Bumping version

* Change the semantic of Index::create_in_dir.

It should return an error if the directory already contains an Index.

* Index::open_or_create is working

* additional test

* Checking that schema matches on open_or_create.

Simplifying unit tests.

* simplifying Eq

.gitignore

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

CHANGELOG.md

@@ -1,10 +1,16 @@
+Tantivy 0.7.1
+=====================
+- 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 = "0.7.2"
 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" }
@@ -33,7 +32,7 @@ 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"] }
+uuid = { version = "0.7", features = ["v4", "serde"] }
 crossbeam = "0.4"
 crossbeam-channel = "0.2"
 futures = "0.1"
@@ -48,24 +47,33 @@ census = "0.1"
 fnv = "1.0.6"
 owned-read = "0.4"
 failure = "0.1"
+htmlescape = "0.3.1"
+fail = "0.2"
 
 [target.'cfg(windows)'.dependencies]
 winapi = "0.2"
 
 [dev-dependencies]
 rand = "0.5"
+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"]
 
 [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

@@ -10,7 +10,6 @@
 // - search for the best document matchings "sea whale"
 // - retrieve the best document original content.
 
-
 extern crate tempdir;
 
 // ---
@@ -231,13 +230,11 @@ fn main() -> tantivy::Result<()> {
     // a title.
 
     for doc_address in doc_addresses {
-        let retrieved_doc = searcher.doc(&doc_address)?;
+        let retrieved_doc = searcher.doc(doc_address)?;
         println!("{}", schema.to_json(&retrieved_doc));
     }
 
-
     Ok(())
 }
 
-
 use tempdir::TempDir;

examples/custom_tokenizer.rs

@@ -3,7 +3,6 @@
 // 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;
@@ -12,7 +11,6 @@ use tantivy::schema::*;
 use tantivy::tokenizer::NgramTokenizer;
 use tantivy::Index;
 
-
 fn main() -> tantivy::Result<()> {
     // # Defining the schema
     //
@@ -111,7 +109,7 @@ fn main() -> tantivy::Result<()> {
 
     let doc_addresses = top_collector.docs();
     for doc_address in doc_addresses {
-        let retrieved_doc = searcher.doc(&doc_address)?;
+        let retrieved_doc = searcher.doc(doc_address)?;
         println!("{}", schema.to_json(&retrieved_doc));
     }
 

examples/deleting_updating_documents.rs

@@ -11,10 +11,9 @@
 #[macro_use]
 extern crate tantivy;
 use tantivy::collector::TopCollector;
+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.
@@ -31,8 +30,8 @@ fn extract_doc_given_isbn(index: &Index, isbn_term: &Term) -> tantivy::Result<Op
     let mut top_collector = TopCollector::with_limit(1);
     searcher.search(&term_query, &mut top_collector)?;
 
-    if let Some(doc_address) =  top_collector.docs().first() {
-        let doc = searcher.doc(doc_address)?;
+    if let Some(doc_address) = top_collector.docs().first() {
+        let doc = searcher.doc(*doc_address)?;
         Ok(Some(doc))
     } else {
         // no doc matching this ID.
@@ -41,7 +40,6 @@ 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
@@ -126,7 +124,6 @@ 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 +140,4 @@ fn main() -> tantivy::Result<()> {
     );
 
     Ok(())
-}
+}

examples/faceted_search.rs

@@ -22,60 +22,60 @@ 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 = SchemaBuilder::default();
 
-  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();
+    searcher.search(&AllQuery, &mut 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)
-    ]
-  );
+    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),
+        ]
+    );
 
-  Ok(())
+    Ok(())
 }
 
 use tempdir::TempDir;

examples/iterating_docs_and_positions.rs

@@ -7,18 +7,15 @@
 // 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();
@@ -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 +1,71 @@
+// # 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::TopCollector;
+use tantivy::query::QueryParser;
+use tantivy::schema::*;
+use tantivy::Index;
+use tantivy::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 = SchemaBuilder::default();
+    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 mut top_collector = TopCollector::with_limit(10);
+    searcher.search(&*query, &mut top_collector)?;
+
+    let snippet_generator = SnippetGenerator::new(&searcher, &*query, body)?;
+
+    let doc_addresses = top_collector.docs();
+    for doc_address in doc_addresses {
+        let doc = searcher.doc(doc_address)?;
+        let snippet = snippet_generator.snippet_from_doc(&doc);
+        println!("title: {}", doc.get_first(title).unwrap().text().unwrap());
+        println!("snippet: {}", snippet.to_html());
+    }
+
+    Ok(())
+}

examples/stop_words.rs

@@ -22,72 +22,71 @@ 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 = SchemaBuilder::default();
 
-  // 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 +94,28 @@ 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\"")?;
 
-  // this will have hits
-  let query = query_parser.parse_query("is")?;
+    let mut top_collector = TopCollector::with_limit(10);
 
-  let mut top_collector = TopCollector::with_limit(10);
+    searcher.search(&*query, &mut top_collector)?;
 
-  searcher.search(&*query, &mut top_collector)?;
+    let doc_addresses = top_collector.docs();
 
-  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));
+    }
 
-  for doc_address in doc_addresses {
-    let retrieved_doc = searcher.doc(&doc_address)?;
-    println!("{}", schema.to_json(&retrieved_doc));
-  }
-
-  Ok(())
+    Ok(())
 }
-
-use tempdir::TempDir;

run-tests.sh

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

src/collector/facet_collector.rs

@@ -342,16 +342,19 @@ impl FacetCollector {
     pub fn harvest(mut self) -> FacetCounts {
         self.finalize_segment();
 
-        let collapsed_facet_ords: Vec<&[u64]> = self.segment_counters
+        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
+        let collapsed_facet_counts: Vec<&[u64]> = self
+            .segment_counters
             .iter()
             .map(|segment_counter| &segment_counter.facet_counts[..])
             .collect();
 
-        let facet_streams = self.segment_counters
+        let facet_streams = self
+            .segment_counters
             .iter()
             .map(|seg_counts| seg_counts.facet_reader.facet_dict().range().into_stream())
             .collect::<Vec<_>>();
@@ -374,10 +377,8 @@ impl FacetCollector {
                             } else {
                                 collapsed_facet_counts[seg_ord][collapsed_term_id]
                             }
-                        })
-                        .unwrap_or(0)
-                })
-                .sum();
+                        }).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
@@ -402,7 +403,8 @@ impl Collector for FacetCollector {
 
     fn collect(&mut self, doc: DocId, _: Score) {
         let facet_reader: &mut FacetReader = unsafe {
-            &mut *self.ff_reader
+            &mut *self
+                .ff_reader
                 .as_ref()
                 .expect("collect() was called before set_segment. This should never happen.")
                 .get()
@@ -476,9 +478,8 @@ 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 {
@@ -526,8 +527,7 @@ 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());
@@ -554,7 +554,8 @@ mod tests {
                     ("/top1/mid1", 50),
                     ("/top1/mid2", 50),
                     ("/top1/mid3", 50),
-                ].iter()
+                ]
+                    .iter()
                     .map(|&(facet_str, count)| (String::from(facet_str), count))
                     .collect::<Vec<_>>()
             );
@@ -618,9 +619,13 @@ 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();
+            }).map(|mut doc| {
+                doc.add_facet(
+                    facet_field,
+                    &format!("/facet/{}", thread_rng().sample(&uniform)),
+                );
+                doc
+            }).collect();
         thread_rng().shuffle(&mut docs[..]);
 
         let mut index_writer = index.writer_with_num_threads(1, 3_000_000).unwrap();

src/collector/mod.rs

@@ -15,7 +15,14 @@ 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::TopScoreCollector;
+#[deprecated]
+pub use self::top_score_collector::TopScoreCollector as TopCollector;
+
+mod top_field_collector;
+pub use self::top_field_collector::TopFieldCollector;
 
 mod facet_collector;
 pub use self::facet_collector::FacetCollector;

src/collector/multi_collector.rs

@@ -100,11 +100,11 @@ impl<'a> Collector for MultiCollector<'a> {
 mod tests {
 
     use super::*;
-    use collector::{Collector, CountCollector, TopCollector};
+    use collector::{Collector, CountCollector, TopScoreCollector};
 
     #[test]
     fn test_multi_collector() {
-        let mut top_collector = TopCollector::with_limit(2);
+        let mut top_collector = TopScoreCollector::with_limit(2);
         let mut count_collector = CountCollector::default();
         {
             let mut collectors =

src/collector/top_collector.rs

@@ -1,115 +1,61 @@
-use super::Collector;
 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.
+/// 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.
 #[derive(Clone, Copy)]
-struct GlobalScoredDoc {
-    score: Score,
+pub struct ComparableDoc<T> {
+    feature: T,
     doc_address: DocAddress,
 }
 
-impl PartialOrd for GlobalScoredDoc {
-    fn partial_cmp(&self, other: &GlobalScoredDoc) -> Option<Ordering> {
+impl<T: PartialOrd> PartialOrd for ComparableDoc<T> {
+    fn partial_cmp(&self, other: &Self) -> Option<Ordering> {
         Some(self.cmp(other))
     }
 }
 
-impl Ord for GlobalScoredDoc {
+impl<T: PartialOrd> Ord for ComparableDoc<T> {
     #[inline]
-    fn cmp(&self, other: &GlobalScoredDoc) -> Ordering {
+    fn cmp(&self, other: &Self) -> Ordering {
         other
-            .score
-            .partial_cmp(&self.score)
+            .feature
+            .partial_cmp(&self.feature)
             .unwrap_or_else(|| other.doc_address.cmp(&self.doc_address))
     }
 }
 
-impl PartialEq for GlobalScoredDoc {
-    fn eq(&self, other: &GlobalScoredDoc) -> bool {
+impl<T: PartialOrd> PartialEq for ComparableDoc<T> {
+    fn eq(&self, other: &Self) -> bool {
         self.cmp(other) == Ordering::Equal
     }
 }
 
-impl Eq for GlobalScoredDoc {}
+impl<T: PartialOrd> Eq for ComparableDoc<T> {}
 
 /// The Top Collector keeps track of the K documents
-/// with the best scores.
+/// 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)`.
-///
-/// ```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 struct TopCollector<T> {
     limit: usize,
-    heap: BinaryHeap<GlobalScoredDoc>,
+    heap: BinaryHeap<ComparableDoc<T>>,
     segment_id: u32,
 }
 
-impl TopCollector {
+impl<T: PartialOrd + Clone> TopCollector<T> {
     /// 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.");
         }
@@ -125,23 +71,27 @@ impl TopCollector {
     /// Calling this method triggers the sort.
     /// The result of the sort is not cached.
     pub fn docs(&self) -> Vec<DocAddress> {
-        self.score_docs()
+        self.top_docs()
             .into_iter()
-            .map(|score_doc| score_doc.1)
+            .map(|(_feature, doc)| doc)
             .collect()
     }
 
-    /// Returns K best ScoredDocument sorted in decreasing order.
+    /// Returns K best FeatureDocuments 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 top_docs(&self) -> Vec<(T, DocAddress)> {
+        let mut feature_docs: Vec<ComparableDoc<T>> = self.heap.iter().cloned().collect();
+        feature_docs.sort();
+        feature_docs
             .into_iter()
-            .map(|GlobalScoredDoc { score, doc_address }| (score, doc_address))
-            .collect()
+            .map(
+                |ComparableDoc {
+                     feature,
+                     doc_address,
+                 }| (feature, doc_address),
+            ).collect()
     }
 
     /// Return true iff at least K documents have gone through
@@ -150,46 +100,45 @@ impl TopCollector {
     pub fn at_capacity(&self) -> bool {
         self.heap.len() >= self.limit
     }
-}
 
-impl Collector for TopCollector {
-    fn set_segment(&mut self, segment_id: SegmentLocalId, _: &SegmentReader) -> Result<()> {
+    /// Sets the segment local ID for the collector
+    pub fn set_segment_id(&mut self, segment_id: SegmentLocalId) {
         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.
+    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
+            let limit_doc: ComparableDoc<T> = self
+                .heap
                 .peek()
-                .expect("Top collector with size 0 is forbidden");
-            if limit_doc.score < score {
-                let mut mut_head = self.heap
+                .expect("Top collector with size 0 is forbidden")
+                .clone();
+            if limit_doc.feature < feature {
+                let mut mut_head = self
+                    .heap
                     .peek_mut()
                     .expect("Top collector with size 0 is forbidden");
-                mut_head.score = score;
+                mut_head.feature = feature;
                 mut_head.doc_address = DocAddress(self.segment_id, doc);
             }
         } else {
-            let wrapped_doc = GlobalScoredDoc {
-                score,
+            let wrapped_doc = ComparableDoc {
+                feature,
                 doc_address: DocAddress(self.segment_id, doc),
             };
             self.heap.push(wrapped_doc);
         }
     }
-
-    fn requires_scoring(&self) -> bool {
-        true
-    }
 }
 
 #[cfg(test)]
 mod tests {
-
     use super::*;
-    use collector::Collector;
     use DocId;
     use Score;
 
@@ -201,7 +150,7 @@ mod tests {
         top_collector.collect(5, 0.3);
         assert!(!top_collector.at_capacity());
         let score_docs: Vec<(Score, DocId)> = top_collector
-            .score_docs()
+            .top_docs()
             .into_iter()
             .map(|(score, doc_address)| (score, doc_address.doc()))
             .collect();
@@ -219,7 +168,7 @@ mod tests {
         assert!(top_collector.at_capacity());
         {
             let score_docs: Vec<(Score, DocId)> = top_collector
-                .score_docs()
+                .top_docs()
                 .into_iter()
                 .map(|(score, doc_address)| (score, doc_address.doc()))
                 .collect();
@@ -238,7 +187,7 @@ mod tests {
     #[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 +1,263 @@
+use super::Collector;
+use collector::top_collector::TopCollector;
+use fastfield::FastFieldReader;
+use fastfield::FastValue;
+use schema::Field;
+use DocAddress;
+use DocId;
+use Result;
+use Score;
+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::{SchemaBuilder, TEXT, FAST};
+/// use tantivy::{Index, Result, DocId};
+/// use tantivy::collector::TopFieldCollector;
+/// 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 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)?;
+///         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()?;
+///     let searcher = index.searcher();
+///
+///     {
+///	        let mut top_collector = TopFieldCollector::with_limit(rating, 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<(u64, DocId)> = top_collector
+///           .top_docs()
+///           .into_iter()
+///           .map(|(field, doc_address)| (field, doc_address.doc()))
+///           .collect();
+///
+///         assert_eq!(score_docs, vec![(97u64, 1), (80, 3)]);
+///     }
+///
+///     Ok(())
+/// }
+/// ```
+pub struct TopFieldCollector<T: FastValue> {
+    field: Field,
+    collector: TopCollector<T>,
+    fast_field: Option<FastFieldReader<T>>,
+}
+
+impl<T: FastValue + PartialOrd + Clone> TopFieldCollector<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 fn with_limit(field: Field, limit: usize) -> Self {
+        TopFieldCollector {
+            field,
+            collector: TopCollector::with_limit(limit),
+            fast_field: None,
+        }
+    }
+
+    /// Returns K best documents sorted the given field name in decreasing order.
+    ///
+    /// Calling this method triggers the sort.
+    /// The result of the sort is not cached.
+    pub fn docs(&self) -> Vec<DocAddress> {
+        self.collector.docs()
+    }
+
+    /// Returns K best FieldDocuments sorted in decreasing order.
+    ///
+    /// Calling this method triggers the sort.
+    /// The result of the sort is not cached.
+    pub fn top_docs(&self) -> Vec<(T, DocAddress)> {
+        self.collector.top_docs()
+    }
+
+    /// Return true iff at least K documents have gone through
+    /// the collector.
+    #[inline]
+    pub fn at_capacity(&self) -> bool {
+        self.collector.at_capacity()
+    }
+}
+
+impl<T: FastValue + PartialOrd + Clone> Collector for TopFieldCollector<T> {
+    fn set_segment(&mut self, segment_id: u32, segment: &SegmentReader) -> Result<()> {
+        self.collector.set_segment_id(segment_id);
+        self.fast_field = Some(segment.fast_field_reader(self.field)?);
+        Ok(())
+    }
+
+    fn collect(&mut self, doc: DocId, _score: Score) {
+        let field_value = self
+            .fast_field
+            .as_ref()
+            .expect("collect() was called before set_segment. This should never happen.")
+            .get(doc);
+        self.collector.collect(doc, field_value);
+    }
+
+    fn requires_scoring(&self) -> bool {
+        false
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+    use query::Query;
+    use query::QueryParser;
+    use schema::Field;
+    use schema::IntOptions;
+    use schema::Schema;
+    use schema::{SchemaBuilder, FAST, TEXT};
+    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 = SchemaBuilder::new();
+        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 mut top_collector = TopFieldCollector::with_limit(size, 4);
+        searcher.search(&*query, &mut top_collector).unwrap();
+        assert!(!top_collector.at_capacity());
+
+        let score_docs: Vec<(u64, DocId)> = top_collector
+            .top_docs()
+            .into_iter()
+            .map(|(field, doc_address)| (field, doc_address.doc()))
+            .collect();
+        assert_eq!(score_docs, vec![(64, 1), (16, 2), (12, 0)]);
+    }
+
+    #[test]
+    #[should_panic]
+    fn test_field_does_not_exist() {
+        let mut schema_builder = SchemaBuilder::new();
+        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 segment = searcher.segment_reader(0);
+        let mut top_collector: TopFieldCollector<u64> = TopFieldCollector::with_limit(Field(2), 4);
+        let _ = top_collector.set_segment(0, segment);
+    }
+
+    #[test]
+    fn test_field_not_fast_field() {
+        let mut schema_builder = SchemaBuilder::new();
+        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 mut top_collector: TopFieldCollector<u64> = TopFieldCollector::with_limit(size, 4);
+        assert_matches!(
+            top_collector.set_segment(0, segment),
+            Err(TantivyError::FastFieldError(_))
+        );
+    }
+
+    #[test]
+    #[should_panic]
+    fn test_collect_before_set_segment() {
+        let mut top_collector: TopFieldCollector<u64> = TopFieldCollector::with_limit(Field(0), 4);
+        top_collector.collect(0, 0f32);
+    }
+
+    #[test]
+    #[should_panic]
+    fn test_top_0() {
+        let _: TopFieldCollector<u64> = TopFieldCollector::with_limit(Field(0), 0);
+    }
+
+    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 +1,187 @@
+use super::Collector;
+use collector::top_collector::TopCollector;
+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::schema::{SchemaBuilder, TEXT};
+/// use tantivy::{Index, Result, DocId, Score};
+/// use tantivy::collector::TopScoreCollector;
+/// 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 = TopScoreCollector::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
+///           .top_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 TopScoreCollector {
+    collector: TopCollector<Score>,
+}
+
+impl TopScoreCollector {
+    /// 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) -> TopScoreCollector {
+        TopScoreCollector {
+            collector: TopCollector::with_limit(limit),
+        }
+    }
+
+    /// Returns K best scored 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.collector.docs()
+    }
+
+    /// Returns K best ScoredDocuments sorted in decreasing order.
+    ///
+    /// Calling this method triggers the sort.
+    /// The result of the sort is not cached.
+    pub fn top_docs(&self) -> Vec<(Score, DocAddress)> {
+        self.collector.top_docs()
+    }
+
+    /// Returns K best ScoredDocuments sorted in decreasing order.
+    ///
+    /// Calling this method triggers the sort.
+    /// The result of the sort is not cached.
+    #[deprecated]
+    pub fn score_docs(&self) -> Vec<(Score, DocAddress)> {
+        self.collector.top_docs()
+    }
+
+    /// Return true iff at least K documents have gone through
+    /// the collector.
+    #[inline]
+    pub fn at_capacity(&self) -> bool {
+        self.collector.at_capacity()
+    }
+}
+
+impl Collector for TopScoreCollector {
+    fn set_segment(&mut self, segment_id: SegmentLocalId, _: &SegmentReader) -> Result<()> {
+        self.collector.set_segment_id(segment_id);
+        Ok(())
+    }
+
+    fn collect(&mut self, doc: DocId, score: Score) {
+        self.collector.collect(doc, score);
+    }
+
+    fn requires_scoring(&self) -> bool {
+        true
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+    use collector::Collector;
+    use DocId;
+    use Score;
+
+    #[test]
+    fn test_top_collector_not_at_capacity() {
+        let mut top_collector = TopScoreCollector::with_limit(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
+            .top_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)]);
+    }
+
+    #[test]
+    fn test_top_collector_at_capacity() {
+        let mut top_collector = TopScoreCollector::with_limit(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
+                .top_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]);
+        }
+    }
+
+    #[test]
+    #[should_panic]
+    fn test_top_0() {
+        TopScoreCollector::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::PerFieldSpaceUsage;
+use space_usage::FieldUsage;
 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 +168,16 @@ 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/index.rs

@@ -1,34 +1,36 @@
-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::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)?;
@@ -47,6 +49,11 @@ pub struct Index {
 }
 
 impl Index {
+    /// Examines the director to see if it contains an index
+    pub fn exists<Dir: Directory>(dir: &Dir) -> bool {
+        dir.exists(&META_FILEPATH)
+    }
+
     /// Creates a new index using the `RAMDirectory`.
     ///
     /// The index will be allocated in anonymous memory.
@@ -63,9 +70,28 @@ 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 +113,8 @@ 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);
@@ -113,6 +141,27 @@ 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 +205,7 @@ 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 +243,8 @@ 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 +279,8 @@ 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 +293,18 @@ 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 +313,7 @@ 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(())
@@ -295,3 +351,73 @@ impl Clone for Index {
         }
     }
 }
+
+#[cfg(test)]
+mod tests {
+    use schema::{Schema, SchemaBuilder, INT_INDEXED, TEXT};
+    use Index;
+    use directory::RAMDirectory;
+
+    #[test]
+    fn test_indexer_for_field() {
+        let mut schema_builder = SchemaBuilder::default();
+        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(), SchemaBuilder::default().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, SchemaBuilder::default().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 = SchemaBuilder::default();
+        let _ = schema_builder.add_u64_field("num_likes", INT_INDEXED);
+        schema_builder.build()
+    }
+}

src/core/index_meta.rs

@@ -58,7 +58,7 @@ mod tests {
         };
         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

@@ -33,10 +33,4 @@ 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

@@ -5,10 +5,12 @@ use query::Query;
 use schema::Document;
 use schema::Schema;
 use schema::{Field, Term};
+use space_usage::SearcherSpaceUsage;
 use std::fmt;
 use std::sync::Arc;
 use termdict::TermMerger;
 use DocAddress;
+use Index;
 use Result;
 
 /// Holds a list of `SegmentReader`s ready for search.
@@ -18,23 +20,35 @@ use Result;
 ///
 pub struct Searcher {
     schema: Schema,
+    index: Index,
     segment_readers: Vec<SegmentReader>,
 }
 
 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 {
         Searcher {
             schema,
+            index,
             segment_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;
+    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)
     }
@@ -48,7 +62,7 @@ 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 +71,9 @@ 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
@@ -78,12 +93,22 @@ impl Searcher {
 
     /// 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 +123,8 @@ 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 +134,8 @@ 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 +4,6 @@ use core::InvertedIndexReader;
 use core::Segment;
 use core::SegmentComponent;
 use core::SegmentId;
-use core::SegmentMeta;
 use error::TantivyError;
 use fastfield::DeleteBitSet;
 use fastfield::FacetReader;
@@ -17,6 +16,7 @@ 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,
@@ -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)
     }
@@ -225,6 +228,8 @@ impl SegmentReader {
         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)?;
 
@@ -260,7 +265,8 @@ 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,
@@ -282,7 +288,8 @@ 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 +313,28 @@ 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,
@@ -372,6 +382,21 @@ 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.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 +416,7 @@ 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,
         }

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;
@@ -19,6 +20,15 @@ pub enum TantivyError {
     /// File already exists, this is a problem when we try to write into a new file.
     #[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)]
     IOError(#[cause] IOError),
@@ -46,48 +56,46 @@ pub enum TantivyError {
 
 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 +106,7 @@ impl From<OpenWriteError> for TantivyError {
                 TantivyError::FileAlreadyExists(filepath)
             }
             OpenWriteError::IOError(io_error) => TantivyError::IOError(io_error),
-        }.into()
+        }
     }
 }
 
@@ -106,11 +114,11 @@ 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 +126,6 @@ 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/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

@@ -370,7 +370,7 @@ mod tests {
     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();
+        let mut permutation: Vec<u64> = (0u64..100_000u64).collect();
         rng.shuffle(&mut permutation);
         permutation
     }

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

@@ -11,7 +11,6 @@ use schema::SchemaBuilder;
 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();

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/functional_test.rs

@@ -1,8 +1,8 @@
 use rand::thread_rng;
 use std::collections::HashSet;
 
-use rand::Rng;
 use rand::distributions::Range;
+use rand::Rng;
 use schema::*;
 use Index;
 use Searcher;

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

@@ -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()));
 
@@ -385,11 +390,9 @@ impl IndexWriter {
             .name(format!(
                 "indexing thread {} for gen {}",
                 self.worker_id, generation
-            ))
-            .spawn(move || {
+            )).spawn(move || {
                 loop {
-                    let mut document_iterator =
-                        document_receiver_clone.clone().into_iter().peekable();
+                    let mut document_iterator = document_receiver_clone.clone().peekable();
 
                     // the peeking here is to avoid
                     // creating a new segment's files
@@ -488,7 +491,8 @@ 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.");
 
@@ -657,11 +661,26 @@ mod tests {
         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::SchemaBuilder::default();
+        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();
@@ -843,4 +862,32 @@ 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::SchemaBuilder::default();
+        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.
@@ -666,8 +664,7 @@ mod tests {
                 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);
@@ -769,24 +766,24 @@ mod tests {
                 );
             }
             {
-                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>| {
@@ -821,8 +818,7 @@ mod tests {
         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);

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(())
@@ -336,8 +336,7 @@ 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()

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"]
 
@@ -96,7 +93,7 @@
 //! // most relevant doc ids...
 //! let doc_addresses = top_collector.docs();
 //! for doc_address in doc_addresses {
-//!     let retrieved_doc = searcher.doc(&doc_address)?;
+//!     let retrieved_doc = searcher.doc(doc_address)?;
 //!     println!("{}", schema.to_json(&retrieved_doc));
 //! }
 //!
@@ -133,7 +130,6 @@ extern crate bit_set;
 extern crate bitpacking;
 extern crate byteorder;
 
-#[macro_use]
 extern crate combine;
 
 extern crate crossbeam;
@@ -143,6 +139,7 @@ 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;
@@ -155,6 +152,8 @@ extern crate tempdir;
 extern crate tempfile;
 extern crate uuid;
 
+
+
 #[cfg(test)]
 #[macro_use]
 extern crate matches;
@@ -165,14 +164,19 @@ 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 +185,10 @@ 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 +213,13 @@ 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;
+
 mod docset;
 pub use self::docset::{DocSet, SkipResult};
 
@@ -261,12 +272,12 @@ 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
     }
 }
@@ -892,11 +903,11 @@ 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]

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)

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,
@@ -71,8 +71,7 @@ pub mod tests {
         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];
@@ -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");

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)
     }
@@ -126,7 +126,6 @@ impl SegmentPostings {
 fn exponential_search(target: u32, arr: &[u32]) -> (usize, usize) {
     let mut start = 0;
     let end = arr.len();
-    debug_assert!(target >= arr[start]);
     debug_assert!(target <= arr[end - 1]);
     let mut jump = 1;
     loop {
@@ -151,7 +150,11 @@ 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 +182,20 @@ 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;
@@ -202,22 +204,25 @@ 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;
         }
 
         // we're in the right block now, start with an exponential search
         let block_docs = self.block_cursor.docs();
+        let new_cur = self
+            .cur
+            .wrapping_add(search_within_block(&block_docs[self.cur..], target));
 
-        debug_assert!(target >= self.doc());
-        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 +234,9 @@ 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 +335,10 @@ 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 +353,7 @@ 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 +361,7 @@ 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 +370,10 @@ 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 +457,6 @@ impl BlockSegmentPostings {
         self.doc_decoder.output_len
     }
 
-
     /// position on a block that may contains `doc_id`.
     /// Always advance the current block.
     ///
@@ -461,9 +467,7 @@ 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 +476,11 @@ 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 +490,9 @@ 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 +522,8 @@ impl BlockSegmentPostings {
                 }
             }
             self.num_vint_docs = 0;
-            return self.docs()
+            return self
+                .docs()
                 .last()
                 .map(|last_doc| {
                     if *last_doc >= target_doc {
@@ -526,8 +531,7 @@ impl BlockSegmentPostings {
                     } else {
                         BlockSegmentPostingsSkipResult::Terminated
                     }
-                })
-                .unwrap_or(BlockSegmentPostingsSkipResult::Terminated);
+                }).unwrap_or(BlockSegmentPostingsSkipResult::Terminated);
         }
         BlockSegmentPostingsSkipResult::Terminated
     }
@@ -538,11 +542,11 @@ 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 +556,9 @@ 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 +598,6 @@ impl BlockSegmentPostings {
             doc_offset: 0,
             doc_freq: 0,
 
-
             remaining_data: OwnedRead::new(vec![]),
             skip_reader: SkipReader::new(OwnedRead::new(vec![]), IndexRecordOption::Basic),
         }
@@ -616,7 +619,9 @@ 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;
@@ -625,10 +630,10 @@ mod tests {
     use schema::IndexRecordOption;
     use schema::SchemaBuilder;
     use schema::Term;
+    use super::exponential_search;
     use schema::INT_INDEXED;
-    use super::BlockSegmentPostingsSkipResult;
     use DocId;
-    use super::search_within_block;
+    use SkipResult;
 
     #[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 +657,22 @@ mod tests {
             .enumerate()
             .filter(|&(_, ref val)| *val >= target)
             .next()
-            .unwrap().0
+            .unwrap()
+            .0
+    }
+
+    #[test]
+    fn test_exponentiel_search() {
+        assert_eq!(exponential_search(0, &[1,2]), (0, 1));
+        assert_eq!(exponential_search(1, &[1,2]), (0, 1));
+        assert_eq!(exponential_search(7, &[1,2,3,4,5,6,7,8,9,10,11]), (3,7));
     }
 
     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]) {
@@ -677,14 +692,14 @@ 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[..]);
         }
     }
 
     #[test]
     fn test_block_segment_postings() {
-        let mut block_segments = build_block_postings((0..100_000).collect::<Vec<u32>>());
+        let mut block_segments = build_block_postings(&(0..100_000).collect::<Vec<u32>>());
         let mut offset: u32 = 0u32;
         // checking that the block before calling advance is empty
         assert!(block_segments.docs().is_empty());
@@ -698,14 +713,45 @@ mod tests {
         }
     }
 
-    fn build_block_postings(docs: Vec<DocId>) -> BlockSegmentPostings {
+
+    #[test]
+    fn test_skip_right_at_new_block() {
+        let mut doc_ids = (0..128).collect::<Vec<u32>>();
+        doc_ids.push(129);
+        doc_ids.push(130);
+        {
+            let block_segments = build_block_postings(&doc_ids);
+            let mut docset = SegmentPostings::from_block_postings(block_segments, None);
+            assert_eq!(docset.skip_next(128), SkipResult::OverStep);
+            assert_eq!(docset.doc(), 129);
+            assert!(docset.advance());
+            assert_eq!(docset.doc(), 130);
+            assert!(!docset.advance());
+        }
+        {
+            let block_segments = build_block_postings(&doc_ids);
+            let mut docset = SegmentPostings::from_block_postings(block_segments, None);
+            assert_eq!(docset.skip_next(129), SkipResult::Reached);
+            assert_eq!(docset.doc(), 129);
+            assert!(docset.advance());
+            assert_eq!(docset.doc(), 130);
+            assert!(!docset.advance());
+        }
+        {
+            let block_segments = build_block_postings(&doc_ids);
+            let mut docset = SegmentPostings::from_block_postings(block_segments, None);
+            assert_eq!(docset.skip_next(131), SkipResult::End);
+        }
+    }
+
+    fn build_block_postings(docs: &[DocId]) -> BlockSegmentPostings {
         let mut schema_builder = SchemaBuilder::default();
         let int_field = schema_builder.add_u64_field("id", INT_INDEXED);
         let schema = schema_builder.build();
         let index = Index::create_in_ram(schema);
         let mut index_writer = index.writer_with_num_threads(1, 40_000_000).unwrap();
         let mut last_doc = 0u32;
-        for doc in docs {
+        for &doc in docs {
             for _ in last_doc..doc {
                 index_writer.add_document(doc!(int_field=>1u64));
             }
@@ -725,30 +771,47 @@ mod tests {
     #[test]
     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);
+            let mut block_postings = build_block_postings(&[3]);
+            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);
+        let mut block_postings = build_block_postings(&[3]);
+        assert_eq!(
+            block_postings.skip_to(4u32),
+            BlockSegmentPostingsSkipResult::Terminated
+        );
     }
 
-
     #[test]
     fn test_block_segment_postings_skip2() {
         let mut docs = vec![0];
         for i in 0..1300 {
             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));
+        let mut block_postings = build_block_postings(&docs[..]);
+        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]

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/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
         }
     }
 

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

@@ -69,7 +69,7 @@ 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 +81,13 @@ 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 +100,7 @@ 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 +108,7 @@ 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));

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
-    }
-}

src/query/fastfield_filter/fastfield_filter_weight.rs

@@ -1,29 +0,0 @@
-use query::Weight;
-use schema::Field;
-use query::fastfield_filter::RangeU64;
-use query::fastfield_filter::FastFieldFilterScorer;
-use SegmentReader;
-use query::Scorer;
-use TantivyError;
-use fastfield::FastFieldReader;
-
-pub struct FastFieldFilterWeight {
-    field: Field,
-    range: RangeU64,
-}
-
-impl FastFieldFilterWeight {
-    pub(crate) fn new(field: Field, range: RangeU64) -> FastFieldFilterWeight {
-        FastFieldFilterWeight {
-            field,
-            range
-        }
-    }
-}
-
-impl Weight for FastFieldFilterWeight {
-    fn scorer(&self, reader: &SegmentReader) -> Result<Box<Scorer>, TantivyError> {
-        let fastfield_reader: FastFieldReader<u64> = reader.fast_field_reader(self.field    )?;
-        Ok(Box::new(FastFieldFilterScorer::new(fastfield_reader, self.range.clone(), reader.max_doc())))
-    }
-}

src/query/fastfield_filter/mod.rs

@@ -1,47 +0,0 @@
-use std::collections::Bound;
-
-mod fastfield_filter_query;
-mod fastfield_filter_weight;
-mod fastfield_filter_scorer;
-
-
-pub use self::fastfield_filter_query::FastFieldFilterQuery;
-use self::fastfield_filter_weight::FastFieldFilterWeight;
-use self::fastfield_filter_scorer::FastFieldFilterScorer;
-
-#[derive(Debug, Clone)]
-pub(crate) struct RangeU64 {
-    pub low: Bound<u64>,
-    pub high: Bound<u64>,
-}
-
-
-impl RangeU64 {
-
-    fn match_high(&self, val: u64) -> bool {
-        match self.high {
-            Bound::Excluded(bound) =>
-                val < bound,
-            Bound::Included(bound) =>
-                val <= bound,
-            Bound::Unbounded =>
-                true
-        }
-    }
-
-    fn match_low(&self, val: u64) -> bool {
-        match self.high {
-            Bound::Excluded(bound) =>
-                bound < val,
-            Bound::Included(bound) =>
-                bound <= val,
-            Bound::Unbounded =>
-                true
-        }
-    }
-
-    pub fn contains(&self, val: u64) -> bool {
-        self.match_low(val) && self.match_high(val)
-    }
-
-}

src/query/fuzzy_query.rs

@@ -10,7 +10,7 @@ lazy_static! {
         let mut lev_builder_cache = HashMap::new();
         // TODO make population lazy on a `(distance, val)` basis
         for distance in 0..3 {
-            for &transposition in [false, true].iter() {
+            for &transposition in &[false, true] {
                 let lev_automaton_builder = LevenshteinAutomatonBuilder::new(distance, transposition);
                 lev_builder_cache.insert((distance, transposition), lev_automaton_builder);
             }
@@ -153,7 +153,7 @@ mod test {
 
             let fuzzy_query = FuzzyTermQuery::new(term, 1, true);
             searcher.search(&fuzzy_query, &mut collector).unwrap();
-            let scored_docs = collector.score_docs();
+            let scored_docs = collector.top_docs();
             assert_eq!(scored_docs.len(), 1, "Expected only 1 document");
             let (score, _) = scored_docs[0];
             assert_nearly_equals(1f32, score);

src/query/intersection.rs

@@ -26,10 +26,11 @@ pub fn intersect_scorers(mut scorers: Vec<Box<Scorer>>) -> Box<Scorer> {
         (Some(single_docset), None) => single_docset,
         (Some(left), Some(right)) => {
             {
-                if [&left, &right].into_iter().all(|scorer| {
+                let all_term_scorers = [&left, &right].into_iter().all(|scorer| {
                     let scorer_ref: &Scorer = (*scorer).borrow();
                     Downcast::<TermScorer>::is_type(scorer_ref)
-                }) {
+                });
+                if all_term_scorers {
                     let left = *Downcast::<TermScorer>::downcast(left).unwrap();
                     let right = *Downcast::<TermScorer>::downcast(right).unwrap();
                     return Box::new(Intersection {
@@ -40,12 +41,12 @@ pub fn intersect_scorers(mut scorers: Vec<Box<Scorer>>) -> Box<Scorer> {
                     });
                 }
             }
-            return Box::new(Intersection {
+            Box::new(Intersection {
                 left,
                 right,
                 others: scorers,
                 num_docsets,
-            });
+            })
         }
         _ => {
             unreachable!();
@@ -99,7 +100,7 @@ impl<TDocSet: DocSet, TOtherDocSet: DocSet> Intersection<TDocSet, TOtherDocSet>
 }
 
 impl<TDocSet: DocSet, TOtherDocSet: DocSet> DocSet for Intersection<TDocSet, TOtherDocSet> {
-    #[allow(never_loop)]
+    #[cfg_attr(feature = "cargo-clippy", allow(clippy::never_loop))]
     fn advance(&mut self) -> bool {
         let (left, right) = (&mut self.left, &mut self.right);
 

src/query/mod.rs

@@ -3,11 +3,11 @@ Query
 */
 
 mod all_query;
-mod empty_query;
 mod automaton_weight;
 mod bitset;
 mod bm25;
 mod boolean_query;
+mod empty_query;
 mod exclude;
 mod fuzzy_query;
 mod intersection;
@@ -22,13 +22,11 @@ mod scorer;
 mod term_query;
 mod union;
 mod weight;
-mod fastfield_filter;
 
 #[cfg(test)]
 mod vec_docset;
 
 pub(crate) mod score_combiner;
-
 pub use self::intersection::Intersection;
 pub use self::union::Union;
 
@@ -36,10 +34,10 @@ pub use self::union::Union;
 pub use self::vec_docset::VecDocSet;
 
 pub use self::all_query::{AllQuery, AllScorer, AllWeight};
-pub use self::empty_query::{EmptyQuery, EmptyWeight, EmptyScorer};
 pub use self::automaton_weight::AutomatonWeight;
 pub use self::bitset::BitSetDocSet;
 pub use self::boolean_query::BooleanQuery;
+pub use self::empty_query::{EmptyQuery, EmptyScorer, EmptyWeight};
 pub use self::exclude::Exclude;
 pub use self::fuzzy_query::FuzzyTermQuery;
 pub use self::intersection::intersect_scorers;
@@ -55,4 +53,53 @@ pub use self::scorer::ConstScorer;
 pub use self::scorer::Scorer;
 pub use self::term_query::TermQuery;
 pub use self::weight::Weight;
-pub use self::fastfield_filter::FastFieldFilterQuery;
+
+#[cfg(test)]
+mod tests {
+    use Index;
+    use schema::{SchemaBuilder, TEXT};
+    use query::QueryParser;
+    use Term;
+    use std::collections::BTreeSet;
+
+    #[test]
+    fn test_query_terms() {
+        let mut schema_builder = SchemaBuilder::default();
+        let text_field = schema_builder.add_text_field("text", TEXT);
+        let schema = schema_builder.build();
+        let index = Index::create_in_ram(schema);
+        let query_parser = QueryParser::for_index(&index, vec![text_field]);
+        let term_a = Term::from_field_text(text_field, "a");
+        let term_b = Term::from_field_text(text_field, "b");
+        {
+            let mut terms_set: BTreeSet<Term> = BTreeSet::new();
+            query_parser.parse_query("a").unwrap().query_terms(&mut terms_set);
+            let terms: Vec<&Term> = terms_set.iter().collect();
+            assert_eq!(vec![&term_a], terms);
+        }
+        {
+            let mut terms_set: BTreeSet<Term> = BTreeSet::new();
+            query_parser.parse_query("a b").unwrap().query_terms(&mut terms_set);
+            let terms: Vec<&Term> = terms_set.iter().collect();
+            assert_eq!(vec![&term_a, &term_b], terms);
+        }
+        {
+            let mut terms_set: BTreeSet<Term> = BTreeSet::new();
+            query_parser.parse_query("\"a b\"").unwrap().query_terms(&mut terms_set);
+            let terms: Vec<&Term> = terms_set.iter().collect();
+            assert_eq!(vec![&term_a, &term_b], terms);
+        }
+        {
+            let mut terms_set: BTreeSet<Term> = BTreeSet::new();
+            query_parser.parse_query("a a a a a").unwrap().query_terms(&mut terms_set);
+            let terms: Vec<&Term> = terms_set.iter().collect();
+            assert_eq!(vec![&term_a], terms);
+        }
+        {
+            let mut terms_set: BTreeSet<Term> = BTreeSet::new();
+            query_parser.parse_query("a -b").unwrap().query_terms(&mut terms_set);
+            let terms: Vec<&Term> = terms_set.iter().collect();
+            assert_eq!(vec![&term_a, &term_b], terms);
+        }
+    }
+}

src/query/occur.rs

@@ -12,3 +12,38 @@ pub enum Occur {
     /// search.
     MustNot,
 }
+
+impl Occur {
+    /// Returns the one-char prefix symbol for this `Occur`.
+    /// - `Should` => '?',
+    /// - `Must` => '+'
+    /// - `Not` => '-'
+    pub fn to_char(self) -> char {
+        match self {
+            Occur::Should => '?',
+            Occur::Must => '+',
+            Occur::MustNot => '-',
+        }
+    }
+}
+
+/// Compose two occur values.
+pub fn compose_occur(left: Occur, right: Occur) -> Occur {
+    match left {
+        Occur::Should => right,
+        Occur::Must => {
+            if right == Occur::MustNot {
+                Occur::MustNot
+            } else {
+                Occur::Must
+            }
+        }
+        Occur::MustNot => {
+            if right == Occur::MustNot {
+                Occur::Must
+            } else {
+                Occur::MustNot
+            }
+        }
+    }
+}

src/query/phrase_query/phrase_query.rs

@@ -5,6 +5,7 @@ use query::bm25::BM25Weight;
 use query::Query;
 use query::Weight;
 use schema::{Field, Term};
+use std::collections::BTreeSet;
 use Result;
 
 /// `PhraseQuery` matches a specific sequence of words.
@@ -107,4 +108,10 @@ impl Query for PhraseQuery {
             )))
         }
     }
+
+    fn query_terms(&self, term_set: &mut BTreeSet<Term>) {
+        for (_, query_term) in &self.phrase_terms {
+            term_set.insert(query_term.clone());
+        }
+    }
 }

src/query/phrase_query/phrase_scorer.rs

@@ -124,7 +124,8 @@ impl<TPostings: Postings> PhraseScorer<TPostings> {
         fieldnorm_reader: FieldNormReader,
         score_needed: bool,
     ) -> PhraseScorer<TPostings> {
-        let max_offset = term_postings.iter()
+        let max_offset = term_postings
+            .iter()
             .map(|&(offset, _)| offset)
             .max()
             .unwrap_or(0);
@@ -133,8 +134,7 @@ impl<TPostings: Postings> PhraseScorer<TPostings> {
             .into_iter()
             .map(|(offset, postings)| {
                 PostingsWithOffset::new(postings, (max_offset - offset) as u32)
-            })
-            .collect::<Vec<_>>();
+            }).collect::<Vec<_>>();
         PhraseScorer {
             intersection_docset: Intersection::new(postings_with_offsets),
             num_docsets,

src/query/query.rs

@@ -2,9 +2,11 @@ use super::Weight;
 use collector::Collector;
 use core::searcher::Searcher;
 use downcast;
+use std::collections::BTreeSet;
 use std::fmt;
 use Result;
 use SegmentLocalId;
+use Term;
 
 /// The `Query` trait defines a set of documents and a scoring method
 /// for those documents.
@@ -58,6 +60,10 @@ pub trait Query: QueryClone + downcast::Any + fmt::Debug {
         Ok(result)
     }
 
+    /// Extract all of the terms associated to the query and insert them in the
+    /// term set given in arguments.
+    fn query_terms(&self, _term_set: &mut BTreeSet<Term>) {}
+
     /// Search works as follows :
     ///
     /// First the weight object associated to the query is created.

src/query/query_parser/query_grammar.rs

@@ -1,6 +1,12 @@
+#![cfg_attr(feature = "cargo-clippy", allow(clippy::unneeded_field_pattern))]
+#![cfg_attr(feature = "cargo-clippy", allow(clippy::toplevel_ref_arg))]
+
 use super::user_input_ast::*;
 use combine::char::*;
+use combine::error::StreamError;
+use combine::stream::StreamErrorFor;
 use combine::*;
+use query::occur::Occur;
 use query::query_parser::user_input_ast::UserInputBound;
 
 parser! {
@@ -17,18 +23,25 @@ parser! {
     fn word[I]()(I) -> String
     where [I: Stream<Item = char>] {
         many1(satisfy(|c: char| c.is_alphanumeric()))
+               .and_then(|s: String| {
+                   match s.as_str() {
+                     "OR" => Err(StreamErrorFor::<I>::unexpected_static_message("OR")),
+                     "AND" => Err(StreamErrorFor::<I>::unexpected_static_message("AND")),
+                     "NOT" => Err(StreamErrorFor::<I>::unexpected_static_message("NOT")),
+                     _ => Ok(s)
+                   }
+               })
     }
 }
 
 parser! {
-    fn literal[I]()(I) -> UserInputAST
+    fn literal[I]()(I) -> UserInputLeaf
     where [I: Stream<Item = char>]
     {
         let term_val = || {
             let phrase = (char('"'), many1(satisfy(|c| c != '"')), char('"')).map(|(_, s, _)| s);
             phrase.or(word())
         };
-
         let term_val_with_field = negative_number().or(term_val());
         let term_query =
             (field(), char(':'), term_val_with_field).map(|(field_name, _, phrase)| UserInputLiteral {
@@ -41,7 +54,7 @@ parser! {
         });
         try(term_query)
             .or(term_default_field)
-            .map(UserInputAST::from)
+            .map(UserInputLeaf::from)
     }
 }
 
@@ -55,7 +68,14 @@ parser! {
 }
 
 parser! {
-    fn range[I]()(I) -> UserInputAST
+    fn spaces1[I]()(I) -> ()
+    where [I: Stream<Item = char>] {
+        skip_many1(space())
+    }
+}
+
+parser! {
+    fn range[I]()(I) -> UserInputLeaf
     where [I: Stream<Item = char>] {
         let term_val = || {
             word().or(negative_number()).or(char('*').map(|_| "*".to_string()))
@@ -77,7 +97,7 @@ parser! {
             string("TO"),
             spaces(),
             upper_bound,
-        ).map(|(field, lower, _, _, _, upper)| UserInputAST::Range {
+        ).map(|(field, lower, _, _, _, upper)| UserInputLeaf::Range {
                 field,
                 lower,
                 upper
@@ -88,13 +108,50 @@ parser! {
 parser! {
     fn leaf[I]()(I) -> UserInputAST
     where [I: Stream<Item = char>] {
-         (char('-'), leaf())
-        .map(|(_, expr)| UserInputAST::Not(Box::new(expr)))
-        .or((char('+'), leaf()).map(|(_, expr)| UserInputAST::Must(Box::new(expr))))
+         (char('-'), leaf()).map(|(_, expr)| expr.unary(Occur::MustNot) )
+        .or((char('+'), leaf()).map(|(_, expr)| expr.unary(Occur::Must) ))
         .or((char('('), parse_to_ast(), char(')')).map(|(_, expr, _)| expr))
-        .or(char('*').map(|_| UserInputAST::All))
-        .or(try(range()))
-        .or(literal())
+        .or(char('*').map(|_| UserInputAST::from(UserInputLeaf::All) ))
+        .or(try(
+            (string("NOT"), spaces1(), leaf()).map(|(_, _, expr)| expr.unary(Occur::MustNot))
+            )
+         )
+        .or(try(
+            range().map(UserInputAST::from)
+            )
+        )
+        .or(literal().map(|leaf| UserInputAST::Leaf(Box::new(leaf))))
+    }
+}
+
+enum BinaryOperand {
+    Or,
+    And,
+}
+
+parser! {
+    fn binary_operand[I]()(I) -> BinaryOperand
+    where [I: Stream<Item = char>] {
+        (spaces1(),
+         (
+            string("AND").map(|_| BinaryOperand::And)
+           .or(string("OR").map(|_| BinaryOperand::Or))
+         ),
+         spaces1()).map(|(_, op,_)| op)
+    }
+}
+
+enum Element {
+    SingleEl(UserInputAST),
+    NormalDisjunctive(Vec<Vec<UserInputAST>>),
+}
+
+impl Element {
+    pub fn into_dnf(self) -> Vec<Vec<UserInputAST>> {
+        match self {
+            Element::NormalDisjunctive(conjunctions) => conjunctions,
+            Element::SingleEl(el) => vec![vec![el]],
+        }
     }
 }
 
@@ -102,14 +159,56 @@ parser! {
     pub fn parse_to_ast[I]()(I) -> UserInputAST
     where [I: Stream<Item = char>]
     {
-        sep_by(leaf(), spaces())
-        .map(|subqueries: Vec<UserInputAST>| {
-            if subqueries.len() == 1 {
-                subqueries.into_iter().next().unwrap()
-            } else {
-                UserInputAST::Clause(subqueries.into_iter().map(Box::new).collect())
-            }
-        })
+        (
+            try(
+                chainl1(
+                    leaf().map(Element::SingleEl),
+                    binary_operand().map(|op: BinaryOperand|
+                        move |left: Element, right: Element| {
+                            let mut dnf = left.into_dnf();
+                            if let Element::SingleEl(el) = right {
+                                match op {
+                                    BinaryOperand::And => {
+                                        if let Some(last) = dnf.last_mut() {
+                                            last.push(el);
+                                        }
+                                    }
+                                    BinaryOperand::Or => {
+                                        dnf.push(vec!(el));
+                                    }
+                                }
+                            } else {
+                                unreachable!("Please report.")
+                            }
+                            Element::NormalDisjunctive(dnf)
+                        }
+                    )
+                )
+                .map(|el| el.into_dnf())
+                .map(|fnd| {
+                    if fnd.len() == 1 {
+                        UserInputAST::and(fnd.into_iter().next().unwrap()) //< safe
+                    } else {
+                        let conjunctions = fnd
+                        .into_iter()
+                        .map(UserInputAST::and)
+                        .collect();
+                        UserInputAST::or(conjunctions)
+                    }
+                })
+            )
+            .or(
+                sep_by(leaf(), spaces())
+                .map(|subqueries: Vec<UserInputAST>| {
+                    if subqueries.len() == 1 {
+                        subqueries.into_iter().next().unwrap()
+                    } else {
+                        UserInputAST::Clause(subqueries.into_iter().collect())
+                    }
+                })
+            )
+        )
+
     }
 }
 
@@ -128,6 +227,40 @@ mod test {
         assert!(parse_to_ast().parse(query).is_err());
     }
 
+    #[test]
+    fn test_parse_query_to_ast_not_op() {
+        assert_eq!(
+            format!("{:?}", parse_to_ast().parse("NOT")),
+            "Err(UnexpectedParse)"
+        );
+        test_parse_query_to_ast_helper("NOTa", "\"NOTa\"");
+        test_parse_query_to_ast_helper("NOT a", "-(\"a\")");
+    }
+
+    #[test]
+    fn test_parse_query_to_ast_binary_op() {
+        test_parse_query_to_ast_helper("a AND b", "(+(\"a\") +(\"b\"))");
+        test_parse_query_to_ast_helper("a OR b", "(?(\"a\") ?(\"b\"))");
+        test_parse_query_to_ast_helper("a OR b AND c", "(?(\"a\") ?((+(\"b\") +(\"c\"))))");
+        test_parse_query_to_ast_helper("a AND b         AND c", "(+(\"a\") +(\"b\") +(\"c\"))");
+        assert_eq!(
+            format!("{:?}", parse_to_ast().parse("a OR b aaa")),
+            "Err(UnexpectedParse)"
+        );
+        assert_eq!(
+            format!("{:?}", parse_to_ast().parse("a AND b aaa")),
+            "Err(UnexpectedParse)"
+        );
+        assert_eq!(
+            format!("{:?}", parse_to_ast().parse("aaa a OR b ")),
+            "Err(UnexpectedParse)"
+        );
+        assert_eq!(
+            format!("{:?}", parse_to_ast().parse("aaa ccc a OR b ")),
+            "Err(UnexpectedParse)"
+        );
+    }
+
     #[test]
     fn test_parse_query_to_ast() {
         test_parse_query_to_ast_helper("+(a b) +d", "(+((\"a\" \"b\")) +(\"d\"))");

src/query/query_parser/query_parser.rs

@@ -1,9 +1,13 @@
 use super::logical_ast::*;
 use super::query_grammar::parse_to_ast;
 use super::user_input_ast::*;
+use combine::Parser;
 use core::Index;
+use query::occur::compose_occur;
+use query::query_parser::logical_ast::LogicalAST;
 use query::AllQuery;
 use query::BooleanQuery;
+use query::EmptyQuery;
 use query::Occur;
 use query::PhraseQuery;
 use query::Query;
@@ -17,9 +21,6 @@ use std::num::ParseIntError;
 use std::ops::Bound;
 use std::str::FromStr;
 use tokenizer::TokenizerManager;
-use combine::Parser;
-use query::EmptyQuery;
-
 
 /// Possible error that may happen when parsing a query.
 #[derive(Debug, PartialEq, Eq)]
@@ -57,6 +58,27 @@ impl From<ParseIntError> for QueryParserError {
     }
 }
 
+/// Recursively remove empty clause from the AST
+///
+/// Returns `None` iff the `logical_ast` ended up being empty.
+fn trim_ast(logical_ast: LogicalAST) -> Option<LogicalAST> {
+    match logical_ast {
+        LogicalAST::Clause(children) => {
+            let trimmed_children = children
+                .into_iter()
+                .flat_map(|(occur, child)| {
+                    trim_ast(child).map(|trimmed_child| (occur, trimmed_child))
+                }).collect::<Vec<_>>();
+            if trimmed_children.is_empty() {
+                None
+            } else {
+                Some(LogicalAST::Clause(trimmed_children))
+            }
+        }
+        _ => Some(logical_ast),
+    }
+}
+
 /// Tantivy's Query parser
 ///
 /// The language covered by the current parser is extremely simple.
@@ -79,12 +101,22 @@ impl From<ParseIntError> for QueryParserError {
 ///
 ///   Switching to a default of `AND` can be done by calling `.set_conjunction_by_default()`.
 ///
+///
+/// * boolean operators `AND`, `OR`. `AND` takes precedence over `OR`, so that `a AND b OR c` is interpreted
+/// as `(a AND b) OR c`.
+///
+/// * In addition to the boolean operators, the `-`, `+` can help define. These operators
+///   are sufficient to axpress all queries using boolean operators. For instance `x AND y OR z` can
+///   be written (`(+x +y) z`). In addition, these operators can help define "required optional"
+///   queries. `(+x y)` matches the same document set as simply `x`, but `y` will help refining the score.
+///
 /// * negative terms: By prepending a term by a `-`, a term can be excluded
 ///   from the search. This is useful for disambiguating a query.
 ///   e.g. `apple -fruit`
 ///
 /// * must terms: By prepending a term by a `+`, a term can be made required for the search.
 ///
+///
 /// * phrase terms: Quoted terms become phrase searches on fields that have positions indexed.
 ///   e.g., `title:"Barack Obama"` will only find documents that have "barack" immediately followed
 ///   by "obama".
@@ -155,8 +187,9 @@ impl QueryParser {
 
     /// Parse the user query into an AST.
     fn parse_query_to_logical_ast(&self, query: &str) -> Result<LogicalAST, QueryParserError> {
-        let (user_input_ast, _remaining) =
-            parse_to_ast().parse(query).map_err(|_| QueryParserError::SyntaxError)?;
+        let (user_input_ast, _remaining) = parse_to_ast()
+            .parse(query)
+            .map_err(|_| QueryParserError::SyntaxError)?;
         self.compute_logical_ast(user_input_ast)
     }
 
@@ -201,14 +234,15 @@ impl QueryParser {
             }
             FieldType::Str(ref str_options) => {
                 if let Some(option) = str_options.get_indexing_options() {
-                    let mut tokenizer = self.tokenizer_manager.get(option.tokenizer()).ok_or_else(
-                        || {
-                            QueryParserError::UnknownTokenizer(
-                                field_entry.name().to_string(),
-                                option.tokenizer().to_string(),
-                            )
-                        },
-                    )?;
+                    let mut tokenizer =
+                        self.tokenizer_manager
+                            .get(option.tokenizer())
+                            .ok_or_else(|| {
+                                QueryParserError::UnknownTokenizer(
+                                    field_entry.name().to_string(),
+                                    option.tokenizer().to_string(),
+                                )
+                            })?;
                     let mut terms: Vec<(usize, Term)> = Vec::new();
                     let mut token_stream = tokenizer.token_stream(phrase);
                     token_stream.process(&mut |token| {
@@ -258,12 +292,9 @@ impl QueryParser {
     ) -> Result<Option<LogicalLiteral>, QueryParserError> {
         let terms = self.compute_terms_for_string(field, phrase)?;
         match &terms[..] {
-            [] =>
-                Ok(None),
-            [(_, term)] =>
-                Ok(Some(LogicalLiteral::Term(term.clone()))),
-            _ =>
-                Ok(Some(LogicalLiteral::Phrase(terms.clone()))),
+            [] => Ok(None),
+            [(_, term)] => Ok(Some(LogicalLiteral::Term(term.clone()))),
+            _ => Ok(Some(LogicalLiteral::Phrase(terms.clone()))),
         }
     }
 
@@ -275,7 +306,11 @@ impl QueryParser {
         }
     }
 
-    fn resolve_bound(&self, field: Field, bound: &UserInputBound) -> Result<Bound<Term>, QueryParserError> {
+    fn resolve_bound(
+        &self,
+        field: Field,
+        bound: &UserInputBound,
+    ) -> Result<Bound<Term>, QueryParserError> {
         if bound.term_str() == "*" {
             return Ok(Bound::Unbounded);
         }
@@ -315,56 +350,30 @@ impl QueryParser {
                 let default_occur = self.default_occur();
                 let mut logical_sub_queries: Vec<(Occur, LogicalAST)> = Vec::new();
                 for sub_query in sub_queries {
-                    let (occur, sub_ast) = self.compute_logical_ast_with_occur(*sub_query)?;
+                    let (occur, sub_ast) = self.compute_logical_ast_with_occur(sub_query)?;
                     let new_occur = compose_occur(default_occur, occur);
                     logical_sub_queries.push((new_occur, sub_ast));
                 }
                 Ok((Occur::Should, LogicalAST::Clause(logical_sub_queries)))
             }
-            UserInputAST::Not(subquery) => {
-                let (occur, logical_sub_queries) = self.compute_logical_ast_with_occur(*subquery)?;
-                Ok((compose_occur(Occur::MustNot, occur), logical_sub_queries))
+            UserInputAST::Unary(left_occur, subquery) => {
+                let (right_occur, logical_sub_queries) =
+                    self.compute_logical_ast_with_occur(*subquery)?;
+                Ok((compose_occur(left_occur, right_occur), logical_sub_queries))
             }
-            UserInputAST::Must(subquery) => {
-                let (occur, logical_sub_queries) = self.compute_logical_ast_with_occur(*subquery)?;
-                Ok((compose_occur(Occur::Must, occur), logical_sub_queries))
-            }
-            UserInputAST::Range {
-                field,
-                lower,
-                upper,
-            } => {
-                let fields = self.resolved_fields(&field)?;
-                let mut clauses = fields
-                    .iter()
-                    .map(|&field| {
-                        let field_entry = self.schema.get_field_entry(field);
-                        let value_type = field_entry.field_type().value_type();
-                        Ok(LogicalAST::Leaf(Box::new(LogicalLiteral::Range {
-                            field,
-                            value_type,
-                            lower: self.resolve_bound(field, &lower)?,
-                            upper: self.resolve_bound(field, &upper)?,
-                        })))
-                    })
-                    .collect::<Result<Vec<_>, QueryParserError>>()?;
-                let result_ast = if clauses.len() == 1 {
-                    clauses.pop().unwrap()
-                } else {
-                    LogicalAST::Clause(
-                        clauses
-                            .into_iter()
-                            .map(|clause| (Occur::Should, clause))
-                            .collect(),
-                    )
-                };
+            UserInputAST::Leaf(leaf) => {
+                let result_ast = self.compute_logical_ast_from_leaf(*leaf)?;
                 Ok((Occur::Should, result_ast))
             }
-            UserInputAST::All => Ok((
-                Occur::Should,
-                LogicalAST::Leaf(Box::new(LogicalLiteral::All)),
-            )),
-            UserInputAST::Leaf(literal) => {
+        }
+    }
+
+    fn compute_logical_ast_from_leaf(
+        &self,
+        leaf: UserInputLeaf,
+    ) -> Result<LogicalAST, QueryParserError> {
+        match leaf {
+            UserInputLeaf::Literal(literal) => {
                 let term_phrases: Vec<(Field, String)> = match literal.field_name {
                     Some(ref field_name) => {
                         let field = self.resolve_field_name(field_name)?;
@@ -387,36 +396,43 @@ impl QueryParser {
                         asts.push(LogicalAST::Leaf(Box::new(ast)));
                     }
                 }
-                let result_ast = if asts.is_empty() {
-                    // this should never happen
-                    return Err(QueryParserError::SyntaxError);
-                } else if asts.len() == 1 {
-                    asts[0].clone()
+                let result_ast: LogicalAST = if asts.len() == 1 {
+                    asts.into_iter().next().unwrap()
                 } else {
                     LogicalAST::Clause(asts.into_iter().map(|ast| (Occur::Should, ast)).collect())
                 };
-                Ok((Occur::Should, result_ast))
+                Ok(result_ast)
             }
-        }
-    }
-}
-
-/// Compose two occur values.
-fn compose_occur(left: Occur, right: Occur) -> Occur {
-    match left {
-        Occur::Should => right,
-        Occur::Must => {
-            if right == Occur::MustNot {
-                Occur::MustNot
-            } else {
-                Occur::Must
-            }
-        }
-        Occur::MustNot => {
-            if right == Occur::MustNot {
-                Occur::Must
-            } else {
-                Occur::MustNot
+            UserInputLeaf::All => Ok(LogicalAST::Leaf(Box::new(LogicalLiteral::All))),
+            UserInputLeaf::Range {
+                field,
+                lower,
+                upper,
+            } => {
+                let fields = self.resolved_fields(&field)?;
+                let mut clauses = fields
+                    .iter()
+                    .map(|&field| {
+                        let field_entry = self.schema.get_field_entry(field);
+                        let value_type = field_entry.field_type().value_type();
+                        Ok(LogicalAST::Leaf(Box::new(LogicalLiteral::Range {
+                            field,
+                            value_type,
+                            lower: self.resolve_bound(field, &lower)?,
+                            upper: self.resolve_bound(field, &upper)?,
+                        })))
+                    }).collect::<Result<Vec<_>, QueryParserError>>()?;
+                let result_ast = if clauses.len() == 1 {
+                    clauses.pop().unwrap()
+                } else {
+                    LogicalAST::Clause(
+                        clauses
+                            .into_iter()
+                            .map(|clause| (Occur::Should, clause))
+                            .collect(),
+                    )
+                };
+                Ok(result_ast)
             }
         }
     }
@@ -425,31 +441,38 @@ fn compose_occur(left: Occur, right: Occur) -> Occur {
 fn convert_literal_to_query(logical_literal: LogicalLiteral) -> Box<Query> {
     match logical_literal {
         LogicalLiteral::Term(term) => Box::new(TermQuery::new(term, IndexRecordOption::WithFreqs)),
-        LogicalLiteral::Phrase(term_with_offsets) => Box::new(PhraseQuery::new_with_offset(term_with_offsets)),
+        LogicalLiteral::Phrase(term_with_offsets) => {
+            Box::new(PhraseQuery::new_with_offset(term_with_offsets))
+        }
         LogicalLiteral::Range {
             field,
             value_type,
             lower,
             upper,
-        } => Box::new(RangeQuery::new_term_bounds(field, value_type, lower, upper)),
+        } => Box::new(RangeQuery::new_term_bounds(
+            field, value_type, &lower, &upper,
+        )),
         LogicalLiteral::All => Box::new(AllQuery),
     }
 }
 
 fn convert_to_query(logical_ast: LogicalAST) -> Box<Query> {
-    match logical_ast {
-        LogicalAST::Clause(clause) => {
-            if clause.is_empty() {
-                Box::new(EmptyQuery)
-            } else {
-                let occur_subqueries = clause
-                    .into_iter()
-                    .map(|(occur, subquery)| (occur, convert_to_query(subquery)))
-                    .collect::<Vec<_>>();
-                Box::new(BooleanQuery::from(occur_subqueries))
-            }
+    match trim_ast(logical_ast) {
+        Some(LogicalAST::Clause(trimmed_clause)) => {
+            let occur_subqueries = trimmed_clause
+                .into_iter()
+                .map(|(occur, subquery)| (occur, convert_to_query(subquery)))
+                .collect::<Vec<_>>();
+            assert!(
+                !occur_subqueries.is_empty(),
+                "Should not be empty after trimming"
+            );
+            Box::new(BooleanQuery::from(occur_subqueries))
         }
-        LogicalAST::Leaf(logical_literal) => convert_literal_to_query(*logical_literal),
+        Some(LogicalAST::Leaf(trimmed_logical_literal)) => {
+            convert_literal_to_query(*trimmed_logical_literal)
+        }
+        None => Box::new(EmptyQuery),
     }
 }
 
@@ -462,12 +485,17 @@ mod test {
     use schema::Field;
     use schema::{IndexRecordOption, TextFieldIndexing, TextOptions};
     use schema::{SchemaBuilder, Term, INT_INDEXED, STORED, STRING, TEXT};
-    use tokenizer::SimpleTokenizer;
-    use tokenizer::TokenizerManager;
+    use tokenizer::{LowerCaser, SimpleTokenizer, StopWordFilter, Tokenizer, TokenizerManager};
     use Index;
 
     fn make_query_parser() -> QueryParser {
         let mut schema_builder = SchemaBuilder::default();
+        let text_field_indexing = TextFieldIndexing::default()
+            .set_tokenizer("en_with_stop_words")
+            .set_index_option(IndexRecordOption::WithFreqsAndPositions);
+        let text_options = TextOptions::default()
+            .set_indexing_options(text_field_indexing)
+            .set_stored();
         let title = schema_builder.add_text_field("title", TEXT);
         let text = schema_builder.add_text_field("text", TEXT);
         schema_builder.add_i64_field("signed", INT_INDEXED);
@@ -476,9 +504,16 @@ mod test {
         schema_builder.add_text_field("notindexed_u64", STORED);
         schema_builder.add_text_field("notindexed_i64", STORED);
         schema_builder.add_text_field("nottokenized", STRING);
+        schema_builder.add_text_field("with_stop_words", text_options);
         let schema = schema_builder.build();
         let default_fields = vec![title, text];
         let tokenizer_manager = TokenizerManager::default();
+        tokenizer_manager.register(
+            "en_with_stop_words",
+            SimpleTokenizer
+                .filter(LowerCaser)
+                .filter(StopWordFilter::remove(vec!["the".to_string()])),
+        );
         QueryParser::new(schema, default_fields, tokenizer_manager)
     }
 
@@ -548,16 +583,8 @@ mod test {
 
     #[test]
     pub fn test_parse_query_empty() {
-        test_parse_query_to_logical_ast_helper(
-            "",
-            "<emptyclause>",
-            false,
-        );
-        test_parse_query_to_logical_ast_helper(
-            " ",
-            "<emptyclause>",
-            false,
-        );
+        test_parse_query_to_logical_ast_helper("", "<emptyclause>", false);
+        test_parse_query_to_logical_ast_helper(" ", "<emptyclause>", false);
         let query_parser = make_query_parser();
         let query_result = query_parser.parse_query("");
         let query = query_result.unwrap();
@@ -670,11 +697,7 @@ mod test {
             "(Excluded(Term([0, 0, 0, 0, 116, 105, 116, 105])) TO Unbounded)",
             false,
         );
-        test_parse_query_to_logical_ast_helper(
-            "*",
-            "*",
-            false,
-        );
+        test_parse_query_to_logical_ast_helper("*", "*", false);
     }
 
     #[test]
@@ -747,6 +770,13 @@ mod test {
         );
     }
 
+    #[test]
+    pub fn test_query_parser_not_empty_but_no_tokens() {
+        let query_parser = make_query_parser();
+        assert!(query_parser.parse_query(" !, ").is_ok());
+        assert!(query_parser.parse_query("with_stop_words:the").is_ok());
+    }
+
     #[test]
     pub fn test_parse_query_to_ast_conjunction() {
         test_parse_query_to_logical_ast_helper(

src/query/query_parser/user_input_ast.rs

@@ -1,4 +1,39 @@
 use std::fmt;
+use std::fmt::{Debug, Formatter};
+
+use query::Occur;
+
+pub enum UserInputLeaf {
+    Literal(UserInputLiteral),
+    All,
+    Range {
+        field: Option<String>,
+        lower: UserInputBound,
+        upper: UserInputBound,
+    },
+}
+
+impl Debug for UserInputLeaf {
+    fn fmt(&self, formatter: &mut Formatter) -> Result<(), fmt::Error> {
+        match self {
+            UserInputLeaf::Literal(literal) => literal.fmt(formatter),
+            UserInputLeaf::Range {
+                ref field,
+                ref lower,
+                ref upper,
+            } => {
+                if let Some(ref field) = field {
+                    write!(formatter, "{}:", field)?;
+                }
+                lower.display_lower(formatter)?;
+                write!(formatter, " TO ")?;
+                upper.display_upper(formatter)?;
+                Ok(())
+            }
+            UserInputLeaf::All => write!(formatter, "*"),
+        }
+    }
+}
 
 pub struct UserInputLiteral {
     pub field_name: Option<String>,
@@ -43,28 +78,93 @@ impl UserInputBound {
 }
 
 pub enum UserInputAST {
-    Clause(Vec<Box<UserInputAST>>),
-    Not(Box<UserInputAST>),
-    Must(Box<UserInputAST>),
-    Range {
-        field: Option<String>,
-        lower: UserInputBound,
-        upper: UserInputBound,
-    },
-    All,
-    Leaf(Box<UserInputLiteral>),
+    Clause(Vec<UserInputAST>),
+    Unary(Occur, Box<UserInputAST>),
+    //    Not(Box<UserInputAST>),
+    //    Should(Box<UserInputAST>),
+    //    Must(Box<UserInputAST>),
+    Leaf(Box<UserInputLeaf>),
 }
 
-impl From<UserInputLiteral> for UserInputAST {
-    fn from(literal: UserInputLiteral) -> UserInputAST {
-        UserInputAST::Leaf(Box::new(literal))
+impl UserInputAST {
+    pub fn unary(self, occur: Occur) -> UserInputAST {
+        UserInputAST::Unary(occur, Box::new(self))
+    }
+
+    fn compose(occur: Occur, asts: Vec<UserInputAST>) -> UserInputAST {
+        assert!(occur != Occur::MustNot);
+        assert!(!asts.is_empty());
+        if asts.len() == 1 {
+            asts.into_iter().next().unwrap() //< safe
+        } else {
+            UserInputAST::Clause(
+                asts.into_iter()
+                    .map(|ast: UserInputAST| ast.unary(occur))
+                    .collect::<Vec<_>>(),
+            )
+        }
+    }
+
+    pub fn and(asts: Vec<UserInputAST>) -> UserInputAST {
+        UserInputAST::compose(Occur::Must, asts)
+    }
+
+    pub fn or(asts: Vec<UserInputAST>) -> UserInputAST {
+        UserInputAST::compose(Occur::Should, asts)
+    }
+}
+
+/*
+impl UserInputAST {
+
+    fn compose_occur(self, occur: Occur) -> UserInputAST {
+        match self {
+            UserInputAST::Not(other) => {
+                let new_occur = compose_occur(Occur::MustNot, occur);
+                other.simplify()
+            }
+            _ => {
+                self
+            }
+        }
+    }
+
+    pub fn simplify(self) -> UserInputAST {
+        match self {
+            UserInputAST::Clause(els) => {
+                if els.len() == 1 {
+                    return els.into_iter().next().unwrap();
+                } else {
+                    return self;
+                }
+            }
+            UserInputAST::Not(els) => {
+                if els.len() == 1 {
+                    return els.into_iter().next().unwrap();
+                } else {
+                    return self;
+                }
+            }
+        }
+    }
+}
+*/
+
+impl From<UserInputLiteral> for UserInputLeaf {
+    fn from(literal: UserInputLiteral) -> UserInputLeaf {
+        UserInputLeaf::Literal(literal)
+    }
+}
+
+impl From<UserInputLeaf> for UserInputAST {
+    fn from(leaf: UserInputLeaf) -> UserInputAST {
+        UserInputAST::Leaf(Box::new(leaf))
     }
 }
 
 impl fmt::Debug for UserInputAST {
     fn fmt(&self, formatter: &mut fmt::Formatter) -> Result<(), fmt::Error> {
         match *self {
-            UserInputAST::Must(ref subquery) => write!(formatter, "+({:?})", subquery),
             UserInputAST::Clause(ref subqueries) => {
                 if subqueries.is_empty() {
                     write!(formatter, "<emptyclause>")?;
@@ -78,21 +178,9 @@ impl fmt::Debug for UserInputAST {
                 }
                 Ok(())
             }
-            UserInputAST::Not(ref subquery) => write!(formatter, "-({:?})", subquery),
-            UserInputAST::Range {
-                ref field,
-                ref lower,
-                ref upper,
-            } => {
-                if let &Some(ref field) = field {
-                    write!(formatter, "{}:", field)?;
-                }
-                lower.display_lower(formatter)?;
-                write!(formatter, " TO ")?;
-                upper.display_upper(formatter)?;
-                Ok(())
+            UserInputAST::Unary(ref occur, ref subquery) => {
+                write!(formatter, "{}({:?})", occur.to_char(), subquery)
             }
-            UserInputAST::All => write!(formatter, "*"),
             UserInputAST::Leaf(ref subquery) => write!(formatter, "{:?}", subquery),
         }
     }

src/query/range_query.rs

@@ -68,7 +68,7 @@ fn map_bound<TFrom, TTo, Transform: Fn(&TFrom) -> TTo>(
 /// let docs_in_the_sixties = RangeQuery::new_u64(year_field, 1960..1970);
 ///
 /// let mut count_collector = CountCollector::default();
-/// docs_in_the_sixties.search(&*searcher, &mut count_collector)?;
+/// docs_in_the_sixties.search(&searcher, &mut count_collector)?;
 ///
 /// let num_60s_books = count_collector.count();
 ///
@@ -96,8 +96,8 @@ impl RangeQuery {
     pub fn new_term_bounds(
         field: Field,
         value_type: Type,
-        left_bound: Bound<Term>,
-        right_bound: Bound<Term>,
+        left_bound: &Bound<Term>,
+        right_bound: &Bound<Term>,
     ) -> RangeQuery {
         let verify_and_unwrap_term = |val: &Term| {
             assert_eq!(field, val.field());
@@ -184,11 +184,7 @@ impl RangeQuery {
     ///
     /// If the field is not of the type `Str`, tantivy
     /// will panic when the `Weight` object is created.
-    pub fn new_str_bounds<'b>(
-        field: Field,
-        left: Bound<&'b str>,
-        right: Bound<&'b str>,
-    ) -> RangeQuery {
+    pub fn new_str_bounds(field: Field, left: Bound<&str>, right: Bound<&str>) -> RangeQuery {
         let make_term_val = |val: &&str| val.as_bytes().to_vec();
         RangeQuery {
             field,
@@ -202,7 +198,7 @@ impl RangeQuery {
     ///
     /// If the field is not of the type `Str`, tantivy
     /// will panic when the `Weight` object is created.
-    pub fn new_str<'b>(field: Field, range: Range<&'b str>) -> RangeQuery {
+    pub fn new_str(field: Field, range: Range<&str>) -> RangeQuery {
         RangeQuery::new_str_bounds(
             field,
             Bound::Included(range.start),
@@ -332,7 +328,7 @@ mod tests {
 
             // ... or `1960..=1969` if inclusive range is enabled.
             let mut count_collector = CountCollector::default();
-            docs_in_the_sixties.search(&*searcher, &mut count_collector)?;
+            docs_in_the_sixties.search(&searcher, &mut count_collector)?;
             assert_eq!(count_collector.count(), 2285);
             Ok(())
         }
@@ -369,9 +365,7 @@ mod tests {
         let searcher = index.searcher();
         let count_multiples = |range_query: RangeQuery| {
             let mut count_collector = CountCollector::default();
-            range_query
-                .search(&*searcher, &mut count_collector)
-                .unwrap();
+            range_query.search(&searcher, &mut count_collector).unwrap();
             count_collector.count()
         };
 

src/query/regex_query.rs

@@ -82,7 +82,7 @@ impl RegexQuery {
         let automaton = Regex::new(&self.regex_pattern)
             .map_err(|_| TantivyError::InvalidArgument(self.regex_pattern.clone()))?;
 
-        Ok(AutomatonWeight::new(self.field.clone(), automaton))
+        Ok(AutomatonWeight::new(self.field, automaton))
     }
 }
 
@@ -123,7 +123,7 @@ mod test {
             let mut collector = TopCollector::with_limit(2);
             let regex_query = RegexQuery::new("jap[ao]n".to_string(), country_field);
             searcher.search(&regex_query, &mut collector).unwrap();
-            let scored_docs = collector.score_docs();
+            let scored_docs = collector.top_docs();
             assert_eq!(scored_docs.len(), 1, "Expected only 1 document");
             let (score, _) = scored_docs[0];
             assert_nearly_equals(1f32, score);
@@ -132,7 +132,7 @@ mod test {
             let mut collector = TopCollector::with_limit(2);
             let regex_query = RegexQuery::new("jap[A-Z]n".to_string(), country_field);
             searcher.search(&regex_query, &mut collector).unwrap();
-            let scored_docs = collector.score_docs();
+            let scored_docs = collector.top_docs();
             assert_eq!(scored_docs.len(), 0, "Expected ZERO document");
         }
     }