SegmentUpdater.add_segment does not need to return true (#693)

* Added handling of pre-tokenized text fields (#642).

* * Updated changelog and examples concerning #642.
* Added tokenized_text method to Value implementation.
* Implemented From<TokenizedString> for TokenizedStream.

* * Removed tokenized flag from TextOptions and code reliance on the flag.
* Changed naming to use word "pre-tokenized" instead of "tokenized".
* Updated example code.
* Fixed comments.

* Minor code refactoring. Test improvements.

.github/FUNDING.yml

@@ -0,0 +1,12 @@
+# These are supported funding model platforms
+
+github: fulmicoton
+patreon: # Replace with a single Patreon username
+open_collective: # Replace with a single Open Collective username
+ko_fi: # Replace with a single Ko-fi username
+tidelift: # Replace with a single Tidelift platform-name/package-name e.g., npm/babel
+community_bridge: # Replace with a single Community Bridge project-name e.g., cloud-foundry
+liberapay: # Replace with a single Liberapay username
+issuehunt: # Replace with a single IssueHunt username
+otechie: # Replace with a single Otechie username
+custom: # Replace with up to 4 custom sponsorship URLs e.g., ['link1', 'link2']

CHANGELOG.md

@@ -9,7 +9,11 @@ Tantivy 0.11.0
 - API change around `Box<BoxableTokenizer>`. See detail in #629
 - Avoid rebuilding Regex automaton whenever a regex query is reused. #639 (@brainlock)
 - Add footer with some metadata to index files. #605 (@fdb-hiroshima)
- 
+- TopDocs collector: ensure stable sorting on equal score. #671 (@brainlock)
+- Added handling of pre-tokenized text fields (#642), which will enable users to
+  load tokens created outside tantivy. See usage in examples/pre_tokenized_text. (@kkoziara)
+- Fix crash when committing multiple times with deleted documents. #681 (@brainlock)
+
 ## How to update?
 
 - `Box<dyn BoxableTokenizer>` has been replaced by a `BoxedTokenizer` struct.

Cargo.toml

@@ -13,7 +13,7 @@ keywords = ["search", "information", "retrieval"]
 edition = "2018"
 
 [dependencies]
-base64 = "0.10.0"
+base64 = "0.11.0"
 byteorder = "1.0"
 crc32fast = "1.2.0"
 once_cell = "1.0"
@@ -34,7 +34,7 @@ itertools = "0.8"
 levenshtein_automata = {version="0.1", features=["fst_automaton"]}
 notify = {version="4", optional=true}
 bit-set = "0.5"
-uuid = { version = "0.7.2", features = ["v4", "serde"] }
+uuid = { version = "0.8", features = ["v4", "serde"] }
 crossbeam = "0.7"
 futures = "0.1"
 futures-cpupool = "0.1"
@@ -50,10 +50,10 @@ owned-read = "0.4"
 failure = "0.1"
 htmlescape = "0.3.1"
 fail = "0.3"
-scoped-pool = "1.0"
 murmurhash32 = "0.2"
 chrono = "0.4"
-smallvec = "0.6"
+smallvec = "1.0"
+rayon = "1"
 
 [target.'cfg(windows)'.dependencies]
 winapi = "0.3"
@@ -64,6 +64,10 @@ maplit = "1"
 matches = "0.1.8"
 time = "0.1.42"
 
+[dev-dependencies.fail]
+version = "0.3"
+features = ["failpoints"]
+
 [profile.release]
 opt-level = 3
 debug = false
@@ -87,10 +91,6 @@ members = ["query-grammar"]
 [badges]
 travis-ci = { repository = "tantivy-search/tantivy" }
 
-[dev-dependencies.fail]
-version = "0.3"
-features = ["failpoints"]
-
 # Following the "fail" crate best practises, we isolate
 # tests that define specific behavior in fail check points
 # in a different binary.

README.md

@@ -21,9 +21,9 @@
 [![Become a patron](https://c5.patreon.com/external/logo/become_a_patron_button.png)](https://www.patreon.com/fulmicoton)
 
 
-**Tantivy** is a **full text search engine library** written in rust.
+**Tantivy** is a **full text search engine library** written in Rust.
 
-It is closer to [Apache Lucene](https://lucene.apache.org/) than to [Elasticsearch](https://www.elastic.co/products/elasticsearch) and [Apache Solr](https://lucene.apache.org/solr/) in the sense it is not
+It is closer to [Apache Lucene](https://lucene.apache.org/) than to [Elasticsearch](https://www.elastic.co/products/elasticsearch) or [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.
 
@@ -31,7 +31,7 @@ Tantivy is, in fact, strongly inspired by Lucene's design.
 
 # Benchmark
 
-Tantivy is typically faster than Lucene, but the results will depend on 
+Tantivy is typically faster than Lucene, but the results depend on 
 the nature of the queries in your workload.
 
 The following [benchmark](https://tantivy-search.github.io/bench/) break downs 
@@ -40,19 +40,19 @@ performance for different type of queries / collection.
 # Features
 
 - Full-text search
-- Configurable tokenizer. (stemming available for 17 latin languages. Third party support for Chinese ([tantivy-jieba](https://crates.io/crates/tantivy-jieba) and [cang-jie](https://crates.io/crates/cang-jie)) and [Japanese](https://crates.io/crates/tantivy-tokenizer-tiny-segmenter)
+- Configurable tokenizer (stemming available for 17 Latin languages with third party support for Chinese ([tantivy-jieba](https://crates.io/crates/tantivy-jieba) and [cang-jie](https://crates.io/crates/cang-jie)) and [Japanese](https://crates.io/crates/tantivy-tokenizer-tiny-segmenter))
 - 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)
-- Natural query language `(michael AND jackson) OR "king of pop"`
-- Phrase queries search (`"michael jackson"`)
+- BM25 scoring (the same as Lucene)
+- Natural query language (e.g. `(michael AND jackson) OR "king of pop"`)
+- Phrase queries search (e.g. `"michael jackson"`)
 - Incremental indexing
 - Multithreaded indexing (indexing English Wikipedia takes < 3 minutes on my desktop)
 - Mmap directory
-- SIMD integer compression when the platform/CPU includes the SSE2 instruction set.
-- Single valued and multivalued u64, i64 and f64 fast fields (equivalent of doc values in Lucene)
+- SIMD integer compression when the platform/CPU includes the SSE2 instruction set
+- Single valued and multivalued u64, i64, and f64 fast fields (equivalent of doc values in Lucene)
 - `&[u8]` fast fields
-- Text, i64, u64, f64, dates and hierarchical facet fields
+- Text, i64, u64, f64, dates, and hierarchical facet fields
 - LZ4 compressed document store
 - Range queries
 - Faceted search
@@ -61,43 +61,42 @@ performance for different type of queries / collection.
 
 # Non-features
 
-- Distributed search is out of the scope of tantivy. That being said, tantivy is meant as a
+- Distributed search is out of the scope of Tantivy. That being said, Tantivy is a
 library upon which one could build a distributed search. Serializable/mergeable collector state for instance, 
-are within the scope of tantivy.
+are within the scope of Tantivy.
 
 # Supported OS and compiler
 
-Tantivy works on stable rust (>= 1.27) and supports Linux, MacOS and Windows.
+Tantivy works on stable Rust (>= 1.27) and supports Linux, MacOS, and Windows.
 
 # Getting started
 
-- [tantivy's simple search example](https://tantivy-search.github.io/examples/basic_search.html)
-- [tantivy-cli and its tutorial](https://github.com/tantivy-search/tantivy-cli).
-`tantivy-cli` is an actual command line interface that makes it easy for you to create a search engine,
-index documents and search via the CLI or a small server with a REST API.
-It will walk you through getting a wikipedia search engine up and running in a few minutes.
-- [reference doc for the last released version](https://docs.rs/tantivy/)
+- [Tantivy's simple search example](https://tantivy-search.github.io/examples/basic_search.html)
+- [tantivy-cli and its tutorial](https://github.com/tantivy-search/tantivy-cli) - `tantivy-cli` is an actual command line interface that makes it easy for you to create a search engine,
+index documents, and search via the CLI or a small server with a REST API.
+It walks you through getting a wikipedia search engine up and running in a few minutes.
+- [Reference doc for the last released version](https://docs.rs/tantivy/)
 
 # How can I support this project?
 
 There are many ways to support this project. 
 
-- Use tantivy and tell us about your experience on [gitter](https://gitter.im/tantivy-search/tantivy) or by email (paul.masurel@gmail.com)
+- Use Tantivy and tell us about your experience on [Gitter](https://gitter.im/tantivy-search/tantivy) or by email (paul.masurel@gmail.com)
 - Report bugs
 - Write a blog post
 - Help with documentation by asking questions or submitting PRs
-- Contribute code (you can join [our gitter](https://gitter.im/tantivy-search/tantivy) )
-- Talk about tantivy around you
+- Contribute code (you can join [our Gitter](https://gitter.im/tantivy-search/tantivy))
+- Talk about Tantivy around you
 - Drop a word on on [![Say Thanks!](https://img.shields.io/badge/Say%20Thanks-!-1EAEDB.svg)](https://saythanks.io/to/fulmicoton) or even [![Become a patron](https://c5.patreon.com/external/logo/become_a_patron_button.png)](https://www.patreon.com/fulmicoton)
 
 # Contributing code
 
-We use the GitHub Pull Request workflow - reference a GitHub ticket and/or include a comprehensive commit message when opening a PR.
+We use the GitHub Pull Request workflow: reference a GitHub ticket and/or include a comprehensive commit message when opening a PR.
 
 ## Clone and build locally
 
-Tantivy compiles on stable rust but requires `Rust >= 1.27`.
-To check out and run tests, you can simply run :
+Tantivy compiles on stable Rust but requires `Rust >= 1.27`.
+To check out and run tests, you can simply run:
 
 ```bash
     git clone https://github.com/tantivy-search/tantivy.git
@@ -108,7 +107,7 @@ To check out and run tests, you can simply run :
 ## Run tests
 
 Some tests will not run with just `cargo test` because of `fail-rs`.
-To run the tests exhaustively, run `./run-tests.sh`
+To run the tests exhaustively, run `./run-tests.sh`.
 
 ## Debug
 
@@ -116,13 +115,13 @@ You might find it useful to step through the programme with a debugger.
 
 ### A failing test
 
-Make sure you haven't run `cargo clean` after the most recent `cargo test` or `cargo build` to guarantee that `target/` dir exists. Use this bash script to find the most name of the most recent debug build of tantivy and run it under rust-gdb.
+Make sure you haven't run `cargo clean` after the most recent `cargo test` or `cargo build` to guarantee that the `target/` directory exists. Use this bash script to find the name of the most recent debug build of Tantivy and run it under `rust-gdb`:
 
 ```bash
 find target/debug/ -maxdepth 1 -executable -type f -name "tantivy*" -printf '%TY-%Tm-%Td %TT %p\n' | sort -r | cut -d " " -f 3 | xargs -I RECENT_DBG_TANTIVY rust-gdb RECENT_DBG_TANTIVY
 ```
 
-Now that you are in rust-gdb, you can set breakpoints on lines and methods that match your source-code and run the debug executable with flags that you normally pass to `cargo test` to like this
+Now that you are in `rust-gdb`, you can set breakpoints on lines and methods that match your source code and run the debug executable with flags that you normally pass to `cargo test` like this:
 
 ```bash
 $gdb run --test-threads 1 --test $NAME_OF_TEST
@@ -130,7 +129,7 @@ $gdb run --test-threads 1 --test $NAME_OF_TEST
 
 ### An example
 
-By default, rustc compiles everything in the `examples/` dir in debug mode. This makes it easy for you to make examples to reproduce bugs.
+By default, `rustc` compiles everything in the `examples/` directory in debug mode. This makes it easy for you to make examples to reproduce bugs:
 
 ```bash
 rust-gdb target/debug/examples/$EXAMPLE_NAME

examples/pre_tokenized_text.rs

@@ -0,0 +1,140 @@
+// # Pre-tokenized text example
+//
+// This example shows how to use pre-tokenized text. Sometimes yout might
+// want to index and search through text which is already split into
+// tokens by some external tool.
+//
+// In this example we will:
+// - use tantivy tokenizer to create tokens and load them directly into tantivy,
+// - import tokenized text straight from json,
+// - perform a search on documents with pre-tokenized text
+
+use tantivy::tokenizer::{PreTokenizedString, SimpleTokenizer, Token, TokenStream, Tokenizer};
+
+use tantivy::collector::{Count, TopDocs};
+use tantivy::query::TermQuery;
+use tantivy::schema::*;
+use tantivy::{doc, Index, ReloadPolicy};
+use tempfile::TempDir;
+
+fn pre_tokenize_text(text: &str) -> Vec<Token> {
+    let mut token_stream = SimpleTokenizer.token_stream(text);
+    let mut tokens = vec![];
+    while token_stream.advance() {
+        tokens.push(token_stream.token().clone());
+    }
+    tokens
+}
+
+fn main() -> tantivy::Result<()> {
+    let index_path = TempDir::new()?;
+
+    let mut schema_builder = Schema::builder();
+
+    schema_builder.add_text_field("title", TEXT | STORED);
+    schema_builder.add_text_field("body", TEXT);
+
+    let schema = schema_builder.build();
+
+    let index = Index::create_in_dir(&index_path, schema.clone())?;
+
+    let mut index_writer = index.writer(50_000_000)?;
+
+    // We can create a document manually, by setting the fields
+    // one by one in a Document object.
+    let title = schema.get_field("title").unwrap();
+    let body = schema.get_field("body").unwrap();
+
+    let title_text = "The Old Man and the Sea";
+    let body_text = "He was an old man who fished alone in a skiff in the Gulf Stream";
+
+    // Content of our first document
+    // We create `PreTokenizedString` which contains original text and vector of tokens
+    let title_tok = PreTokenizedString {
+        text: String::from(title_text),
+        tokens: pre_tokenize_text(title_text),
+    };
+
+    println!(
+        "Original text: \"{}\" and tokens: {:?}",
+        title_tok.text, title_tok.tokens
+    );
+
+    let body_tok = PreTokenizedString {
+        text: String::from(body_text),
+        tokens: pre_tokenize_text(body_text),
+    };
+
+    // Now lets create a document and add our `PreTokenizedString` using
+    // `add_pre_tokenized_text` method of `Document`
+    let mut old_man_doc = Document::default();
+    old_man_doc.add_pre_tokenized_text(title, &title_tok);
+    old_man_doc.add_pre_tokenized_text(body, &body_tok);
+
+    // ... now let's just add it to the IndexWriter
+    index_writer.add_document(old_man_doc);
+
+    // Pretokenized text can also be fed as JSON
+    let short_man_json = r#"{
+        "title":[{
+            "text":"The Old Man",
+            "tokens":[
+                {"offset_from":0,"offset_to":3,"position":0,"text":"The","position_length":1},
+                {"offset_from":4,"offset_to":7,"position":1,"text":"Old","position_length":1},
+                {"offset_from":8,"offset_to":11,"position":2,"text":"Man","position_length":1}
+            ]
+        }]
+    }"#;
+
+    let short_man_doc = schema.parse_document(&short_man_json)?;
+
+    index_writer.add_document(short_man_doc);
+
+    // Let's commit changes
+    index_writer.commit()?;
+
+    // ... and now is the time to query our index
+
+    let reader = index
+        .reader_builder()
+        .reload_policy(ReloadPolicy::OnCommit)
+        .try_into()?;
+
+    let searcher = reader.searcher();
+
+    // We want to get documents with token "Man", we will use TermQuery to do it
+    // Using PreTokenizedString means the tokens are stored as is avoiding stemming
+    // and lowercasing, which preserves full words in their original form
+    let query = TermQuery::new(
+        Term::from_field_text(title, "Man"),
+        IndexRecordOption::Basic,
+    );
+
+    let (top_docs, count) = searcher
+        .search(&query, &(TopDocs::with_limit(2), Count))
+        .unwrap();
+
+    assert_eq!(count, 2);
+
+    for (_score, doc_address) in top_docs {
+        let retrieved_doc = searcher.doc(doc_address)?;
+        println!("Document: {}", schema.to_json(&retrieved_doc));
+    }
+
+    // In contrary to the previous query, when we search for the "man" term we
+    // should get no results, as it's not one of the indexed tokens. SimpleTokenizer
+    // only splits text on whitespace / punctuation.
+
+    let query = TermQuery::new(
+        Term::from_field_text(title, "man"),
+        IndexRecordOption::Basic,
+    );
+
+    let (_top_docs, count) = searcher
+        .search(&query, &(TopDocs::with_limit(2), Count))
+        .unwrap();
+
+    assert_eq!(count, 0);
+
+    Ok(())
+}

query-grammar/src/occur.rs

@@ -2,7 +2,7 @@ use std::fmt;
 use std::fmt::Write;
 
 /// Defines whether a term in a query must be present,
-/// should be present or must not be present.
+/// should be present or must be not present.
 #[derive(Debug, Clone, Hash, Copy, Eq, PartialEq)]
 pub enum Occur {
     /// For a given document to be considered for scoring,

src/collector/count_collector.rs

@@ -123,5 +123,4 @@ mod tests {
             assert_eq!(count_collector.harvest(), 2);
         }
     }
-
 }

src/collector/facet_collector.rs

@@ -515,7 +515,7 @@ mod tests {
     #[should_panic(expected = "Tried to add a facet which is a descendant of \
                                an already added facet.")]
     fn test_misused_facet_collector() {
-        let mut facet_collector = FacetCollector::for_field(Field(0));
+        let mut facet_collector = FacetCollector::for_field(Field::from_field_id(0));
         facet_collector.add_facet(Facet::from("/country"));
         facet_collector.add_facet(Facet::from("/country/europe"));
     }
@@ -546,7 +546,7 @@ mod tests {
 
     #[test]
     fn test_non_used_facet_collector() {
-        let mut facet_collector = FacetCollector::for_field(Field(0));
+        let mut facet_collector = FacetCollector::for_field(Field::from_field_id(0));
         facet_collector.add_facet(Facet::from("/country"));
         facet_collector.add_facet(Facet::from("/countryeurope"));
     }
@@ -599,19 +599,18 @@ mod tests {
             );
         }
     }
-
 }
 
 #[cfg(all(test, feature = "unstable"))]
 mod bench {
 
-    use collector::FacetCollector;
-    use query::AllQuery;
-    use rand::{thread_rng, Rng};
-    use schema::Facet;
-    use schema::Schema;
+    use crate::collector::FacetCollector;
+    use crate::query::AllQuery;
+    use crate::schema::{Facet, Schema};
+    use crate::Index;
+    use rand::seq::SliceRandom;
+    use rand::thread_rng;
     use test::Bencher;
-    use Index;
 
     #[bench]
     fn bench_facet_collector(b: &mut Bencher) {
@@ -628,7 +627,7 @@ mod bench {
             }
         }
         // 40425 docs
-        thread_rng().shuffle(&mut docs[..]);
+        docs[..].shuffle(&mut thread_rng());
 
         let mut index_writer = index.writer_with_num_threads(1, 3_000_000).unwrap();
         for doc in docs {
@@ -637,7 +636,7 @@ mod bench {
         index_writer.commit().unwrap();
         let reader = index.reader().unwrap();
         b.iter(|| {
-            let searcher = index.searcher();
+            let searcher = reader.searcher();
             let facet_collector = FacetCollector::for_field(facet_field);
             searcher.search(&AllQuery, &facet_collector).unwrap();
         });

src/collector/top_collector.rs

@@ -12,6 +12,9 @@ use std::collections::BinaryHeap;
 /// 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.
 ///
+/// Additionally, it guarantees stable sorting: in case of a tie on the feature, the document
+/// address is used.
+///
 /// WARNING: equality is not what you would expect here.
 /// Two elements are equal if their feature is equal, and regardless of whether `doc`
 /// is equal. This should be perfectly fine for this usage, but let's make sure this
@@ -21,29 +24,37 @@ struct ComparableDoc<T, D> {
     doc: D,
 }
 
-impl<T: PartialOrd, D> PartialOrd for ComparableDoc<T, D> {
+impl<T: PartialOrd, D: PartialOrd> PartialOrd for ComparableDoc<T, D> {
     fn partial_cmp(&self, other: &Self) -> Option<Ordering> {
         Some(self.cmp(other))
     }
 }
 
-impl<T: PartialOrd, D> Ord for ComparableDoc<T, D> {
+impl<T: PartialOrd, D: PartialOrd> Ord for ComparableDoc<T, D> {
     #[inline]
     fn cmp(&self, other: &Self) -> Ordering {
-        other
+        // Reversed to make BinaryHeap work as a min-heap
+        let by_feature = other
             .feature
             .partial_cmp(&self.feature)
-            .unwrap_or_else(|| Ordering::Equal)
+            .unwrap_or(Ordering::Equal);
+
+        let lazy_by_doc_address = || self.doc.partial_cmp(&other.doc).unwrap_or(Ordering::Equal);
+
+        // In case of a tie on the feature, we sort by ascending
+        // `DocAddress` in order to ensure a stable sorting of the
+        // documents.
+        by_feature.then_with(lazy_by_doc_address)
     }
 }
 
-impl<T: PartialOrd, D> PartialEq for ComparableDoc<T, D> {
+impl<T: PartialOrd, D: PartialOrd> PartialEq for ComparableDoc<T, D> {
     fn eq(&self, other: &Self) -> bool {
         self.cmp(other) == Ordering::Equal
     }
 }
 
-impl<T: PartialOrd, D> Eq for ComparableDoc<T, D> {}
+impl<T: PartialOrd, D: PartialOrd> Eq for ComparableDoc<T, D> {}
 
 pub(crate) struct TopCollector<T> {
     limit: usize,
@@ -214,4 +225,94 @@ mod tests {
             ]
         );
     }
+
+    #[test]
+    fn test_top_segment_collector_stable_ordering_for_equal_feature() {
+        // given that the documents are collected in ascending doc id order,
+        // when harvesting we have to guarantee stable sorting in case of a tie
+        // on the score
+        let doc_ids_collection = [4, 5, 6];
+        let score = 3.14;
+
+        let mut top_collector_limit_2 = TopSegmentCollector::new(0, 2);
+        for id in &doc_ids_collection {
+            top_collector_limit_2.collect(*id, score);
+        }
+
+        let mut top_collector_limit_3 = TopSegmentCollector::new(0, 3);
+        for id in &doc_ids_collection {
+            top_collector_limit_3.collect(*id, score);
+        }
+
+        assert_eq!(
+            top_collector_limit_2.harvest(),
+            top_collector_limit_3.harvest()[..2].to_vec(),
+        );
+    }
+}
+
+#[cfg(all(test, feature = "unstable"))]
+mod bench {
+    use super::TopSegmentCollector;
+    use test::Bencher;
+
+    #[bench]
+    fn bench_top_segment_collector_collect_not_at_capacity(b: &mut Bencher) {
+        let mut top_collector = TopSegmentCollector::new(0, 400);
+
+        b.iter(|| {
+            for i in 0..100 {
+                top_collector.collect(i, 0.8);
+            }
+        });
+    }
+
+    #[bench]
+    fn bench_top_segment_collector_collect_at_capacity(b: &mut Bencher) {
+        let mut top_collector = TopSegmentCollector::new(0, 100);
+
+        for i in 0..100 {
+            top_collector.collect(i, 0.8);
+        }
+
+        b.iter(|| {
+            for i in 0..100 {
+                top_collector.collect(i, 0.8);
+            }
+        });
+    }
+
+    #[bench]
+    fn bench_top_segment_collector_collect_and_harvest_many_ties(b: &mut Bencher) {
+        b.iter(|| {
+            let mut top_collector = TopSegmentCollector::new(0, 100);
+
+            for i in 0..100 {
+                top_collector.collect(i, 0.8);
+            }
+
+            // it would be nice to be able to do the setup N times but still
+            // measure only harvest(). We can't since harvest() consumes
+            // the top_collector.
+            top_collector.harvest()
+        });
+    }
+
+    #[bench]
+    fn bench_top_segment_collector_collect_and_harvest_no_tie(b: &mut Bencher) {
+        b.iter(|| {
+            let mut top_collector = TopSegmentCollector::new(0, 100);
+            let mut score = 1.0;
+
+            for i in 0..100 {
+                score += 1.0;
+                top_collector.collect(i, score);
+            }
+
+            // it would be nice to be able to do the setup N times but still
+            // measure only harvest(). We can't since harvest() consumes
+            // the top_collector.
+            top_collector.harvest()
+        });
+    }
 }

src/collector/top_score_collector.rs

@@ -15,13 +15,16 @@ use crate::SegmentLocalId;
 use crate::SegmentReader;
 use std::fmt;
 
-/// The Top Score Collector keeps track of the K documents
+/// The `TopDocs` collector keeps track of the top `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)`.
 ///
+/// This collector guarantees a stable sorting in case of a tie on the
+/// document score. As such, it is suitable to implement pagination.
+///
 /// ```rust
 /// use tantivy::collector::TopDocs;
 /// use tantivy::query::QueryParser;
@@ -428,12 +431,13 @@ impl SegmentCollector for TopScoreSegmentCollector {
 mod tests {
     use super::TopDocs;
     use crate::collector::Collector;
-    use crate::query::{Query, QueryParser};
+    use crate::query::{AllQuery, Query, QueryParser};
     use crate::schema::{Field, Schema, FAST, STORED, TEXT};
     use crate::DocAddress;
     use crate::Index;
     use crate::IndexWriter;
     use crate::Score;
+    use itertools::Itertools;
 
     fn make_index() -> Index {
         let mut schema_builder = Schema::builder();
@@ -494,6 +498,29 @@ mod tests {
         );
     }
 
+    #[test]
+    fn test_top_collector_stable_sorting() {
+        let index = make_index();
+
+        // using AllQuery to get a constant score
+        let searcher = index.reader().unwrap().searcher();
+
+        let page_1 = searcher.search(&AllQuery, &TopDocs::with_limit(2)).unwrap();
+
+        let page_2 = searcher.search(&AllQuery, &TopDocs::with_limit(3)).unwrap();
+
+        // precondition for the test to be meaningful: we did get documents
+        // with the same score
+        assert!(page_1.iter().map(|result| result.0).all_equal());
+        assert!(page_2.iter().map(|result| result.0).all_equal());
+
+        // sanity check since we're relying on make_index()
+        assert_eq!(page_1.len(), 2);
+        assert_eq!(page_2.len(), 3);
+
+        assert_eq!(page_1, &page_2[..page_1.len()]);
+    }
+
     #[test]
     #[should_panic]
     fn test_top_0() {
@@ -551,7 +578,7 @@ mod tests {
             ));
         });
         let searcher = index.reader().unwrap().searcher();
-        let top_collector = TopDocs::with_limit(4).order_by_u64_field(Field(2));
+        let top_collector = TopDocs::with_limit(4).order_by_u64_field(Field::from_field_id(2));
         let segment_reader = searcher.segment_reader(0u32);
         top_collector
             .for_segment(0, segment_reader)
@@ -592,5 +619,4 @@ mod tests {
         let query = query_parser.parse_query(query).unwrap();
         (index, query)
     }
-
 }

src/common/composite_file.rs

@@ -199,13 +199,13 @@ mod test {
             let w = directory.open_write(path).unwrap();
             let mut composite_write = CompositeWrite::wrap(w);
             {
-                let mut write_0 = composite_write.for_field(Field(0u32));
+                let mut write_0 = composite_write.for_field(Field::from_field_id(0u32));
                 VInt(32431123u64).serialize(&mut write_0).unwrap();
                 write_0.flush().unwrap();
             }
 
             {
-                let mut write_4 = composite_write.for_field(Field(4u32));
+                let mut write_4 = composite_write.for_field(Field::from_field_id(4u32));
                 VInt(2).serialize(&mut write_4).unwrap();
                 write_4.flush().unwrap();
             }
@@ -215,14 +215,18 @@ mod test {
             let r = directory.open_read(path).unwrap();
             let composite_file = CompositeFile::open(&r).unwrap();
             {
-                let file0 = composite_file.open_read(Field(0u32)).unwrap();
+                let file0 = composite_file
+                    .open_read(Field::from_field_id(0u32))
+                    .unwrap();
                 let mut file0_buf = file0.as_slice();
                 let payload_0 = VInt::deserialize(&mut file0_buf).unwrap().0;
                 assert_eq!(file0_buf.len(), 0);
                 assert_eq!(payload_0, 32431123u64);
             }
             {
-                let file4 = composite_file.open_read(Field(4u32)).unwrap();
+                let file4 = composite_file
+                    .open_read(Field::from_field_id(4u32))
+                    .unwrap();
                 let mut file4_buf = file4.as_slice();
                 let payload_4 = VInt::deserialize(&mut file4_buf).unwrap().0;
                 assert_eq!(file4_buf.len(), 0);
@@ -230,5 +234,4 @@ mod test {
             }
         }
     }
-
 }

src/common/serialize.rs

@@ -199,10 +199,7 @@ pub mod test {
     fn test_serialize_string() {
         assert_eq!(serialize_test(String::from("")), 1);
         assert_eq!(serialize_test(String::from("ぽよぽよ")), 1 + 3 * 4);
-        assert_eq!(
-            serialize_test(String::from("富士さん見える。")),
-            1 + 3 * 8
-        );
+        assert_eq!(serialize_test(String::from("富士さん見える。")), 1 + 3 * 8);
     }
 
     #[test]

src/core/executor.rs

@@ -1,6 +1,6 @@
 use crate::Result;
 use crossbeam::channel;
-use scoped_pool::{Pool, ThreadConfig};
+use rayon::{ThreadPool, ThreadPoolBuilder};
 
 /// Search executor whether search request are single thread or multithread.
 ///
@@ -11,7 +11,7 @@ use scoped_pool::{Pool, ThreadConfig};
 /// used by the client. Second, we may stop using rayon in the future.
 pub enum Executor {
     SingleThread,
-    ThreadPool(Pool),
+    ThreadPool(ThreadPool),
 }
 
 impl Executor {
@@ -21,10 +21,12 @@ impl Executor {
     }
 
     // Creates an Executor that dispatches the tasks in a thread pool.
-    pub fn multi_thread(num_threads: usize, prefix: &'static str) -> Executor {
-        let thread_config = ThreadConfig::new().prefix(prefix);
-        let pool = Pool::with_thread_config(num_threads, thread_config);
-        Executor::ThreadPool(pool)
+    pub fn multi_thread(num_threads: usize, prefix: &'static str) -> Result<Executor> {
+        let pool = ThreadPoolBuilder::new()
+            .num_threads(num_threads)
+            .thread_name(move |num| format!("{}{}", prefix, num))
+            .build()?;
+        Ok(Executor::ThreadPool(pool))
     }
 
     // Perform a map in the thread pool.
@@ -48,9 +50,9 @@ impl Executor {
                 let num_fruits = args_with_indices.len();
                 let fruit_receiver = {
                     let (fruit_sender, fruit_receiver) = channel::unbounded();
-                    pool.scoped(|scope| {
+                    pool.scope(|scope| {
                         for arg_with_idx in args_with_indices {
-                            scope.execute(|| {
+                            scope.spawn(|_| {
                                 let (idx, arg) = arg_with_idx;
                                 let fruit = f(arg);
                                 if let Err(err) = fruit_sender.send((idx, fruit)) {
@@ -103,6 +105,7 @@ mod tests {
     #[should_panic] //< unfortunately the panic message is not propagated
     fn test_panic_propagates_multi_thread() {
         let _result: Vec<usize> = Executor::multi_thread(1, "search-test")
+            .unwrap()
             .map(
                 |_| {
                     panic!("panic should propagate");
@@ -126,6 +129,7 @@ mod tests {
     #[test]
     fn test_map_multithread() {
         let result: Vec<usize> = Executor::multi_thread(3, "search-test")
+            .unwrap()
             .map(|i| Ok(i * 2), 0..10)
             .unwrap();
         assert_eq!(result.len(), 10);

src/core/index.rs

@@ -73,15 +73,16 @@ impl Index {
 
     /// Replace the default single thread search executor pool
     /// by a thread pool with a given number of threads.
-    pub fn set_multithread_executor(&mut self, num_threads: usize) {
-        self.executor = Arc::new(Executor::multi_thread(num_threads, "thrd-tantivy-search-"));
+    pub fn set_multithread_executor(&mut self, num_threads: usize) -> Result<()> {
+        self.executor = Arc::new(Executor::multi_thread(num_threads, "thrd-tantivy-search-")?);
+        Ok(())
     }
 
     /// Replace the default single thread search executor pool
     /// by a thread pool with a given number of threads.
-    pub fn set_default_multithread_executor(&mut self) {
+    pub fn set_default_multithread_executor(&mut self) -> Result<()> {
         let default_num_threads = num_cpus::get();
-        self.set_multithread_executor(default_num_threads);
+        self.set_multithread_executor(default_num_threads)
     }
 
     /// Creates a new index using the `RAMDirectory`.
@@ -601,5 +602,4 @@ mod tests {
         assert_eq!(searcher.num_docs(), 8_000);
         assert!(mem_right_after_merge_finished < mem_right_after_commit);
     }
-
 }

src/core/index_meta.rs

@@ -150,6 +150,21 @@ impl SegmentMeta {
         self.num_deleted_docs() > 0
     }
 
+    /// Updates the max_doc value from the `SegmentMeta`.
+    ///
+    /// This method is only used when updating `max_doc` from 0
+    /// as we finalize a fresh new segment.
+    pub(crate) fn with_max_doc(self, max_doc: u32) -> SegmentMeta {
+        assert_eq!(self.tracked.max_doc, 0);
+        assert!(self.tracked.deletes.is_none());
+        let tracked = self.tracked.map(move |inner_meta| InnerSegmentMeta {
+            segment_id: inner_meta.segment_id,
+            max_doc,
+            deletes: None,
+        });
+        SegmentMeta { tracked }
+    }
+
     #[doc(hidden)]
     pub fn with_delete_meta(self, num_deleted_docs: u32, opstamp: Opstamp) -> SegmentMeta {
         let delete_meta = DeleteMeta {

src/core/segment.rs

@@ -50,6 +50,17 @@ impl Segment {
         &self.meta
     }
 
+    /// Updates the max_doc value from the `SegmentMeta`.
+    ///
+    /// This method is only used when updating `max_doc` from 0
+    /// as we finalize a fresh new segment.
+    pub(crate) fn with_max_doc(self, max_doc: u32) -> Segment {
+        Segment {
+            index: self.index,
+            meta: self.meta.with_max_doc(max_doc),
+        }
+    }
+
     #[doc(hidden)]
     pub fn with_delete_meta(self, num_deleted_docs: u32, opstamp: Opstamp) -> Segment {
         Segment {

src/core/segment_id.rs

@@ -76,7 +76,7 @@ impl SegmentId {
 }
 
 /// Error type used when parsing a `SegmentId` from a string fails.
-pub struct SegmentIdParseError(uuid::parser::ParseError);
+pub struct SegmentIdParseError(uuid::Error);
 
 impl Error for SegmentIdParseError {}
 

src/directory/managed_directory.rs

@@ -327,8 +327,7 @@ mod tests_mmap_specific {
                 .unwrap();
             assert!(managed_directory.exists(test_path1));
             assert!(managed_directory.exists(test_path2));
-            let living_files: HashSet<PathBuf> =
-                [test_path1.to_owned()].into_iter().cloned().collect();
+            let living_files: HashSet<PathBuf> = [test_path1.to_owned()].iter().cloned().collect();
             managed_directory.garbage_collect(|| living_files);
             assert!(managed_directory.exists(test_path1));
             assert!(!managed_directory.exists(test_path2));

src/directory/watch_event_router.rs

@@ -152,5 +152,4 @@ mod tests {
         thread::sleep(Duration::from_millis(WAIT_TIME));
         assert_eq!(2, counter.load(Ordering::SeqCst));
     }
-
 }

src/error.rs

@@ -170,3 +170,9 @@ impl From<serde_json::Error> for TantivyError {
         TantivyError::IOError(io_err.into())
     }
 }
+
+impl From<rayon::ThreadPoolBuildError> for TantivyError {
+    fn from(error: rayon::ThreadPoolBuildError) -> TantivyError {
+        TantivyError::SystemError(error.to_string())
+    }
+}

src/fastfield/delete.rs

@@ -10,11 +10,14 @@ use std::io::Write;
 /// Write a delete `BitSet`
 ///
 /// where `delete_bitset` is the set of deleted `DocId`.
-pub fn write_delete_bitset(delete_bitset: &BitSet, writer: &mut WritePtr) -> io::Result<()> {
-    let max_doc = delete_bitset.capacity();
+pub fn write_delete_bitset(
+    delete_bitset: &BitSet,
+    max_doc: u32,
+    writer: &mut WritePtr,
+) -> io::Result<()> {
     let mut byte = 0u8;
     let mut shift = 0u8;
-    for doc in 0..max_doc {
+    for doc in 0..(max_doc as usize) {
         if delete_bitset.contains(doc) {
             byte |= 1 << shift;
         }
@@ -86,18 +89,17 @@ mod tests {
     use bit_set::BitSet;
     use std::path::PathBuf;
 
-    fn test_delete_bitset_helper(bitset: &BitSet) {
+    fn test_delete_bitset_helper(bitset: &BitSet, max_doc: u32) {
         let test_path = PathBuf::from("test");
         let mut directory = RAMDirectory::create();
         {
             let mut writer = directory.open_write(&*test_path).unwrap();
-            write_delete_bitset(bitset, &mut writer).unwrap();
+            write_delete_bitset(bitset, max_doc, &mut writer).unwrap();
         }
         {
             let source = directory.open_read(&test_path).unwrap();
             let delete_bitset = DeleteBitSet::open(source);
-            let n = bitset.capacity();
-            for doc in 0..n {
+            for doc in 0..max_doc as usize {
                 assert_eq!(bitset.contains(doc), delete_bitset.is_deleted(doc as DocId));
             }
             assert_eq!(delete_bitset.len(), bitset.len());
@@ -110,7 +112,7 @@ mod tests {
             let mut bitset = BitSet::with_capacity(10);
             bitset.insert(1);
             bitset.insert(9);
-            test_delete_bitset_helper(&bitset);
+            test_delete_bitset_helper(&bitset, 10);
         }
         {
             let mut bitset = BitSet::with_capacity(8);
@@ -119,7 +121,7 @@ mod tests {
             bitset.insert(3);
             bitset.insert(5);
             bitset.insert(7);
-            test_delete_bitset_helper(&bitset);
+            test_delete_bitset_helper(&bitset, 8);
         }
     }
 }

src/fastfield/mod.rs

@@ -429,7 +429,6 @@ mod tests {
             }
         }
     }
-
 }
 
 #[cfg(all(test, feature = "unstable"))]
@@ -437,9 +436,9 @@ mod bench {
     use super::tests::FIELD;
     use super::tests::{generate_permutation, SCHEMA};
     use super::*;
-    use common::CompositeFile;
-    use directory::{Directory, RAMDirectory, WritePtr};
-    use fastfield::FastFieldReader;
+    use crate::common::CompositeFile;
+    use crate::directory::{Directory, RAMDirectory, WritePtr};
+    use crate::fastfield::FastFieldReader;
     use std::collections::HashMap;
     use std::path::Path;
     use test::{self, Bencher};
@@ -537,5 +536,4 @@ mod bench {
             });
         }
     }
-
 }

src/fastfield/readers.rs

@@ -59,8 +59,7 @@ impl FastFieldReaders {
             fast_bytes: Default::default(),
             fast_fields_composite: fast_fields_composite.clone(),
         };
-        for (field_id, field_entry) in schema.fields().iter().enumerate() {
-            let field = Field(field_id as u32);
+        for (field, field_entry) in schema.fields() {
             let field_type = field_entry.field_type();
             if field_type == &FieldType::Bytes {
                 let idx_reader = fast_fields_composite

src/fastfield/writer.rs

@@ -24,8 +24,7 @@ impl FastFieldsWriter {
         let mut multi_values_writers = Vec::new();
         let mut bytes_value_writers = Vec::new();
 
-        for (field_id, field_entry) in schema.fields().iter().enumerate() {
-            let field = Field(field_id as u32);
+        for (field, field_entry) in schema.fields() {
             let default_value = match *field_entry.field_type() {
                 FieldType::I64(_) => common::i64_to_u64(0i64),
                 FieldType::F64(_) => common::f64_to_u64(0.0f64),

src/fieldnorm/writer.rs

@@ -22,11 +22,14 @@ impl FieldNormsWriter {
     pub(crate) fn fields_with_fieldnorm(schema: &Schema) -> Vec<Field> {
         schema
             .fields()
-            .iter()
-            .enumerate()
-            .filter(|&(_, field_entry)| field_entry.is_indexed())
-            .map(|(field, _)| Field(field as u32))
-            .collect::<Vec<Field>>()
+            .filter_map(|(field, field_entry)| {
+                if field_entry.is_indexed() {
+                    Some(field)
+                } else {
+                    None
+                }
+            })
+            .collect::<Vec<_>>()
     }
 
     /// Initialize with state for tracking the field norm fields
@@ -35,7 +38,7 @@ impl FieldNormsWriter {
         let fields = FieldNormsWriter::fields_with_fieldnorm(schema);
         let max_field = fields
             .iter()
-            .map(|field| field.0)
+            .map(Field::field_id)
             .max()
             .map(|max_field_id| max_field_id as usize + 1)
             .unwrap_or(0);
@@ -50,8 +53,8 @@ impl FieldNormsWriter {
     ///
     /// Will extend with 0-bytes for documents that have not been seen.
     pub fn fill_up_to_max_doc(&mut self, max_doc: DocId) {
-        for &field in self.fields.iter() {
-            self.fieldnorms_buffer[field.0 as usize].resize(max_doc as usize, 0u8);
+        for field in self.fields.iter() {
+            self.fieldnorms_buffer[field.field_id() as usize].resize(max_doc as usize, 0u8);
         }
     }
 
@@ -64,7 +67,7 @@ impl FieldNormsWriter {
     /// * field     - the field being set
     /// * fieldnorm - the number of terms present in document `doc` in field `field`
     pub fn record(&mut self, doc: DocId, field: Field, fieldnorm: u32) {
-        let fieldnorm_buffer: &mut Vec<u8> = &mut self.fieldnorms_buffer[field.0 as usize];
+        let fieldnorm_buffer: &mut Vec<u8> = &mut self.fieldnorms_buffer[field.field_id() as usize];
         assert!(
             fieldnorm_buffer.len() <= doc as usize,
             "Cannot register a given fieldnorm twice"
@@ -77,7 +80,7 @@ impl FieldNormsWriter {
     /// Serialize the seen fieldnorm values to the serializer for all fields.
     pub fn serialize(&self, fieldnorms_serializer: &mut FieldNormsSerializer) -> io::Result<()> {
         for &field in self.fields.iter() {
-            let fieldnorm_values: &[u8] = &self.fieldnorms_buffer[field.0 as usize][..];
+            let fieldnorm_values: &[u8] = &self.fieldnorms_buffer[field.field_id() as usize][..];
             fieldnorms_serializer.serialize_field(field, fieldnorm_values)?;
         }
         Ok(())

src/indexer/delete_queue.rs

@@ -258,7 +258,7 @@ mod tests {
         let delete_queue = DeleteQueue::new();
 
         let make_op = |i: usize| {
-            let field = Field(1u32);
+            let field = Field::from_field_id(1u32);
             DeleteOperation {
                 opstamp: i as u64,
                 term: Term::from_field_u64(field, i as u64),

src/indexer/index_writer.rs

@@ -148,7 +148,6 @@ pub(crate) fn advance_deletes(
         };
 
         let delete_cursor = segment_entry.delete_cursor();
-
         compute_deleted_bitset(
             &mut delete_bitset,
             &segment_reader,
@@ -168,7 +167,7 @@ pub(crate) fn advance_deletes(
         if num_deleted_docs > 0 {
             segment = segment.with_delete_meta(num_deleted_docs as u32, target_opstamp);
             let mut delete_file = segment.open_write(SegmentComponent::DELETE)?;
-            write_delete_bitset(&delete_bitset, &mut delete_file)?;
+            write_delete_bitset(&delete_bitset, max_doc, &mut delete_file)?;
             delete_file.terminate()?;
         }
     }
@@ -178,13 +177,13 @@ pub(crate) fn advance_deletes(
 
 fn index_documents(
     memory_budget: usize,
-    segment: &Segment,
+    segment: Segment,
     grouped_document_iterator: &mut dyn Iterator<Item = OperationGroup>,
     segment_updater: &mut SegmentUpdater,
     mut delete_cursor: DeleteCursor,
 ) -> Result<bool> {
     let schema = segment.schema();
-    let segment_id = segment.id();
+
     let mut segment_writer = SegmentWriter::for_segment(memory_budget, segment.clone(), &schema)?;
     for document_group in grouped_document_iterator {
         for doc in document_group {
@@ -204,22 +203,32 @@ fn index_documents(
         return Ok(false);
     }
 
-    let num_docs = segment_writer.max_doc();
+    let max_doc = segment_writer.max_doc();
 
     // this is ensured by the call to peek before starting
     // the worker thread.
-    assert!(num_docs > 0);
+    assert!(max_doc > 0);
 
     let doc_opstamps: Vec<Opstamp> = segment_writer.finalize()?;
-    let segment_meta = segment.index().new_segment_meta(segment_id, num_docs);
+
+    let segment_with_max_doc = segment.with_max_doc(max_doc);
 
     let last_docstamp: Opstamp = *(doc_opstamps.last().unwrap());
 
-    let delete_bitset_opt =
-        apply_deletes(&segment, &mut delete_cursor, &doc_opstamps, last_docstamp)?;
+    let delete_bitset_opt = apply_deletes(
+        &segment_with_max_doc,
+        &mut delete_cursor,
+        &doc_opstamps,
+        last_docstamp,
+    )?;
 
-    let segment_entry = SegmentEntry::new(segment_meta, delete_cursor, delete_bitset_opt);
-    Ok(segment_updater.add_segment(segment_entry))
+    let segment_entry = SegmentEntry::new(
+        segment_with_max_doc.meta().clone(),
+        delete_cursor,
+        delete_bitset_opt,
+    );
+    segment_updater.add_segment(segment_entry);
+    Ok(true)
 }
 
 fn apply_deletes(
@@ -235,7 +244,9 @@ fn apply_deletes(
     }
     let segment_reader = SegmentReader::open(segment)?;
     let doc_to_opstamps = DocToOpstampMapping::from(doc_opstamps);
-    let mut deleted_bitset = BitSet::with_capacity(segment_reader.max_doc() as usize);
+
+    let max_doc = segment.meta().max_doc();
+    let mut deleted_bitset = BitSet::with_capacity(max_doc as usize);
     let may_have_deletes = compute_deleted_bitset(
         &mut deleted_bitset,
         &segment_reader,
@@ -407,7 +418,7 @@ impl IndexWriter {
                     let segment = index.new_segment();
                     index_documents(
                         mem_budget,
-                        &segment,
+                        segment,
                         &mut document_iterator,
                         &mut segment_updater,
                         delete_cursor.clone(),
@@ -1179,5 +1190,4 @@ mod tests {
         assert!(clear_again.is_ok());
         assert!(commit_again.is_ok());
     }
-
 }

src/indexer/merger.rs

@@ -190,8 +190,7 @@ impl IndexMerger {
         fast_field_serializer: &mut FastFieldSerializer,
         mut term_ord_mappings: HashMap<Field, TermOrdinalMapping>,
     ) -> Result<()> {
-        for (field_id, field_entry) in self.schema.fields().iter().enumerate() {
-            let field = Field(field_id as u32);
+        for (field, field_entry) in self.schema.fields() {
             let field_type = field_entry.field_type();
             match *field_type {
                 FieldType::HierarchicalFacet => {
@@ -649,15 +648,12 @@ impl IndexMerger {
         serializer: &mut InvertedIndexSerializer,
     ) -> Result<HashMap<Field, TermOrdinalMapping>> {
         let mut term_ordinal_mappings = HashMap::new();
-        for (field_ord, field_entry) in self.schema.fields().iter().enumerate() {
+        for (field, field_entry) in self.schema.fields() {
             if field_entry.is_indexed() {
-                let indexed_field = Field(field_ord as u32);
-                if let Some(term_ordinal_mapping) = self.write_postings_for_field(
-                    indexed_field,
-                    field_entry.field_type(),
-                    serializer,
-                )? {
-                    term_ordinal_mappings.insert(indexed_field, term_ordinal_mapping);
+                if let Some(term_ordinal_mapping) =
+                    self.write_postings_for_field(field, field_entry.field_type(), serializer)?
+                {
+                    term_ordinal_mappings.insert(field, term_ordinal_mapping);
                 }
             }
         }

src/indexer/mod.rs

@@ -28,3 +28,25 @@ pub use self::segment_writer::SegmentWriter;
 
 /// Alias for the default merge policy, which is the `LogMergePolicy`.
 pub type DefaultMergePolicy = LogMergePolicy;
+
+#[cfg(test)]
+mod tests {
+    use crate::schema::{self, Schema};
+    use crate::{Index, Term};
+    #[test]
+    fn test_advance_delete_bug() {
+        let mut schema_builder = Schema::builder();
+        let text_field = schema_builder.add_text_field("text", schema::TEXT);
+        let index = Index::create_from_tempdir(schema_builder.build()).unwrap();
+        let mut index_writer = index.writer_with_num_threads(1, 3_000_000).unwrap();
+        // there must be one deleted document in the segment
+        index_writer.add_document(doc!(text_field=>"b"));
+        index_writer.delete_term(Term::from_field_text(text_field, "b"));
+        // we need enough data to trigger the bug (at least 32 documents)
+        for _ in 0..32 {
+            index_writer.add_document(doc!(text_field=>"c"));
+        }
+        index_writer.commit().unwrap();
+        index_writer.commit().unwrap();
+    }
+}

src/indexer/segment_register.rs

@@ -134,5 +134,4 @@ mod tests {
         }
         assert_eq!(segment_ids(&segment_register), vec![segment_id_merged]);
     }
-
 }

src/indexer/segment_updater.rs

@@ -199,14 +199,12 @@ impl SegmentUpdater {
         self.0.pool.spawn_fn(move || Ok(f(me_clone)))
     }
 
-    pub fn add_segment(&self, segment_entry: SegmentEntry) -> bool {
+    pub fn add_segment(&self, segment_entry: SegmentEntry) {
         self.run_async(|segment_updater| {
             segment_updater.0.segment_manager.add_segment(segment_entry);
             segment_updater.consider_merge_options();
-            true
         })
         .forget();
-        true
     }
 
     /// Orders `SegmentManager` to remove all segments

src/indexer/segment_writer.rs

@@ -6,14 +6,15 @@ use crate::fieldnorm::FieldNormsWriter;
 use crate::indexer::segment_serializer::SegmentSerializer;
 use crate::postings::compute_table_size;
 use crate::postings::MultiFieldPostingsWriter;
-use crate::schema::FieldEntry;
 use crate::schema::FieldType;
 use crate::schema::Schema;
 use crate::schema::Term;
 use crate::schema::Value;
+use crate::schema::{Field, FieldEntry};
 use crate::tokenizer::BoxedTokenizer;
 use crate::tokenizer::FacetTokenizer;
-use crate::tokenizer::{TokenStream, Tokenizer};
+use crate::tokenizer::PreTokenizedStream;
+use crate::tokenizer::{TokenStream, TokenStreamChain, Tokenizer};
 use crate::DocId;
 use crate::Opstamp;
 use crate::Result;
@@ -70,12 +71,10 @@ impl SegmentWriter {
         let table_num_bits = initial_table_size(memory_budget)?;
         let segment_serializer = SegmentSerializer::for_segment(&mut segment)?;
         let multifield_postings = MultiFieldPostingsWriter::new(schema, table_num_bits);
-        let tokenizers =
-            schema
-                .fields()
-                .iter()
-                .map(FieldEntry::field_type)
-                .map(|field_type| match *field_type {
+        let tokenizers = schema
+            .fields()
+            .map(
+                |(_, field_entry): (Field, &FieldEntry)| match field_entry.field_type() {
                     FieldType::Str(ref text_options) => text_options
                         .get_indexing_options()
                         .and_then(|text_index_option| {
@@ -83,8 +82,9 @@ impl SegmentWriter {
                             segment.index().tokenizers().get(tokenizer_name)
                         }),
                     _ => None,
-                })
-                .collect();
+                },
+            )
+            .collect();
         Ok(SegmentWriter {
             max_doc: 0,
             multifield_postings,
@@ -159,26 +159,44 @@ impl SegmentWriter {
                     }
                 }
                 FieldType::Str(_) => {
-                    let num_tokens = if let Some(ref mut tokenizer) =
-                        self.tokenizers[field.0 as usize]
-                    {
-                        let texts: Vec<&str> = field_values
-                            .iter()
-                            .flat_map(|field_value| match *field_value.value() {
-                                Value::Str(ref text) => Some(text.as_str()),
-                                _ => None,
-                            })
-                            .collect();
-                        if texts.is_empty() {
-                            0
-                        } else {
-                            let mut token_stream = tokenizer.token_stream_texts(&texts[..]);
-                            self.multifield_postings
-                                .index_text(doc_id, field, &mut token_stream)
+                    let mut token_streams: Vec<Box<dyn TokenStream>> = vec![];
+                    let mut offsets = vec![];
+                    let mut total_offset = 0;
+
+                    for field_value in field_values {
+                        match field_value.value() {
+                            Value::PreTokStr(tok_str) => {
+                                offsets.push(total_offset);
+                                if let Some(last_token) = tok_str.tokens.last() {
+                                    total_offset += last_token.offset_to;
+                                }
+
+                                token_streams
+                                    .push(Box::new(PreTokenizedStream::from(tok_str.clone())));
+                            }
+                            Value::Str(ref text) => {
+                                if let Some(ref mut tokenizer) =
+                                    self.tokenizers[field.field_id() as usize]
+                                {
+                                    offsets.push(total_offset);
+                                    total_offset += text.len();
+
+                                    token_streams.push(tokenizer.token_stream(text));
+                                }
+                            }
+                            _ => (),
                         }
-                    } else {
+                    }
+
+                    let num_tokens = if token_streams.is_empty() {
                         0
+                    } else {
+                        let mut token_stream: Box<dyn TokenStream> =
+                            Box::new(TokenStreamChain::new(offsets, token_streams));
+                        self.multifield_postings
+                            .index_text(doc_id, field, &mut token_stream)
                     };
+
                     self.fieldnorms_writer.record(doc_id, field, num_tokens);
                 }
                 FieldType::U64(ref int_option) => {
@@ -296,5 +314,4 @@ mod tests {
         assert_eq!(initial_table_size(10_000_000).unwrap(), 17);
         assert_eq!(initial_table_size(1_000_000_000).unwrap(), 19);
     }
-
 }

src/lib.rs

@@ -212,15 +212,13 @@ pub type Score = f32;
 pub type SegmentLocalId = u32;
 
 impl DocAddress {
-    /// Return the segment ordinal.
-    /// The segment ordinal is an id identifying the segment
-    /// hosting the document. It is only meaningful, in the context
-    /// of a searcher.
+    /// Return the segment ordinal id that identifies the segment
+    /// hosting the document in the `Searcher` it is called from.
     pub fn segment_ord(self) -> SegmentLocalId {
         self.0
     }
 
-    /// Return the segment local `DocId`
+    /// Return the segment-local `DocId`
     pub fn doc(self) -> DocId {
         self.1
     }
@@ -229,11 +227,11 @@ impl DocAddress {
 /// `DocAddress` contains all the necessary information
 /// to identify a document given a `Searcher` object.
 ///
-/// It consists in an id identifying its segment, and
-/// its segment-local `DocId`.
+/// It consists of an id identifying its segment, and
+/// a segment-local `DocId`.
 ///
 /// The id used for the segment is actually an ordinal
-/// in the list of segment hold by a `Searcher`.
+/// in the list of `Segment`s held by a `Searcher`.
 #[derive(Debug, Clone, Copy, PartialEq, Eq, PartialOrd, Ord)]
 pub struct DocAddress(pub SegmentLocalId, pub DocId);
 

src/postings/compression/mod.rs

@@ -274,13 +274,15 @@ pub mod tests {
 mod bench {
 
     use super::*;
+    use rand::rngs::StdRng;
+    use rand::Rng;
     use rand::SeedableRng;
-    use rand::{Rng, XorShiftRng};
     use test::Bencher;
 
     fn generate_array_with_seed(n: usize, ratio: f64, seed_val: u8) -> Vec<u32> {
-        let seed: &[u8; 16] = &[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, seed_val];
-        let mut rng: XorShiftRng = XorShiftRng::from_seed(*seed);
+        let mut seed: [u8; 32] = [0; 32];
+        seed[31] = seed_val;
+        let mut rng = StdRng::from_seed(seed);
         (0u32..).filter(|_| rng.gen_bool(ratio)).take(n).collect()
     }
 

src/postings/mod.rs

@@ -356,9 +356,9 @@ pub mod tests {
 
     #[test]
     fn test_skip_next() {
-        let term_0 = Term::from_field_u64(Field(0), 0);
-        let term_1 = Term::from_field_u64(Field(0), 1);
-        let term_2 = Term::from_field_u64(Field(0), 2);
+        let term_0 = Term::from_field_u64(Field::from_field_id(0), 0);
+        let term_1 = Term::from_field_u64(Field::from_field_id(0), 1);
+        let term_2 = Term::from_field_u64(Field::from_field_id(0), 2);
 
         let num_docs = 300u32;
 
@@ -511,19 +511,19 @@ pub mod tests {
     }
 
     pub static TERM_A: Lazy<Term> = Lazy::new(|| {
-        let field = Field(0);
+        let field = Field::from_field_id(0);
         Term::from_field_text(field, "a")
     });
     pub static TERM_B: Lazy<Term> = Lazy::new(|| {
-        let field = Field(0);
+        let field = Field::from_field_id(0);
         Term::from_field_text(field, "b")
     });
     pub static TERM_C: Lazy<Term> = Lazy::new(|| {
-        let field = Field(0);
+        let field = Field::from_field_id(0);
         Term::from_field_text(field, "c")
     });
     pub static TERM_D: Lazy<Term> = Lazy::new(|| {
-        let field = Field(0);
+        let field = Field::from_field_id(0);
         Term::from_field_text(field, "d")
     });
 
@@ -622,23 +622,23 @@ pub mod tests {
             assert!(!postings_unopt.advance());
         }
     }
-
 }
 
 #[cfg(all(test, feature = "unstable"))]
 mod bench {
 
     use super::tests::*;
-    use docset::SkipResult;
-    use query::Intersection;
-    use schema::IndexRecordOption;
+    use crate::docset::SkipResult;
+    use crate::query::Intersection;
+    use crate::schema::IndexRecordOption;
+    use crate::tests;
+    use crate::DocSet;
     use test::{self, Bencher};
-    use tests;
-    use DocSet;
 
     #[bench]
     fn bench_segment_postings(b: &mut Bencher) {
-        let searcher = INDEX.searcher();
+        let reader = INDEX.reader().unwrap();
+        let searcher = reader.searcher();
         let segment_reader = searcher.segment_reader(0);
 
         b.iter(|| {
@@ -652,7 +652,8 @@ mod bench {
 
     #[bench]
     fn bench_segment_intersection(b: &mut Bencher) {
-        let searcher = INDEX.searcher();
+        let reader = INDEX.reader().unwrap();
+        let searcher = reader.searcher();
         let segment_reader = searcher.segment_reader(0);
         b.iter(|| {
             let segment_postings_a = segment_reader
@@ -682,7 +683,8 @@ mod bench {
     }
 
     fn bench_skip_next(p: f64, b: &mut Bencher) {
-        let searcher = INDEX.searcher();
+        let reader = INDEX.reader().unwrap();
+        let searcher = reader.searcher();
         let segment_reader = searcher.segment_reader(0);
         let docs = tests::sample(segment_reader.num_docs(), p);
 
@@ -737,7 +739,8 @@ mod bench {
 
     #[bench]
     fn bench_iterate_segment_postings(b: &mut Bencher) {
-        let searcher = INDEX.searcher();
+        let reader = INDEX.reader().unwrap();
+        let searcher = reader.searcher();
         let segment_reader = searcher.segment_reader(0);
         b.iter(|| {
             let n: u32 = test::black_box(17);

src/postings/postings_writer.rs

@@ -61,12 +61,12 @@ fn make_field_partition(
         .iter()
         .map(|(key, _, _)| Term::wrap(key).field())
         .enumerate();
-    let mut prev_field = Field(u32::max_value());
+    let mut prev_field_opt = None;
     let mut fields = vec![];
     let mut offsets = vec![];
     for (offset, field) in term_offsets_it {
-        if field != prev_field {
-            prev_field = field;
+        if Some(field) != prev_field_opt {
+            prev_field_opt = Some(field);
             fields.push(field);
             offsets.push(offset);
         }
@@ -86,8 +86,7 @@ impl MultiFieldPostingsWriter {
         let term_index = TermHashMap::new(table_bits);
         let per_field_postings_writers: Vec<_> = schema
             .fields()
-            .iter()
-            .map(|field_entry| posting_from_field_entry(field_entry))
+            .map(|(_, field_entry)| posting_from_field_entry(field_entry))
             .collect();
         MultiFieldPostingsWriter {
             heap: MemoryArena::new(),
@@ -107,7 +106,8 @@ impl MultiFieldPostingsWriter {
         field: Field,
         token_stream: &mut dyn TokenStream,
     ) -> u32 {
-        let postings_writer = self.per_field_postings_writers[field.0 as usize].deref_mut();
+        let postings_writer =
+            self.per_field_postings_writers[field.field_id() as usize].deref_mut();
         postings_writer.index_text(
             &mut self.term_index,
             doc,
@@ -118,7 +118,8 @@ impl MultiFieldPostingsWriter {
     }
 
     pub fn subscribe(&mut self, doc: DocId, term: &Term) -> UnorderedTermId {
-        let postings_writer = self.per_field_postings_writers[term.field().0 as usize].deref_mut();
+        let postings_writer =
+            self.per_field_postings_writers[term.field().field_id() as usize].deref_mut();
         postings_writer.subscribe(&mut self.term_index, doc, 0u32, term, &mut self.heap)
     }
 
@@ -160,7 +161,7 @@ impl MultiFieldPostingsWriter {
                 FieldType::Bytes => {}
             }
 
-            let postings_writer = &self.per_field_postings_writers[field.0 as usize];
+            let postings_writer = &self.per_field_postings_writers[field.field_id() as usize];
             let mut field_serializer =
                 serializer.new_field(field, postings_writer.total_num_tokens())?;
             postings_writer.serialize(

src/postings/stacker/expull.rs

@@ -310,6 +310,7 @@ mod bench {
     use super::super::MemoryArena;
     use super::ExpUnrolledLinkedList;
     use byteorder::{NativeEndian, WriteBytesExt};
+    use std::iter;
     use test::Bencher;
 
     const NUM_STACK: usize = 10_000;
@@ -335,11 +336,10 @@ mod bench {
     fn bench_push_stack(bench: &mut Bencher) {
         bench.iter(|| {
             let mut heap = MemoryArena::new();
-            let mut stacks = Vec::with_capacity(100);
-            for _ in 0..NUM_STACK {
-                let mut stack = ExpUnrolledLinkedList::new();
-                stacks.push(stack);
-            }
+            let mut stacks: Vec<ExpUnrolledLinkedList> =
+                iter::repeat_with(ExpUnrolledLinkedList::new)
+                    .take(NUM_STACK)
+                    .collect();
             for s in 0..NUM_STACK {
                 for i in 0u32..STACK_SIZE {
                     let t = s * 392017 % NUM_STACK;

src/query/all_query.rs

@@ -130,5 +130,4 @@ mod tests {
             assert!(!scorer.advance());
         }
     }
-
 }

src/query/bitset/mod.rs

@@ -216,7 +216,6 @@ mod tests {
             assert!(!docset.advance());
         }
     }
-
 }
 
 #[cfg(all(test, feature = "unstable"))]
@@ -224,13 +223,12 @@ mod bench {
 
     use super::BitSet;
     use super::BitSetDocSet;
-    use test;
-    use tests;
-    use DocSet;
+    use crate::test;
+    use crate::tests;
+    use crate::DocSet;
 
     #[bench]
     fn bench_bitset_1pct_insert(b: &mut test::Bencher) {
-        use tests;
         let els = tests::generate_nonunique_unsorted(1_000_000u32, 10_000);
         b.iter(|| {
             let mut bitset = BitSet::with_max_value(1_000_000);
@@ -242,7 +240,6 @@ mod bench {
 
     #[bench]
     fn bench_bitset_1pct_clone(b: &mut test::Bencher) {
-        use tests;
         let els = tests::generate_nonunique_unsorted(1_000_000u32, 10_000);
         let mut bitset = BitSet::with_max_value(1_000_000);
         for el in els {

src/query/bm25.rs

@@ -137,5 +137,4 @@ mod tests {
     fn test_idf() {
         assert_nearly_equals(idf(1, 2), 0.6931472);
     }
-
 }

src/query/boolean_query/boolean_query.rs

@@ -9,7 +9,8 @@ use crate::Result;
 use crate::Searcher;
 use std::collections::BTreeSet;
 
-/// The boolean query combines a set of queries
+/// The boolean query returns a set of documents
+/// that matches the Boolean combination of constituent subqueries.
 ///
 /// The documents matched by the boolean query are
 /// those which
@@ -19,6 +20,113 @@ use std::collections::BTreeSet;
 /// `MustNot` occurence.
 /// * match at least one of the subqueries that is not
 /// a `MustNot` occurence.
+///
+///
+/// You can combine other query types and their `Occur`ances into one `BooleanQuery`
+///
+/// ```rust
+///use tantivy::collector::Count;
+///use tantivy::doc;
+///use tantivy::query::{BooleanQuery, Occur, PhraseQuery, Query, TermQuery};
+///use tantivy::schema::{IndexRecordOption, Schema, TEXT};
+///use tantivy::Term;
+///use tantivy::{Index, Result};
+///
+///fn main() -> Result<()> {
+///    let mut schema_builder = Schema::builder();
+///    let title = schema_builder.add_text_field("title", TEXT);
+///    let body = schema_builder.add_text_field("body", TEXT);
+///    let schema = schema_builder.build();
+///    let index = Index::create_in_ram(schema);
+///    {
+///        let mut index_writer = index.writer(3_000_000)?;
+///        index_writer.add_document(doc!(
+///            title => "The Name of the Wind",
+///        ));
+///        index_writer.add_document(doc!(
+///            title => "The Diary of Muadib",
+///        ));
+///        index_writer.add_document(doc!(
+///            title => "A Dairy Cow",
+///            body => "hidden",
+///        ));
+///        index_writer.add_document(doc!(
+///            title => "A Dairy Cow",
+///            body => "found",
+///        ));
+///        index_writer.add_document(doc!(
+///            title => "The Diary of a Young Girl",
+///        ));
+///        index_writer.commit().unwrap();
+///    }
+///
+///    let reader = index.reader()?;
+///    let searcher = reader.searcher();
+///
+///    // Make TermQuery's for "girl" and "diary" in the title
+///    let girl_term_query: Box<dyn Query> = Box::new(TermQuery::new(
+///        Term::from_field_text(title, "girl"),
+///        IndexRecordOption::Basic,
+///    ));
+///    let diary_term_query: Box<dyn Query> = Box::new(TermQuery::new(
+///        Term::from_field_text(title, "diary"),
+///        IndexRecordOption::Basic,
+///    ));
+///    // A TermQuery with "found" in the body
+///    let body_term_query: Box<dyn Query> = Box::new(TermQuery::new(
+///        Term::from_field_text(body, "found"),
+///        IndexRecordOption::Basic,
+///    ));
+///    // TermQuery "diary" must and "girl" must not be present
+///    let queries_with_occurs1 = vec![
+///        (Occur::Must, diary_term_query.box_clone()),
+///        (Occur::MustNot, girl_term_query),
+///    ];
+///    // Make a BooleanQuery equivalent to
+///    // title:+diary title:-girl
+///    let diary_must_and_girl_mustnot = BooleanQuery::from(queries_with_occurs1);
+///    let count1 = searcher.search(&diary_must_and_girl_mustnot, &Count)?;
+///    assert_eq!(count1, 1);
+///
+///    // TermQuery for "cow" in the title
+///    let cow_term_query: Box<dyn Query> = Box::new(TermQuery::new(
+///        Term::from_field_text(title, "cow"),
+///        IndexRecordOption::Basic,
+///    ));
+///    // "title:diary OR title:cow"
+///    let title_diary_or_cow = BooleanQuery::from(vec![
+///        (Occur::Should, diary_term_query.box_clone()),
+///        (Occur::Should, cow_term_query),
+///    ]);
+///    let count2 = searcher.search(&title_diary_or_cow, &Count)?;
+///    assert_eq!(count2, 4);
+///
+///    // Make a `PhraseQuery` from a vector of `Term`s
+///    let phrase_query: Box<dyn Query> = Box::new(PhraseQuery::new(vec![
+///        Term::from_field_text(title, "dairy"),
+///        Term::from_field_text(title, "cow"),
+///    ]));
+///    // You can combine subqueries of different types into 1 BooleanQuery:
+///    // `TermQuery` and `PhraseQuery`
+///    // "title:diary OR "dairy cow"
+///    let term_of_phrase_query = BooleanQuery::from(vec![
+///        (Occur::Should, diary_term_query.box_clone()),
+///        (Occur::Should, phrase_query.box_clone()),
+///    ]);
+///    let count3 = searcher.search(&term_of_phrase_query, &Count)?;
+///    assert_eq!(count3, 4);
+///
+///    // You can nest one BooleanQuery inside another
+///    // body:found AND ("title:diary OR "dairy cow")
+///    let nested_query = BooleanQuery::from(vec![
+///        (Occur::Must, body_term_query),
+///        (Occur::Must, Box::new(term_of_phrase_query))
+///    ]);
+///    let count4 = searcher.search(&nested_query, &Count)?;
+///    assert_eq!(count4, 1);
+///    Ok(())
+///}
+/// ```
 #[derive(Debug)]
 pub struct BooleanQuery {
     subqueries: Vec<(Occur, Box<dyn Query>)>,

src/query/boolean_query/mod.rs

@@ -247,9 +247,7 @@ mod tests {
         let reader = index.reader().unwrap();
         let searcher = reader.searcher();
         let query_parser = QueryParser::for_index(&index, vec![title, text]);
-        let query = query_parser
-            .parse_query("Оксана Лифенко")
-            .unwrap();
+        let query = query_parser.parse_query("Оксана Лифенко").unwrap();
         let weight = query.weight(&searcher, true).unwrap();
         let mut scorer = weight.scorer(searcher.segment_reader(0u32)).unwrap();
         scorer.advance();

src/query/exclude.rs

@@ -175,5 +175,4 @@ mod tests {
             sample_skip,
         );
     }
-
 }

src/query/phrase_query/phrase_query.rs

@@ -40,7 +40,7 @@ impl PhraseQuery {
         PhraseQuery::new_with_offset(terms_with_offset)
     }
 
-    /// Creates a new `PhraseQuery` given a list of terms and there offsets.
+    /// Creates a new `PhraseQuery` given a list of terms and their offsets.
     ///
     /// Can be used to provide custom offset for each term.
     pub fn new_with_offset(mut terms: Vec<(usize, Term)>) -> PhraseQuery {
@@ -73,7 +73,7 @@ impl PhraseQuery {
             .collect::<Vec<Term>>()
     }
 
-    /// Returns the `PhraseWeight` for the given phrase query given a specific `searcher`.  
+    /// Returns the `PhraseWeight` for the given phrase query given a specific `searcher`.
     ///
     /// This function is the same as `.weight(...)` except it returns
     /// a specialized type `PhraseWeight` instead of a Boxed trait.

src/query/query_parser/query_parser.rs

@@ -674,13 +674,19 @@ mod test {
 
         test_parse_query_to_logical_ast_helper(
             "signed:-2324",
-            &format!("{:?}", Term::from_field_i64(Field(2u32), -2324)),
+            &format!(
+                "{:?}",
+                Term::from_field_i64(Field::from_field_id(2u32), -2324)
+            ),
             false,
         );
 
         test_parse_query_to_logical_ast_helper(
             "float:2.5",
-            &format!("{:?}", Term::from_field_f64(Field(10u32), 2.5)),
+            &format!(
+                "{:?}",
+                Term::from_field_f64(Field::from_field_id(10u32), 2.5)
+            ),
             false,
         );
     }

src/query/range_query.rs

@@ -479,5 +479,4 @@ mod tests {
             91
         );
     }
-
 }

src/query/reqopt_scorer.rs

@@ -190,5 +190,4 @@ mod tests {
             skip_docs,
         );
     }
-
 }

src/query/term_query/mod.rs

@@ -118,7 +118,7 @@ mod tests {
     #[test]
     fn test_term_query_debug() {
         let term_query = TermQuery::new(
-            Term::from_field_text(Field(1), "hello"),
+            Term::from_field_text(Field::from_field_id(1), "hello"),
             IndexRecordOption::WithFreqs,
         );
         assert_eq!(

src/query/union.rs

@@ -409,20 +409,17 @@ mod tests {
             vec![1, 2, 3, 7, 8, 9, 99, 100, 101, 500, 20000],
         );
     }
-
 }
 
 #[cfg(all(test, feature = "unstable"))]
 mod bench {
 
-    use query::score_combiner::DoNothingCombiner;
-    use query::ConstScorer;
-    use query::Union;
-    use query::VecDocSet;
+    use crate::query::score_combiner::DoNothingCombiner;
+    use crate::query::{ConstScorer, Union, VecDocSet};
+    use crate::tests;
+    use crate::DocId;
+    use crate::DocSet;
     use test::Bencher;
-    use tests;
-    use DocId;
-    use DocSet;
 
     #[bench]
     fn bench_union_3_high(bench: &mut Bencher) {

src/query/vec_docset.rs

@@ -82,5 +82,4 @@ pub mod tests {
         }
         assert_eq!(postings.fill_buffer(&mut buffer[..]), 9);
     }
-
 }

src/schema/document.rs

@@ -1,6 +1,7 @@
 use super::*;
 use crate::common::BinarySerializable;
 use crate::common::VInt;
+use crate::tokenizer::PreTokenizedString;
 use crate::DateTime;
 use itertools::Itertools;
 use std::io::{self, Read, Write};
@@ -29,8 +30,8 @@ impl From<Vec<FieldValue>> for Document {
 impl PartialEq for Document {
     fn eq(&self, other: &Document) -> bool {
         // super slow, but only here for tests
-        let mut self_field_values = self.field_values.clone();
-        let mut other_field_values = other.field_values.clone();
+        let mut self_field_values: Vec<&_> = self.field_values.iter().collect();
+        let mut other_field_values: Vec<&_> = other.field_values.iter().collect();
         self_field_values.sort();
         other_field_values.sort();
         self_field_values.eq(&other_field_values)
@@ -78,6 +79,16 @@ impl Document {
         self.add(FieldValue::new(field, value));
     }
 
+    /// Add a pre-tokenized text field.
+    pub fn add_pre_tokenized_text(
+        &mut self,
+        field: Field,
+        pre_tokenized_text: &PreTokenizedString,
+    ) {
+        let value = Value::PreTokStr(pre_tokenized_text.clone());
+        self.add(FieldValue::new(field, value));
+    }
+
     /// Add a u64 field
     pub fn add_u64(&mut self, field: Field, value: u64) {
         self.add(FieldValue::new(field, Value::U64(value)));
@@ -178,5 +189,4 @@ mod tests {
         doc.add_text(text_field, "My title");
         assert_eq!(doc.field_values().len(), 1);
     }
-
 }

src/schema/field.rs

@@ -3,14 +3,23 @@ use std::io;
 use std::io::Read;
 use std::io::Write;
 
-/// `Field` is actually a `u8` identifying a `Field`
-/// The schema is in charge of holding mapping between field names
-/// to `Field` objects.
-///
-/// Because the field id is a `u8`, tantivy can only have at most `255` fields.
-/// Value 255 is reserved.
+/// `Field` is represented by an unsigned 32-bit integer type
+/// The schema holds the mapping between field names and `Field` objects.
 #[derive(Copy, Clone, Debug, PartialEq, PartialOrd, Eq, Ord, Hash, Serialize, Deserialize)]
-pub struct Field(pub u32);
+pub struct Field(u32);
+
+impl Field {
+    /// Create a new field object for the given FieldId.
+    pub fn from_field_id(field_id: u32) -> Field {
+        Field(field_id)
+    }
+
+    /// Returns a u32 identifying uniquely a field within a schema.
+    #[allow(clippy::trivially_copy_pass_by_ref)]
+    pub fn field_id(&self) -> u32 {
+        self.0
+    }
+}
 
 impl BinarySerializable for Field {
     fn serialize<W: Write>(&self, writer: &mut W) -> io::Result<()> {

src/schema/field_type.rs

@@ -1,11 +1,11 @@
 use base64::decode;
 
-use crate::schema::{IntOptions, TextOptions};
-
 use crate::schema::Facet;
 use crate::schema::IndexRecordOption;
 use crate::schema::TextFieldIndexing;
 use crate::schema::Value;
+use crate::schema::{IntOptions, TextOptions};
+use crate::tokenizer::PreTokenizedString;
 use serde_json::Value as JsonValue;
 
 /// Possible error that may occur while parsing a field value
@@ -169,6 +169,28 @@ impl FieldType {
                     Err(ValueParsingError::TypeError(msg))
                 }
             },
+            JsonValue::Object(_) => match *self {
+                FieldType::Str(_) => {
+                    if let Ok(tok_str_val) =
+                        serde_json::from_value::<PreTokenizedString>(json.clone())
+                    {
+                        Ok(Value::PreTokStr(tok_str_val))
+                    } else {
+                        let msg = format!(
+                            "Json value {:?} cannot be translated to PreTokenizedString.",
+                            json
+                        );
+                        Err(ValueParsingError::TypeError(msg))
+                    }
+                }
+                _ => {
+                    let msg = format!(
+                        "Json value not supported error {:?}. Expected {:?}",
+                        json, self
+                    );
+                    Err(ValueParsingError::TypeError(msg))
+                }
+            },
             _ => {
                 let msg = format!(
                     "Json value not supported error {:?}. Expected {:?}",
@@ -184,7 +206,9 @@ impl FieldType {
 mod tests {
     use super::FieldType;
     use crate::schema::field_type::ValueParsingError;
+    use crate::schema::TextOptions;
     use crate::schema::Value;
+    use crate::tokenizer::{PreTokenizedString, Token};
 
     #[test]
     fn test_bytes_value_from_json() {
@@ -205,4 +229,71 @@ mod tests {
             _ => panic!("Expected parse failure for invalid base64"),
         }
     }
+
+    #[test]
+    fn test_pre_tok_str_value_from_json() {
+        let pre_tokenized_string_json = r#"{
+  "text": "The Old Man",
+  "tokens": [
+    {
+      "offset_from": 0,
+      "offset_to": 3,
+      "position": 0,
+      "text": "The",
+      "position_length": 1
+    },
+    {
+      "offset_from": 4,
+      "offset_to": 7,
+      "position": 1,
+      "text": "Old",
+      "position_length": 1
+    },
+    {
+      "offset_from": 8,
+      "offset_to": 11,
+      "position": 2,
+      "text": "Man",
+      "position_length": 1
+    }
+  ]
+}"#;
+
+        let expected_value = Value::PreTokStr(PreTokenizedString {
+            text: String::from("The Old Man"),
+            tokens: vec![
+                Token {
+                    offset_from: 0,
+                    offset_to: 3,
+                    position: 0,
+                    text: String::from("The"),
+                    position_length: 1,
+                },
+                Token {
+                    offset_from: 4,
+                    offset_to: 7,
+                    position: 1,
+                    text: String::from("Old"),
+                    position_length: 1,
+                },
+                Token {
+                    offset_from: 8,
+                    offset_to: 11,
+                    position: 2,
+                    text: String::from("Man"),
+                    position_length: 1,
+                },
+            ],
+        });
+
+        let deserialized_value = FieldType::Str(TextOptions::default())
+            .value_from_json(&serde_json::from_str(pre_tokenized_string_json).unwrap())
+            .unwrap();
+
+        assert_eq!(deserialized_value, expected_value);
+
+        let serialized_value_json = serde_json::to_string_pretty(&expected_value).unwrap();
+
+        assert_eq!(serialized_value_json, pre_tokenized_string_json);
+    }
 }

src/schema/mod.rs

@@ -174,5 +174,4 @@ mod tests {
         assert!(!is_valid_field_name("シャボン玉"));
         assert!(is_valid_field_name("my_text_field"));
     }
-
 }

src/schema/schema.rs

@@ -167,7 +167,7 @@ impl SchemaBuilder {
 
     /// Adds a field entry to the schema in build.
     fn add_field(&mut self, field_entry: FieldEntry) -> Field {
-        let field = Field(self.fields.len() as u32);
+        let field = Field::from_field_id(self.fields.len() as u32);
         let field_name = field_entry.name().to_string();
         self.fields.push(field_entry);
         self.fields_map.insert(field_name, field);
@@ -223,7 +223,7 @@ pub struct Schema(Arc<InnerSchema>);
 impl Schema {
     /// Return the `FieldEntry` associated to a `Field`.
     pub fn get_field_entry(&self, field: Field) -> &FieldEntry {
-        &self.0.fields[field.0 as usize]
+        &self.0.fields[field.field_id() as usize]
     }
 
     /// Return the field name for a given `Field`.
@@ -232,8 +232,12 @@ impl Schema {
     }
 
     /// Return the list of all the `Field`s.
-    pub fn fields(&self) -> &[FieldEntry] {
-        &self.0.fields
+    pub fn fields(&self) -> impl Iterator<Item = (Field, &FieldEntry)> {
+        self.0
+            .fields
+            .iter()
+            .enumerate()
+            .map(|(field_id, field_entry)| (Field::from_field_id(field_id as u32), field_entry))
     }
 
     /// Creates a new builder.
@@ -485,13 +489,32 @@ mod tests {
 
         let schema: Schema = serde_json::from_str(expected).unwrap();
 
-        let mut fields = schema.fields().iter();
-
-        assert_eq!("title", fields.next().unwrap().name());
-        assert_eq!("author", fields.next().unwrap().name());
-        assert_eq!("count", fields.next().unwrap().name());
-        assert_eq!("popularity", fields.next().unwrap().name());
-        assert_eq!("score", fields.next().unwrap().name());
+        let mut fields = schema.fields();
+        {
+            let (field, field_entry) = fields.next().unwrap();
+            assert_eq!("title", field_entry.name());
+            assert_eq!(0, field.field_id());
+        }
+        {
+            let (field, field_entry) = fields.next().unwrap();
+            assert_eq!("author", field_entry.name());
+            assert_eq!(1, field.field_id());
+        }
+        {
+            let (field, field_entry) = fields.next().unwrap();
+            assert_eq!("count", field_entry.name());
+            assert_eq!(2, field.field_id());
+        }
+        {
+            let (field, field_entry) = fields.next().unwrap();
+            assert_eq!("popularity", field_entry.name());
+            assert_eq!(3, field.field_id());
+        }
+        {
+            let (field, field_entry) = fields.next().unwrap();
+            assert_eq!("score", field_entry.name());
+            assert_eq!(4, field.field_id());
+        }
         assert!(fields.next().is_none());
     }
 

src/schema/term.rs

@@ -105,7 +105,7 @@ impl Term {
         if self.0.len() < 4 {
             self.0.resize(4, 0u8);
         }
-        BigEndian::write_u32(&mut self.0[0..4], field.0);
+        BigEndian::write_u32(&mut self.0[0..4], field.field_id());
     }
 
     /// Sets a u64 value in the term.
@@ -157,7 +157,7 @@ where
 
     /// Returns the field.
     pub fn field(&self) -> Field {
-        Field(BigEndian::read_u32(&self.0.as_ref()[..4]))
+        Field::from_field_id(BigEndian::read_u32(&self.0.as_ref()[..4]))
     }
 
     /// Returns the `u64` value stored in a term.
@@ -227,7 +227,7 @@ impl fmt::Debug for Term {
         write!(
             f,
             "Term(field={},bytes={:?})",
-            self.field().0,
+            self.field().field_id(),
             self.value_bytes()
         )
     }

src/schema/value.rs

@@ -1,4 +1,5 @@
 use crate::schema::Facet;
+use crate::tokenizer::PreTokenizedString;
 use crate::DateTime;
 use serde::de::Visitor;
 use serde::{Deserialize, Deserializer, Serialize, Serializer};
@@ -10,6 +11,8 @@ use std::{cmp::Ordering, fmt};
 pub enum Value {
     /// The str type is used for any text information.
     Str(String),
+    /// Pre-tokenized str type,
+    PreTokStr(PreTokenizedString),
     /// Unsigned 64-bits Integer `u64`
     U64(u64),
     /// Signed 64-bits Integer `i64`
@@ -29,6 +32,7 @@ impl Ord for Value {
     fn cmp(&self, other: &Self) -> Ordering {
         match (self, other) {
             (Value::Str(l), Value::Str(r)) => l.cmp(r),
+            (Value::PreTokStr(l), Value::PreTokStr(r)) => l.cmp(r),
             (Value::U64(l), Value::U64(r)) => l.cmp(r),
             (Value::I64(l), Value::I64(r)) => l.cmp(r),
             (Value::Date(l), Value::Date(r)) => l.cmp(r),
@@ -44,6 +48,8 @@ impl Ord for Value {
             }
             (Value::Str(_), _) => Ordering::Less,
             (_, Value::Str(_)) => Ordering::Greater,
+            (Value::PreTokStr(_), _) => Ordering::Less,
+            (_, Value::PreTokStr(_)) => Ordering::Greater,
             (Value::U64(_), _) => Ordering::Less,
             (_, Value::U64(_)) => Ordering::Greater,
             (Value::I64(_), _) => Ordering::Less,
@@ -65,6 +71,7 @@ impl Serialize for Value {
     {
         match *self {
             Value::Str(ref v) => serializer.serialize_str(v),
+            Value::PreTokStr(ref v) => v.serialize(serializer),
             Value::U64(u) => serializer.serialize_u64(u),
             Value::I64(u) => serializer.serialize_i64(u),
             Value::F64(u) => serializer.serialize_f64(u),
@@ -124,6 +131,15 @@ impl Value {
         }
     }
 
+    /// Returns the tokenized text, provided the value is of the `PreTokStr` type.
+    /// (Returns None if the value is not of the `PreTokStr` type).
+    pub fn tokenized_text(&self) -> Option<&PreTokenizedString> {
+        match *self {
+            Value::PreTokStr(ref tok_text) => Some(tok_text),
+            _ => None,
+        }
+    }
+
     /// Returns the u64-value, provided the value is of the `U64` type.
     ///
     /// # Panics
@@ -221,6 +237,7 @@ mod binary_serialize {
     use super::Value;
     use crate::common::{f64_to_u64, u64_to_f64, BinarySerializable};
     use crate::schema::Facet;
+    use crate::tokenizer::PreTokenizedString;
     use chrono::{TimeZone, Utc};
     use std::io::{self, Read, Write};
 
@@ -231,6 +248,11 @@ mod binary_serialize {
     const BYTES_CODE: u8 = 4;
     const DATE_CODE: u8 = 5;
     const F64_CODE: u8 = 6;
+    const EXT_CODE: u8 = 7;
+
+    // extended types
+
+    const TOK_STR_CODE: u8 = 0;
 
     impl BinarySerializable for Value {
         fn serialize<W: Write>(&self, writer: &mut W) -> io::Result<()> {
@@ -239,6 +261,18 @@ mod binary_serialize {
                     TEXT_CODE.serialize(writer)?;
                     text.serialize(writer)
                 }
+                Value::PreTokStr(ref tok_str) => {
+                    EXT_CODE.serialize(writer)?;
+                    TOK_STR_CODE.serialize(writer)?;
+                    if let Ok(text) = serde_json::to_string(tok_str) {
+                        text.serialize(writer)
+                    } else {
+                        Err(io::Error::new(
+                            io::ErrorKind::Other,
+                            "Failed to dump Value::PreTokStr(_) to json.",
+                        ))
+                    }
+                }
                 Value::U64(ref val) => {
                     U64_CODE.serialize(writer)?;
                     val.serialize(writer)
@@ -290,6 +324,30 @@ mod binary_serialize {
                 }
                 HIERARCHICAL_FACET_CODE => Ok(Value::Facet(Facet::deserialize(reader)?)),
                 BYTES_CODE => Ok(Value::Bytes(Vec::<u8>::deserialize(reader)?)),
+                EXT_CODE => {
+                    let ext_type_code = u8::deserialize(reader)?;
+                    match ext_type_code {
+                        TOK_STR_CODE => {
+                            let str_val = String::deserialize(reader)?;
+                            if let Ok(value) = serde_json::from_str::<PreTokenizedString>(&str_val)
+                            {
+                                Ok(Value::PreTokStr(value))
+                            } else {
+                                Err(io::Error::new(
+                                    io::ErrorKind::Other,
+                                    "Failed to parse string data as Value::PreTokStr(_).",
+                                ))
+                            }
+                        }
+                        _ => Err(io::Error::new(
+                            io::ErrorKind::InvalidData,
+                            format!(
+                                "No extened field type is associated with code {:?}",
+                                ext_type_code
+                            ),
+                        )),
+                    }
+                }
                 _ => Err(io::Error::new(
                     io::ErrorKind::InvalidData,
                     format!("No field type is associated with code {:?}", type_code),

src/store/mod.rs

@@ -120,17 +120,16 @@ pub mod tests {
             );
         }
     }
-
 }
 
 #[cfg(all(test, feature = "unstable"))]
 mod bench {
 
     use super::tests::write_lorem_ipsum_store;
-    use directory::Directory;
-    use directory::RAMDirectory;
+    use crate::directory::Directory;
+    use crate::directory::RAMDirectory;
+    use crate::store::StoreReader;
     use std::path::Path;
-    use store::StoreReader;
     use test::Bencher;
 
     #[bench]

src/store/skiplist/mod.rs

@@ -165,5 +165,4 @@ mod tests {
         assert_eq!(output.len(), 65);
         assert_eq!(output[0], 128u8 + 3u8);
     }
-
 }

src/termdict/term_info_store.rs

@@ -328,5 +328,4 @@ mod tests {
             assert_eq!(term_info_store.get(i as u64), term_infos[i]);
         }
     }
-
 }

src/tokenizer/lower_caser.rs

@@ -98,10 +98,6 @@ mod tests {
     #[test]
     fn test_lowercaser() {
         assert_eq!(lowercase_helper("Tree"), vec!["tree".to_string()]);
-        assert_eq!(
-            lowercase_helper("Русский"),
-            vec!["русский".to_string()]
-        );
+        assert_eq!(lowercase_helper("Русский"), vec!["русский".to_string()]);
     }
-
 }

src/tokenizer/mod.rs

@@ -136,6 +136,7 @@ mod simple_tokenizer;
 mod stemmer;
 mod stop_word_filter;
 mod token_stream_chain;
+mod tokenized_string;
 mod tokenizer;
 mod tokenizer_manager;
 
@@ -152,7 +153,9 @@ pub use self::stop_word_filter::StopWordFilter;
 pub(crate) use self::token_stream_chain::TokenStreamChain;
 pub use self::tokenizer::BoxedTokenizer;
 
+pub use self::tokenized_string::{PreTokenizedStream, PreTokenizedString};
 pub use self::tokenizer::{Token, TokenFilter, TokenStream, Tokenizer};
+
 pub use self::tokenizer_manager::TokenizerManager;
 
 /// Maximum authorized len (in bytes) for a token.
@@ -281,5 +284,4 @@ pub mod tests {
             assert!(tokens.is_empty());
         }
     }
-
 }

src/tokenizer/ngram_tokenizer.rs

@@ -460,5 +460,4 @@ mod tests {
         assert_eq!(it.next(), Some((8, 9)));
         assert_eq!(it.next(), None);
     }
-
 }

src/tokenizer/token_stream_chain.rs

@@ -97,5 +97,4 @@ mod tests {
 
         assert!(!token_chain.advance());
     }
-
 }

src/tokenizer/tokenized_string.rs

@@ -0,0 +1,189 @@
+use crate::tokenizer::{Token, TokenStream, TokenStreamChain};
+use std::cmp::Ordering;
+
+/// Struct representing pre-tokenized text
+#[derive(Debug, Clone, Serialize, Deserialize, Eq, PartialEq)]
+pub struct PreTokenizedString {
+    /// Original text
+    pub text: String,
+    /// Tokens derived from the text
+    pub tokens: Vec<Token>,
+}
+
+impl Ord for PreTokenizedString {
+    fn cmp(&self, other: &Self) -> Ordering {
+        self.text.cmp(&other.text)
+    }
+}
+
+impl PartialOrd for PreTokenizedString {
+    fn partial_cmp(&self, other: &Self) -> Option<Ordering> {
+        Some(self.cmp(other))
+    }
+}
+
+/// TokenStream implementation which wraps PreTokenizedString
+pub struct PreTokenizedStream {
+    tokenized_string: PreTokenizedString,
+    current_token: i64,
+}
+
+impl From<PreTokenizedString> for PreTokenizedStream {
+    fn from(s: PreTokenizedString) -> PreTokenizedStream {
+        PreTokenizedStream {
+            tokenized_string: s,
+            current_token: -1,
+        }
+    }
+}
+
+impl PreTokenizedStream {
+    /// Creates a TokenStream from PreTokenizedString array
+    pub fn chain_tokenized_strings<'a>(
+        tok_strings: &'a [&'a PreTokenizedString],
+    ) -> Box<dyn TokenStream + 'a> {
+        if tok_strings.len() == 1 {
+            Box::new(PreTokenizedStream::from((*tok_strings[0]).clone()))
+        } else {
+            let mut offsets = vec![];
+            let mut total_offset = 0;
+            for &tok_string in tok_strings {
+                offsets.push(total_offset);
+                if let Some(last_token) = tok_string.tokens.last() {
+                    total_offset += last_token.offset_to;
+                }
+            }
+            let token_streams: Vec<_> = tok_strings
+                .iter()
+                .map(|tok_string| PreTokenizedStream::from((*tok_string).clone()))
+                .collect();
+            Box::new(TokenStreamChain::new(offsets, token_streams))
+        }
+    }
+}
+
+impl TokenStream for PreTokenizedStream {
+    fn advance(&mut self) -> bool {
+        self.current_token += 1;
+        self.current_token < self.tokenized_string.tokens.len() as i64
+    }
+
+    fn token(&self) -> &Token {
+        assert!(
+            self.current_token >= 0,
+            "TokenStream not initialized. You should call advance() at least once."
+        );
+        &self.tokenized_string.tokens[self.current_token as usize]
+    }
+
+    fn token_mut(&mut self) -> &mut Token {
+        assert!(
+            self.current_token >= 0,
+            "TokenStream not initialized. You should call advance() at least once."
+        );
+        &mut self.tokenized_string.tokens[self.current_token as usize]
+    }
+}
+
+#[cfg(test)]
+mod tests {
+
+    use super::*;
+
+    use crate::tokenizer::Token;
+
+    #[test]
+    fn test_tokenized_stream() {
+        let tok_text = PreTokenizedString {
+            text: String::from("A a"),
+            tokens: vec![
+                Token {
+                    offset_from: 0,
+                    offset_to: 1,
+                    position: 0,
+                    text: String::from("A"),
+                    position_length: 1,
+                },
+                Token {
+                    offset_from: 2,
+                    offset_to: 3,
+                    position: 1,
+                    text: String::from("a"),
+                    position_length: 1,
+                },
+            ],
+        };
+
+        let mut token_stream = PreTokenizedStream::from(tok_text.clone());
+
+        for expected_token in tok_text.tokens {
+            assert!(token_stream.advance());
+            assert_eq!(token_stream.token(), &expected_token);
+        }
+        assert!(!token_stream.advance());
+    }
+
+    #[test]
+    fn test_chain_tokenized_strings() {
+        let tok_text = PreTokenizedString {
+            text: String::from("A a"),
+            tokens: vec![
+                Token {
+                    offset_from: 0,
+                    offset_to: 1,
+                    position: 0,
+                    text: String::from("A"),
+                    position_length: 1,
+                },
+                Token {
+                    offset_from: 2,
+                    offset_to: 3,
+                    position: 1,
+                    text: String::from("a"),
+                    position_length: 1,
+                },
+            ],
+        };
+
+        let chain_parts = vec![&tok_text, &tok_text];
+
+        let mut token_stream = PreTokenizedStream::chain_tokenized_strings(&chain_parts[..]);
+
+        let expected_tokens = vec![
+            Token {
+                offset_from: 0,
+                offset_to: 1,
+                position: 0,
+                text: String::from("A"),
+                position_length: 1,
+            },
+            Token {
+                offset_from: 2,
+                offset_to: 3,
+                position: 1,
+                text: String::from("a"),
+                position_length: 1,
+            },
+            Token {
+                offset_from: 3,
+                offset_to: 4,
+                position: 3,
+                text: String::from("A"),
+                position_length: 1,
+            },
+            Token {
+                offset_from: 5,
+                offset_to: 6,
+                position: 4,
+                text: String::from("a"),
+                position_length: 1,
+            },
+        ];
+
+        for expected_token in expected_tokens {
+            assert!(token_stream.advance());
+            assert_eq!(token_stream.token(), &expected_token);
+        }
+        assert!(!token_stream.advance());
+    }
+}

src/tokenizer/tokenizer.rs

@@ -4,7 +4,7 @@ use crate::tokenizer::TokenStreamChain;
 use std::borrow::{Borrow, BorrowMut};
 
 /// Token
-#[derive(Debug, Clone)]
+#[derive(Debug, Clone, Serialize, Deserialize, Eq, PartialEq)]
 pub struct Token {
     /// Offset (byte index) of the first character of the token.
     /// Offsets shall not be modified by token filters.

tests/failpoints/mod.rs

@@ -1,5 +1,4 @@
 use fail;
-use std::io::Write;
 use std::path::Path;
 use tantivy::directory::{Directory, ManagedDirectory, RAMDirectory, TerminatingWrite};
 use tantivy::doc;