();
                 match self.freq_reading_option {
@@ -486,9 +492,9 @@ impl BlockSegmentPostings {
                         self.remaining_data.advance(num_bytes_to_skip);
                     }
                     FreqReadingOption::ReadFreq => {
-                        let num_consumed_bytes = self.freq_decoder
-                            .uncompress_block_unsorted(self.remaining_data.as_ref(),
-                                                       tf_num_bits);
+                        let num_consumed_bytes = self
+                            .freq_decoder
+                            .uncompress_block_unsorted(self.remaining_data.as_ref(), tf_num_bits);
                         self.remaining_data.advance(num_consumed_bytes);
                     }
                 }
@@ -518,7 +524,8 @@ impl BlockSegmentPostings {
                 }
             }
             self.num_vint_docs = 0;
-            return self.docs()
+            return self
+                .docs()
                 .last()
                 .map(|last_doc| {
                     if *last_doc >= target_doc {
@@ -526,8 +533,7 @@ impl BlockSegmentPostings {
                     } else {
                         BlockSegmentPostingsSkipResult::Terminated
                     }
-                })
-                .unwrap_or(BlockSegmentPostingsSkipResult::Terminated);
+                }).unwrap_or(BlockSegmentPostingsSkipResult::Terminated);
         }
         BlockSegmentPostingsSkipResult::Terminated
     }
@@ -538,11 +544,11 @@ impl BlockSegmentPostings {
     pub fn advance(&mut self) -> bool {
         if self.skip_reader.advance() {
             let num_bits = self.skip_reader.doc_num_bits();
-            let num_consumed_bytes = self.doc_decoder
-                .uncompress_block_sorted(
-                    self.remaining_data.as_ref(),
-                        self.doc_offset,
-                        num_bits);
+            let num_consumed_bytes = self.doc_decoder.uncompress_block_sorted(
+                self.remaining_data.as_ref(),
+                self.doc_offset,
+                num_bits,
+            );
             self.remaining_data.advance(num_consumed_bytes);
             let tf_num_bits = self.skip_reader.tf_num_bits();
             match self.freq_reading_option {
@@ -552,9 +558,9 @@ impl BlockSegmentPostings {
                     self.remaining_data.advance(num_bytes_to_skip);
                 }
                 FreqReadingOption::ReadFreq => {
-                    let num_consumed_bytes = self.freq_decoder
-                        .uncompress_block_unsorted(self.remaining_data.as_ref(),
-                                                   tf_num_bits);
+                    let num_consumed_bytes = self
+                        .freq_decoder
+                        .uncompress_block_unsorted(self.remaining_data.as_ref(), tf_num_bits);
                     self.remaining_data.advance(num_consumed_bytes);
                 }
             }
@@ -594,7 +600,6 @@ impl BlockSegmentPostings {
             doc_offset: 0,
             doc_freq: 0,
 
-
             remaining_data: OwnedRead::new(vec![]),
             skip_reader: SkipReader::new(OwnedRead::new(vec![]), IndexRecordOption::Basic),
         }
@@ -616,19 +621,19 @@ impl<'b> Streamer<'b> for BlockSegmentPostings {
 #[cfg(test)]
 mod tests {
 
+    use super::search_within_block;
     use super::BlockSegmentPostings;
+    use super::BlockSegmentPostingsSkipResult;
     use super::SegmentPostings;
     use common::HasLen;
     use core::Index;
     use docset::DocSet;
     use fst::Streamer;
     use schema::IndexRecordOption;
-    use schema::SchemaBuilder;
+    use schema::Schema;
     use schema::Term;
     use schema::INT_INDEXED;
-    use super::BlockSegmentPostingsSkipResult;
     use DocId;
-    use super::search_within_block;
 
     #[test]
     fn test_empty_segment_postings() {
@@ -645,7 +650,6 @@ mod tests {
         assert_eq!(postings.doc_freq(), 0);
     }
 
-
     fn search_within_block_trivial_but_slow(block: &[u32], target: u32) -> usize {
         block
             .iter()
@@ -653,11 +657,15 @@ mod tests {
             .enumerate()
             .filter(|&(_, ref val)| *val >= target)
             .next()
-            .unwrap().0
+            .unwrap()
+            .0
     }
 
     fn util_test_search_within_block(block: &[u32], target: u32) {
-        assert_eq!(search_within_block(block, target), search_within_block_trivial_but_slow(block, target));
+        assert_eq!(
+            search_within_block(block, target),
+            search_within_block_trivial_but_slow(block, target)
+        );
     }
 
     fn util_test_search_within_block_all(block: &[u32]) {
@@ -677,7 +685,7 @@ mod tests {
     #[test]
     fn test_search_within_block() {
         for len in 1u32..128u32 {
-            let v: Vec<u32> = (0..len).map(|i| i*2).collect();
+            let v: Vec<u32> = (0..len).map(|i| i * 2).collect();
             util_test_search_within_block_all(&v[..]);
         }
     }
@@ -699,7 +707,7 @@ mod tests {
     }
 
     fn build_block_postings(docs: Vec<DocId>) -> BlockSegmentPostings {
-        let mut schema_builder = SchemaBuilder::default();
+        let mut schema_builder = Schema::builder();
         let int_field = schema_builder.add_u64_field("id", INT_INDEXED);
         let schema = schema_builder.build();
         let index = Index::create_in_ram(schema);
@@ -726,14 +734,22 @@ mod tests {
     fn test_block_segment_postings_skip() {
         for i in 0..4 {
             let mut block_postings = build_block_postings(vec![3]);
-            assert_eq!(block_postings.skip_to(i), BlockSegmentPostingsSkipResult::Success(0u32));
-            assert_eq!(block_postings.skip_to(i), BlockSegmentPostingsSkipResult::Terminated);
+            assert_eq!(
+                block_postings.skip_to(i),
+                BlockSegmentPostingsSkipResult::Success(0u32)
+            );
+            assert_eq!(
+                block_postings.skip_to(i),
+                BlockSegmentPostingsSkipResult::Terminated
+            );
         }
         let mut block_postings = build_block_postings(vec![3]);
-        assert_eq!(block_postings.skip_to(4u32), BlockSegmentPostingsSkipResult::Terminated);
+        assert_eq!(
+            block_postings.skip_to(4u32),
+            BlockSegmentPostingsSkipResult::Terminated
+        );
     }
 
-
     #[test]
     fn test_block_segment_postings_skip2() {
         let mut docs = vec![0];
@@ -741,19 +757,28 @@ mod tests {
             docs.push((i * i / 100) + i);
         }
         let mut block_postings = build_block_postings(docs.clone());
-        for i in vec![0,  424, 10000] {
-            assert_eq!(block_postings.skip_to(i), BlockSegmentPostingsSkipResult::Success(0u32));
+        for i in vec![0, 424, 10000] {
+            assert_eq!(
+                block_postings.skip_to(i),
+                BlockSegmentPostingsSkipResult::Success(0u32)
+            );
             let docs = block_postings.docs();
             assert!(docs[0] <= i);
             assert!(docs.last().cloned().unwrap_or(0u32) >= i);
         }
-        assert_eq!(block_postings.skip_to(100_000), BlockSegmentPostingsSkipResult::Terminated);
-        assert_eq!(block_postings.skip_to(101_000), BlockSegmentPostingsSkipResult::Terminated);
+        assert_eq!(
+            block_postings.skip_to(100_000),
+            BlockSegmentPostingsSkipResult::Terminated
+        );
+        assert_eq!(
+            block_postings.skip_to(101_000),
+            BlockSegmentPostingsSkipResult::Terminated
+        );
     }
 
     #[test]
     fn test_reset_block_segment_postings() {
-        let mut schema_builder = SchemaBuilder::default();
+        let mut schema_builder = Schema::builder();
         let int_field = schema_builder.add_u64_field("id", INT_INDEXED);
         let schema = schema_builder.build();
         let index = Index::create_in_ram(schema);

src/postings/serializer.rs

@@ -1,18 +1,18 @@
 use super::TermInfo;
-use common::{VInt, BinarySerializable};
+use common::{BinarySerializable, VInt};
 use common::{CompositeWrite, CountingWriter};
-use postings::compression::{VIntEncoder, BlockEncoder, COMPRESSION_BLOCK_SIZE};
 use core::Segment;
 use directory::WritePtr;
+use positions::PositionSerializer;
+use postings::compression::{BlockEncoder, VIntEncoder, COMPRESSION_BLOCK_SIZE};
+use postings::skip::SkipSerializer;
+use postings::USE_SKIP_INFO_LIMIT;
 use schema::Schema;
 use schema::{Field, FieldEntry, FieldType};
 use std::io::{self, Write};
 use termdict::{TermDictionaryBuilder, TermOrdinal};
 use DocId;
 use Result;
-use postings::USE_SKIP_INFO_LIMIT;
-use postings::skip::SkipSerializer;
-use positions::PositionSerializer;
 
 /// `PostingsSerializer` is in charge of serializing
 /// postings on disk, in the
@@ -100,11 +100,11 @@ impl InvertedIndexSerializer {
         let positionsidx_write = self.positionsidx_write.for_field(field);
         let field_type: FieldType = (*field_entry.field_type()).clone();
         FieldSerializer::new(
-            field_type,
+            &field_type,
             term_dictionary_write,
             postings_write,
             positions_write,
-            positionsidx_write
+            positionsidx_write,
         )
     }
 
@@ -131,11 +131,11 @@ pub struct FieldSerializer<'a> {
 
 impl<'a> FieldSerializer<'a> {
     fn new(
-        field_type: FieldType,
+        field_type: &FieldType,
         term_dictionary_write: &'a mut CountingWriter<WritePtr>,
         postings_write: &'a mut CountingWriter<WritePtr>,
         positions_write: &'a mut CountingWriter<WritePtr>,
-        positionsidx_write: &'a mut CountingWriter<WritePtr>
+        positionsidx_write: &'a mut CountingWriter<WritePtr>,
     ) -> io::Result<FieldSerializer<'a>> {
         let (term_freq_enabled, position_enabled): (bool, bool) = match field_type {
             FieldType::Str(ref text_options) => {
@@ -152,8 +152,9 @@ impl<'a> FieldSerializer<'a> {
             _ => (false, false),
         };
         let term_dictionary_builder =
-            TermDictionaryBuilder::new(term_dictionary_write, field_type)?;
-        let postings_serializer = PostingsSerializer::new(postings_write, term_freq_enabled, position_enabled);
+            TermDictionaryBuilder::new(term_dictionary_write, &field_type)?;
+        let postings_serializer =
+            PostingsSerializer::new(postings_write, term_freq_enabled, position_enabled);
         let positions_serializer_opt = if position_enabled {
             Some(PositionSerializer::new(positions_write, positionsidx_write))
         } else {
@@ -171,14 +172,15 @@ impl<'a> FieldSerializer<'a> {
     }
 
     fn current_term_info(&self) -> TermInfo {
-        let positions_idx = self.positions_serializer_opt
+        let positions_idx = self
+            .positions_serializer_opt
             .as_ref()
             .map(|positions_serializer| positions_serializer.positions_idx())
             .unwrap_or(0u64);
         TermInfo {
             doc_freq: 0,
             postings_offset: self.postings_serializer.addr(),
-            positions_idx
+            positions_idx,
         }
     }
 
@@ -253,7 +255,7 @@ impl<'a> FieldSerializer<'a> {
 struct Block {
     doc_ids: [DocId; COMPRESSION_BLOCK_SIZE],
     term_freqs: [u32; COMPRESSION_BLOCK_SIZE],
-    len: usize
+    len: usize,
 }
 
 impl Block {
@@ -261,7 +263,7 @@ impl Block {
         Block {
             doc_ids: [0u32; COMPRESSION_BLOCK_SIZE],
             term_freqs: [0u32; COMPRESSION_BLOCK_SIZE],
-            len: 0
+            len: 0,
         }
     }
 
@@ -312,9 +314,12 @@ pub struct PostingsSerializer<W: Write> {
     termfreq_sum_enabled: bool,
 }
 
-
 impl<W: Write> PostingsSerializer<W> {
-    pub fn new(write: W, termfreq_enabled: bool, termfreq_sum_enabled: bool) -> PostingsSerializer<W> {
+    pub fn new(
+        write: W,
+        termfreq_enabled: bool,
+        termfreq_sum_enabled: bool,
+    ) -> PostingsSerializer<W> {
         PostingsSerializer {
             output_write: CountingWriter::wrap(write),
 
@@ -337,14 +342,16 @@ impl<W: Write> PostingsSerializer<W> {
                 .block_encoder
                 .compress_block_sorted(&self.block.doc_ids(), self.last_doc_id_encoded);
             self.last_doc_id_encoded = self.block.last_doc();
-            self.skip_write.write_doc(self.last_doc_id_encoded, num_bits);
+            self.skip_write
+                .write_doc(self.last_doc_id_encoded, num_bits);
             // last el block 0, offset block 1,
             self.postings_write.extend(block_encoded);
         }
         if self.termfreq_enabled {
             // encode the term_freqs
-            let (num_bits, block_encoded): (u8, &[u8]) =
-                self.block_encoder.compress_block_unsorted(&self.block.term_freqs());
+            let (num_bits, block_encoded): (u8, &[u8]) = self
+                .block_encoder
+                .compress_block_unsorted(&self.block.term_freqs());
             self.postings_write.extend(block_encoded);
             self.skip_write.write_term_freq(num_bits);
             if self.termfreq_sum_enabled {
@@ -375,13 +382,15 @@ impl<W: Write> PostingsSerializer<W> {
             // In that case, the remaining part is encoded
             // using variable int encoding.
             {
-                let block_encoded = self.block_encoder
+                let block_encoded = self
+                    .block_encoder
                     .compress_vint_sorted(&self.block.doc_ids(), self.last_doc_id_encoded);
                 self.postings_write.write_all(block_encoded)?;
             }
             // ... Idem for term frequencies
             if self.termfreq_enabled {
-                let block_encoded = self.block_encoder
+                let block_encoded = self
+                    .block_encoder
                     .compress_vint_unsorted(self.block.term_freqs());
                 self.postings_write.write_all(block_encoded)?;
             }
@@ -392,7 +401,6 @@ impl<W: Write> PostingsSerializer<W> {
             VInt(skip_data.len() as u64).serialize(&mut self.output_write)?;
             self.output_write.write_all(skip_data)?;
             self.output_write.write_all(&self.postings_write[..])?;
-
         } else {
             self.output_write.write_all(&self.postings_write[..])?;
         }

src/postings/skip.rs

@@ -1,8 +1,8 @@
-use DocId;
 use common::BinarySerializable;
 use owned_read::OwnedRead;
 use postings::compression::COMPRESSION_BLOCK_SIZE;
 use schema::IndexRecordOption;
+use DocId;
 
 pub struct SkipSerializer {
     buffer: Vec<u8>,
@@ -18,8 +18,11 @@ impl SkipSerializer {
     }
 
     pub fn write_doc(&mut self, last_doc: DocId, doc_num_bits: u8) {
-        assert!(last_doc > self.prev_doc, "write_doc(...) called with non-increasing doc ids. \
-                Did you forget to call clear maybe?");
+        assert!(
+            last_doc > self.prev_doc,
+            "write_doc(...) called with non-increasing doc ids. \
+             Did you forget to call clear maybe?"
+        );
         let delta_doc = last_doc - self.prev_doc;
         self.prev_doc = last_doc;
         delta_doc.serialize(&mut self.buffer).unwrap();
@@ -30,9 +33,10 @@ impl SkipSerializer {
         self.buffer.push(tf_num_bits);
     }
 
-
     pub fn write_total_term_freq(&mut self, tf_sum: u32) {
-        tf_sum.serialize(&mut self.buffer).expect("Should never fail");
+        tf_sum
+            .serialize(&mut self.buffer)
+            .expect("Should never fail");
     }
 
     pub fn data(&self) -> &[u8] {
@@ -103,33 +107,32 @@ impl SkipReader {
         } else {
             let doc_delta = u32::deserialize(&mut self.owned_read).expect("Skip data corrupted");
             self.doc += doc_delta as DocId;
-            self.doc_num_bits =  self.owned_read.get(0);
+            self.doc_num_bits = self.owned_read.get(0);
             match self.skip_info {
                 IndexRecordOption::Basic => {
                     self.owned_read.advance(1);
                 }
-                IndexRecordOption::WithFreqs=> {
+                IndexRecordOption::WithFreqs => {
                     self.tf_num_bits = self.owned_read.get(1);
                     self.owned_read.advance(2);
                 }
                 IndexRecordOption::WithFreqsAndPositions => {
                     self.tf_num_bits = self.owned_read.get(1);
                     self.owned_read.advance(2);
-                    self.tf_sum = u32::deserialize(&mut self.owned_read)
-                        .expect("Failed reading tf_sum");
+                    self.tf_sum =
+                        u32::deserialize(&mut self.owned_read).expect("Failed reading tf_sum");
                 }
             }
             true
         }
-
     }
 }
 
 #[cfg(test)]
 mod tests {
 
-    use super::{SkipReader, SkipSerializer};
     use super::IndexRecordOption;
+    use super::{SkipReader, SkipSerializer};
     use owned_read::OwnedRead;
 
     #[test]
@@ -171,4 +174,4 @@ mod tests {
         assert_eq!(skip_reader.doc_num_bits(), 5u8);
         assert!(!skip_reader.advance());
     }
-}
+}

src/postings/stacker/expull.rs

@@ -174,8 +174,8 @@ mod tests {
 
 #[cfg(all(test, feature = "unstable"))]
 mod bench {
+    use super::super::MemoryArena;
     use super::ExpUnrolledLinkedList;
-    use tantivy_memory_arena::MemoryArena;
     use test::Bencher;
 
     const NUM_STACK: usize = 10_000;
@@ -199,20 +199,19 @@ mod bench {
 
     #[bench]
     fn bench_push_stack(bench: &mut Bencher) {
-        let heap = MemoryArena::new();
         bench.iter(|| {
+            let mut heap = MemoryArena::new();
             let mut stacks = Vec::with_capacity(100);
             for _ in 0..NUM_STACK {
-                let (_, stack) = heap.allocate_object::<ExpUnrolledLinkedList>();
+                let mut stack = ExpUnrolledLinkedList::new(&mut heap);
                 stacks.push(stack);
             }
             for s in 0..NUM_STACK {
                 for i in 0u32..STACK_SIZE {
                     let t = s * 392017 % NUM_STACK;
-                    stacks[t].push(i, &heap);
+                    stacks[t].push(i, &mut heap);
                 }
             }
-            heap.clear();
         });
     }
 }

src/postings/stacker/memory_arena.rs

@@ -47,7 +47,7 @@ impl Addr {
     }
 
     /// Returns the `Addr` object for `addr + offset`
-    pub fn offset(&self, offset: u32) -> Addr {
+    pub fn offset(self, offset: u32) -> Addr {
         Addr(self.0.wrapping_add(offset))
     }
 
@@ -55,16 +55,16 @@ impl Addr {
         Addr((page_id << NUM_BITS_PAGE_ADDR | local_addr) as u32)
     }
 
-    fn page_id(&self) -> usize {
+    fn page_id(self) -> usize {
         (self.0 as usize) >> NUM_BITS_PAGE_ADDR
     }
 
-    fn page_local_addr(&self) -> usize {
+    fn page_local_addr(self) -> usize {
         (self.0 as usize) & (PAGE_SIZE - 1)
     }
 
     /// Returns true if and only if the `Addr` is null.
-    pub fn is_null(&self) -> bool {
+    pub fn is_null(self) -> bool {
         self.0 == u32::max_value()
     }
 }
@@ -233,12 +233,12 @@ impl Page {
 
     #[inline(always)]
     pub(crate) unsafe fn get_ptr(&self, addr: usize) -> *const u8 {
-        self.data.as_ptr().offset(addr as isize)
+        self.data.as_ptr().add(addr)
     }
 
     #[inline(always)]
     pub(crate) unsafe fn get_mut_ptr(&mut self, addr: usize) -> *mut u8 {
-        self.data.as_mut_ptr().offset(addr as isize)
+        self.data.as_mut_ptr().add(addr)
     }
 }
 

src/postings/stacker/murmurhash2.rs

@@ -4,6 +4,7 @@ const M: u32 = 0x5bd1_e995;
 
 #[inline(always)]
 pub fn murmurhash2(key: &[u8]) -> u32 {
+    #[cfg_attr(feature = "cargo-clippy", allow(clippy::cast_ptr_alignment))]
     let mut key_ptr: *const u32 = key.as_ptr() as *const u32;
     let len = key.len() as u32;
     let mut h: u32 = SEED ^ len;

src/postings/stacker/term_hashmap.rs

@@ -61,7 +61,7 @@ impl Default for KeyValue {
 }
 
 impl KeyValue {
-    fn is_empty(&self) -> bool {
+    fn is_empty(self) -> bool {
         self.key_value_addr.is_null()
     }
 }

src/query/all_query.rs

@@ -59,10 +59,10 @@ impl DocSet for AllScorer {
             }
         }
         if self.doc < self.max_doc {
-            return true;
+            true
         } else {
             self.state = State::Finished;
-            return false;
+            false
         }
     }
 
@@ -86,12 +86,12 @@ mod tests {
 
     use super::AllQuery;
     use query::Query;
-    use schema::{SchemaBuilder, TEXT};
+    use schema::{Schema, TEXT};
     use Index;
 
     #[test]
     fn test_all_query() {
-        let mut schema_builder = SchemaBuilder::default();
+        let mut schema_builder = Schema::builder();
         let field = schema_builder.add_text_field("text", TEXT);
         let schema = schema_builder.build();
         let index = Index::create_in_ram(schema);

src/query/automaton_weight.rs

@@ -11,7 +11,7 @@ use Result;
 /// A weight struct for Fuzzy Term and Regex Queries
 pub struct AutomatonWeight<A>
 where
-    A: Automaton,
+    A: Automaton + Send + Sync + 'static,
 {
     field: Field,
     automaton: A,
@@ -19,7 +19,7 @@ where
 
 impl<A> AutomatonWeight<A>
 where
-    A: Automaton,
+    A: Automaton + Send + Sync + 'static,
 {
     /// Create a new AutomationWeight
     pub fn new(field: Field, automaton: A) -> AutomatonWeight<A> {
@@ -34,7 +34,7 @@ where
 
 impl<A> Weight for AutomatonWeight<A>
 where
-    A: Automaton,
+    A: Automaton + Send + Sync + 'static,
 {
     fn scorer(&self, reader: &SegmentReader) -> Result<Box<Scorer>> {
         let max_doc = reader.max_doc();

src/query/bm25.rs

@@ -17,9 +17,9 @@ fn cached_tf_component(fieldnorm: u32, average_fieldnorm: f32) -> f32 {
 
 fn compute_tf_cache(average_fieldnorm: f32) -> [f32; 256] {
     let mut cache = [0f32; 256];
-    for fieldnorm_id in 0..256 {
+    for (fieldnorm_id, cache_mut) in cache.iter_mut().enumerate() {
         let fieldnorm = FieldNormReader::id_to_fieldnorm(fieldnorm_id as u8);
-        cache[fieldnorm_id] = cached_tf_component(fieldnorm, average_fieldnorm);
+        *cache_mut = cached_tf_component(fieldnorm, average_fieldnorm);
     }
     cache
 }
@@ -54,7 +54,7 @@ impl BM25Weight {
         for segment_reader in searcher.segment_readers() {
             let inverted_index = segment_reader.inverted_index(field);
             total_num_tokens += inverted_index.total_num_tokens();
-            total_num_docs += segment_reader.max_doc() as u64;
+            total_num_docs += u64::from(segment_reader.max_doc());
         }
         let average_fieldnorm = total_num_tokens as f32 / total_num_docs as f32;
 
@@ -63,8 +63,7 @@ impl BM25Weight {
             .map(|term| {
                 let term_doc_freq = searcher.doc_freq(term);
                 idf(term_doc_freq, total_num_docs)
-            })
-            .sum::<f32>();
+            }).sum::<f32>();
         BM25Weight::new(idf, average_fieldnorm)
     }
 

src/query/boolean_query/boolean_query.rs

@@ -5,6 +5,7 @@ use query::TermQuery;
 use query::Weight;
 use schema::IndexRecordOption;
 use schema::Term;
+use std::collections::BTreeSet;
 use Result;
 use Searcher;
 
@@ -27,7 +28,7 @@ impl Clone for BooleanQuery {
     fn clone(&self) -> Self {
         self.subqueries
             .iter()
-            .map(|(x, y)| (x.clone(), y.box_clone()))
+            .map(|(occur, subquery)| (*occur, subquery.box_clone()))
             .collect::<Vec<_>>()
             .into()
     }
@@ -41,14 +42,20 @@ impl From<Vec<(Occur, Box<Query>)>> for BooleanQuery {
 
 impl Query for BooleanQuery {
     fn weight(&self, searcher: &Searcher, scoring_enabled: bool) -> Result<Box<Weight>> {
-        let sub_weights = self.subqueries
+        let sub_weights = self
+            .subqueries
             .iter()
             .map(|&(ref occur, ref subquery)| {
                 Ok((*occur, subquery.weight(searcher, scoring_enabled)?))
-            })
-            .collect::<Result<_>>()?;
+            }).collect::<Result<_>>()?;
         Ok(Box::new(BooleanWeight::new(sub_weights, scoring_enabled)))
     }
+
+    fn query_terms(&self, term_set: &mut BTreeSet<Term>) {
+        for (_occur, subquery) in &self.subqueries {
+            subquery.query_terms(term_set);
+        }
+    }
 }
 
 impl BooleanQuery {
@@ -61,8 +68,7 @@ impl BooleanQuery {
                 let term_query: Box<Query> =
                     Box::new(TermQuery::new(term, IndexRecordOption::WithFreqs));
                 (Occur::Should, term_query)
-            })
-            .collect();
+            }).collect();
         BooleanQuery::from(occur_term_queries)
     }
 

src/query/boolean_query/boolean_weight.rs

@@ -39,7 +39,7 @@ where
     }
 
     let scorer: Box<Scorer> = Box::new(Union::<_, TScoreCombiner>::from(scorers));
-    return scorer;
+    scorer
 }
 
 pub struct BooleanWeight {

src/query/boolean_query/mod.rs

@@ -19,10 +19,11 @@ mod tests {
     use query::Scorer;
     use query::TermQuery;
     use schema::*;
+    use DocId;
     use Index;
 
     fn aux_test_helper() -> (Index, Field) {
-        let mut schema_builder = SchemaBuilder::default();
+        let mut schema_builder = Schema::builder();
         let text_field = schema_builder.add_text_field("text", TEXT);
         let schema = schema_builder.build();
         let index = Index::create_in_ram(schema);
@@ -69,7 +70,7 @@ mod tests {
         let query_parser = QueryParser::for_index(&index, vec![text_field]);
         let query = query_parser.parse_query("+a").unwrap();
         let searcher = index.searcher();
-        let weight = query.weight(&*searcher, true).unwrap();
+        let weight = query.weight(&searcher, true).unwrap();
         let scorer = weight.scorer(searcher.segment_reader(0u32)).unwrap();
         assert!(Downcast::<TermScorer>::is_type(&*scorer));
     }
@@ -81,13 +82,13 @@ mod tests {
         let searcher = index.searcher();
         {
             let query = query_parser.parse_query("+a +b +c").unwrap();
-            let weight = query.weight(&*searcher, true).unwrap();
+            let weight = query.weight(&searcher, true).unwrap();
             let scorer = weight.scorer(searcher.segment_reader(0u32)).unwrap();
             assert!(Downcast::<Intersection<TermScorer>>::is_type(&*scorer));
         }
         {
             let query = query_parser.parse_query("+a +(b c)").unwrap();
-            let weight = query.weight(&*searcher, true).unwrap();
+            let weight = query.weight(&searcher, true).unwrap();
             let scorer = weight.scorer(searcher.segment_reader(0u32)).unwrap();
             assert!(Downcast::<Intersection<Box<Scorer>>>::is_type(&*scorer));
         }
@@ -100,7 +101,7 @@ mod tests {
         let searcher = index.searcher();
         {
             let query = query_parser.parse_query("+a b").unwrap();
-            let weight = query.weight(&*searcher, true).unwrap();
+            let weight = query.weight(&searcher, true).unwrap();
             let scorer = weight.scorer(searcher.segment_reader(0u32)).unwrap();
             assert!(Downcast::<
                 RequiredOptionalScorer<Box<Scorer>, Box<Scorer>, SumWithCoordsCombiner>,
@@ -108,7 +109,7 @@ mod tests {
         }
         {
             let query = query_parser.parse_query("+a b").unwrap();
-            let weight = query.weight(&*searcher, false).unwrap();
+            let weight = query.weight(&searcher, false).unwrap();
             let scorer = weight.scorer(searcher.segment_reader(0u32)).unwrap();
             println!("{:?}", scorer.type_name());
             assert!(Downcast::<TermScorer>::is_type(&*scorer));
@@ -130,9 +131,13 @@ mod tests {
 
         let matching_docs = |boolean_query: &Query| {
             let searcher = index.searcher();
-            let mut test_collector = TestCollector::default();
-            searcher.search(boolean_query, &mut test_collector).unwrap();
-            test_collector.docs()
+            let test_docs = searcher.search(boolean_query, &TestCollector).unwrap();
+            test_docs
+                .docs()
+                .iter()
+                .cloned()
+                .map(|doc| doc.1)
+                .collect::<Vec<DocId>>()
         };
 
         {
@@ -186,9 +191,8 @@ mod tests {
 
         let score_docs = |boolean_query: &Query| {
             let searcher = index.searcher();
-            let mut test_collector = TestCollector::default();
-            searcher.search(boolean_query, &mut test_collector).unwrap();
-            test_collector.scores()
+            let fruit = searcher.search(boolean_query, &TestCollector).unwrap();
+            fruit.scores().to_vec()
         };
 
         {

src/query/empty_query.rs

@@ -1,11 +1,11 @@
 use super::Scorer;
-use DocSet;
-use Score;
-use DocId;
 use query::Query;
-use Result;
-use Searcher;
 use query::Weight;
+use DocId;
+use DocSet;
+use Result;
+use Score;
+use Searcher;
 use SegmentReader;
 
 /// `EmptyQuery` is a dummy `Query` in which no document matches.

src/query/fastfield_filter/fastfield_filter_query.rs

@@ -1,96 +0,0 @@
-use super::FastFieldFilterWeight;
-use query::Query;
-use query::Weight;
-use Result;
-use Searcher;
-use schema::Field;
-use super::RangeU64;
-use std::collections::Bound;
-use common::i64_to_u64;
-use schema::Schema;
-use schema::FieldEntry;
-use TantivyError;
-use schema::Type;
-
-#[derive(Debug, Copy, Clone)]
-enum TypeInt {
-    U64, I64
-}
-
-impl TypeInt {
-    fn value_type(self) -> Type {
-        match self {
-            TypeInt::I64 => Type::I64,
-            TypeInt::U64 => Type::U64
-        }
-    }
-}
-
-//< TODO i64 range Debug string will not look good in the
-// current implementation. Defer conversion to the scorer, or
-// back convert values for Debug.
-#[derive(Debug, Clone)]
-pub struct FastFieldFilterQuery {
-    field: Field,
-    range: RangeU64,
-    int_type: TypeInt, //< just here to check the schema at runtime, as we call `.weight`
-}
-
-fn convert_bound_to_u64(bound: Bound<i64>) -> Bound<u64> {
-    match bound {
-        Bound::Included(val) =>
-            Bound::Excluded(i64_to_u64(val)),
-        Bound::Excluded(val) =>
-            Bound::Excluded(i64_to_u64(val)),
-        Bound::Unbounded => Bound::Unbounded
-    }
-}
-
-impl FastFieldFilterQuery {
-
-    pub fn new_u64(field: Field, low: Bound<u64>, high: Bound<u64>) -> FastFieldFilterQuery {
-        FastFieldFilterQuery {
-            field: field,
-            range: RangeU64 { low, high },
-            int_type: TypeInt::U64
-        }
-    }
-
-    pub fn new_i64(field: Field, low: Bound<i64>, high: Bound<i64>) -> FastFieldFilterQuery {
-        FastFieldFilterQuery {
-            field: field,
-            range: RangeU64 {
-                low: convert_bound_to_u64(low),
-                high: convert_bound_to_u64(high)
-            },
-            int_type: TypeInt::I64
-        }
-    }
-
-
-    fn validate_schema(&self, schema: &Schema) -> Result<()> {
-        let field_entry: &FieldEntry = schema.get_field_entry(self.field);
-        if !field_entry.is_int_fast() {
-            return Err(TantivyError::SchemaError(format!(
-                "Field {:?} is not an int fast field",
-                field_entry.name()
-            )));
-        }
-        let expected_value_type = self.int_type.value_type();
-        if field_entry.field_type().value_type() != self.int_type.value_type() {
-            return Err(TantivyError::SchemaError(format!(
-                "Field {:?} is not a {:?}",
-                field_entry.name(),
-                expected_value_type
-            )));
-        }
-        Ok(())
-    }
-}
-
-impl Query for FastFieldFilterQuery {
-    fn weight(&self, searcher: &Searcher, _scoring_enabled: bool) -> Result<Box<Weight>> {
-        self.validate_schema(searcher.schema())?;
-        Ok(Box::new(FastFieldFilterWeight::new(self.field, self.range.clone())))
-    }
-}

src/query/fastfield_filter/fastfield_filter_scorer.rs

@@ -1,58 +0,0 @@
-use query::Scorer;
-use fastfield::FastFieldReader;
-use DocId;
-use DocSet;
-use query::fastfield_filter::RangeU64;
-
-pub(crate) struct FastFieldFilterScorer {
-    fastfield_reader: FastFieldReader<u64>,
-    range: RangeU64,
-    max_doc: DocId,
-    doc: DocId,
-}
-
-impl FastFieldFilterScorer {
-    pub fn new(fastfield_reader: FastFieldReader<u64>,
-           range: RangeU64,
-           max_doc: DocId) -> FastFieldFilterScorer {
-        FastFieldFilterScorer {
-            fastfield_reader,
-            range,
-            max_doc,
-            doc: 0u32,
-        }
-    }
-
-    fn within_range(&self, doc: DocId) -> bool {
-        let val = self.fastfield_reader.get(doc);
-        self.range.contains(val)
-    }
-
-}
-
-impl DocSet for FastFieldFilterScorer {
-    fn advance(&mut self) -> bool {
-        for doc in (self.doc + 1)..self.max_doc {
-            if self.within_range(doc) {
-                self.doc = doc;
-                return true;
-            }
-        }
-        self.doc = self.max_doc;
-        return false;
-    }
-
-    fn doc(&self) -> u32 {
-        self.doc
-    }
-
-    fn size_hint(&self) -> u32 {
-        self.max_doc
-    }
-}
-
-impl Scorer for FastFieldFilterScorer {
-    fn score(&mut self) -> f32 {
-        1f32
-    }
-}