Removed temporary comment

CHANGELOG.md

@@ -1,32 +1,9 @@
-Tantivy 0.13.0
-======================
-- Bugfix in `FuzzyTermQuery` not matching terms by prefix when it should (@Peachball)
-- Relaxed constraints on the custom/tweak score functions. At the segment level, they can be mut, and they are not required to be Sync + Send.
-- `MMapDirectory::open` does not return a `Result` anymore.
-- Change in the DocSet and Scorer API. (@fulmicoton). 
-A freshly created DocSet point directly to their first doc. A sentinel value called TERMINATED marks the end of a DocSet.
-`.advance()` returns the new DocId. `Scorer::skip(target)` has been replaced by `Scorer::seek(target)` and returns the resulting DocId.
-As a result, iterating through DocSet now looks as follows
-```rust
-let mut doc = docset.doc();
-while doc != TERMINATED {
-   // ...
-   doc = docset.advance();
-}
-```
-The change made it possible to greatly simplify a lot of the docset's code.
-- Misc internal optimization and introduction of the `Scorer::for_each_pruning` function. (@fulmicoton)
-- Added an offset option to the Top(.*)Collectors. (@robyoung)
-
-
 Tantivy 0.12.0
 ======================
 - Removing static dispatch in tokenizers for simplicity. (#762)
 - Added backward iteration for `TermDictionary` stream. (@halvorboe)
 - Fixed a performance issue when searching for the posting lists of a missing term (@audunhalland)
 - Added a configurable maximum number of docs (10M by default) for a segment to be considered for merge (@hntd187, landed by @halvorboe #713) 
-- Important Bugfix #777, causing tantivy to retain memory mapping. (diagnosed by @poljar)
-- Added support for field boosting. (#547, @fulmicoton)
 
 ## How to update?
 

Cargo.toml

@@ -5,7 +5,7 @@ authors = ["Paul Masurel <paul.masurel@gmail.com>"]
 license = "MIT"
 categories = ["database-implementations", "data-structures"]
 description = """Search engine library"""
-documentation = "https://docs.rs/tantivy/"
+documentation = "https://tantivy-search.github.io/tantivy/tantivy/index.html"
 homepage = "https://github.com/tantivy-search/tantivy"
 repository = "https://github.com/tantivy-search/tantivy"
 readme = "README.md"
@@ -13,23 +13,25 @@ keywords = ["search", "information", "retrieval"]
 edition = "2018"
 
 [dependencies]
-base64 = "0.12.0"
+base64 = "0.11.0"
 byteorder = "1.0"
 crc32fast = "1.2.0"
 once_cell = "1.0"
 regex ={version = "1.3.0", default-features = false, features = ["std"]}
-tantivy-fst = "0.3"
+tantivy-fst = "0.2.1"
 memmap = {version = "0.7", optional=true}
 lz4 = {version="1.20", optional=true}
-snap = "1"
+snap = {version="0.2"}
 atomicwrites = {version="0.2.2", optional=true}
 tempfile = "3.0"
 log = "0.4"
-serde = {version="1.0", features=["derive"]}
+serde = "1.0"
+serde_derive = "1.0"
 serde_json = "1.0"
 num_cpus = "1.2"
 fs2={version="0.4", optional=true}
-levenshtein_automata = "0.2"
+itertools = "0.8"
+levenshtein_automata = {version="0.1", features=["fst_automaton"]}
 notify = {version="4", optional=true}
 uuid = { version = "0.8", features = ["v4", "serde"] }
 crossbeam = "0.7"
@@ -38,14 +40,14 @@ owning_ref = "0.4"
 stable_deref_trait = "1.0.0"
 rust-stemmers = "1.2"
 downcast-rs = { version="1.0" }
-tantivy-query-grammar = { version="0.13", path="./query-grammar" }
+tantivy-query-grammar = { version="0.11", path="./query-grammar" }
 bitpacking = {version="0.8", default-features = false, features=["bitpacker4x"]}
 census = "0.4"
 fnv = "1.0.6"
 owned-read = "0.4"
 failure = "0.1"
 htmlescape = "0.3.1"
-fail = "0.4"
+fail = "0.3"
 murmurhash32 = "0.2"
 chrono = "0.4"
 smallvec = "1.0"
@@ -60,7 +62,7 @@ maplit = "1"
 matches = "0.1.8"
 
 [dev-dependencies.fail]
-version = "0.4"
+version = "0.3"
 features = ["failpoints"]
 
 [profile.release]

README.md

@@ -31,20 +31,16 @@ Tantivy is, in fact, strongly inspired by Lucene's design.
 
 # Benchmark
 
+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 
 performance for different type of queries / collection.
 
-
-In general, Tantivy tends to be 
-- slower than Lucene on union with a Top-K due to Block-WAND optimization.
-- faster than Lucene on intersection and phrase queries. 
-
-Your mileage WILL vary depending on the nature of queries and their load.
-
 # Features
 
 - Full-text search
-- 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)), Japanese ([lindera](https://github.com/lindera-morphology/lindera-tantivy) and [tantivy-tokenizer-tiny-segmente](https://crates.io/crates/tantivy-tokenizer-tiny-segmenter)) and Korean ([lindera](https://github.com/lindera-morphology/lindera-tantivy) + [lindera-ko-dic-builder](https://github.com/lindera-morphology/lindera-ko-dic-builder))
+- 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)
@@ -63,17 +59,18 @@ Your mileage WILL vary depending on the nature of queries and their load.
 - Configurable indexing (optional term frequency and position indexing)
 - Cheesy logo with a horse
 
-## Non-features
+# Non-features
 
 - 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.
 
-
-# Getting started
+# Supported OS and compiler
 
 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.

appveyor.yml

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

examples/faceted_search_with_tweaked_score.rs

@@ -1,98 +0,0 @@
-use std::collections::HashSet;
-use tantivy::collector::TopDocs;
-use tantivy::doc;
-use tantivy::query::BooleanQuery;
-use tantivy::schema::*;
-use tantivy::{DocId, Index, Score, SegmentReader};
-
-fn main() -> tantivy::Result<()> {
-    let mut schema_builder = Schema::builder();
-
-    let title = schema_builder.add_text_field("title", STORED);
-    let ingredient = schema_builder.add_facet_field("ingredient");
-
-    let schema = schema_builder.build();
-    let index = Index::create_in_ram(schema.clone());
-
-    let mut index_writer = index.writer(30_000_000)?;
-
-    index_writer.add_document(doc!(
-        title => "Fried egg",
-        ingredient => Facet::from("/ingredient/egg"),
-        ingredient => Facet::from("/ingredient/oil"),
-    ));
-    index_writer.add_document(doc!(
-        title => "Scrambled egg",
-        ingredient => Facet::from("/ingredient/egg"),
-        ingredient => Facet::from("/ingredient/butter"),
-        ingredient => Facet::from("/ingredient/milk"),
-        ingredient => Facet::from("/ingredient/salt"),
-    ));
-    index_writer.add_document(doc!(
-        title => "Egg rolls",
-        ingredient => Facet::from("/ingredient/egg"),
-        ingredient => Facet::from("/ingredient/garlic"),
-        ingredient => Facet::from("/ingredient/salt"),
-        ingredient => Facet::from("/ingredient/oil"),
-        ingredient => Facet::from("/ingredient/tortilla-wrap"),
-        ingredient => Facet::from("/ingredient/mushroom"),
-    ));
-    index_writer.commit()?;
-
-    let reader = index.reader()?;
-    let searcher = reader.searcher();
-    {
-        let facets = vec![
-            Facet::from("/ingredient/egg"),
-            Facet::from("/ingredient/oil"),
-            Facet::from("/ingredient/garlic"),
-            Facet::from("/ingredient/mushroom"),
-        ];
-        let query = BooleanQuery::new_multiterms_query(
-            facets
-                .iter()
-                .map(|key| Term::from_facet(ingredient, &key))
-                .collect(),
-        );
-        let top_docs_by_custom_score =
-            TopDocs::with_limit(2).tweak_score(move |segment_reader: &SegmentReader| {
-                let mut ingredient_reader = segment_reader.facet_reader(ingredient).unwrap();
-                let facet_dict = ingredient_reader.facet_dict();
-
-                let query_ords: HashSet<u64> = facets
-                    .iter()
-                    .filter_map(|key| facet_dict.term_ord(key.encoded_str()))
-                    .collect();
-
-                let mut facet_ords_buffer: Vec<u64> = Vec::with_capacity(20);
-
-                move |doc: DocId, original_score: Score| {
-                    ingredient_reader.facet_ords(doc, &mut facet_ords_buffer);
-                    let missing_ingredients = facet_ords_buffer
-                        .iter()
-                        .filter(|ord| !query_ords.contains(ord))
-                        .count();
-                    let tweak = 1.0 / 4_f32.powi(missing_ingredients as i32);
-
-                    original_score * tweak
-                }
-            });
-        let top_docs = searcher.search(&query, &top_docs_by_custom_score)?;
-
-        let titles: Vec<String> = top_docs
-            .iter()
-            .map(|(_, doc_id)| {
-                searcher
-                    .doc(*doc_id)
-                    .unwrap()
-                    .get_first(title)
-                    .unwrap()
-                    .text()
-                    .unwrap()
-                    .to_owned()
-            })
-            .collect();
-        assert_eq!(titles, vec!["Fried egg", "Egg rolls"]);
-    }
-    Ok(())
-}

examples/iterating_docs_and_positions.rs

@@ -10,7 +10,7 @@
 // ---
 // Importing tantivy...
 use tantivy::schema::*;
-use tantivy::{doc, DocSet, Index, Postings, TERMINATED};
+use tantivy::{doc, DocId, DocSet, Index, Postings};
 
 fn main() -> tantivy::Result<()> {
     // We first create a schema for the sake of the
@@ -62,11 +62,12 @@ fn main() -> tantivy::Result<()> {
         {
             // this buffer will be used to request for positions
             let mut positions: Vec<u32> = Vec::with_capacity(100);
-            let mut doc_id = segment_postings.doc();
-            while doc_id != TERMINATED {
+            while segment_postings.advance() {
+                // the number of time the term appears in the document.
+                let doc_id: DocId = segment_postings.doc(); //< do not try to access this before calling advance once.
+
                 // This MAY contains deleted documents as well.
                 if segment_reader.is_deleted(doc_id) {
-                    doc_id = segment_postings.advance();
                     continue;
                 }
 
@@ -85,7 +86,6 @@ fn main() -> tantivy::Result<()> {
                 // Doc 2: TermFreq 1: [0]
                 // ```
                 println!("Doc {}: TermFreq {}: {:?}", doc_id, term_freq, positions);
-                doc_id = segment_postings.advance();
             }
         }
     }

examples/pre_tokenized_text.rs

@@ -9,10 +9,11 @@
 // - import tokenized text straight from json,
 // - perform a search on documents with pre-tokenized text
 
+use tantivy::tokenizer::{PreTokenizedString, SimpleTokenizer, Token, Tokenizer};
+
 use tantivy::collector::{Count, TopDocs};
 use tantivy::query::TermQuery;
 use tantivy::schema::*;
-use tantivy::tokenizer::{PreTokenizedString, SimpleTokenizer, Token, Tokenizer};
 use tantivy::{doc, Index, ReloadPolicy};
 use tempfile::TempDir;
 

query-grammar/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "tantivy-query-grammar"
-version = "0.13.0"
+version = "0.11.0"
 authors = ["Paul Masurel <paul.masurel@gmail.com>"]
 license = "MIT"
 categories = ["database-implementations", "data-structures"]
@@ -13,4 +13,4 @@ keywords = ["search", "information", "retrieval"]
 edition = "2018"
 
 [dependencies]
-combine = {version="4", default-features=false, features=[] }
+combine = ">=3.6.0,<4.0.0"

query-grammar/src/lib.rs

@@ -1,3 +1,5 @@
+#![recursion_limit = "100"]
+
 mod occur;
 mod query_grammar;
 mod user_input_ast;

query-grammar/src/query_grammar.rs

@@ -1,209 +1,171 @@
-use super::user_input_ast::{UserInputAST, UserInputBound, UserInputLeaf, UserInputLiteral};
+use super::user_input_ast::*;
 use crate::Occur;
-use combine::error::StringStreamError;
-use combine::parser::char::{char, digit, letter, space, spaces, string};
-use combine::parser::Parser;
-use combine::{
-    attempt, choice, eof, many, many1, one_of, optional, parser, satisfy, skip_many1, value,
-};
+use combine::char::*;
+use combine::error::StreamError;
+use combine::stream::StreamErrorFor;
+use combine::*;
 
-fn field<'a>() -> impl Parser<&'a str, Output = String> {
-    (
-        letter(),
-        many(satisfy(|c: char| c.is_alphanumeric() || c == '_')),
-    )
-        .skip(char(':'))
+parser! {
+    fn field[I]()(I) -> String
+    where [I: Stream<Item = char>] {
+        (
+            letter(),
+            many(satisfy(|c: char| c.is_alphanumeric() || c == '_')),
+        ).skip(char(':')).map(|(s1, s2): (char, String)| format!("{}{}", s1, s2))
+    }
+}
+
+parser! {
+    fn word[I]()(I) -> String
+    where [I: Stream<Item = char>] {
+        (
+            satisfy(|c: char| !c.is_whitespace() && !['-', '`', ':', '{', '}', '"', '[', ']', '(',')'].contains(&c) ),
+            many(satisfy(|c: char| !c.is_whitespace() && ![':', '{', '}', '"', '[', ']', '(',')'].contains(&c)))
+        )
         .map(|(s1, s2): (char, String)| format!("{}{}", s1, s2))
+        .and_then(|s: String|
+           match s.as_str() {
+             "OR" => Err(StreamErrorFor::<I>::unexpected_static_message("OR")),
+             "AND" => Err(StreamErrorFor::<I>::unexpected_static_message("AND")),
+             "NOT" => Err(StreamErrorFor::<I>::unexpected_static_message("NOT")),
+             _ => Ok(s)
+           })
+    }
 }
 
-fn word<'a>() -> impl Parser<&'a str, Output = String> {
-    (
-        satisfy(|c: char| {
-            !c.is_whitespace()
-                && !['-', '^', '`', ':', '{', '}', '"', '[', ']', '(', ')'].contains(&c)
-        }),
-        many(satisfy(|c: char| {
-            !c.is_whitespace() && ![':', '^', '{', '}', '"', '[', ']', '(', ')'].contains(&c)
-        })),
-    )
-        .map(|(s1, s2): (char, String)| format!("{}{}", s1, s2))
-        .and_then(|s: String| match s.as_str() {
-            "OR" | "AND " | "NOT" => Err(StringStreamError::UnexpectedParse),
-            _ => Ok(s),
+parser! {
+    fn literal[I]()(I) -> UserInputLeaf
+    where [I: Stream<Item = char>]
+    {
+        let term_val = || {
+            let phrase = char('"').with(many1(satisfy(|c| c != '"'))).skip(char('"'));
+            phrase.or(word())
+        };
+        let term_val_with_field = negative_number().or(term_val());
+        let term_query =
+            (field(), term_val_with_field)
+            .map(|(field_name, phrase)| UserInputLiteral {
+                field_name: Some(field_name),
+                phrase,
+            });
+        let term_default_field = term_val().map(|phrase| UserInputLiteral {
+            field_name: None,
+            phrase,
+        });
+        attempt(term_query)
+            .or(term_default_field)
+            .map(UserInputLeaf::from)
+    }
+}
+
+parser! {
+    fn negative_number[I]()(I) -> String
+    where [I: Stream<Item = char>]
+    {
+        (char('-'), many1(satisfy(char::is_numeric)),
+         optional((char('.'), many1(satisfy(char::is_numeric)))))
+            .map(|(s1, s2, s3): (char, String, Option<(char, String)>)| {
+                if let Some(('.', s3)) = s3 {
+                    format!("{}{}.{}", s1, s2, s3)
+                } else {
+                    format!("{}{}", s1, s2)
+                }
+            })
+    }
+}
+
+parser! {
+    fn spaces1[I]()(I) -> ()
+    where [I: Stream<Item = char>] {
+        skip_many1(space())
+    }
+}
+
+parser! {
+    /// Function that parses a range out of a Stream
+    /// Supports ranges like:
+    /// [5 TO 10], {5 TO 10}, [* TO 10], [10 TO *], {10 TO *], >5, <=10
+    /// [a TO *], [a TO c], [abc TO bcd}
+    fn range[I]()(I) -> UserInputLeaf
+    where [I: Stream<Item = char>] {
+        let range_term_val = || {
+            word().or(negative_number()).or(char('*').with(value("*".to_string())))
+        };
+
+        // check for unbounded range in the form of <5, <=10, >5, >=5
+        let elastic_unbounded_range = (choice([attempt(string(">=")),
+                                               attempt(string("<=")),
+                                               attempt(string("<")),
+                                               attempt(string(">"))])
+                                       .skip(spaces()),
+                                       range_term_val()).
+            map(|(comparison_sign, bound): (&str, String)|
+                match comparison_sign {
+                    ">=" => (UserInputBound::Inclusive(bound), UserInputBound::Unbounded),
+                    "<=" => (UserInputBound::Unbounded, UserInputBound::Inclusive(bound)),
+                    "<" => (UserInputBound::Unbounded, UserInputBound::Exclusive(bound)),
+                    ">" => (UserInputBound::Exclusive(bound), UserInputBound::Unbounded),
+                    // default case
+                    _ => (UserInputBound::Unbounded, UserInputBound::Unbounded)
+                });
+        let lower_bound = (one_of("{[".chars()), range_term_val())
+            .map(|(boundary_char, lower_bound): (char, String)|
+                 if lower_bound == "*" {
+                     UserInputBound::Unbounded
+                 } else if boundary_char == '{' {
+                         UserInputBound::Exclusive(lower_bound)
+                 } else {
+                     UserInputBound::Inclusive(lower_bound)
+                 });
+        let upper_bound = (range_term_val(), one_of("}]".chars()))
+            .map(|(higher_bound, boundary_char): (String, char)|
+                 if higher_bound == "*" {
+                     UserInputBound::Unbounded
+                 } else if boundary_char == '}' {
+                     UserInputBound::Exclusive(higher_bound)
+                 } else {
+                     UserInputBound::Inclusive(higher_bound)
+                 });
+         // return only lower and upper
+        let lower_to_upper = (lower_bound.
+                                    skip((spaces(),
+                                          string("TO"),
+                                          spaces())),
+                                    upper_bound);
+
+        (optional(field()).skip(spaces()),
+         // try elastic first, if it matches, the range is unbounded
+         attempt(elastic_unbounded_range).or(lower_to_upper))
+            .map(|(field, (lower, upper))|
+                 // Construct the leaf from extracted field (optional)
+                 // and bounds
+                 UserInputLeaf::Range {
+                     field,
+                     lower,
+                     upper
         })
-}
-
-fn term_val<'a>() -> impl Parser<&'a str, Output = String> {
-    let phrase = char('"').with(many1(satisfy(|c| c != '"'))).skip(char('"'));
-    phrase.or(word())
-}
-
-fn term_query<'a>() -> impl Parser<&'a str, Output = UserInputLiteral> {
-    let term_val_with_field = negative_number().or(term_val());
-    (field(), term_val_with_field).map(|(field_name, phrase)| UserInputLiteral {
-        field_name: Some(field_name),
-        phrase,
-    })
-}
-
-fn literal<'a>() -> impl Parser<&'a str, Output = UserInputLeaf> {
-    let term_default_field = term_val().map(|phrase| UserInputLiteral {
-        field_name: None,
-        phrase,
-    });
-    attempt(term_query())
-        .or(term_default_field)
-        .map(UserInputLeaf::from)
-}
-
-fn negative_number<'a>() -> impl Parser<&'a str, Output = String> {
-    (
-        char('-'),
-        many1(digit()),
-        optional((char('.'), many1(digit()))),
-    )
-        .map(|(s1, s2, s3): (char, String, Option<(char, String)>)| {
-            if let Some(('.', s3)) = s3 {
-                format!("{}{}.{}", s1, s2, s3)
-            } else {
-                format!("{}{}", s1, s2)
-            }
-        })
-}
-
-fn spaces1<'a>() -> impl Parser<&'a str, Output = ()> {
-    skip_many1(space())
-}
-
-/// Function that parses a range out of a Stream
-/// Supports ranges like:
-/// [5 TO 10], {5 TO 10}, [* TO 10], [10 TO *], {10 TO *], >5, <=10
-/// [a TO *], [a TO c], [abc TO bcd}
-fn range<'a>() -> impl Parser<&'a str, Output = UserInputLeaf> {
-    let range_term_val = || {
-        word()
-            .or(negative_number())
-            .or(char('*').with(value("*".to_string())))
-    };
-
-    // check for unbounded range in the form of <5, <=10, >5, >=5
-    let elastic_unbounded_range = (
-        choice([
-            attempt(string(">=")),
-            attempt(string("<=")),
-            attempt(string("<")),
-            attempt(string(">")),
-        ])
-        .skip(spaces()),
-        range_term_val(),
-    )
-        .map(
-            |(comparison_sign, bound): (&str, String)| match comparison_sign {
-                ">=" => (UserInputBound::Inclusive(bound), UserInputBound::Unbounded),
-                "<=" => (UserInputBound::Unbounded, UserInputBound::Inclusive(bound)),
-                "<" => (UserInputBound::Unbounded, UserInputBound::Exclusive(bound)),
-                ">" => (UserInputBound::Exclusive(bound), UserInputBound::Unbounded),
-                // default case
-                _ => (UserInputBound::Unbounded, UserInputBound::Unbounded),
-            },
-        );
-    let lower_bound = (one_of("{[".chars()), range_term_val()).map(
-        |(boundary_char, lower_bound): (char, String)| {
-            if lower_bound == "*" {
-                UserInputBound::Unbounded
-            } else if boundary_char == '{' {
-                UserInputBound::Exclusive(lower_bound)
-            } else {
-                UserInputBound::Inclusive(lower_bound)
-            }
-        },
-    );
-    let upper_bound = (range_term_val(), one_of("}]".chars())).map(
-        |(higher_bound, boundary_char): (String, char)| {
-            if higher_bound == "*" {
-                UserInputBound::Unbounded
-            } else if boundary_char == '}' {
-                UserInputBound::Exclusive(higher_bound)
-            } else {
-                UserInputBound::Inclusive(higher_bound)
-            }
-        },
-    );
-    // return only lower and upper
-    let lower_to_upper = (
-        lower_bound.skip((spaces(), string("TO"), spaces())),
-        upper_bound,
-    );
-
-    (
-        optional(field()).skip(spaces()),
-        // try elastic first, if it matches, the range is unbounded
-        attempt(elastic_unbounded_range).or(lower_to_upper),
-    )
-        .map(|(field, (lower, upper))|
-             // Construct the leaf from extracted field (optional)
-             // and bounds
-             UserInputLeaf::Range {
-                 field,
-                 lower,
-                 upper
-    })
+    }
 }
 
 fn negate(expr: UserInputAST) -> UserInputAST {
     expr.unary(Occur::MustNot)
 }
 
-fn leaf<'a>() -> impl Parser<&'a str, Output = UserInputAST> {
-    parser(|input| {
-        char('(')
-            .with(ast())
-            .skip(char(')'))
-            .or(char('*').map(|_| UserInputAST::from(UserInputLeaf::All)))
-            .or(attempt(
-                string("NOT").skip(spaces1()).with(leaf()).map(negate),
-            ))
-            .or(attempt(range().map(UserInputAST::from)))
-            .or(literal().map(UserInputAST::from))
-            .parse_stream(input)
-            .into_result()
-    })
+fn must(expr: UserInputAST) -> UserInputAST {
+    expr.unary(Occur::Must)
 }
 
-fn occur_symbol<'a>() -> impl Parser<&'a str, Output = Occur> {
-    char('-')
-        .map(|_| Occur::MustNot)
-        .or(char('+').map(|_| Occur::Must))
-}
-
-fn occur_leaf<'a>() -> impl Parser<&'a str, Output = (Option<Occur>, UserInputAST)> {
-    (optional(occur_symbol()), boosted_leaf())
-}
-
-fn positive_float_number<'a>() -> impl Parser<&'a str, Output = f32> {
-    (many1(digit()), optional((char('.'), many1(digit())))).map(
-        |(int_part, decimal_part_opt): (String, Option<(char, String)>)| {
-            let mut float_str = int_part;
-            if let Some((chr, decimal_str)) = decimal_part_opt {
-                float_str.push(chr);
-                float_str.push_str(&decimal_str);
-            }
-            float_str.parse::<f32>().unwrap()
-        },
-    )
-}
-
-fn boost<'a>() -> impl Parser<&'a str, Output = f32> {
-    (char('^'), positive_float_number()).map(|(_, boost)| boost)
-}
-
-fn boosted_leaf<'a>() -> impl Parser<&'a str, Output = UserInputAST> {
-    (leaf(), optional(boost())).map(|(leaf, boost_opt)| match boost_opt {
-        Some(boost) if (boost - 1.0).abs() > std::f32::EPSILON => {
-            UserInputAST::Boost(Box::new(leaf), boost)
-        }
-        _ => leaf,
-    })
+parser! {
+    fn leaf[I]()(I) -> UserInputAST
+    where [I: Stream<Item = char>] {
+            char('-').with(leaf()).map(negate)
+        .or(char('+').with(leaf()).map(must))
+        .or(char('(').with(ast()).skip(char(')')))
+        .or(char('*').map(|_| UserInputAST::from(UserInputLeaf::All)))
+        .or(attempt(string("NOT").skip(spaces1()).with(leaf()).map(negate)))
+        .or(attempt(range().map(UserInputAST::from)))
+        .or(literal().map(UserInputAST::from))
+    }
 }
 
 #[derive(Clone, Copy)]
@@ -212,10 +174,13 @@ enum BinaryOperand {
     And,
 }
 
-fn binary_operand<'a>() -> impl Parser<&'a str, Output = BinaryOperand> {
-    string("AND")
-        .with(value(BinaryOperand::And))
-        .or(string("OR").with(value(BinaryOperand::Or)))
+parser! {
+    fn binary_operand[I]()(I) -> BinaryOperand
+    where [I: Stream<Item = char>]
+    {
+       string("AND").with(value(BinaryOperand::And))
+       .or(string("OR").with(value(BinaryOperand::Or)))
+    }
 }
 
 fn aggregate_binary_expressions(
@@ -243,81 +208,37 @@ fn aggregate_binary_expressions(
     }
 }
 
-fn operand_leaf<'a>() -> impl Parser<&'a str, Output = (BinaryOperand, UserInputAST)> {
-    (
-        binary_operand().skip(spaces()),
-        boosted_leaf().skip(spaces()),
-    )
-}
-
-pub fn ast<'a>() -> impl Parser<&'a str, Output = UserInputAST> {
-    let boolean_expr = (boosted_leaf().skip(spaces()), many1(operand_leaf()))
-        .map(|(left, right)| aggregate_binary_expressions(left, right));
-    let whitespace_separated_leaves = many1(occur_leaf().skip(spaces().silent())).map(
-        |subqueries: Vec<(Option<Occur>, UserInputAST)>| {
+parser! {
+    pub fn ast[I]()(I) -> UserInputAST
+    where [I: Stream<Item = char>]
+    {
+        let operand_leaf = (binary_operand().skip(spaces()), leaf().skip(spaces()));
+        let boolean_expr = (leaf().skip(spaces().silent()), many1(operand_leaf)).map(
+            |(left, right)| aggregate_binary_expressions(left,right));
+        let whitespace_separated_leaves = many1(leaf().skip(spaces().silent()))
+        .map(|subqueries: Vec<UserInputAST>|
             if subqueries.len() == 1 {
-                let (occur_opt, ast) = subqueries.into_iter().next().unwrap();
-                match occur_opt.unwrap_or(Occur::Should) {
-                    Occur::Must | Occur::Should => ast,
-                    Occur::MustNot => UserInputAST::Clause(vec![(Some(Occur::MustNot), ast)]),
-                }
+                subqueries.into_iter().next().unwrap()
             } else {
                 UserInputAST::Clause(subqueries.into_iter().collect())
-            }
-        },
-    );
-    let expr = attempt(boolean_expr).or(whitespace_separated_leaves);
-    spaces().with(expr).skip(spaces())
+            });
+        let expr = attempt(boolean_expr).or(whitespace_separated_leaves);
+        spaces().with(expr).skip(spaces())
+    }
 }
 
-pub fn parse_to_ast<'a>() -> impl Parser<&'a str, Output = UserInputAST> {
-    spaces()
-        .with(optional(ast()).skip(eof()))
-        .map(|opt_ast| opt_ast.unwrap_or_else(UserInputAST::empty_query))
+parser! {
+    pub fn parse_to_ast[I]()(I) -> UserInputAST
+    where [I: Stream<Item = char>]
+    {
+        spaces().with(optional(ast()).skip(eof())).map(|opt_ast| opt_ast.unwrap_or_else(UserInputAST::empty_query))
+    }
 }
 
 #[cfg(test)]
 mod test {
 
     use super::*;
-    use combine::parser::Parser;
-
-    pub fn nearly_equals(a: f32, b: f32) -> bool {
-        (a - b).abs() < 0.0005 * (a + b).abs()
-    }
-
-    fn assert_nearly_equals(expected: f32, val: f32) {
-        assert!(
-            nearly_equals(val, expected),
-            "Got {}, expected {}.",
-            val,
-            expected
-        );
-    }
-
-    #[test]
-    fn test_occur_symbol() {
-        assert_eq!(super::occur_symbol().parse("-"), Ok((Occur::MustNot, "")));
-        assert_eq!(super::occur_symbol().parse("+"), Ok((Occur::Must, "")));
-    }
-
-    #[test]
-    fn test_positive_float_number() {
-        fn valid_parse(float_str: &str, expected_val: f32, expected_remaining: &str) {
-            let (val, remaining) = positive_float_number().parse(float_str).unwrap();
-            assert_eq!(remaining, expected_remaining);
-            assert_nearly_equals(val, expected_val);
-        }
-        fn error_parse(float_str: &str) {
-            assert!(positive_float_number().parse(float_str).is_err());
-        }
-        valid_parse("1.0", 1.0f32, "");
-        valid_parse("1", 1.0f32, "");
-        valid_parse("0.234234 aaa", 0.234234f32, " aaa");
-        error_parse(".3332");
-        error_parse("1.");
-        error_parse("-1.");
-    }
 
     fn test_parse_query_to_ast_helper(query: &str, expected: &str) {
         let query = parse_to_ast().parse(query).unwrap().0;
@@ -348,24 +269,15 @@ mod test {
             "Err(UnexpectedParse)"
         );
         test_parse_query_to_ast_helper("NOTa", "\"NOTa\"");
-        test_parse_query_to_ast_helper("NOT a", "(-\"a\")");
-    }
-
-    #[test]
-    fn test_boosting() {
-        assert!(parse_to_ast().parse("a^2^3").is_err());
-        assert!(parse_to_ast().parse("a^2^").is_err());
-        test_parse_query_to_ast_helper("a^3", "(\"a\")^3");
-        test_parse_query_to_ast_helper("a^3 b^2", "(*(\"a\")^3 *(\"b\")^2)");
-        test_parse_query_to_ast_helper("a^1", "\"a\"");
+        test_parse_query_to_ast_helper("NOT a", "-(\"a\")");
     }
 
     #[test]
     fn test_parse_query_to_ast_binary_op() {
-        test_parse_query_to_ast_helper("a AND b", "(+\"a\" +\"b\")");
-        test_parse_query_to_ast_helper("a OR b", "(?\"a\" ?\"b\")");
-        test_parse_query_to_ast_helper("a OR b AND c", "(?\"a\" ?(+\"b\" +\"c\"))");
-        test_parse_query_to_ast_helper("a AND b         AND c", "(+\"a\" +\"b\" +\"c\")");
+        test_parse_query_to_ast_helper("a AND b", "(+(\"a\") +(\"b\"))");
+        test_parse_query_to_ast_helper("a OR b", "(?(\"a\") ?(\"b\"))");
+        test_parse_query_to_ast_helper("a OR b AND c", "(?(\"a\") ?((+(\"b\") +(\"c\"))))");
+        test_parse_query_to_ast_helper("a AND b         AND c", "(+(\"a\") +(\"b\") +(\"c\"))");
         assert_eq!(
             format!("{:?}", parse_to_ast().parse("a OR b aaa")),
             "Err(UnexpectedParse)"
@@ -403,13 +315,6 @@ mod test {
         test_parse_query_to_ast_helper("weight: <= 70.5", "weight:{\"*\" TO \"70.5\"]");
     }
 
-    #[test]
-    fn test_occur_leaf() {
-        let ((occur, ast), _) = super::occur_leaf().parse("+abc").unwrap();
-        assert_eq!(occur, Some(Occur::Must));
-        assert_eq!(format!("{:?}", ast), "\"abc\"");
-    }
-
     #[test]
     fn test_range_parser() {
         // testing the range() parser separately
@@ -438,67 +343,32 @@ mod test {
     fn test_parse_query_to_triming_spaces() {
         test_parse_query_to_ast_helper("   abc", "\"abc\"");
         test_parse_query_to_ast_helper("abc ", "\"abc\"");
-        test_parse_query_to_ast_helper("(  a OR abc)", "(?\"a\" ?\"abc\")");
-        test_parse_query_to_ast_helper("(a  OR abc)", "(?\"a\" ?\"abc\")");
-        test_parse_query_to_ast_helper("(a OR  abc)", "(?\"a\" ?\"abc\")");
-        test_parse_query_to_ast_helper("a OR abc ", "(?\"a\" ?\"abc\")");
-        test_parse_query_to_ast_helper("(a OR abc )", "(?\"a\" ?\"abc\")");
-        test_parse_query_to_ast_helper("(a OR  abc) ", "(?\"a\" ?\"abc\")");
+        test_parse_query_to_ast_helper("(  a OR abc)", "(?(\"a\") ?(\"abc\"))");
+        test_parse_query_to_ast_helper("(a  OR abc)", "(?(\"a\") ?(\"abc\"))");
+        test_parse_query_to_ast_helper("(a OR  abc)", "(?(\"a\") ?(\"abc\"))");
+        test_parse_query_to_ast_helper("a OR abc ", "(?(\"a\") ?(\"abc\"))");
+        test_parse_query_to_ast_helper("(a OR abc )", "(?(\"a\") ?(\"abc\"))");
+        test_parse_query_to_ast_helper("(a OR  abc) ", "(?(\"a\") ?(\"abc\"))");
     }
 
     #[test]
-    fn test_parse_query_single_term() {
+    fn test_parse_query_to_ast() {
         test_parse_query_to_ast_helper("abc", "\"abc\"");
-    }
-
-    #[test]
-    fn test_parse_query_default_clause() {
-        test_parse_query_to_ast_helper("a b", "(*\"a\" *\"b\")");
-    }
-
-    #[test]
-    fn test_parse_query_must_default_clause() {
-        test_parse_query_to_ast_helper("+(a b)", "(*\"a\" *\"b\")");
-    }
-
-    #[test]
-    fn test_parse_query_must_single_term() {
-        test_parse_query_to_ast_helper("+d", "\"d\"");
-    }
-
-    #[test]
-    fn test_single_term_with_field() {
+        test_parse_query_to_ast_helper("a b", "(\"a\" \"b\")");
+        test_parse_query_to_ast_helper("+(a b)", "+((\"a\" \"b\"))");
+        test_parse_query_to_ast_helper("+d", "+(\"d\")");
+        test_parse_query_to_ast_helper("+(a b) +d", "(+((\"a\" \"b\")) +(\"d\"))");
+        test_parse_query_to_ast_helper("(+a +b) d", "((+(\"a\") +(\"b\")) \"d\")");
+        test_parse_query_to_ast_helper("(+a)", "+(\"a\")");
+        test_parse_query_to_ast_helper("(+a +b)", "(+(\"a\") +(\"b\"))");
         test_parse_query_to_ast_helper("abc:toto", "abc:\"toto\"");
-    }
-
-    #[test]
-    fn test_single_term_with_float() {
         test_parse_query_to_ast_helper("abc:1.1", "abc:\"1.1\"");
-    }
-
-    #[test]
-    fn test_must_clause() {
-        test_parse_query_to_ast_helper("(+a +b)", "(+\"a\" +\"b\")");
-    }
-
-    #[test]
-    fn test_parse_test_query_plus_a_b_plus_d() {
-        test_parse_query_to_ast_helper("+(a b) +d", "(+(*\"a\" *\"b\") +\"d\")");
-    }
-
-    #[test]
-    fn test_parse_test_query_other() {
-        test_parse_query_to_ast_helper("(+a +b) d", "(*(+\"a\" +\"b\") *\"d\")");
-        test_parse_query_to_ast_helper("+abc:toto", "abc:\"toto\"");
-        test_parse_query_to_ast_helper("(+abc:toto -titi)", "(+abc:\"toto\" -\"titi\")");
-        test_parse_query_to_ast_helper("-abc:toto", "(-abc:\"toto\")");
-        test_parse_query_to_ast_helper("abc:a b", "(*abc:\"a\" *\"b\")");
+        test_parse_query_to_ast_helper("+abc:toto", "+(abc:\"toto\")");
+        test_parse_query_to_ast_helper("(+abc:toto -titi)", "(+(abc:\"toto\") -(\"titi\"))");
+        test_parse_query_to_ast_helper("-abc:toto", "-(abc:\"toto\")");
+        test_parse_query_to_ast_helper("abc:a b", "(abc:\"a\" \"b\")");
         test_parse_query_to_ast_helper("abc:\"a b\"", "abc:\"a b\"");
         test_parse_query_to_ast_helper("foo:[1 TO 5]", "foo:[\"1\" TO \"5\"]");
-    }
-
-    #[test]
-    fn test_parse_query_with_range() {
         test_parse_query_to_ast_helper("[1 TO 5]", "[\"1\" TO \"5\"]");
         test_parse_query_to_ast_helper("foo:{a TO z}", "foo:{\"a\" TO \"z\"}");
         test_parse_query_to_ast_helper("foo:[1 TO toto}", "foo:[\"1\" TO \"toto\"}");

query-grammar/src/user_input_ast.rs

@@ -85,14 +85,14 @@ impl UserInputBound {
 }
 
 pub enum UserInputAST {
-    Clause(Vec<(Option<Occur>, UserInputAST)>),
+    Clause(Vec<UserInputAST>),
+    Unary(Occur, Box<UserInputAST>),
     Leaf(Box<UserInputLeaf>),
-    Boost(Box<UserInputAST>, f32),
 }
 
 impl UserInputAST {
     pub fn unary(self, occur: Occur) -> UserInputAST {
-        UserInputAST::Clause(vec![(Some(occur), self)])
+        UserInputAST::Unary(occur, Box::new(self))
     }
 
     fn compose(occur: Occur, asts: Vec<UserInputAST>) -> UserInputAST {
@@ -103,7 +103,7 @@ impl UserInputAST {
         } else {
             UserInputAST::Clause(
                 asts.into_iter()
-                    .map(|ast: UserInputAST| (Some(occur), ast))
+                    .map(|ast: UserInputAST| ast.unary(occur))
                     .collect::<Vec<_>>(),
             )
         }
@@ -134,38 +134,26 @@ impl From<UserInputLeaf> for UserInputAST {
     }
 }
 
-fn print_occur_ast(
-    occur_opt: Option<Occur>,
-    ast: &UserInputAST,
-    formatter: &mut fmt::Formatter,
-) -> fmt::Result {
-    if let Some(occur) = occur_opt {
-        write!(formatter, "{}{:?}", occur, ast)?;
-    } else {
-        write!(formatter, "*{:?}", ast)?;
-    }
-    Ok(())
-}
-
 impl fmt::Debug for UserInputAST {
-    fn fmt(&self, formatter: &mut fmt::Formatter) -> fmt::Result {
+    fn fmt(&self, formatter: &mut fmt::Formatter<'_>) -> Result<(), fmt::Error> {
         match *self {
             UserInputAST::Clause(ref subqueries) => {
                 if subqueries.is_empty() {
                     write!(formatter, "<emptyclause>")?;
                 } else {
                     write!(formatter, "(")?;
-                    print_occur_ast(subqueries[0].0, &subqueries[0].1, formatter)?;
+                    write!(formatter, "{:?}", &subqueries[0])?;
                     for subquery in &subqueries[1..] {
-                        write!(formatter, " ")?;
-                        print_occur_ast(subquery.0, &subquery.1, formatter)?;
+                        write!(formatter, " {:?}", subquery)?;
                     }
                     write!(formatter, ")")?;
                 }
                 Ok(())
             }
+            UserInputAST::Unary(ref occur, ref subquery) => {
+                write!(formatter, "{}({:?})", occur, subquery)
+            }
             UserInputAST::Leaf(ref subquery) => write!(formatter, "{:?}", subquery),
-            UserInputAST::Boost(ref leaf, boost) => write!(formatter, "({:?})^{}", leaf, boost),
         }
     }
 }

src/collector/custom_score_top_collector.rs

@@ -11,13 +11,13 @@ impl<TCustomScorer, TScore> CustomScoreTopCollector<TCustomScorer, TScore>
 where
     TScore: Clone + PartialOrd,
 {
-    pub(crate) fn new(
+    pub fn new(
         custom_scorer: TCustomScorer,
-        collector: TopCollector<TScore>,
+        limit: usize,
     ) -> CustomScoreTopCollector<TCustomScorer, TScore> {
         CustomScoreTopCollector {
             custom_scorer,
-            collector,
+            collector: TopCollector::with_limit(limit),
         }
     }
 }
@@ -28,7 +28,7 @@ where
 /// It is the segment local version of the [`CustomScorer`](./trait.CustomScorer.html).
 pub trait CustomSegmentScorer<TScore>: 'static {
     /// Computes the score of a specific `doc`.
-    fn score(&mut self, doc: DocId) -> TScore;
+    fn score(&self, doc: DocId) -> TScore;
 }
 
 /// `CustomScorer` makes it possible to define any kind of score.
@@ -117,9 +117,9 @@ where
 
 impl<F, TScore> CustomSegmentScorer<TScore> for F
 where
-    F: 'static + FnMut(DocId) -> TScore,
+    F: 'static + Sync + Send + Fn(DocId) -> TScore,
 {
-    fn score(&mut self, doc: DocId) -> TScore {
+    fn score(&self, doc: DocId) -> TScore {
         (self)(doc)
     }
 }

src/collector/facet_collector.rs

@@ -1,5 +1,6 @@
 use crate::collector::Collector;
 use crate::collector::SegmentCollector;
+use crate::docset::SkipResult;
 use crate::fastfield::FacetReader;
 use crate::schema::Facet;
 use crate::schema::Field;
@@ -187,11 +188,6 @@ pub struct FacetSegmentCollector {
     collapse_facet_ords: Vec<u64>,
 }
 
-enum SkipResult {
-    Found,
-    NotFound,
-}
-
 fn skip<'a, I: Iterator<Item = &'a Facet>>(
     target: &[u8],
     collapse_it: &mut Peekable<I>,
@@ -201,14 +197,14 @@ fn skip<'a, I: Iterator<Item = &'a Facet>>(
             Some(facet_bytes) => match facet_bytes.encoded_str().as_bytes().cmp(target) {
                 Ordering::Less => {}
                 Ordering::Greater => {
-                    return SkipResult::NotFound;
+                    return SkipResult::OverStep;
                 }
                 Ordering::Equal => {
-                    return SkipResult::Found;
+                    return SkipResult::Reached;
                 }
             },
             None => {
-                return SkipResult::NotFound;
+                return SkipResult::End;
             }
         }
         collapse_it.next();
@@ -285,7 +281,7 @@ impl Collector for FacetCollector {
                     // is positionned on a term that has not been processed yet.
                     let skip_result = skip(facet_streamer.key(), &mut collapse_facet_it);
                     match skip_result {
-                        SkipResult::Found => {
+                        SkipResult::Reached => {
                             // we reach a facet we decided to collapse.
                             let collapse_depth = facet_depth(facet_streamer.key());
                             let mut collapsed_id = 0;
@@ -305,7 +301,7 @@ impl Collector for FacetCollector {
                             }
                             break;
                         }
-                        SkipResult::NotFound => {
+                        SkipResult::End | SkipResult::OverStep => {
                             collapse_mapping.push(0);
                             if !facet_streamer.advance() {
                                 break;

src/collector/mod.rs

@@ -109,7 +109,6 @@ pub use self::tweak_score_top_collector::{ScoreSegmentTweaker, ScoreTweaker};
 
 mod facet_collector;
 pub use self::facet_collector::FacetCollector;
-use crate::query::Scorer;
 
 /// `Fruit` is the type for the result of our collection.
 /// e.g. `usize` for the `Count` collector.
@@ -155,28 +154,6 @@ pub trait Collector: Sync {
     /// Combines the fruit associated to the collection of each segments
     /// into one fruit.
     fn merge_fruits(&self, segment_fruits: Vec<Self::Fruit>) -> crate::Result<Self::Fruit>;
-
-    /// Created a segment collector and
-    fn collect_segment(
-        &self,
-        scorer: &mut dyn Scorer,
-        segment_ord: u32,
-        segment_reader: &SegmentReader,
-    ) -> crate::Result<<Self::Child as SegmentCollector>::Fruit> {
-        let mut segment_collector = self.for_segment(segment_ord as u32, segment_reader)?;
-        if let Some(delete_bitset) = segment_reader.delete_bitset() {
-            scorer.for_each(&mut |doc, score| {
-                if delete_bitset.is_alive(doc) {
-                    segment_collector.collect(doc, score);
-                }
-            });
-        } else {
-            scorer.for_each(&mut |doc, score| {
-                segment_collector.collect(doc, score);
-            })
-        }
-        Ok(segment_collector.harvest())
-    }
 }
 
 /// The `SegmentCollector` is the trait in charge of defining the

src/collector/top_collector.rs

@@ -18,9 +18,9 @@ use std::collections::BinaryHeap;
 /// Two elements are equal if their feature is equal, and regardless of whether `doc`
 /// is equal. This should be perfectly fine for this usage, but let's make sure this
 /// struct is never public.
-pub(crate) struct ComparableDoc<T, D> {
-    pub feature: T,
-    pub doc: D,
+struct ComparableDoc<T, D> {
+    feature: T,
+    doc: D,
 }
 
 impl<T: PartialOrd, D: PartialOrd> PartialOrd for ComparableDoc<T, D> {
@@ -56,8 +56,7 @@ impl<T: PartialOrd, D: PartialOrd> PartialEq for ComparableDoc<T, D> {
 impl<T: PartialOrd, D: PartialOrd> Eq for ComparableDoc<T, D> {}
 
 pub(crate) struct TopCollector<T> {
-    pub limit: usize,
-    pub offset: usize,
+    limit: usize,
     _marker: PhantomData<T>,
 }
 
@@ -73,20 +72,14 @@ where
         if limit < 1 {
             panic!("Limit must be strictly greater than 0.");
         }
-        Self {
+        TopCollector {
             limit,
-            offset: 0,
             _marker: PhantomData,
         }
     }
 
-    /// Skip the first "offset" documents when collecting.
-    ///
-    /// This is equivalent to `OFFSET` in MySQL or PostgreSQL and `start` in
-    /// Lucene's TopDocsCollector.
-    pub fn and_offset(mut self, offset: usize) -> TopCollector<T> {
-        self.offset = offset;
-        self
+    pub fn limit(&self) -> usize {
+        self.limit
     }
 
     pub fn merge_fruits(
@@ -99,7 +92,7 @@ where
         let mut top_collector = BinaryHeap::new();
         for child_fruit in children {
             for (feature, doc) in child_fruit {
-                if top_collector.len() < (self.limit + self.offset) {
+                if top_collector.len() < self.limit {
                     top_collector.push(ComparableDoc { feature, doc });
                 } else if let Some(mut head) = top_collector.peek_mut() {
                     if head.feature < feature {
@@ -111,7 +104,6 @@ where
         Ok(top_collector
             .into_sorted_vec()
             .into_iter()
-            .skip(self.offset)
             .map(|cdoc| (cdoc.feature, cdoc.doc))
             .collect())
     }
@@ -121,23 +113,7 @@ where
         segment_id: SegmentLocalId,
         _: &SegmentReader,
     ) -> crate::Result<TopSegmentCollector<F>> {
-        Ok(TopSegmentCollector::new(
-            segment_id,
-            self.limit + self.offset,
-        ))
-    }
-
-    /// Create a new TopCollector with the same limit and offset.
-    ///
-    /// Ideally we would use Into but the blanket implementation seems to cause the Scorer traits
-    /// to fail.
-    #[doc(hidden)]
-    pub(crate) fn into_tscore<TScore: PartialOrd + Clone>(self) -> TopCollector<TScore> {
-        TopCollector {
-            limit: self.limit,
-            offset: self.offset,
-            _marker: PhantomData,
-        }
+        Ok(TopSegmentCollector::new(segment_id, self.limit))
     }
 }
 
@@ -211,7 +187,7 @@ impl<T: PartialOrd + Clone> TopSegmentCollector<T> {
 
 #[cfg(test)]
 mod tests {
-    use super::{TopCollector, TopSegmentCollector};
+    use super::TopSegmentCollector;
     use crate::DocAddress;
 
     #[test]
@@ -272,48 +248,6 @@ mod tests {
             top_collector_limit_3.harvest()[..2].to_vec(),
         );
     }
-
-    #[test]
-    fn test_top_collector_with_limit_and_offset() {
-        let collector = TopCollector::with_limit(2).and_offset(1);
-
-        let results = collector
-            .merge_fruits(vec![vec![
-                (0.9, DocAddress(0, 1)),
-                (0.8, DocAddress(0, 2)),
-                (0.7, DocAddress(0, 3)),
-                (0.6, DocAddress(0, 4)),
-                (0.5, DocAddress(0, 5)),
-            ]])
-            .unwrap();
-
-        assert_eq!(
-            results,
-            vec![(0.8, DocAddress(0, 2)), (0.7, DocAddress(0, 3)),]
-        );
-    }
-
-    #[test]
-    fn test_top_collector_with_limit_larger_than_set_and_offset() {
-        let collector = TopCollector::with_limit(2).and_offset(1);
-
-        let results = collector
-            .merge_fruits(vec![vec![(0.9, DocAddress(0, 1)), (0.8, DocAddress(0, 2))]])
-            .unwrap();
-
-        assert_eq!(results, vec![(0.8, DocAddress(0, 2)),]);
-    }
-
-    #[test]
-    fn test_top_collector_with_limit_and_offset_larger_than_set() {
-        let collector = TopCollector::with_limit(2).and_offset(20);
-
-        let results = collector
-            .merge_fruits(vec![vec![(0.9, DocAddress(0, 1)), (0.8, DocAddress(0, 2))]])
-            .unwrap();
-
-        assert_eq!(results, vec![]);
-    }
 }
 
 #[cfg(all(test, feature = "unstable"))]

src/collector/top_score_collector.rs

@@ -1,21 +1,18 @@
 use super::Collector;
 use crate::collector::custom_score_top_collector::CustomScoreTopCollector;
+use crate::collector::top_collector::TopCollector;
 use crate::collector::top_collector::TopSegmentCollector;
-use crate::collector::top_collector::{ComparableDoc, TopCollector};
 use crate::collector::tweak_score_top_collector::TweakedScoreTopCollector;
 use crate::collector::{
     CustomScorer, CustomSegmentScorer, ScoreSegmentTweaker, ScoreTweaker, SegmentCollector,
 };
-use crate::docset::TERMINATED;
 use crate::fastfield::FastFieldReader;
-use crate::query::Scorer;
 use crate::schema::Field;
 use crate::DocAddress;
 use crate::DocId;
 use crate::Score;
 use crate::SegmentLocalId;
 use crate::SegmentReader;
-use std::collections::BinaryHeap;
 use std::fmt;
 
 /// The `TopDocs` collector keeps track of the top `K` documents
@@ -60,11 +57,7 @@ pub struct TopDocs(TopCollector<Score>);
 
 impl fmt::Debug for TopDocs {
     fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
-        write!(
-            f,
-            "TopDocs(limit={}, offset={})",
-            self.0.limit, self.0.offset
-        )
+        write!(f, "TopDocs({})", self.0.limit())
     }
 }
 
@@ -73,7 +66,7 @@ struct ScorerByFastFieldReader {
 }
 
 impl CustomSegmentScorer<u64> for ScorerByFastFieldReader {
-    fn score(&mut self, doc: DocId) -> u64 {
+    fn score(&self, doc: DocId) -> u64 {
         self.ff_reader.get_u64(u64::from(doc))
     }
 }
@@ -91,8 +84,7 @@ impl CustomScorer<u64> for ScorerByField {
             .u64(self.field)
             .ok_or_else(|| {
                 crate::TantivyError::SchemaError(format!(
-                    "Field requested ({:?}) is not a i64/u64 fast field.",
-                    self.field
+                    "Field requested is not a i64/u64 fast field."
                 ))
             })?;
         Ok(ScorerByFastFieldReader { ff_reader })
@@ -108,45 +100,6 @@ impl TopDocs {
         TopDocs(TopCollector::with_limit(limit))
     }
 
-    /// Skip the first "offset" documents when collecting.
-    ///
-    /// This is equivalent to `OFFSET` in MySQL or PostgreSQL and `start` in
-    /// Lucene's TopDocsCollector.
-    ///
-    /// ```rust
-    /// use tantivy::collector::TopDocs;
-    /// use tantivy::query::QueryParser;
-    /// use tantivy::schema::{Schema, TEXT};
-    /// use tantivy::{doc, DocAddress, Index};
-    ///
-    /// let mut schema_builder = Schema::builder();
-    /// let title = schema_builder.add_text_field("title", TEXT);
-    /// let schema = schema_builder.build();
-    /// let index = Index::create_in_ram(schema);
-    ///
-    /// let mut index_writer = index.writer_with_num_threads(1, 3_000_000).unwrap();
-    /// index_writer.add_document(doc!(title => "The Name of the Wind"));
-    /// index_writer.add_document(doc!(title => "The Diary of Muadib"));
-    /// index_writer.add_document(doc!(title => "A Dairy Cow"));
-    /// index_writer.add_document(doc!(title => "The Diary of a Young Girl"));
-    /// index_writer.add_document(doc!(title => "The Diary of Lena Mukhina"));
-    /// assert!(index_writer.commit().is_ok());
-    ///
-    /// let reader = index.reader().unwrap();
-    /// let searcher = reader.searcher();
-    ///
-    /// let query_parser = QueryParser::for_index(&index, vec![title]);
-    /// let query = query_parser.parse_query("diary").unwrap();
-    /// let top_docs = searcher.search(&query, &TopDocs::with_limit(2).and_offset(1)).unwrap();
-    ///
-    /// assert_eq!(top_docs.len(), 2);
-    /// assert_eq!(&top_docs[0], &(0.5204813, DocAddress(0, 4)));
-    /// assert_eq!(&top_docs[1], &(0.4793185, DocAddress(0, 3)));
-    /// ```
-    pub fn and_offset(self, offset: usize) -> TopDocs {
-        TopDocs(self.0.and_offset(offset))
-    }
-
     /// Set top-K to rank documents by a given fast field.
     ///
     /// ```rust
@@ -327,7 +280,7 @@ impl TopDocs {
         TScoreSegmentTweaker: ScoreSegmentTweaker<TScore> + 'static,
         TScoreTweaker: ScoreTweaker<TScore, Child = TScoreSegmentTweaker>,
     {
-        TweakedScoreTopCollector::new(score_tweaker, self.0.into_tscore())
+        TweakedScoreTopCollector::new(score_tweaker, self.0.limit())
     }
 
     /// Ranks the documents using a custom score.
@@ -441,7 +394,7 @@ impl TopDocs {
         TCustomSegmentScorer: CustomSegmentScorer<TScore> + 'static,
         TCustomScorer: CustomScorer<TScore, Child = TCustomSegmentScorer>,
     {
-        CustomScoreTopCollector::new(custom_score, self.0.into_tscore())
+        CustomScoreTopCollector::new(custom_score, self.0.limit())
     }
 }
 
@@ -469,58 +422,6 @@ impl Collector for TopDocs {
     ) -> crate::Result<Self::Fruit> {
         self.0.merge_fruits(child_fruits)
     }
-
-    fn collect_segment(
-        &self,
-        scorer: &mut dyn Scorer,
-        segment_ord: u32,
-        segment_reader: &SegmentReader,
-    ) -> crate::Result<<Self::Child as SegmentCollector>::Fruit> {
-        let mut heap: BinaryHeap<ComparableDoc<Score, DocId>> =
-            BinaryHeap::with_capacity(self.0.limit + self.0.offset);
-        // first we fill the heap with the first `limit` elements.
-        let mut doc = scorer.doc();
-        while doc != TERMINATED && heap.len() < (self.0.limit + self.0.offset) {
-            if !segment_reader.is_deleted(doc) {
-                let score = scorer.score();
-                heap.push(ComparableDoc {
-                    feature: score,
-                    doc,
-                });
-            }
-            doc = scorer.advance();
-        }
-
-        let threshold = heap.peek().map(|el| el.feature).unwrap_or(std::f32::MIN);
-
-        if let Some(delete_bitset) = segment_reader.delete_bitset() {
-            scorer.for_each_pruning(threshold, &mut |doc, score| {
-                if delete_bitset.is_alive(doc) {
-                    *heap.peek_mut().unwrap() = ComparableDoc {
-                        feature: score,
-                        doc,
-                    };
-                }
-                heap.peek().map(|el| el.feature).unwrap_or(std::f32::MIN)
-            });
-        } else {
-            scorer.for_each_pruning(threshold, &mut |doc, score| {
-                *heap.peek_mut().unwrap() = ComparableDoc {
-                    feature: score,
-                    doc,
-                };
-                heap.peek().map(|el| el.feature).unwrap_or(std::f32::MIN)
-            });
-        }
-
-        let fruit = heap
-            .into_sorted_vec()
-            .into_iter()
-            .map(|cid| (cid.feature, DocAddress(segment_ord, cid.doc)))
-            .collect();
-
-        Ok(fruit)
-    }
 }
 
 /// Segment Collector associated to `TopDocs`.
@@ -530,7 +431,7 @@ impl SegmentCollector for TopScoreSegmentCollector {
     type Fruit = Vec<(Score, DocAddress)>;
 
     fn collect(&mut self, doc: DocId, score: Score) {
-        self.0.collect(doc, score);
+        self.0.collect(doc, score)
     }
 
     fn harvest(self) -> Vec<(Score, DocAddress)> {
@@ -544,10 +445,11 @@ mod tests {
     use crate::collector::Collector;
     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 crate::{DocAddress, DocId, SegmentReader};
+    use itertools::Itertools;
 
     fn make_index() -> Index {
         let mut schema_builder = Schema::builder();
@@ -587,21 +489,6 @@ mod tests {
         );
     }
 
-    #[test]
-    fn test_top_collector_not_at_capacity_with_offset() {
-        let index = make_index();
-        let field = index.schema().get_field("text").unwrap();
-        let query_parser = QueryParser::for_index(&index, vec![field]);
-        let text_query = query_parser.parse_query("droopy tax").unwrap();
-        let score_docs: Vec<(Score, DocAddress)> = index
-            .reader()
-            .unwrap()
-            .searcher()
-            .search(&text_query, &TopDocs::with_limit(4).and_offset(2))
-            .unwrap();
-        assert_eq!(score_docs, vec![(0.48527452, DocAddress(0, 0))]);
-    }
-
     #[test]
     fn test_top_collector_at_capacity() {
         let index = make_index();
@@ -623,27 +510,6 @@ mod tests {
         );
     }
 
-    #[test]
-    fn test_top_collector_at_capacity_with_offset() {
-        let index = make_index();
-        let field = index.schema().get_field("text").unwrap();
-        let query_parser = QueryParser::for_index(&index, vec![field]);
-        let text_query = query_parser.parse_query("droopy tax").unwrap();
-        let score_docs: Vec<(Score, DocAddress)> = index
-            .reader()
-            .unwrap()
-            .searcher()
-            .search(&text_query, &TopDocs::with_limit(2).and_offset(1))
-            .unwrap();
-        assert_eq!(
-            score_docs,
-            vec![
-                (0.5376842, DocAddress(0u32, 2)),
-                (0.48527452, DocAddress(0, 0))
-            ]
-        );
-    }
-
     #[test]
     fn test_top_collector_stable_sorting() {
         let index = make_index();
@@ -657,8 +523,8 @@ mod tests {
 
         // precondition for the test to be meaningful: we did get documents
         // with the same score
-        assert!(page_1.iter().all(|result| result.0 == page_1[0].0));
-        assert!(page_2.iter().all(|result| result.0 == page_2[0].0));
+        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);
@@ -748,59 +614,12 @@ mod tests {
         let top_collector = TopDocs::with_limit(4).order_by_u64_field(size);
         let err = top_collector.for_segment(0, segment);
         if let Err(crate::TantivyError::SchemaError(msg)) = err {
-            assert_eq!(
-                msg,
-                "Field requested (Field(1)) is not a i64/u64 fast field."
-            );
+            assert_eq!(msg, "Field requested is not a i64/u64 fast field.");
         } else {
             assert!(false);
         }
     }
 
-    #[test]
-    fn test_tweak_score_top_collector_with_offset() {
-        let index = make_index();
-        let field = index.schema().get_field("text").unwrap();
-        let query_parser = QueryParser::for_index(&index, vec![field]);
-        let text_query = query_parser.parse_query("droopy tax").unwrap();
-        let collector = TopDocs::with_limit(2).and_offset(1).tweak_score(
-            move |_segment_reader: &SegmentReader| move |doc: DocId, _original_score: Score| doc,
-        );
-        let score_docs: Vec<(u32, DocAddress)> = index
-            .reader()
-            .unwrap()
-            .searcher()
-            .search(&text_query, &collector)
-            .unwrap();
-
-        assert_eq!(
-            score_docs,
-            vec![(1, DocAddress(0, 1)), (0, DocAddress(0, 0)),]
-        );
-    }
-
-    #[test]
-    fn test_custom_score_top_collector_with_offset() {
-        let index = make_index();
-        let field = index.schema().get_field("text").unwrap();
-        let query_parser = QueryParser::for_index(&index, vec![field]);
-        let text_query = query_parser.parse_query("droopy tax").unwrap();
-        let collector = TopDocs::with_limit(2)
-            .and_offset(1)
-            .custom_score(move |_segment_reader: &SegmentReader| move |doc: DocId| doc);
-        let score_docs: Vec<(u32, DocAddress)> = index
-            .reader()
-            .unwrap()
-            .searcher()
-            .search(&text_query, &collector)
-            .unwrap();
-
-        assert_eq!(
-            score_docs,
-            vec![(1, DocAddress(0, 1)), (0, DocAddress(0, 0)),]
-        );
-    }
-
     fn index(
         query: &str,
         query_field: Field,

src/collector/tweak_score_top_collector.rs

@@ -14,11 +14,11 @@ where
 {
     pub fn new(
         score_tweaker: TScoreTweaker,
-        collector: TopCollector<TScore>,
+        limit: usize,
     ) -> TweakedScoreTopCollector<TScoreTweaker, TScore> {
         TweakedScoreTopCollector {
             score_tweaker,
-            collector,
+            collector: TopCollector::with_limit(limit),
         }
     }
 }
@@ -29,7 +29,7 @@ where
 /// It is the segment local version of the [`ScoreTweaker`](./trait.ScoreTweaker.html).
 pub trait ScoreSegmentTweaker<TScore>: 'static {
     /// Tweak the given `score` for the document `doc`.
-    fn score(&mut self, doc: DocId, score: Score) -> TScore;
+    fn score(&self, doc: DocId, score: Score) -> TScore;
 }
 
 /// `ScoreTweaker` makes it possible to tweak the score
@@ -121,9 +121,9 @@ where
 
 impl<F, TScore> ScoreSegmentTweaker<TScore> for F
 where
-    F: 'static + FnMut(DocId, Score) -> TScore,
+    F: 'static + Sync + Send + Fn(DocId, Score) -> TScore,
 {
-    fn score(&mut self, doc: DocId, score: Score) -> TScore {
+    fn score(&self, doc: DocId, score: Score) -> TScore {
         (self)(doc, score)
     }
 }

src/common/bitset.rs

@@ -33,10 +33,6 @@ impl TinySet {
         TinySet(0u64)
     }
 
-    pub fn clear(&mut self) {
-        self.0 = 0u64;
-    }
-
     /// Returns the complement of the set in `[0, 64[`.
     fn complement(self) -> TinySet {
         TinySet(!self.0)
@@ -47,11 +43,6 @@ impl TinySet {
         !self.intersect(TinySet::singleton(el)).is_empty()
     }
 
-    /// Returns the number of elements in the TinySet.
-    pub fn len(self) -> u32 {
-        self.0.count_ones()
-    }
-
     /// Returns the intersection of `self` and `other`
     pub fn intersect(self, other: TinySet) -> TinySet {
         TinySet(self.0 & other.0)
@@ -118,12 +109,22 @@ impl TinySet {
     pub fn range_greater_or_equal(from_included: u32) -> TinySet {
         TinySet::range_lower(from_included).complement()
     }
+
+    pub fn clear(&mut self) {
+        self.0 = 0u64;
+    }
+
+    pub fn len(self) -> u32 {
+        self.0.count_ones()
+    }
 }
 
 #[derive(Clone)]
 pub struct BitSet {
     tinysets: Box<[TinySet]>,
-    len: usize,
+    len: usize, //< Technically it should be u32, but we
+    // count multiple inserts.
+    // `usize` guards us from overflow.
     max_value: u32,
 }
 
@@ -203,7 +204,7 @@ mod tests {
 
     use super::BitSet;
     use super::TinySet;
-    use crate::docset::{DocSet, TERMINATED};
+    use crate::docset::DocSet;
     use crate::query::BitSetDocSet;
     use crate::tests;
     use crate::tests::generate_nonunique_unsorted;
@@ -277,13 +278,11 @@ mod tests {
         }
         assert_eq!(btreeset.len(), bitset.len());
         let mut bitset_docset = BitSetDocSet::from(bitset);
-        let mut remaining = true;
         for el in btreeset.into_iter() {
-            assert!(remaining);
+            bitset_docset.advance();
             assert_eq!(bitset_docset.doc(), el);
-            remaining = bitset_docset.advance() != TERMINATED;
         }
-        assert!(!remaining);
+        assert!(!bitset_docset.advance());
     }
 
     #[test]

src/common/mod.rs

@@ -18,19 +18,6 @@ pub use byteorder::LittleEndian as Endianness;
 /// We do not allow segments with more than
 pub const MAX_DOC_LIMIT: u32 = 1 << 31;
 
-pub fn minmax<I, T>(mut vals: I) -> Option<(T, T)>
-where
-    I: Iterator<Item = T>,
-    T: Copy + Ord,
-{
-    if let Some(first_el) = vals.next() {
-        return Some(vals.fold((first_el, first_el), |(min_val, max_val), el| {
-            (min_val.min(el), max_val.max(el))
-        }));
-    }
-    None
-}
-
 /// Computes the number of bits that will be used for bitpacking.
 ///
 /// In general the target is the minimum number of bits
@@ -147,7 +134,6 @@ pub fn u64_to_f64(val: u64) -> f64 {
 #[cfg(test)]
 pub(crate) mod test {
 
-    pub use super::minmax;
     pub use super::serialize::test::fixed_size_test;
     use super::{compute_num_bits, f64_to_u64, i64_to_u64, u64_to_f64, u64_to_i64};
     use std::f64;
@@ -213,21 +199,4 @@ pub(crate) mod test {
         assert!(((super::MAX_DOC_LIMIT - 1) as i32) >= 0);
         assert!((super::MAX_DOC_LIMIT as i32) < 0);
     }
-
-    #[test]
-    fn test_minmax_empty() {
-        let vals: Vec<u32> = vec![];
-        assert_eq!(minmax(vals.into_iter()), None);
-    }
-
-    #[test]
-    fn test_minmax_one() {
-        assert_eq!(minmax(vec![1].into_iter()), Some((1, 1)));
-    }
-
-    #[test]
-    fn test_minmax_two() {
-        assert_eq!(minmax(vec![1, 2].into_iter()), Some((1, 2)));
-        assert_eq!(minmax(vec![2, 1].into_iter()), Some((1, 2)));
-    }
 }

src/core/index.rs

@@ -1,3 +1,4 @@
+use super::segment::create_segment;
 use super::segment::Segment;
 use crate::core::Executor;
 use crate::core::IndexMeta;
@@ -21,6 +22,7 @@ use crate::schema::FieldType;
 use crate::schema::Schema;
 use crate::tokenizer::{TextAnalyzer, TokenizerManager};
 use crate::IndexWriter;
+use num_cpus;
 use std::borrow::BorrowMut;
 use std::collections::HashSet;
 use std::fmt;
@@ -335,7 +337,7 @@ impl Index {
 
     #[doc(hidden)]
     pub fn segment(&self, segment_meta: SegmentMeta) -> Segment {
-        Segment::for_index(self.clone(), segment_meta)
+        create_segment(self.clone(), segment_meta)
     }
 
     /// Creates a new segment.

src/core/index_meta.rs

@@ -3,7 +3,8 @@ use crate::core::SegmentId;
 use crate::schema::Schema;
 use crate::Opstamp;
 use census::{Inventory, TrackedObject};
-use serde::{Deserialize, Serialize};
+use serde;
+use serde_json;
 use std::collections::HashSet;
 use std::fmt;
 use std::path::PathBuf;

src/core/searcher.rs

@@ -1,8 +1,11 @@
 use crate::collector::Collector;
+use crate::collector::SegmentCollector;
 use crate::core::Executor;
 use crate::core::InvertedIndexReader;
 use crate::core::SegmentReader;
 use crate::query::Query;
+use crate::query::Scorer;
+use crate::query::Weight;
 use crate::schema::Document;
 use crate::schema::Schema;
 use crate::schema::{Field, Term};
@@ -14,6 +17,26 @@ use crate::Index;
 use std::fmt;
 use std::sync::Arc;
 
+fn collect_segment<C: Collector>(
+    collector: &C,
+    weight: &dyn Weight,
+    segment_ord: u32,
+    segment_reader: &SegmentReader,
+) -> crate::Result<C::Fruit> {
+    let mut scorer = weight.scorer(segment_reader)?;
+    let mut segment_collector = collector.for_segment(segment_ord as u32, segment_reader)?;
+    if let Some(delete_bitset) = segment_reader.delete_bitset() {
+        scorer.for_each(&mut |doc, score| {
+            if delete_bitset.is_alive(doc) {
+                segment_collector.collect(doc, score);
+            }
+        });
+    } else {
+        scorer.for_each(&mut |doc, score| segment_collector.collect(doc, score));
+    }
+    Ok(segment_collector.harvest())
+}
+
 /// Holds a list of `SegmentReader`s ready for search.
 ///
 /// It guarantees that the `Segment` will not be removed before
@@ -140,8 +163,12 @@ impl Searcher {
         let segment_readers = self.segment_readers();
         let fruits = executor.map(
             |(segment_ord, segment_reader)| {
-                let mut scorer = weight.scorer(segment_reader, 1.0f32)?;
-                collector.collect_segment(scorer.as_mut(), segment_ord as u32, segment_reader)
+                collect_segment(
+                    collector,
+                    weight.as_ref(),
+                    segment_ord as u32,
+                    segment_reader,
+                )
             },
             segment_readers.iter().enumerate(),
         )?;

src/core/segment.rs

@@ -24,12 +24,15 @@ impl fmt::Debug for Segment {
     }
 }
 
-impl Segment {
-    /// Creates a new segment given an `Index` and a `SegmentId`
-    pub(crate) fn for_index(index: Index, meta: SegmentMeta) -> Segment {
-        Segment { index, meta }
-    }
+/// Creates a new segment given an `Index` and a `SegmentId`
+///
+/// The function is here to make it private outside `tantivy`.
+/// #[doc(hidden)]
+pub fn create_segment(index: Index, meta: SegmentMeta) -> Segment {
+    Segment { index, meta }
+}
 
+impl Segment {
     /// Returns the index the segment belongs to.
     pub fn index(&self) -> &Index {
         &self.index

src/core/segment_id.rs

@@ -4,7 +4,6 @@ use uuid::Uuid;
 
 #[cfg(test)]
 use once_cell::sync::Lazy;
-use serde::{Deserialize, Serialize};
 use std::error::Error;
 use std::str::FromStr;
 #[cfg(test)]

src/directory/footer.rs

@@ -8,8 +8,6 @@ use crc32fast::Hasher;
 use std::io;
 use std::io::Write;
 
-const FOOTER_MAX_LEN: usize = 10_000;
-
 type CrcHashU32 = u32;
 
 #[derive(Debug, Clone, PartialEq)]
@@ -145,23 +143,12 @@ impl BinarySerializable for VersionedFooter {
             }
         }
         BinarySerializable::serialize(&VInt(buf.len() as u64), writer)?;
-        assert!(buf.len() <= FOOTER_MAX_LEN);
         writer.write_all(&buf[..])?;
         Ok(())
     }
 
     fn deserialize<R: io::Read>(reader: &mut R) -> io::Result<Self> {
         let len = VInt::deserialize(reader)?.0 as usize;
-        if len > FOOTER_MAX_LEN {
-            return Err(io::Error::new(
-                io::ErrorKind::InvalidData,
-                format!(
-                    "Footer seems invalid as it suggests a footer len of {}. File is corrupted, \
-            or the index was created with a different & old version of tantivy.",
-                    len
-                ),
-            ));
-        }
         let mut buf = vec![0u8; len];
         reader.read_exact(&mut buf[..])?;
         let mut cursor = &buf[..];
@@ -234,12 +221,11 @@ mod tests {
 
     use super::CrcHashU32;
     use super::FooterProxy;
-    use crate::common::{BinarySerializable, VInt};
+    use crate::common::BinarySerializable;
     use crate::directory::footer::{Footer, VersionedFooter};
     use crate::directory::TerminatingWrite;
     use byteorder::{ByteOrder, LittleEndian};
     use regex::Regex;
-    use std::io;
 
     #[test]
     fn test_versioned_footer() {
@@ -350,20 +336,4 @@ mod tests {
         let res = footer.is_compatible();
         assert!(res.is_err());
     }
-
-    #[test]
-    fn test_deserialize_too_large_footer() {
-        let mut buf = vec![];
-        assert!(FooterProxy::new(&mut buf).terminate().is_ok());
-        let mut long_len_buf = [0u8; 10];
-        let num_bytes = VInt(super::FOOTER_MAX_LEN as u64 + 1u64).serialize_into(&mut long_len_buf);
-        buf[0..num_bytes].copy_from_slice(&long_len_buf[..num_bytes]);
-        let err = Footer::deserialize(&mut &buf[..]).unwrap_err();
-        assert_eq!(err.kind(), io::ErrorKind::InvalidData);
-        assert_eq!(
-            err.to_string(),
-            "Footer seems invalid as it suggests a footer len of 10001. File is corrupted, \
-            or the index was created with a different & old version of tantivy."
-        );
-    }
 }

src/directory/managed_directory.rs

@@ -11,6 +11,7 @@ use crate::error::DataCorruption;
 use crate::Directory;
 
 use crc32fast::Hasher;
+use serde_json;
 use std::collections::HashSet;
 use std::io;
 use std::io::Write;

src/directory/mmap_directory.rs

@@ -1,3 +1,10 @@
+use fs2;
+use notify;
+
+use self::fs2::FileExt;
+use self::notify::RawEvent;
+use self::notify::RecursiveMode;
+use self::notify::Watcher;
 use crate::core::META_FILEPATH;
 use crate::directory::error::LockError;
 use crate::directory::error::{
@@ -13,12 +20,8 @@ use crate::directory::WatchCallback;
 use crate::directory::WatchCallbackList;
 use crate::directory::WatchHandle;
 use crate::directory::{TerminatingWrite, WritePtr};
-use fs2::FileExt;
+use atomicwrites;
 use memmap::Mmap;
-use notify::RawEvent;
-use notify::RecursiveMode;
-use notify::Watcher;
-use serde::{Deserialize, Serialize};
 use std::collections::HashMap;
 use std::convert::From;
 use std::fmt;
@@ -139,7 +142,7 @@ impl MmapCache {
 }
 
 struct WatcherWrapper {
-    _watcher: Mutex<notify::RecommendedWatcher>,
+    _watcher: Mutex<notify::PollWatcher>,
     watcher_router: Arc<WatchCallbackList>,
 }
 
@@ -147,7 +150,7 @@ impl WatcherWrapper {
     pub fn new(path: &Path) -> Result<Self, OpenDirectoryError> {
         let (tx, watcher_recv): (Sender<RawEvent>, Receiver<RawEvent>) = channel();
         // We need to initialize the
-        let watcher = notify::raw_watcher(tx)
+        let watcher = notify::poll::PollWatcher::with_delay_ms(tx, 1)
             .and_then(|mut watcher| {
                 watcher.watch(path, RecursiveMode::Recursive)?;
                 Ok(watcher)
@@ -220,13 +223,17 @@ struct MmapDirectoryInner {
 }
 
 impl MmapDirectoryInner {
-    fn new(root_path: PathBuf, temp_directory: Option<TempDir>) -> MmapDirectoryInner {
-        MmapDirectoryInner {
+    fn new(
+        root_path: PathBuf,
+        temp_directory: Option<TempDir>,
+    ) -> Result<MmapDirectoryInner, OpenDirectoryError> {
+        let mmap_directory_inner = MmapDirectoryInner {
             root_path,
             mmap_cache: Default::default(),
             _temp_directory: temp_directory,
             watcher: RwLock::new(None),
-        }
+        };
+        Ok(mmap_directory_inner)
     }
 
     fn watch(&self, watch_callback: WatchCallback) -> crate::Result<WatchHandle> {
@@ -260,11 +267,14 @@ impl fmt::Debug for MmapDirectory {
 }
 
 impl MmapDirectory {
-    fn new(root_path: PathBuf, temp_directory: Option<TempDir>) -> MmapDirectory {
-        let inner = MmapDirectoryInner::new(root_path, temp_directory);
-        MmapDirectory {
+    fn new(
+        root_path: PathBuf,
+        temp_directory: Option<TempDir>,
+    ) -> Result<MmapDirectory, OpenDirectoryError> {
+        let inner = MmapDirectoryInner::new(root_path, temp_directory)?;
+        Ok(MmapDirectory {
             inner: Arc::new(inner),
-        }
+        })
     }
 
     /// Creates a new MmapDirectory in a temporary directory.
@@ -274,7 +284,7 @@ impl MmapDirectory {
     pub fn create_from_tempdir() -> Result<MmapDirectory, OpenDirectoryError> {
         let tempdir = TempDir::new().map_err(OpenDirectoryError::IoError)?;
         let tempdir_path = PathBuf::from(tempdir.path());
-        Ok(MmapDirectory::new(tempdir_path, Some(tempdir)))
+        MmapDirectory::new(tempdir_path, Some(tempdir))
     }
 
     /// Opens a MmapDirectory in a directory.
@@ -292,7 +302,7 @@ impl MmapDirectory {
                 directory_path,
             )))
         } else {
-            Ok(MmapDirectory::new(PathBuf::from(directory_path), None))
+            Ok(MmapDirectory::new(PathBuf::from(directory_path), None)?)
         }
     }
 

src/directory/ram_directory.rs

@@ -144,22 +144,6 @@ impl RAMDirectory {
     pub fn total_mem_usage(&self) -> usize {
         self.fs.read().unwrap().total_mem_usage()
     }
-
-    /// Write a copy of all of the files saved in the RAMDirectory in the target `Directory`.
-    ///
-    /// Files are all written using the `Directory::write` meaning, even if they were
-    /// written using the `atomic_write` api.
-    ///
-    /// If an error is encounterred, files may be persisted partially.
-    pub fn persist(&self, dest: &mut dyn Directory) -> crate::Result<()> {
-        let wlock = self.fs.write().unwrap();
-        for (path, source) in wlock.fs.iter() {
-            let mut dest_wrt = dest.open_write(path)?;
-            dest_wrt.write_all(source.as_slice())?;
-            dest_wrt.terminate()?;
-        }
-        Ok(())
-    }
 }
 
 impl Directory for RAMDirectory {
@@ -220,28 +204,3 @@ impl Directory for RAMDirectory {
         Ok(self.fs.write().unwrap().watch(watch_callback))
     }
 }
-
-#[cfg(test)]
-mod tests {
-    use super::RAMDirectory;
-    use crate::Directory;
-    use std::io::Write;
-    use std::path::Path;
-
-    #[test]
-    fn test_persist() {
-        let msg_atomic: &'static [u8] = b"atomic is the way";
-        let msg_seq: &'static [u8] = b"sequential is the way";
-        let path_atomic: &'static Path = Path::new("atomic");
-        let path_seq: &'static Path = Path::new("seq");
-        let mut directory = RAMDirectory::create();
-        assert!(directory.atomic_write(path_atomic, msg_atomic).is_ok());
-        let mut wrt = directory.open_write(path_seq).unwrap();
-        assert!(wrt.write_all(msg_seq).is_ok());
-        assert!(wrt.flush().is_ok());
-        let mut directory_copy = RAMDirectory::create();
-        assert!(directory.persist(&mut directory_copy).is_ok());
-        assert_eq!(directory_copy.atomic_read(path_atomic).unwrap(), msg_atomic);
-        assert_eq!(directory_copy.atomic_read(path_seq).unwrap(), msg_seq);
-    }
-}

src/docset.rs

@@ -1,47 +1,58 @@
+use crate::common::BitSet;
 use crate::fastfield::DeleteBitSet;
 use crate::DocId;
 use std::borrow::Borrow;
 use std::borrow::BorrowMut;
+use std::cmp::Ordering;
 
-/// Sentinel value returned when a DocSet has been entirely consumed.
-///
-/// This is not u32::MAX as one would have expected, due to the lack of SSE2 instructions
-/// to compare [u32; 4].
-pub const TERMINATED: DocId = std::i32::MAX as u32;
+/// Expresses the outcome of a call to `DocSet`'s `.skip_next(...)`.
+#[derive(PartialEq, Eq, Debug)]
+pub enum SkipResult {
+    /// target was in the docset
+    Reached,
+    /// target was not in the docset, skipping stopped as a greater element was found
+    OverStep,
+    /// the docset was entirely consumed without finding the target, nor any
+    /// element greater than the target.
+    End,
+}
 
 /// Represents an iterable set of sorted doc ids.
 pub trait DocSet {
     /// Goes to the next element.
-    ///
-    /// The DocId of the next element is returned.
-    /// In other words we should always have :
-    /// ```ignore
-    /// let doc = docset.advance();
-    /// assert_eq!(doc, docset.doc());
-    /// ```
-    ///
-    /// If we reached the end of the DocSet, TERMINATED should be returned.
-    ///
-    /// Calling `.advance()` on a terminated DocSet should be supported, and TERMINATED should
-    /// be returned.
-    /// TODO Test existing docsets.
-    fn advance(&mut self) -> DocId;
+    /// `.advance(...)` needs to be called a first time to point to the correct
+    /// element.
+    fn advance(&mut self) -> bool;
 
-    /// Advances the DocSet forward until reaching the target, or going to the
-    /// lowest DocId greater than the target.
+    /// After skipping, position the iterator in such a way that `.doc()`
+    /// will return a value greater than or equal to target.
     ///
-    /// If the end of the DocSet is reached, TERMINATED is returned.
+    /// SkipResult expresses whether the `target value` was reached, overstepped,
+    /// or if the `DocSet` was entirely consumed without finding any value
+    /// greater or equal to the `target`.
     ///
-    /// Calling `.seek(target)` on a terminated DocSet is legal. Implementation
-    /// of DocSet should support it.
+    /// WARNING: Calling skip always advances the docset.
+    /// More specifically, if the docset is already positionned on the target
+    /// skipping will advance to the next position and return SkipResult::Overstep.
     ///
-    /// Calling `seek(TERMINATED)` is also legal and is the normal way to consume a DocSet.
-    fn seek(&mut self, target: DocId) -> DocId {
-        let mut doc = self.doc();
-        while doc < target {
-            doc = self.advance();
+    /// If `.skip_next()` oversteps, then the docset must be positionned correctly
+    /// on an existing document. In other words, `.doc()` should return the first document
+    /// greater than `DocId`.
+    fn skip_next(&mut self, target: DocId) -> SkipResult {
+        if !self.advance() {
+            return SkipResult::End;
+        }
+        loop {
+            match self.doc().cmp(&target) {
+                Ordering::Less => {
+                    if !self.advance() {
+                        return SkipResult::End;
+                    }
+                }
+                Ordering::Equal => return SkipResult::Reached,
+                Ordering::Greater => return SkipResult::OverStep,
+            }
         }
-        doc
     }
 
     /// Fills a given mutable buffer with the next doc ids from the
@@ -60,38 +71,38 @@ pub trait DocSet {
     /// use case where batching. The normal way to
     /// go through the `DocId`'s is to call `.advance()`.
     fn fill_buffer(&mut self, buffer: &mut [DocId]) -> usize {
-        if self.doc() == TERMINATED {
-            return 0;
-        }
         for (i, buffer_val) in buffer.iter_mut().enumerate() {
-            *buffer_val = self.doc();
-            if self.advance() == TERMINATED {
-                return i + 1;
+            if self.advance() {
+                *buffer_val = self.doc();
+            } else {
+                return i;
             }
         }
         buffer.len()
     }
 
     /// Returns the current document
-    /// Right after creating a new DocSet, the docset points to the first document.
-    ///
-    /// If the DocSet is empty, .doc() should return `TERMINATED`.
     fn doc(&self) -> DocId;
 
     /// Returns a best-effort hint of the
     /// length of the docset.
     fn size_hint(&self) -> u32;
 
+    /// Appends all docs to a `bitset`.
+    fn append_to_bitset(&mut self, bitset: &mut BitSet) {
+        while self.advance() {
+            bitset.insert(self.doc());
+        }
+    }
+
     /// Returns the number documents matching.
     /// Calling this method consumes the `DocSet`.
     fn count(&mut self, delete_bitset: &DeleteBitSet) -> u32 {
         let mut count = 0u32;
-        let mut doc = self.doc();
-        while doc != TERMINATED {
-            if !delete_bitset.is_deleted(doc) {
+        while self.advance() {
+            if !delete_bitset.is_deleted(self.doc()) {
                 count += 1u32;
             }
-            doc = self.advance();
         }
         count
     }
@@ -103,42 +114,22 @@ pub trait DocSet {
     /// given by `count()`.
     fn count_including_deleted(&mut self) -> u32 {
         let mut count = 0u32;
-        let mut doc = self.doc();
-        while doc != TERMINATED {
+        while self.advance() {
             count += 1u32;
-            doc = self.advance();
         }
         count
     }
 }
 
-impl<'a> DocSet for &'a mut dyn DocSet {
-    fn advance(&mut self) -> u32 {
-        (**self).advance()
-    }
-
-    fn seek(&mut self, target: DocId) -> DocId {
-        (**self).seek(target)
-    }
-
-    fn doc(&self) -> u32 {
-        (**self).doc()
-    }
-
-    fn size_hint(&self) -> u32 {
-        (**self).size_hint()
-    }
-}
-
 impl<TDocSet: DocSet + ?Sized> DocSet for Box<TDocSet> {
-    fn advance(&mut self) -> DocId {
+    fn advance(&mut self) -> bool {
         let unboxed: &mut TDocSet = self.borrow_mut();
         unboxed.advance()
     }
 
-    fn seek(&mut self, target: DocId) -> DocId {
+    fn skip_next(&mut self, target: DocId) -> SkipResult {
         let unboxed: &mut TDocSet = self.borrow_mut();
-        unboxed.seek(target)
+        unboxed.skip_next(target)
     }
 
     fn doc(&self) -> DocId {
@@ -160,4 +151,9 @@ impl<TDocSet: DocSet + ?Sized> DocSet for Box<TDocSet> {
         let unboxed: &mut TDocSet = self.borrow_mut();
         unboxed.count_including_deleted()
     }
+
+    fn append_to_bitset(&mut self, bitset: &mut BitSet) {
+        let unboxed: &mut TDocSet = self.borrow_mut();
+        unboxed.append_to_bitset(bitset);
+    }
 }

src/error.rs

@@ -7,6 +7,7 @@ use crate::directory::error::{Incompatibility, LockError};
 use crate::fastfield::FastFieldNotAvailableError;
 use crate::query;
 use crate::schema;
+use serde_json;
 use std::fmt;
 use std::path::PathBuf;
 use std::sync::PoisonError;

src/fastfield/multivalued/writer.rs

@@ -6,6 +6,7 @@ use crate::schema::{Document, Field};
 use crate::termdict::TermOrdinal;
 use crate::DocId;
 use fnv::FnvHashMap;
+use itertools::Itertools;
 use std::io;
 
 /// Writer for multi-valued (as in, more than one value per document)
@@ -150,8 +151,8 @@ impl MultiValueIntFastFieldWriter {
                     }
                 }
                 None => {
-                    let val_min_max = crate::common::minmax(self.vals.iter().cloned());
-                    let (val_min, val_max) = val_min_max.unwrap_or((0u64, 0u64));
+                    let val_min_max = self.vals.iter().cloned().minmax();
+                    let (val_min, val_max) = val_min_max.into_option().unwrap_or((0u64, 0u64));
                     value_serializer =
                         serializer.new_u64_fast_field_with_idx(self.field, val_min, val_max, 1)?;
                     for &val in &self.vals {

src/indexer/index_writer.rs

@@ -10,7 +10,7 @@ use crate::core::SegmentMeta;
 use crate::core::SegmentReader;
 use crate::directory::TerminatingWrite;
 use crate::directory::{DirectoryLock, GarbageCollectionResult};
-use crate::docset::{DocSet, TERMINATED};
+use crate::docset::DocSet;
 use crate::error::TantivyError;
 use crate::fastfield::write_delete_bitset;
 use crate::indexer::delete_queue::{DeleteCursor, DeleteQueue};
@@ -112,15 +112,15 @@ fn compute_deleted_bitset(
         if let Some(mut docset) =
             inverted_index.read_postings(&delete_op.term, IndexRecordOption::Basic)
         {
-            let mut deleted_doc = docset.doc();
-            while deleted_doc != TERMINATED {
+            while docset.advance() {
+                let deleted_doc = docset.doc();
                 if deleted_doc < limit_doc {
                     delete_bitset.insert(deleted_doc);
                     might_have_changed = true;
                 }
-                deleted_doc = docset.advance();
             }
         }
+
         delete_cursor.advance();
     }
     Ok(might_have_changed)
@@ -155,8 +155,6 @@ pub(crate) fn advance_deletes(
         None => BitSet::with_max_value(max_doc),
     };
 
-    let num_deleted_docs_before = segment.meta().num_deleted_docs();
-
     compute_deleted_bitset(
         &mut delete_bitset,
         &segment_reader,
@@ -166,8 +164,6 @@ pub(crate) fn advance_deletes(
     )?;
 
     // TODO optimize
-    // It should be possible to do something smarter by manipulation bitsets directly
-    // to compute this union.
     if let Some(seg_delete_bitset) = segment_reader.delete_bitset() {
         for doc in 0u32..max_doc {
             if seg_delete_bitset.is_deleted(doc) {
@@ -176,9 +172,8 @@ pub(crate) fn advance_deletes(
         }
     }
 
-    let num_deleted_docs: u32 = delete_bitset.len() as u32;
-    if num_deleted_docs > num_deleted_docs_before {
-        // There are new deletes. We need to write a new delete file.
+    let num_deleted_docs = delete_bitset.len();
+    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, max_doc, &mut delete_file)?;
@@ -808,46 +803,6 @@ mod tests {
         assert_eq!(batch_opstamp1, 2u64);
     }
 
-    #[test]
-    fn test_no_need_to_rewrite_delete_file_if_no_new_deletes() {
-        let mut schema_builder = schema::Schema::builder();
-        let text_field = schema_builder.add_text_field("text", schema::TEXT);
-        let index = Index::create_in_ram(schema_builder.build());
-
-        let mut index_writer = index.writer_with_num_threads(1, 3_000_000).unwrap();
-        index_writer.add_document(doc!(text_field => "hello1"));
-        index_writer.add_document(doc!(text_field => "hello2"));
-        assert!(index_writer.commit().is_ok());
-
-        let reader = index.reader().unwrap();
-        let searcher = reader.searcher();
-        assert_eq!(searcher.segment_readers().len(), 1);
-        assert_eq!(searcher.segment_reader(0u32).num_deleted_docs(), 0);
-
-        index_writer.delete_term(Term::from_field_text(text_field, "hello1"));
-        assert!(index_writer.commit().is_ok());
-
-        assert!(reader.reload().is_ok());
-        let searcher = reader.searcher();
-        assert_eq!(searcher.segment_readers().len(), 1);
-        assert_eq!(searcher.segment_reader(0u32).num_deleted_docs(), 1);
-
-        let previous_delete_opstamp = index.load_metas().unwrap().segments[0].delete_opstamp();
-
-        // All docs containing hello1 have been already removed.
-        // We should not update the delete meta.
-        index_writer.delete_term(Term::from_field_text(text_field, "hello1"));
-        assert!(index_writer.commit().is_ok());
-
-        assert!(reader.reload().is_ok());
-        let searcher = reader.searcher();
-        assert_eq!(searcher.segment_readers().len(), 1);
-        assert_eq!(searcher.segment_reader(0u32).num_deleted_docs(), 1);
-
-        let after_delete_opstamp = index.load_metas().unwrap().segments[0].delete_opstamp();
-        assert_eq!(after_delete_opstamp, previous_delete_opstamp);
-    }
-
     #[test]
     fn test_ordered_batched_operations() {
         // * one delete for `doc!(field=>"a")`

src/indexer/merger.rs

@@ -2,7 +2,7 @@ use crate::common::MAX_DOC_LIMIT;
 use crate::core::Segment;
 use crate::core::SegmentReader;
 use crate::core::SerializableSegment;
-use crate::docset::{DocSet, TERMINATED};
+use crate::docset::DocSet;
 use crate::fastfield::BytesFastFieldReader;
 use crate::fastfield::DeleteBitSet;
 use crate::fastfield::FastFieldReader;
@@ -21,6 +21,7 @@ use crate::store::StoreWriter;
 use crate::termdict::TermMerger;
 use crate::termdict::TermOrdinal;
 use crate::DocId;
+use itertools::Itertools;
 use std::cmp;
 use std::collections::HashMap;
 
@@ -69,11 +70,11 @@ fn compute_min_max_val(
             Some(delete_bitset) => {
                 // some deleted documents,
                 // we need to recompute the max / min
-                crate::common::minmax(
-                    (0..max_doc)
-                        .filter(|doc_id| delete_bitset.is_alive(*doc_id))
-                        .map(|doc_id| u64_reader.get(doc_id)),
-                )
+                (0..max_doc)
+                    .filter(|doc_id| delete_bitset.is_alive(*doc_id))
+                    .map(|doc_id| u64_reader.get(doc_id))
+                    .minmax()
+                    .into_option()
             }
             None => {
                 // no deleted documents,
@@ -574,12 +575,10 @@ impl IndexMerger {
                     let inverted_index = segment_reader.inverted_index(indexed_field);
                     let mut segment_postings = inverted_index
                         .read_postings_from_terminfo(term_info, segment_postings_option);
-                    let mut doc = segment_postings.doc();
-                    while doc != TERMINATED {
-                        if !segment_reader.is_deleted(doc) {
+                    while segment_postings.advance() {
+                        if !segment_reader.is_deleted(segment_postings.doc()) {
                             return Some((segment_ord, segment_postings));
                         }
-                        doc = segment_postings.advance();
                     }
                     None
                 })
@@ -606,9 +605,17 @@ impl IndexMerger {
                 // postings serializer.
                 for (segment_ord, mut segment_postings) in segment_postings {
                     let old_to_new_doc_id = &merged_doc_id_map[segment_ord];
+                    loop {
+                        let doc = segment_postings.doc();
+
+                        // `.advance()` has been called once before the loop.
+                        //
+                        // It was required to make sure we only consider segments
+                        // that effectively contain at least one non-deleted document
+                        // and remove terms that do not have documents associated.
+                        //
+                        //  For this reason, we cannot use a `while segment_postings.advance()` loop.
 
-                    let mut doc = segment_postings.doc();
-                    while doc != TERMINATED {
                         // deleted doc are skipped as they do not have a `remapped_doc_id`.
                         if let Some(remapped_doc_id) = old_to_new_doc_id[doc as usize] {
                             // we make sure to only write the term iff
@@ -623,8 +630,9 @@ impl IndexMerger {
                                 delta_positions,
                             )?;
                         }
-
-                        doc = segment_postings.advance();
+                        if !segment_postings.advance() {
+                            break;
+                        }
                     }
                 }
 

src/indexer/segment_updater.rs

@@ -23,6 +23,7 @@ use futures::channel::oneshot;
 use futures::executor::{ThreadPool, ThreadPoolBuilder};
 use futures::future::Future;
 use futures::future::TryFutureExt;
+use serde_json;
 use std::borrow::BorrowMut;
 use std::collections::HashSet;
 use std::io::Write;

src/lib.rs

@@ -98,6 +98,9 @@
 //! [literate programming](https://tantivy-search.github.io/examples/basic_search.html) /
 //! [source code](https://github.com/tantivy-search/tantivy/blob/master/examples/basic_search.rs))
 
+#[macro_use]
+extern crate serde_derive;
+
 #[cfg_attr(test, macro_use)]
 extern crate serde_json;
 
@@ -156,7 +159,7 @@ mod snippet;
 pub use self::snippet::{Snippet, SnippetGenerator};
 
 mod docset;
-pub use self::docset::{DocSet, TERMINATED};
+pub use self::docset::{DocSet, SkipResult};
 pub use crate::common::{f64_to_u64, i64_to_u64, u64_to_f64, u64_to_i64};
 pub use crate::core::{Executor, SegmentComponent};
 pub use crate::core::{Index, IndexMeta, Searcher, Segment, SegmentId, SegmentMeta};
@@ -170,7 +173,6 @@ pub use crate::schema::{Document, Term};
 use std::fmt;
 
 use once_cell::sync::Lazy;
-use serde::{Deserialize, Serialize};
 
 /// Index format version.
 const INDEX_FORMAT_VERSION: u32 = 1;
@@ -285,7 +287,7 @@ mod tests {
 
     use crate::collector::tests::TEST_COLLECTOR_WITH_SCORE;
     use crate::core::SegmentReader;
-    use crate::docset::{DocSet, TERMINATED};
+    use crate::docset::DocSet;
     use crate::query::BooleanQuery;
     use crate::schema::*;
     use crate::DocAddress;
@@ -381,12 +383,19 @@ mod tests {
             index_writer.commit().unwrap();
         }
         {
-            index_writer.add_document(doc!(text_field=>"a"));
-            index_writer.add_document(doc!(text_field=>"a a"));
+            {
+                let doc = doc!(text_field=>"a");
+                index_writer.add_document(doc);
+            }
+            {
+                let doc = doc!(text_field=>"a a");
+                index_writer.add_document(doc);
+            }
             index_writer.commit().unwrap();
         }
         {
-            index_writer.add_document(doc!(text_field=>"c"));
+            let doc = doc!(text_field=>"c");
+            index_writer.add_document(doc);
             index_writer.commit().unwrap();
         }
         {
@@ -465,12 +474,10 @@ mod tests {
     }
 
     fn advance_undeleted(docset: &mut dyn DocSet, reader: &SegmentReader) -> bool {
-        let mut doc = docset.advance();
-        while doc != TERMINATED {
-            if !reader.is_deleted(doc) {
+        while docset.advance() {
+            if !reader.is_deleted(docset.doc()) {
                 return true;
             }
-            doc = docset.advance();
         }
         false
     }
@@ -636,8 +643,9 @@ mod tests {
             .inverted_index(term.field())
             .read_postings(&term, IndexRecordOption::Basic)
             .unwrap();
+        assert!(postings.advance());
         assert_eq!(postings.doc(), 0);
-        assert_eq!(postings.advance(), TERMINATED);
+        assert!(!postings.advance());
     }
 
     #[test]
@@ -659,8 +667,9 @@ mod tests {
             .inverted_index(term.field())
             .read_postings(&term, IndexRecordOption::Basic)
             .unwrap();
+        assert!(postings.advance());
         assert_eq!(postings.doc(), 0);
-        assert_eq!(postings.advance(), TERMINATED);
+        assert!(!postings.advance());
     }
 
     #[test]
@@ -682,8 +691,9 @@ mod tests {
             .inverted_index(term.field())
             .read_postings(&term, IndexRecordOption::Basic)
             .unwrap();
+        assert!(postings.advance());
         assert_eq!(postings.doc(), 0);
-        assert_eq!(postings.advance(), TERMINATED);
+        assert!(!postings.advance());
     }
 
     #[test]
@@ -752,8 +762,10 @@ mod tests {
         {
             // writing the segment
             let mut index_writer = index.writer_with_num_threads(1, 3_000_000).unwrap();
-            let doc = doc!(text_field=>"af af af bc bc");
-            index_writer.add_document(doc);
+            {
+                let doc = doc!(text_field=>"af af af bc bc");
+                index_writer.add_document(doc);
+            }
             index_writer.commit().unwrap();
         }
         {
@@ -769,9 +781,10 @@ mod tests {
             let mut postings = inverted_index
                 .read_postings(&term_af, IndexRecordOption::WithFreqsAndPositions)
                 .unwrap();
+            assert!(postings.advance());
             assert_eq!(postings.doc(), 0);
             assert_eq!(postings.term_freq(), 3);
-            assert_eq!(postings.advance(), TERMINATED);
+            assert!(!postings.advance());
         }
     }
 

src/postings/block_search.rs

@@ -106,7 +106,7 @@ impl BlockSearcher {
     /// the target.
     ///
     /// The results should be equivalent to
-    /// ```compile_fail
+    /// ```ignore
     /// block[..]
     //       .iter()
     //       .take_while(|&&val| val < target)
@@ -129,23 +129,23 @@ impl BlockSearcher {
     ///
     /// If SSE2 instructions are available in the `(platform, running CPU)`,
     /// then we use a different implementation that does an exhaustive linear search over
-    /// the block regardless of whether the block is full or not.
-    ///
-    /// Indeed, if the block is not full, the remaining items are TERMINATED.
-    /// It is surprisingly faster, most likely because of the lack of branch misprediction.
+    /// the full block whenever the block is full (`len == 128`). It is surprisingly faster, most likely because of the lack
+    /// of branch.
     pub(crate) fn search_in_block(
         self,
         block_docs: &AlignedBuffer,
+        len: usize,
         start: usize,
         target: u32,
     ) -> usize {
         #[cfg(target_arch = "x86_64")]
         {
-            if self == BlockSearcher::SSE2 {
+            use crate::postings::compression::COMPRESSION_BLOCK_SIZE;
+            if self == BlockSearcher::SSE2 && len == COMPRESSION_BLOCK_SIZE {
                 return sse2::linear_search_sse2_128(block_docs, target);
             }
         }
-        start + galloping(&block_docs.0[start..], target)
+        start + galloping(&block_docs.0[start..len], target)
     }
 }
 
@@ -166,7 +166,6 @@ mod tests {
     use super::exponential_search;
     use super::linear_search;
     use super::BlockSearcher;
-    use crate::docset::TERMINATED;
     use crate::postings::compression::{AlignedBuffer, COMPRESSION_BLOCK_SIZE};
 
     #[test]
@@ -197,11 +196,16 @@ mod tests {
     fn util_test_search_in_block(block_searcher: BlockSearcher, block: &[u32], target: u32) {
         let cursor = search_in_block_trivial_but_slow(block, target);
         assert!(block.len() < COMPRESSION_BLOCK_SIZE);
-        let mut output_buffer = [TERMINATED; COMPRESSION_BLOCK_SIZE];
+        let mut output_buffer = [u32::max_value(); COMPRESSION_BLOCK_SIZE];
         output_buffer[..block.len()].copy_from_slice(block);
         for i in 0..cursor {
             assert_eq!(
-                block_searcher.search_in_block(&AlignedBuffer(output_buffer), i, target),
+                block_searcher.search_in_block(
+                    &AlignedBuffer(output_buffer),
+                    block.len(),
+                    i,
+                    target
+                ),
                 cursor
             );
         }

src/postings/compression/mod.rs

@@ -1,5 +1,4 @@
 use crate::common::FixedSize;
-use crate::docset::TERMINATED;
 use bitpacking::{BitPacker, BitPacker4x};
 
 pub const COMPRESSION_BLOCK_SIZE: usize = BitPacker4x::BLOCK_LEN;
@@ -91,18 +90,14 @@ impl BlockDecoder {
     }
 
     #[inline]
-    pub(crate) fn output_aligned(&self) -> &AlignedBuffer {
-        &self.output
+    pub(crate) fn output_aligned(&self) -> (&AlignedBuffer, usize) {
+        (&self.output, self.output_len)
     }
 
     #[inline]
     pub fn output(&self, idx: usize) -> u32 {
         self.output.0[idx]
     }
-
-    pub fn clear(&mut self) {
-        self.output.0.iter_mut().for_each(|el| *el = TERMINATED);
-    }
 }
 
 pub trait VIntEncoder {
@@ -139,9 +134,9 @@ pub trait VIntDecoder {
     /// For instance, if delta encoded are `1, 3, 9`, and the
     /// `offset` is 5, then the output will be:
     /// `5 + 1 = 6, 6 + 3= 9, 9 + 9 = 18`
-    fn uncompress_vint_sorted(
+    fn uncompress_vint_sorted<'a>(
         &mut self,
-        compressed_data: &[u8],
+        compressed_data: &'a [u8],
         offset: u32,
         num_els: usize,
     ) -> usize;
@@ -151,7 +146,7 @@ pub trait VIntDecoder {
     ///
     /// The method takes a number of int to decompress, and returns
     /// the amount of bytes that were read to decompress them.
-    fn uncompress_vint_unsorted(&mut self, compressed_data: &[u8], num_els: usize) -> usize;
+    fn uncompress_vint_unsorted<'a>(&mut self, compressed_data: &'a [u8], num_els: usize) -> usize;
 }
 
 impl VIntEncoder for BlockEncoder {
@@ -165,9 +160,9 @@ impl VIntEncoder for BlockEncoder {
 }
 
 impl VIntDecoder for BlockDecoder {
-    fn uncompress_vint_sorted(
+    fn uncompress_vint_sorted<'a>(
         &mut self,
-        compressed_data: &[u8],
+        compressed_data: &'a [u8],
         offset: u32,
         num_els: usize,
     ) -> usize {

src/postings/compression/vint.rs

@@ -42,7 +42,7 @@ pub(crate) fn compress_unsorted<'a>(input: &[u32], output: &'a mut [u8]) -> &'a
 }
 
 #[inline(always)]
-pub fn uncompress_sorted(compressed_data: &[u8], output: &mut [u32], offset: u32) -> usize {
+pub fn uncompress_sorted<'a>(compressed_data: &'a [u8], output: &mut [u32], offset: u32) -> usize {
     let mut read_byte = 0;
     let mut result = offset;
     for output_mut in output.iter_mut() {

src/postings/mod.rs

@@ -51,7 +51,7 @@ pub mod tests {
     use crate::core::Index;
     use crate::core::SegmentComponent;
     use crate::core::SegmentReader;
-    use crate::docset::{DocSet, TERMINATED};
+    use crate::docset::{DocSet, SkipResult};
     use crate::fieldnorm::FieldNormReader;
     use crate::indexer::operation::AddOperation;
     use crate::indexer::SegmentWriter;
@@ -115,12 +115,29 @@ pub mod tests {
             let mut postings = inverted_index
                 .read_postings(&term, IndexRecordOption::WithFreqsAndPositions)
                 .unwrap();
-            assert_eq!(postings.doc(), 0);
+            postings.advance();
             postings.positions(&mut positions);
             assert_eq!(&[0, 1, 2], &positions[..]);
             postings.positions(&mut positions);
             assert_eq!(&[0, 1, 2], &positions[..]);
-            assert_eq!(postings.advance(), 1);
+            postings.advance();
+            postings.positions(&mut positions);
+            assert_eq!(&[0, 5], &positions[..]);
+        }
+        {
+            let mut postings = inverted_index
+                .read_postings(&term, IndexRecordOption::WithFreqsAndPositions)
+                .unwrap();
+            postings.advance();
+            postings.advance();
+            postings.positions(&mut positions);
+            assert_eq!(&[0, 5], &positions[..]);
+        }
+        {
+            let mut postings = inverted_index
+                .read_postings(&term, IndexRecordOption::WithFreqsAndPositions)
+                .unwrap();
+            assert_eq!(postings.skip_next(1), SkipResult::Reached);
             assert_eq!(postings.doc(), 1);
             postings.positions(&mut positions);
             assert_eq!(&[0, 5], &positions[..]);
@@ -129,25 +146,7 @@ pub mod tests {
             let mut postings = inverted_index
                 .read_postings(&term, IndexRecordOption::WithFreqsAndPositions)
                 .unwrap();
-            assert_eq!(postings.doc(), 0);
-            assert_eq!(postings.advance(), 1);
-            postings.positions(&mut positions);
-            assert_eq!(&[0, 5], &positions[..]);
-        }
-        {
-            let mut postings = inverted_index
-                .read_postings(&term, IndexRecordOption::WithFreqsAndPositions)
-                .unwrap();
-            assert_eq!(postings.seek(1), 1);
-            assert_eq!(postings.doc(), 1);
-            postings.positions(&mut positions);
-            assert_eq!(&[0, 5], &positions[..]);
-        }
-        {
-            let mut postings = inverted_index
-                .read_postings(&term, IndexRecordOption::WithFreqsAndPositions)
-                .unwrap();
-            assert_eq!(postings.seek(1002), 1002);
+            assert_eq!(postings.skip_next(1002), SkipResult::Reached);
             assert_eq!(postings.doc(), 1002);
             postings.positions(&mut positions);
             assert_eq!(&[0, 5], &positions[..]);
@@ -156,8 +155,8 @@ pub mod tests {
             let mut postings = inverted_index
                 .read_postings(&term, IndexRecordOption::WithFreqsAndPositions)
                 .unwrap();
-            assert_eq!(postings.seek(100), 100);
-            assert_eq!(postings.seek(1002), 1002);
+            assert_eq!(postings.skip_next(100), SkipResult::Reached);
+            assert_eq!(postings.skip_next(1002), SkipResult::Reached);
             assert_eq!(postings.doc(), 1002);
             postings.positions(&mut positions);
             assert_eq!(&[0, 5], &positions[..]);
@@ -282,21 +281,22 @@ pub mod tests {
                     .read_postings(&term_a, IndexRecordOption::WithFreqsAndPositions)
                     .unwrap();
                 assert_eq!(postings_a.len(), 1000);
+                assert!(postings_a.advance());
                 assert_eq!(postings_a.doc(), 0);
                 assert_eq!(postings_a.term_freq(), 6);
                 postings_a.positions(&mut positions);
                 assert_eq!(&positions[..], [0, 2, 4, 6, 7, 13]);
-                assert_eq!(postings_a.advance(), 1u32);
+                assert!(postings_a.advance());
                 assert_eq!(postings_a.doc(), 1u32);
                 assert_eq!(postings_a.term_freq(), 1);
                 for i in 2u32..1000u32 {
-                    assert_eq!(postings_a.advance(), i);
+                    assert!(postings_a.advance());
                     assert_eq!(postings_a.term_freq(), 1);
                     postings_a.positions(&mut positions);
                     assert_eq!(&positions[..], [i]);
                     assert_eq!(postings_a.doc(), i);
                 }
-                assert_eq!(postings_a.advance(), TERMINATED);
+                assert!(!postings_a.advance());
             }
             {
                 let term_e = Term::from_field_text(text_field, "e");
@@ -306,6 +306,7 @@ pub mod tests {
                     .unwrap();
                 assert_eq!(postings_e.len(), 1000 - 2);
                 for i in 2u32..1000u32 {
+                    assert!(postings_e.advance());
                     assert_eq!(postings_e.term_freq(), i);
                     postings_e.positions(&mut positions);
                     assert_eq!(positions.len(), i as usize);
@@ -313,9 +314,8 @@ pub mod tests {
                         assert_eq!(positions[j], (j as u32));
                     }
                     assert_eq!(postings_e.doc(), i);
-                    postings_e.advance();
                 }
-                assert_eq!(postings_e.doc(), TERMINATED);
+                assert!(!postings_e.advance());
             }
         }
     }
@@ -329,8 +329,16 @@ pub mod tests {
         let index = Index::create_in_ram(schema);
         {
             let mut index_writer = index.writer_with_num_threads(1, 3_000_000).unwrap();
-            index_writer.add_document(doc!(text_field => "g b b d c g c"));
-            index_writer.add_document(doc!(text_field => "g a b b a d c g c"));
+            {
+                let mut doc = Document::default();
+                doc.add_text(text_field, "g b b d c g c");
+                index_writer.add_document(doc);
+            }
+            {
+                let mut doc = Document::default();
+                doc.add_text(text_field, "g a b b a d c g c");
+                index_writer.add_document(doc);
+            }
             assert!(index_writer.commit().is_ok());
         }
         let term_a = Term::from_field_text(text_field, "a");
@@ -340,6 +348,7 @@ pub mod tests {
             .inverted_index(text_field)
             .read_postings(&term_a, IndexRecordOption::WithFreqsAndPositions)
             .unwrap();
+        assert!(postings.advance());
         assert_eq!(postings.doc(), 1u32);
         postings.positions(&mut positions);
         assert_eq!(&positions[..], &[1u32, 4]);
@@ -361,8 +370,11 @@ pub mod tests {
             let index = Index::create_in_ram(schema);
             {
                 let mut index_writer = index.writer_with_num_threads(1, 3_000_000).unwrap();
-                for i in 0u64..num_docs as u64 {
-                    let doc = doc!(value_field => 2u64, value_field => i % 2u64);
+                for i in 0..num_docs {
+                    let mut doc = Document::default();
+                    doc.add_u64(value_field, 2);
+                    doc.add_u64(value_field, (i % 2) as u64);
+
                     index_writer.add_document(doc);
                 }
                 assert!(index_writer.commit().is_ok());
@@ -379,10 +391,11 @@ pub mod tests {
                     .inverted_index(term_2.field())
                     .read_postings(&term_2, IndexRecordOption::Basic)
                     .unwrap();
-                assert_eq!(segment_postings.seek(i), i);
+
+                assert_eq!(segment_postings.skip_next(i), SkipResult::Reached);
                 assert_eq!(segment_postings.doc(), i);
 
-                assert_eq!(segment_postings.seek(j), j);
+                assert_eq!(segment_postings.skip_next(j), SkipResult::Reached);
                 assert_eq!(segment_postings.doc(), j);
             }
         }
@@ -394,16 +407,17 @@ pub mod tests {
                 .unwrap();
 
             // check that `skip_next` advances the iterator
+            assert!(segment_postings.advance());
             assert_eq!(segment_postings.doc(), 0);
 
-            assert_eq!(segment_postings.seek(1), 1);
+            assert_eq!(segment_postings.skip_next(1), SkipResult::Reached);
             assert_eq!(segment_postings.doc(), 1);
 
-            assert_eq!(segment_postings.seek(1), 1);
-            assert_eq!(segment_postings.doc(), 1);
+            assert_eq!(segment_postings.skip_next(1), SkipResult::OverStep);
+            assert_eq!(segment_postings.doc(), 2);
 
             // check that going beyond the end is handled
-            assert_eq!(segment_postings.seek(num_docs), TERMINATED);
+            assert_eq!(segment_postings.skip_next(num_docs), SkipResult::End);
         }
 
         // check that filtering works
@@ -414,7 +428,7 @@ pub mod tests {
                 .unwrap();
 
             for i in 0..num_docs / 2 {
-                assert_eq!(segment_postings.seek(i * 2), i * 2);
+                assert_eq!(segment_postings.skip_next(i * 2), SkipResult::Reached);
                 assert_eq!(segment_postings.doc(), i * 2);
             }
 
@@ -424,7 +438,7 @@ pub mod tests {
                 .unwrap();
 
             for i in 0..num_docs / 2 - 1 {
-                assert!(segment_postings.seek(i * 2 + 1) > (i * 1) * 2);
+                assert_eq!(segment_postings.skip_next(i * 2 + 1), SkipResult::OverStep);
                 assert_eq!(segment_postings.doc(), (i + 1) * 2);
             }
         }
@@ -436,7 +450,6 @@ pub mod tests {
             assert!(index_writer.commit().is_ok());
         }
         let searcher = index.reader().unwrap().searcher();
-        assert_eq!(searcher.segment_readers().len(), 1);
         let segment_reader = searcher.segment_reader(0);
 
         // make sure seeking still works
@@ -447,11 +460,11 @@ pub mod tests {
                 .unwrap();
 
             if i % 2 == 0 {
-                assert_eq!(segment_postings.seek(i), i);
+                assert_eq!(segment_postings.skip_next(i), SkipResult::Reached);
                 assert_eq!(segment_postings.doc(), i);
                 assert!(segment_reader.is_deleted(i));
             } else {
-                assert_eq!(segment_postings.seek(i), i);
+                assert_eq!(segment_postings.skip_next(i), SkipResult::Reached);
                 assert_eq!(segment_postings.doc(), i);
             }
         }
@@ -466,16 +479,12 @@ pub mod tests {
             let mut last = 2; // start from 5 to avoid seeking to 3 twice
             let mut cur = 3;
             loop {
-                let seek = segment_postings.seek(cur);
-                if seek == TERMINATED {
-                    break;
-                }
-                assert_eq!(seek, segment_postings.doc());
-                if seek == cur {
-                    assert_eq!(segment_postings.doc(), cur);
-                } else {
-                    assert_eq!(segment_postings.doc(), cur + 1);
+                match segment_postings.skip_next(cur) {
+                    SkipResult::End => break,
+                    SkipResult::Reached => assert_eq!(segment_postings.doc(), cur),
+                    SkipResult::OverStep => assert_eq!(segment_postings.doc(), cur + 1),
                 }
+
                 let next = cur + last;
                 last = cur;
                 cur = next;
@@ -561,7 +570,7 @@ pub mod tests {
     }
 
     impl<TDocSet: DocSet> DocSet for UnoptimizedDocSet<TDocSet> {
-        fn advance(&mut self) -> DocId {
+        fn advance(&mut self) -> bool {
             self.0.advance()
         }
 
@@ -587,22 +596,30 @@ pub mod tests {
         for target in targets {
             let mut postings_opt = postings_factory();
             let mut postings_unopt = UnoptimizedDocSet::wrap(postings_factory());
-            let skip_result_opt = postings_opt.seek(target);
-            let skip_result_unopt = postings_unopt.seek(target);
+            let skip_result_opt = postings_opt.skip_next(target);
+            let skip_result_unopt = postings_unopt.skip_next(target);
             assert_eq!(
                 skip_result_unopt, skip_result_opt,
                 "Failed while skipping to {}",
                 target
             );
-            assert!(skip_result_opt >= target);
-            assert_eq!(skip_result_opt, postings_opt.doc());
-            if skip_result_opt == TERMINATED {
-                return;
+            match skip_result_opt {
+                SkipResult::Reached => assert_eq!(postings_opt.doc(), target),
+                SkipResult::OverStep => assert!(postings_opt.doc() > target),
+                SkipResult::End => {
+                    return;
+                }
             }
-            while postings_opt.doc() != TERMINATED {
-                assert_eq!(postings_opt.doc(), postings_unopt.doc());
-                assert_eq!(postings_opt.advance(), postings_unopt.advance());
+            while postings_opt.advance() {
+                assert!(postings_unopt.advance());
+                assert_eq!(
+                    postings_opt.doc(),
+                    postings_unopt.doc(),
+                    "Failed while skipping to {}",
+                    target
+                );
             }
+            assert!(!postings_unopt.advance());
         }
     }
 }
@@ -611,7 +628,7 @@ pub mod tests {
 mod bench {
 
     use super::tests::*;
-    use crate::docset::TERMINATED;
+    use crate::docset::SkipResult;
     use crate::query::Intersection;
     use crate::schema::IndexRecordOption;
     use crate::tests;
@@ -629,7 +646,7 @@ mod bench {
                 .inverted_index(TERM_A.field())
                 .read_postings(&*TERM_A, IndexRecordOption::Basic)
                 .unwrap();
-            while segment_postings.advance() != TERMINATED {}
+            while segment_postings.advance() {}
         });
     }
 
@@ -661,7 +678,7 @@ mod bench {
                 segment_postings_c,
                 segment_postings_d,
             ]);
-            while intersection.advance() != TERMINATED {}
+            while intersection.advance() {}
         });
     }
 
@@ -677,10 +694,11 @@ mod bench {
             .unwrap();
 
         let mut existing_docs = Vec::new();
+        segment_postings.advance();
         for doc in &docs {
             if *doc >= segment_postings.doc() {
                 existing_docs.push(*doc);
-                if segment_postings.seek(*doc) == TERMINATED {
+                if segment_postings.skip_next(*doc) == SkipResult::End {
                     break;
                 }
             }
@@ -692,7 +710,7 @@ mod bench {
                 .read_postings(&*TERM_A, IndexRecordOption::Basic)
                 .unwrap();
             for doc in &existing_docs {
-                if segment_postings.seek(*doc) == TERMINATED {
+                if segment_postings.skip_next(*doc) == SkipResult::End {
                     break;
                 }
             }
@@ -731,9 +749,8 @@ mod bench {
                 .read_postings(&*TERM_A, IndexRecordOption::Basic)
                 .unwrap();
             let mut s = 0u32;
-            while segment_postings.doc() != TERMINATED {
+            while segment_postings.advance() {
                 s += (segment_postings.doc() & n) % 1024;
-                segment_postings.advance()
             }
             s
         });

src/postings/segment_postings.rs

@@ -1,6 +1,7 @@
+use crate::common::BitSet;
 use crate::common::HasLen;
 use crate::common::{BinarySerializable, VInt};
-use crate::docset::{DocSet, TERMINATED};
+use crate::docset::{DocSet, SkipResult};
 use crate::positions::PositionReader;
 use crate::postings::compression::{compressed_block_size, AlignedBuffer};
 use crate::postings::compression::{BlockDecoder, VIntDecoder, COMPRESSION_BLOCK_SIZE};
@@ -13,6 +14,7 @@ use crate::postings::USE_SKIP_INFO_LIMIT;
 use crate::schema::IndexRecordOption;
 use crate::DocId;
 use owned_read::OwnedRead;
+use std::cmp::Ordering;
 use tantivy_fst::Streamer;
 
 struct PositionComputer {
@@ -66,14 +68,12 @@ impl SegmentPostings {
     /// Returns an empty segment postings object
     pub fn empty() -> Self {
         let empty_block_cursor = BlockSegmentPostings::empty();
-        let mut segment_postings = SegmentPostings {
+        SegmentPostings {
             block_cursor: empty_block_cursor,
             cur: COMPRESSION_BLOCK_SIZE,
             position_computer: None,
             block_searcher: BlockSearcher::default(),
-        };
-        segment_postings.advance();
-        segment_postings
+        }
     }
 
     /// Creates a segment postings object with the given documents
@@ -116,14 +116,12 @@ impl SegmentPostings {
         segment_block_postings: BlockSegmentPostings,
         positions_stream_opt: Option<PositionReader>,
     ) -> SegmentPostings {
-        let mut postings = SegmentPostings {
+        SegmentPostings {
             block_cursor: segment_block_postings,
             cur: COMPRESSION_BLOCK_SIZE, // cursor within the block
             position_computer: positions_stream_opt.map(PositionComputer::new),
             block_searcher: BlockSearcher::default(),
-        };
-        postings.advance();
-        postings
+        }
     }
 }
 
@@ -131,7 +129,7 @@ impl DocSet for SegmentPostings {
     // goes to the next element.
     // next needs to be called a first time to point to the correct element.
     #[inline]
-    fn advance(&mut self) -> DocId {
+    fn advance(&mut self) -> bool {
         if self.position_computer.is_some() && self.cur < COMPRESSION_BLOCK_SIZE {
             let term_freq = self.term_freq() as usize;
             if let Some(position_computer) = self.position_computer.as_mut() {
@@ -140,19 +138,29 @@ impl DocSet for SegmentPostings {
         }
         self.cur += 1;
         if self.cur >= self.block_cursor.block_len() {
-            if self.block_cursor.advance() {
-                self.cur = 0;
-            } else {
-                self.cur = COMPRESSION_BLOCK_SIZE - 1;
-                return TERMINATED;
+            self.cur = 0;
+            if !self.block_cursor.advance() {
+                self.cur = COMPRESSION_BLOCK_SIZE;
+                return false;
             }
         }
-        self.doc()
+        true
     }
 
-    fn seek(&mut self, target: DocId) -> DocId {
-        if self.doc() >= target {
-            return self.doc();
+    fn skip_next(&mut self, target: DocId) -> SkipResult {
+        if !self.advance() {
+            return SkipResult::End;
+        }
+        match self.doc().cmp(&target) {
+            Ordering::Equal => {
+                return SkipResult::Reached;
+            }
+            Ordering::Greater => {
+                return SkipResult::OverStep;
+            }
+            _ => {
+                // ...
+            }
         }
 
         // In the following, thanks to the call to advance above,
@@ -162,44 +170,44 @@ impl DocSet for SegmentPostings {
         // skip blocks until one that might contain the target
         // check if we need to go to the next block
         let mut sum_freqs_skipped: u32 = 0;
-        if self
+        if !self
             .block_cursor
             .docs()
             .last()
-            .map(|&doc| doc < target)
-            .unwrap_or(true)
+            .map(|doc| *doc >= target)
+            .unwrap_or(false)
+        // there should always be at least a document in the block
+        // since advance returned.
         {
-            // We are not in the right block.
+            // we are not in the right block.
+            //
+            // First compute all of the freqs skipped from the current block.
             if self.position_computer.is_some() {
-                // First compute all of the freqs skipped from the current block.
-                sum_freqs_skipped = self.block_cursor.freqs()[self.cur..].iter().sum::<u32>();
+                sum_freqs_skipped = self.block_cursor.freqs()[self.cur..].iter().sum();
                 match self.block_cursor.skip_to(target) {
                     BlockSegmentPostingsSkipResult::Success(block_skip_freqs) => {
                         sum_freqs_skipped += block_skip_freqs;
                     }
                     BlockSegmentPostingsSkipResult::Terminated => {
-                        self.block_cursor.doc_decoder.clear();
-                        self.cur = 0;
-                        return TERMINATED;
+                        return SkipResult::End;
                     }
                 }
             } else if self.block_cursor.skip_to(target)
                 == BlockSegmentPostingsSkipResult::Terminated
             {
                 // no positions needed. no need to sum freqs.
-                self.block_cursor.doc_decoder.clear();
-                self.cur = 0;
-                return TERMINATED;
+                return SkipResult::End;
             }
             self.cur = 0;
         }
 
-        // At this point we are on the block, that might contain our document.
-
         let cur = self.cur;
 
-        let output = self.block_cursor.docs_aligned();
-        let new_cur = self.block_searcher.search_in_block(&output, cur, target);
+        // we're in the right block now, start with an exponential search
+        let (output, len) = self.block_cursor.docs_aligned();
+        let new_cur = self
+            .block_searcher
+            .search_in_block(&output, len, cur, target);
         if let Some(position_computer) = self.position_computer.as_mut() {
             sum_freqs_skipped += self.block_cursor.freqs()[cur..new_cur].iter().sum::<u32>();
             position_computer.add_skip(sum_freqs_skipped as usize);
@@ -209,7 +217,11 @@ impl DocSet for SegmentPostings {
         // `doc` is now the first element >= `target`
         let doc = output.0[new_cur];
         debug_assert!(doc >= target);
-        doc
+        if doc == target {
+            SkipResult::Reached
+        } else {
+            SkipResult::OverStep
+        }
     }
 
     /// Return the current document's `DocId`.
@@ -219,14 +231,18 @@ impl DocSet for SegmentPostings {
     /// Will panics if called without having called advance before.
     #[inline]
     fn doc(&self) -> DocId {
-        self.block_cursor.doc(self.cur)
+        let docs = self.block_cursor.docs();
+        debug_assert!(
+            self.cur < docs.len(),
+            "Have you forgotten to call `.advance()` at least once before calling `.doc()`                                      ."
+        );
+        docs[self.cur]
     }
 
     fn size_hint(&self) -> u32 {
         self.len() as u32
     }
 
-    /*
     fn append_to_bitset(&mut self, bitset: &mut BitSet) {
         // finish the current block
         if self.advance() {
@@ -241,7 +257,6 @@ impl DocSet for SegmentPostings {
             }
         }
     }
-     */
 }
 
 impl HasLen for SegmentPostings {
@@ -309,14 +324,15 @@ fn split_into_skips_and_postings(
     doc_freq: u32,
     mut data: OwnedRead,
 ) -> (Option<OwnedRead>, OwnedRead) {
-    if doc_freq < USE_SKIP_INFO_LIMIT {
-        return (None, data);
+    if doc_freq >= USE_SKIP_INFO_LIMIT {
+        let skip_len = VInt::deserialize(&mut data).expect("Data corrupted").0 as usize;
+        let mut postings_data = data.clone();
+        postings_data.advance(skip_len);
+        data.clip(skip_len);
+        (Some(data), postings_data)
+    } else {
+        (None, data)
     }
-    let skip_len = VInt::deserialize(&mut data).expect("Data corrupted").0 as usize;
-    let mut postings_data = data.clone();
-    postings_data.advance(skip_len);
-    data.clip(skip_len);
-    (Some(data), postings_data)
 }
 
 #[derive(Debug, Eq, PartialEq)]
@@ -398,7 +414,7 @@ impl BlockSegmentPostings {
         self.doc_decoder.output_array()
     }
 
-    pub(crate) fn docs_aligned(&self) -> &AlignedBuffer {
+    pub(crate) fn docs_aligned(&self) -> (&AlignedBuffer, usize) {
         self.doc_decoder.output_aligned()
     }
 
@@ -479,36 +495,35 @@ impl BlockSegmentPostings {
             }
         }
 
-        self.doc_decoder.clear();
-
-        if self.num_vint_docs == 0 {
-            return BlockSegmentPostingsSkipResult::Terminated;
-        }
         // we are now on the last, incomplete, variable encoded block.
-        let num_compressed_bytes = self.doc_decoder.uncompress_vint_sorted(
-            self.remaining_data.as_ref(),
-            self.doc_offset,
-            self.num_vint_docs,
-        );
-        self.remaining_data.advance(num_compressed_bytes);
-        match self.freq_reading_option {
-            FreqReadingOption::NoFreq | FreqReadingOption::SkipFreq => {}
-            FreqReadingOption::ReadFreq => {
-                self.freq_decoder
-                    .uncompress_vint_unsorted(self.remaining_data.as_ref(), self.num_vint_docs);
-            }
-        }
-        self.num_vint_docs = 0;
-        self.docs()
-            .last()
-            .map(|last_doc| {
-                if *last_doc >= target_doc {
-                    BlockSegmentPostingsSkipResult::Success(skip_freqs)
-                } else {
-                    BlockSegmentPostingsSkipResult::Terminated
+        if self.num_vint_docs > 0 {
+            let num_compressed_bytes = self.doc_decoder.uncompress_vint_sorted(
+                self.remaining_data.as_ref(),
+                self.doc_offset,
+                self.num_vint_docs,
+            );
+            self.remaining_data.advance(num_compressed_bytes);
+            match self.freq_reading_option {
+                FreqReadingOption::NoFreq | FreqReadingOption::SkipFreq => {}
+                FreqReadingOption::ReadFreq => {
+                    self.freq_decoder
+                        .uncompress_vint_unsorted(self.remaining_data.as_ref(), self.num_vint_docs);
                 }
-            })
-            .unwrap_or(BlockSegmentPostingsSkipResult::Terminated)
+            }
+            self.num_vint_docs = 0;
+            return self
+                .docs()
+                .last()
+                .map(|last_doc| {
+                    if *last_doc >= target_doc {
+                        BlockSegmentPostingsSkipResult::Success(skip_freqs)
+                    } else {
+                        BlockSegmentPostingsSkipResult::Terminated
+                    }
+                })
+                .unwrap_or(BlockSegmentPostingsSkipResult::Terminated);
+        }
+        BlockSegmentPostingsSkipResult::Terminated
     }
 
     /// Advance to the next block.
@@ -539,27 +554,26 @@ impl BlockSegmentPostings {
             }
             // it will be used as the next offset.
             self.doc_offset = self.doc_decoder.output(COMPRESSION_BLOCK_SIZE - 1);
-            return true;
-        }
-        self.doc_decoder.clear();
-        if self.num_vint_docs == 0 {
-            return false;
-        }
-        let num_compressed_bytes = self.doc_decoder.uncompress_vint_sorted(
-            self.remaining_data.as_ref(),
-            self.doc_offset,
-            self.num_vint_docs,
-        );
-        self.remaining_data.advance(num_compressed_bytes);
-        match self.freq_reading_option {
-            FreqReadingOption::NoFreq | FreqReadingOption::SkipFreq => {}
-            FreqReadingOption::ReadFreq => {
-                self.freq_decoder
-                    .uncompress_vint_unsorted(self.remaining_data.as_ref(), self.num_vint_docs);
+            true
+        } else if self.num_vint_docs > 0 {
+            let num_compressed_bytes = self.doc_decoder.uncompress_vint_sorted(
+                self.remaining_data.as_ref(),
+                self.doc_offset,
+                self.num_vint_docs,
+            );
+            self.remaining_data.advance(num_compressed_bytes);
+            match self.freq_reading_option {
+                FreqReadingOption::NoFreq | FreqReadingOption::SkipFreq => {}
+                FreqReadingOption::ReadFreq => {
+                    self.freq_decoder
+                        .uncompress_vint_unsorted(self.remaining_data.as_ref(), self.num_vint_docs);
+                }
             }
+            self.num_vint_docs = 0;
+            true
+        } else {
+            false
         }
-        self.num_vint_docs = 0;
-        true
     }
 
     /// Returns an empty segment postings object
@@ -599,34 +613,34 @@ mod tests {
     use super::SegmentPostings;
     use crate::common::HasLen;
     use crate::core::Index;
-    use crate::docset::{DocSet, TERMINATED};
+    use crate::docset::DocSet;
     use crate::postings::postings::Postings;
     use crate::schema::IndexRecordOption;
     use crate::schema::Schema;
     use crate::schema::Term;
     use crate::schema::INDEXED;
     use crate::DocId;
+    use crate::SkipResult;
     use tantivy_fst::Streamer;
 
     #[test]
     fn test_empty_segment_postings() {
         let mut postings = SegmentPostings::empty();
-        assert_eq!(postings.advance(), TERMINATED);
-        assert_eq!(postings.advance(), TERMINATED);
+        assert!(!postings.advance());
+        assert!(!postings.advance());
         assert_eq!(postings.len(), 0);
     }
 
     #[test]
-    fn test_empty_postings_doc_returns_terminated() {
-        let mut postings = SegmentPostings::empty();
-        assert_eq!(postings.doc(), TERMINATED);
-        assert_eq!(postings.advance(), TERMINATED);
+    #[should_panic(expected = "Have you forgotten to call `.advance()`")]
+    fn test_panic_if_doc_called_before_advance() {
+        SegmentPostings::empty().doc();
     }
 
     #[test]
-    fn test_empty_postings_doc_term_freq_returns_0() {
-        let postings = SegmentPostings::empty();
-        assert_eq!(postings.term_freq(), 1);
+    #[should_panic(expected = "Have you forgotten to call `.advance()`")]
+    fn test_panic_if_freq_called_before_advance() {
+        SegmentPostings::empty().term_freq();
     }
 
     #[test]
@@ -660,27 +674,25 @@ mod tests {
         {
             let block_segments = build_block_postings(&doc_ids);
             let mut docset = SegmentPostings::from_block_postings(block_segments, None);
-            assert_eq!(docset.seek(128), 129);
+            assert_eq!(docset.skip_next(128), SkipResult::OverStep);
             assert_eq!(docset.doc(), 129);
-            assert_eq!(docset.advance(), 130);
+            assert!(docset.advance());
             assert_eq!(docset.doc(), 130);
-            assert_eq!(docset.advance(), TERMINATED);
+            assert!(!docset.advance());
         }
         {
             let block_segments = build_block_postings(&doc_ids);
             let mut docset = SegmentPostings::from_block_postings(block_segments, None);
-            assert_eq!(docset.seek(129), 129);
+            assert_eq!(docset.skip_next(129), SkipResult::Reached);
             assert_eq!(docset.doc(), 129);
-            assert_eq!(docset.advance(), 130);
+            assert!(docset.advance());
             assert_eq!(docset.doc(), 130);
-            assert_eq!(docset.advance(), TERMINATED);
+            assert!(!docset.advance());
         }
         {
             let block_segments = build_block_postings(&doc_ids);
             let mut docset = SegmentPostings::from_block_postings(block_segments, None);
-            assert_eq!(docset.doc(), 0);
-            assert_eq!(docset.seek(131), TERMINATED);
-            assert_eq!(docset.doc(), TERMINATED);
+            assert_eq!(docset.skip_next(131), SkipResult::End);
         }
     }
 

src/postings/stacker/term_hashmap.rs

@@ -1,4 +1,6 @@
-use murmurhash32::murmurhash2;
+use murmurhash32;
+
+use self::murmurhash32::murmurhash2;
 
 use super::{Addr, MemoryArena};
 use crate::postings::stacker::memory_arena::store;

src/query/all_query.rs

@@ -1,7 +1,6 @@
 use crate::core::Searcher;
 use crate::core::SegmentReader;
-use crate::docset::{DocSet, TERMINATED};
-use crate::query::boost_query::BoostScorer;
+use crate::docset::DocSet;
 use crate::query::explanation::does_not_match;
 use crate::query::{Explanation, Query, Scorer, Weight};
 use crate::DocId;
@@ -23,12 +22,12 @@ impl Query for AllQuery {
 pub struct AllWeight;
 
 impl Weight for AllWeight {
-    fn scorer(&self, reader: &SegmentReader, boost: f32) -> crate::Result<Box<dyn Scorer>> {
-        let all_scorer = AllScorer {
+    fn scorer(&self, reader: &SegmentReader) -> crate::Result<Box<dyn Scorer>> {
+        Ok(Box::new(AllScorer {
+            state: State::NotStarted,
             doc: 0u32,
             max_doc: reader.max_doc(),
-        };
-        Ok(Box::new(BoostScorer::new(all_scorer, boost)))
+        }))
     }
 
     fn explain(&self, reader: &SegmentReader, doc: DocId) -> crate::Result<Explanation> {
@@ -39,20 +38,39 @@ impl Weight for AllWeight {
     }
 }
 
+enum State {
+    NotStarted,
+    Started,
+    Finished,
+}
+
 /// Scorer associated to the `AllQuery` query.
 pub struct AllScorer {
+    state: State,
     doc: DocId,
     max_doc: DocId,
 }
 
 impl DocSet for AllScorer {
-    fn advance(&mut self) -> DocId {
-        if self.doc + 1 >= self.max_doc {
-            self.doc = TERMINATED;
-            return TERMINATED;
+    fn advance(&mut self) -> bool {
+        match self.state {
+            State::NotStarted => {
+                self.state = State::Started;
+                self.doc = 0;
+            }
+            State::Started => {
+                self.doc += 1u32;
+            }
+            State::Finished => {
+                return false;
+            }
+        }
+        if self.doc < self.max_doc {
+            true
+        } else {
+            self.state = State::Finished;
+            false
         }
-        self.doc += 1;
-        self.doc
     }
 
     fn doc(&self) -> DocId {
@@ -72,13 +90,14 @@ impl Scorer for AllScorer {
 
 #[cfg(test)]
 mod tests {
+
     use super::AllQuery;
-    use crate::docset::TERMINATED;
     use crate::query::Query;
     use crate::schema::{Schema, TEXT};
     use crate::Index;
 
-    fn create_test_index() -> Index {
+    #[test]
+    fn test_all_query() {
         let mut schema_builder = Schema::builder();
         let field = schema_builder.add_text_field("text", TEXT);
         let schema = schema_builder.build();
@@ -89,47 +108,25 @@ mod tests {
         index_writer.commit().unwrap();
         index_writer.add_document(doc!(field=>"ccc"));
         index_writer.commit().unwrap();
-        index
-    }
-
-    #[test]
-    fn test_all_query() {
-        let index = create_test_index();
         let reader = index.reader().unwrap();
+        reader.reload().unwrap();
         let searcher = reader.searcher();
         let weight = AllQuery.weight(&searcher, false).unwrap();
         {
             let reader = searcher.segment_reader(0);
-            let mut scorer = weight.scorer(reader, 1.0f32).unwrap();
+            let mut scorer = weight.scorer(reader).unwrap();
+            assert!(scorer.advance());
             assert_eq!(scorer.doc(), 0u32);
-            assert_eq!(scorer.advance(), 1u32);
+            assert!(scorer.advance());
             assert_eq!(scorer.doc(), 1u32);
-            assert_eq!(scorer.advance(), TERMINATED);
+            assert!(!scorer.advance());
         }
         {
             let reader = searcher.segment_reader(1);
-            let mut scorer = weight.scorer(reader, 1.0f32).unwrap();
+            let mut scorer = weight.scorer(reader).unwrap();
+            assert!(scorer.advance());
             assert_eq!(scorer.doc(), 0u32);
-            assert_eq!(scorer.advance(), TERMINATED);
-        }
-    }
-
-    #[test]
-    fn test_all_query_with_boost() {
-        let index = create_test_index();
-        let reader = index.reader().unwrap();
-        let searcher = reader.searcher();
-        let weight = AllQuery.weight(&searcher, false).unwrap();
-        let reader = searcher.segment_reader(0);
-        {
-            let mut scorer = weight.scorer(reader, 2.0f32).unwrap();
-            assert_eq!(scorer.doc(), 0u32);
-            assert_eq!(scorer.score(), 2.0f32);
-        }
-        {
-            let mut scorer = weight.scorer(reader, 1.5f32).unwrap();
-            assert_eq!(scorer.doc(), 0u32);
-            assert_eq!(scorer.score(), 1.5f32);
+            assert!(!scorer.advance());
         }
     }
 }

src/query/automaton_weight.rs

@@ -6,8 +6,8 @@ use crate::query::{Scorer, Weight};
 use crate::schema::{Field, IndexRecordOption};
 use crate::termdict::{TermDictionary, TermStreamer};
 use crate::DocId;
-use crate::Result;
 use crate::TantivyError;
+use crate::{Result, SkipResult};
 use std::sync::Arc;
 use tantivy_fst::Automaton;
 
@@ -40,7 +40,7 @@ impl<A> Weight for AutomatonWeight<A>
 where
     A: Automaton + Send + Sync + 'static,
 {
-    fn scorer(&self, reader: &SegmentReader, boost: f32) -> Result<Box<dyn Scorer>> {
+    fn scorer(&self, reader: &SegmentReader) -> Result<Box<dyn Scorer>> {
         let max_doc = reader.max_doc();
         let mut doc_bitset = BitSet::with_max_value(max_doc);
 
@@ -58,13 +58,12 @@ where
             }
         }
         let doc_bitset = BitSetDocSet::from(doc_bitset);
-        let const_scorer = ConstScorer::new(doc_bitset, boost);
-        Ok(Box::new(const_scorer))
+        Ok(Box::new(ConstScorer::new(doc_bitset)))
     }
 
     fn explain(&self, reader: &SegmentReader, doc: DocId) -> Result<Explanation> {
-        let mut scorer = self.scorer(reader, 1.0f32)?;
-        if scorer.seek(doc) == doc {
+        let mut scorer = self.scorer(reader)?;
+        if scorer.skip_next(doc) == SkipResult::Reached {
             Ok(Explanation::new("AutomatonScorer", 1.0f32))
         } else {
             Err(TantivyError::InvalidArgument(
@@ -73,94 +72,3 @@ where
         }
     }
 }
-
-#[cfg(test)]
-mod tests {
-    use super::AutomatonWeight;
-    use crate::docset::TERMINATED;
-    use crate::query::Weight;
-    use crate::schema::{Schema, STRING};
-    use crate::Index;
-    use tantivy_fst::Automaton;
-
-    fn create_index() -> Index {
-        let mut schema = Schema::builder();
-        let title = schema.add_text_field("title", STRING);
-        let index = Index::create_in_ram(schema.build());
-        let mut index_writer = index.writer_with_num_threads(1, 3_000_000).unwrap();
-        index_writer.add_document(doc!(title=>"abc"));
-        index_writer.add_document(doc!(title=>"bcd"));
-        index_writer.add_document(doc!(title=>"abcd"));
-        assert!(index_writer.commit().is_ok());
-        index
-    }
-
-    enum State {
-        Start,
-        NotMatching,
-        AfterA,
-    }
-
-    struct PrefixedByA;
-
-    impl Automaton for PrefixedByA {
-        type State = State;
-
-        fn start(&self) -> Self::State {
-            State::Start
-        }
-
-        fn is_match(&self, state: &Self::State) -> bool {
-            match *state {
-                State::AfterA => true,
-                _ => false,
-            }
-        }
-
-        fn accept(&self, state: &Self::State, byte: u8) -> Self::State {
-            match *state {
-                State::Start => {
-                    if byte == b'a' {
-                        State::AfterA
-                    } else {
-                        State::NotMatching
-                    }
-                }
-                State::AfterA => State::AfterA,
-                State::NotMatching => State::NotMatching,
-            }
-        }
-    }
-
-    #[test]
-    fn test_automaton_weight() {
-        let index = create_index();
-        let field = index.schema().get_field("title").unwrap();
-        let automaton_weight = AutomatonWeight::new(field, PrefixedByA);
-        let reader = index.reader().unwrap();
-        let searcher = reader.searcher();
-        let mut scorer = automaton_weight
-            .scorer(searcher.segment_reader(0u32), 1.0f32)
-            .unwrap();
-        assert_eq!(scorer.doc(), 0u32);
-        assert_eq!(scorer.score(), 1.0f32);
-        assert_eq!(scorer.advance(), 2u32);
-        assert_eq!(scorer.doc(), 2u32);
-        assert_eq!(scorer.score(), 1.0f32);
-        assert_eq!(scorer.advance(), TERMINATED);
-    }
-
-    #[test]
-    fn test_automaton_weight_boost() {
-        let index = create_index();
-        let field = index.schema().get_field("title").unwrap();
-        let automaton_weight = AutomatonWeight::new(field, PrefixedByA);
-        let reader = index.reader().unwrap();
-        let searcher = reader.searcher();
-        let mut scorer = automaton_weight
-            .scorer(searcher.segment_reader(0u32), 1.32f32)
-            .unwrap();
-        assert_eq!(scorer.doc(), 0u32);
-        assert_eq!(scorer.score(), 1.32f32);
-    }
-}

src/query/bitset/mod.rs

@@ -1,6 +1,7 @@
 use crate::common::{BitSet, TinySet};
-use crate::docset::{DocSet, TERMINATED};
+use crate::docset::{DocSet, SkipResult};
 use crate::DocId;
+use std::cmp::Ordering;
 
 /// A `BitSetDocSet` makes it possible to iterate through a bitset as if it was a `DocSet`.
 ///
@@ -32,50 +33,75 @@ impl From<BitSet> for BitSetDocSet {
         } else {
             docs.tinyset(0)
         };
-        let mut docset = BitSetDocSet {
+        BitSetDocSet {
             docs,
             cursor_bucket: 0,
             cursor_tinybitset: first_tiny_bitset,
             doc: 0u32,
-        };
-        docset.advance();
-        docset
+        }
     }
 }
 
 impl DocSet for BitSetDocSet {
-    fn advance(&mut self) -> DocId {
+    fn advance(&mut self) -> bool {
         if let Some(lower) = self.cursor_tinybitset.pop_lowest() {
             self.doc = (self.cursor_bucket as u32 * 64u32) | lower;
-            return self.doc;
+            return true;
         }
         if let Some(cursor_bucket) = self.docs.first_non_empty_bucket(self.cursor_bucket + 1) {
             self.go_to_bucket(cursor_bucket);
             let lower = self.cursor_tinybitset.pop_lowest().unwrap();
             self.doc = (cursor_bucket * 64u32) | lower;
-            self.doc
+            true
         } else {
-            self.doc = TERMINATED;
-            TERMINATED
+            false
         }
     }
 
-    fn seek(&mut self, target: DocId) -> DocId {
+    fn skip_next(&mut self, target: DocId) -> SkipResult {
+        // skip is required to advance.
+        if !self.advance() {
+            return SkipResult::End;
+        }
         let target_bucket = target / 64u32;
 
         // Mask for all of the bits greater or equal
         // to our target document.
-        if target_bucket > self.cursor_bucket {
-            self.go_to_bucket(target_bucket);
-            let greater_filter: TinySet = TinySet::range_greater_or_equal(target);
-            self.cursor_tinybitset = self.cursor_tinybitset.intersect(greater_filter);
-            self.advance();
+        match target_bucket.cmp(&self.cursor_bucket) {
+            Ordering::Greater => {
+                self.go_to_bucket(target_bucket);
+                let greater_filter: TinySet = TinySet::range_greater_or_equal(target);
+                self.cursor_tinybitset = self.cursor_tinybitset.intersect(greater_filter);
+                if !self.advance() {
+                    SkipResult::End
+                } else if self.doc() == target {
+                    SkipResult::Reached
+                } else {
+                    debug_assert!(self.doc() > target);
+                    SkipResult::OverStep
+                }
+            }
+            Ordering::Equal => loop {
+                match self.doc().cmp(&target) {
+                    Ordering::Less => {
+                        if !self.advance() {
+                            return SkipResult::End;
+                        }
+                    }
+                    Ordering::Equal => {
+                        return SkipResult::Reached;
+                    }
+                    Ordering::Greater => {
+                        debug_assert!(self.doc() > target);
+                        return SkipResult::OverStep;
+                    }
+                }
+            },
+            Ordering::Less => {
+                debug_assert!(self.doc() > target);
+                SkipResult::OverStep
+            }
         }
-        let mut doc = self.doc();
-        while doc < target {
-            doc = self.advance();
-        }
-        doc
     }
 
     /// Returns the current document
@@ -96,7 +122,7 @@ impl DocSet for BitSetDocSet {
 mod tests {
     use super::BitSetDocSet;
     use crate::common::BitSet;
-    use crate::docset::{DocSet, TERMINATED};
+    use crate::docset::{DocSet, SkipResult};
     use crate::DocId;
 
     fn create_docbitset(docs: &[DocId], max_doc: DocId) -> BitSetDocSet {
@@ -107,24 +133,19 @@ mod tests {
         BitSetDocSet::from(docset)
     }
 
-    #[test]
-    fn test_empty() {
-        let bitset = BitSet::with_max_value(1000);
-        let mut empty = BitSetDocSet::from(bitset);
-        assert_eq!(empty.advance(), TERMINATED)
-    }
-
     fn test_go_through_sequential(docs: &[DocId]) {
         let mut docset = create_docbitset(docs, 1_000u32);
         for &doc in docs {
+            assert!(docset.advance());
             assert_eq!(doc, docset.doc());
-            docset.advance();
         }
-        assert_eq!(docset.advance(), TERMINATED);
+        assert!(!docset.advance());
+        assert!(!docset.advance());
     }
 
     #[test]
     fn test_docbitset_sequential() {
+        test_go_through_sequential(&[]);
         test_go_through_sequential(&[1, 2, 3]);
         test_go_through_sequential(&[1, 2, 3, 4, 5, 63, 64, 65]);
         test_go_through_sequential(&[63, 64, 65]);
@@ -135,64 +156,64 @@ mod tests {
     fn test_docbitset_skip() {
         {
             let mut docset = create_docbitset(&[1, 5, 6, 7, 5112], 10_000);
-            assert_eq!(docset.seek(7), 7);
+            assert_eq!(docset.skip_next(7), SkipResult::Reached);
             assert_eq!(docset.doc(), 7);
-            assert_eq!(docset.advance(), 5112);
+            assert!(docset.advance(), 7);
             assert_eq!(docset.doc(), 5112);
-            assert_eq!(docset.advance(), TERMINATED);
+            assert!(!docset.advance());
         }
         {
             let mut docset = create_docbitset(&[1, 5, 6, 7, 5112], 10_000);
-            assert_eq!(docset.seek(3), 5);
+            assert_eq!(docset.skip_next(3), SkipResult::OverStep);
             assert_eq!(docset.doc(), 5);
-            assert_eq!(docset.advance(), 6);
+            assert!(docset.advance());
         }
         {
             let mut docset = create_docbitset(&[5112], 10_000);
-            assert_eq!(docset.seek(5112), 5112);
+            assert_eq!(docset.skip_next(5112), SkipResult::Reached);
             assert_eq!(docset.doc(), 5112);
-            assert_eq!(docset.advance(), TERMINATED);
+            assert!(!docset.advance());
         }
         {
             let mut docset = create_docbitset(&[5112], 10_000);
-            assert_eq!(docset.seek(5113), TERMINATED);
-            assert_eq!(docset.advance(), TERMINATED);
+            assert_eq!(docset.skip_next(5113), SkipResult::End);
+            assert!(!docset.advance());
         }
         {
             let mut docset = create_docbitset(&[5112], 10_000);
-            assert_eq!(docset.seek(5111), 5112);
+            assert_eq!(docset.skip_next(5111), SkipResult::OverStep);
             assert_eq!(docset.doc(), 5112);
-            assert_eq!(docset.advance(), TERMINATED);
+            assert!(!docset.advance());
         }
         {
             let mut docset = create_docbitset(&[1, 5, 6, 7, 5112, 5500, 6666], 10_000);
-            assert_eq!(docset.seek(5112), 5112);
+            assert_eq!(docset.skip_next(5112), SkipResult::Reached);
             assert_eq!(docset.doc(), 5112);
-            assert_eq!(docset.advance(), 5500);
+            assert!(docset.advance());
             assert_eq!(docset.doc(), 5500);
-            assert_eq!(docset.advance(), 6666);
+            assert!(docset.advance());
             assert_eq!(docset.doc(), 6666);
-            assert_eq!(docset.advance(), TERMINATED);
+            assert!(!docset.advance());
         }
         {
             let mut docset = create_docbitset(&[1, 5, 6, 7, 5112, 5500, 6666], 10_000);
-            assert_eq!(docset.seek(5111), 5112);
+            assert_eq!(docset.skip_next(5111), SkipResult::OverStep);
             assert_eq!(docset.doc(), 5112);
-            assert_eq!(docset.advance(), 5500);
+            assert!(docset.advance());
             assert_eq!(docset.doc(), 5500);
-            assert_eq!(docset.advance(), 6666);
+            assert!(docset.advance());
             assert_eq!(docset.doc(), 6666);
-            assert_eq!(docset.advance(), TERMINATED);
+            assert!(!docset.advance());
         }
         {
             let mut docset = create_docbitset(&[1, 5, 6, 7, 5112, 5513, 6666], 10_000);
-            assert_eq!(docset.seek(5111), 5112);
+            assert_eq!(docset.skip_next(5111), SkipResult::OverStep);
             assert_eq!(docset.doc(), 5112);
-            assert_eq!(docset.advance(), 5513);
+            assert!(docset.advance());
             assert_eq!(docset.doc(), 5513);
-            assert_eq!(docset.advance(), 6666);
+            assert!(docset.advance());
             assert_eq!(docset.doc(), 6666);
-            assert_eq!(docset.advance(), TERMINATED);
+            assert!(!docset.advance());
         }
     }
 }
@@ -202,7 +223,6 @@ mod bench {
 
     use super::BitSet;
     use super::BitSetDocSet;
-    use crate::docset::TERMINATED;
     use crate::test;
     use crate::tests;
     use crate::DocSet;
@@ -237,7 +257,7 @@ mod bench {
         }
         b.iter(|| {
             let mut docset = BitSetDocSet::from(bitset.clone());
-            while docset.advance() != TERMINATED {}
+            while docset.advance() {}
         });
     }
 }

src/query/bm25.rs

@@ -25,6 +25,7 @@ fn compute_tf_cache(average_fieldnorm: f32) -> [f32; 256] {
     cache
 }
 
+#[derive(Clone)]
 pub struct BM25Weight {
     idf_explain: Explanation,
     weight: f32,
@@ -33,15 +34,6 @@ pub struct BM25Weight {
 }
 
 impl BM25Weight {
-    pub fn boost_by(&self, boost: f32) -> BM25Weight {
-        BM25Weight {
-            idf_explain: self.idf_explain.clone(),
-            weight: self.weight * boost,
-            cache: self.cache,
-            average_fieldnorm: self.average_fieldnorm,
-        }
-    }
-
     pub fn for_terms(searcher: &Searcher, terms: &[Term]) -> BM25Weight {
         assert!(!terms.is_empty(), "BM25 requires at least one term");
         let field = terms[0].field();

src/query/boolean_query/boolean_weight.rs

@@ -10,7 +10,7 @@ use crate::query::Scorer;
 use crate::query::Union;
 use crate::query::Weight;
 use crate::query::{intersect_scorers, Explanation};
-use crate::DocId;
+use crate::{DocId, SkipResult};
 use std::collections::HashMap;
 
 fn scorer_union<TScoreCombiner>(scorers: Vec<Box<dyn Scorer>>) -> Box<dyn Scorer>
@@ -55,11 +55,10 @@ impl BooleanWeight {
     fn per_occur_scorers(
         &self,
         reader: &SegmentReader,
-        boost: f32,
     ) -> crate::Result<HashMap<Occur, Vec<Box<dyn Scorer>>>> {
         let mut per_occur_scorers: HashMap<Occur, Vec<Box<dyn Scorer>>> = HashMap::new();
         for &(ref occur, ref subweight) in &self.weights {
-            let sub_scorer: Box<dyn Scorer> = subweight.scorer(reader, boost)?;
+            let sub_scorer: Box<dyn Scorer> = subweight.scorer(reader)?;
             per_occur_scorers
                 .entry(*occur)
                 .or_insert_with(Vec::new)
@@ -71,9 +70,8 @@ impl BooleanWeight {
     fn complex_scorer<TScoreCombiner: ScoreCombiner>(
         &self,
         reader: &SegmentReader,
-        boost: f32,
     ) -> crate::Result<Box<dyn Scorer>> {
-        let mut per_occur_scorers = self.per_occur_scorers(reader, boost)?;
+        let mut per_occur_scorers = self.per_occur_scorers(reader)?;
 
         let should_scorer_opt: Option<Box<dyn Scorer>> = per_occur_scorers
             .remove(&Occur::Should)
@@ -114,7 +112,7 @@ impl BooleanWeight {
 }
 
 impl Weight for BooleanWeight {
-    fn scorer(&self, reader: &SegmentReader, boost: f32) -> crate::Result<Box<dyn Scorer>> {
+    fn scorer(&self, reader: &SegmentReader) -> crate::Result<Box<dyn Scorer>> {
         if self.weights.is_empty() {
             Ok(Box::new(EmptyScorer))
         } else if self.weights.len() == 1 {
@@ -122,18 +120,18 @@ impl Weight for BooleanWeight {
             if occur == Occur::MustNot {
                 Ok(Box::new(EmptyScorer))
             } else {
-                weight.scorer(reader, boost)
+                weight.scorer(reader)
             }
         } else if self.scoring_enabled {
-            self.complex_scorer::<SumWithCoordsCombiner>(reader, boost)
+            self.complex_scorer::<SumWithCoordsCombiner>(reader)
         } else {
-            self.complex_scorer::<DoNothingCombiner>(reader, boost)
+            self.complex_scorer::<DoNothingCombiner>(reader)
         }
     }
 
     fn explain(&self, reader: &SegmentReader, doc: DocId) -> crate::Result<Explanation> {
-        let mut scorer = self.scorer(reader, 1.0f32)?;
-        if scorer.seek(doc) != doc {
+        let mut scorer = self.scorer(reader)?;
+        if scorer.skip_next(doc) != SkipResult::Reached {
             return Err(does_not_match(doc));
         }
         if !self.scoring_enabled {

src/query/boolean_query/mod.rs

@@ -18,7 +18,6 @@ mod tests {
     use crate::query::Scorer;
     use crate::query::TermQuery;
     use crate::schema::*;
-    use crate::tests::assert_nearly_equals;
     use crate::Index;
     use crate::{DocAddress, DocId};
 
@@ -31,11 +30,24 @@ mod tests {
             // writing the segment
             let mut index_writer = index.writer_with_num_threads(1, 3_000_000).unwrap();
             {
-                index_writer.add_document(doc!(text_field => "a b c"));
-                index_writer.add_document(doc!(text_field => "a c"));
-                index_writer.add_document(doc!(text_field => "b c"));
-                index_writer.add_document(doc!(text_field => "a b c d"));
-                index_writer.add_document(doc!(text_field => "d"));
+                let doc = doc!(text_field => "a b c");
+                index_writer.add_document(doc);
+            }
+            {
+                let doc = doc!(text_field => "a c");
+                index_writer.add_document(doc);
+            }
+            {
+                let doc = doc!(text_field => "b c");
+                index_writer.add_document(doc);
+            }
+            {
+                let doc = doc!(text_field => "a b c d");
+                index_writer.add_document(doc);
+            }
+            {
+                let doc = doc!(text_field => "d");
+                index_writer.add_document(doc);
             }
             assert!(index_writer.commit().is_ok());
         }
@@ -58,9 +70,7 @@ mod tests {
         let query = query_parser.parse_query("+a").unwrap();
         let searcher = index.reader().unwrap().searcher();
         let weight = query.weight(&searcher, true).unwrap();
-        let scorer = weight
-            .scorer(searcher.segment_reader(0u32), 1.0f32)
-            .unwrap();
+        let scorer = weight.scorer(searcher.segment_reader(0u32)).unwrap();
         assert!(scorer.is::<TermScorer>());
     }
 
@@ -72,17 +82,13 @@ mod tests {
         {
             let query = query_parser.parse_query("+a +b +c").unwrap();
             let weight = query.weight(&searcher, true).unwrap();
-            let scorer = weight
-                .scorer(searcher.segment_reader(0u32), 1.0f32)
-                .unwrap();
+            let scorer = weight.scorer(searcher.segment_reader(0u32)).unwrap();
             assert!(scorer.is::<Intersection<TermScorer>>());
         }
         {
             let query = query_parser.parse_query("+a +(b c)").unwrap();
             let weight = query.weight(&searcher, true).unwrap();
-            let scorer = weight
-                .scorer(searcher.segment_reader(0u32), 1.0f32)
-                .unwrap();
+            let scorer = weight.scorer(searcher.segment_reader(0u32)).unwrap();
             assert!(scorer.is::<Intersection<Box<dyn Scorer>>>());
         }
     }
@@ -95,9 +101,7 @@ mod tests {
         {
             let query = query_parser.parse_query("+a b").unwrap();
             let weight = query.weight(&searcher, true).unwrap();
-            let scorer = weight
-                .scorer(searcher.segment_reader(0u32), 1.0f32)
-                .unwrap();
+            let scorer = weight.scorer(searcher.segment_reader(0u32)).unwrap();
             assert!(scorer.is::<RequiredOptionalScorer<
                 Box<dyn Scorer>,
                 Box<dyn Scorer>,
@@ -107,9 +111,7 @@ mod tests {
         {
             let query = query_parser.parse_query("+a b").unwrap();
             let weight = query.weight(&searcher, false).unwrap();
-            let scorer = weight
-                .scorer(searcher.segment_reader(0u32), 1.0f32)
-                .unwrap();
+            let scorer = weight.scorer(searcher.segment_reader(0u32)).unwrap();
             assert!(scorer.is::<TermScorer>());
         }
     }
@@ -177,48 +179,6 @@ mod tests {
         }
     }
 
-    #[test]
-    pub fn test_boolean_query_with_weight() {
-        let mut schema_builder = Schema::builder();
-        let text_field = schema_builder.add_text_field("text", TEXT);
-        let schema = schema_builder.build();
-        let index = Index::create_in_ram(schema);
-        {
-            let mut index_writer = index.writer_with_num_threads(1, 3_000_000).unwrap();
-            index_writer.add_document(doc!(text_field => "a b c"));
-            index_writer.add_document(doc!(text_field => "a c"));
-            index_writer.add_document(doc!(text_field => "b c"));
-            assert!(index_writer.commit().is_ok());
-        }
-        let term_a: Box<dyn Query> = Box::new(TermQuery::new(
-            Term::from_field_text(text_field, "a"),
-            IndexRecordOption::WithFreqs,
-        ));
-        let term_b: Box<dyn Query> = Box::new(TermQuery::new(
-            Term::from_field_text(text_field, "b"),
-            IndexRecordOption::WithFreqs,
-        ));
-        let reader = index.reader().unwrap();
-        let searcher = reader.searcher();
-        let boolean_query =
-            BooleanQuery::from(vec![(Occur::Should, term_a), (Occur::Should, term_b)]);
-        let boolean_weight = boolean_query.weight(&searcher, true).unwrap();
-        {
-            let mut boolean_scorer = boolean_weight
-                .scorer(searcher.segment_reader(0u32), 1.0f32)
-                .unwrap();
-            assert_eq!(boolean_scorer.doc(), 0u32);
-            assert_nearly_equals(boolean_scorer.score(), 0.84163445f32);
-        }
-        {
-            let mut boolean_scorer = boolean_weight
-                .scorer(searcher.segment_reader(0u32), 2.0f32)
-                .unwrap();
-            assert_eq!(boolean_scorer.doc(), 0u32);
-            assert_nearly_equals(boolean_scorer.score(), 1.6832689f32);
-        }
-    }
-
     #[test]
     pub fn test_intersection_score() {
         let (index, text_field) = aux_test_helper();
@@ -289,9 +249,7 @@ mod tests {
         let query_parser = QueryParser::for_index(&index, vec![title, text]);
         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), 1.0f32)
-            .unwrap();
+        let mut scorer = weight.scorer(searcher.segment_reader(0u32)).unwrap();
         scorer.advance();
 
         let explanation = query.explain(&searcher, DocAddress(0u32, 0u32)).unwrap();

src/query/boost_query.rs

@@ -1,159 +0,0 @@
-use crate::fastfield::DeleteBitSet;
-use crate::query::explanation::does_not_match;
-use crate::query::{Explanation, Query, Scorer, Weight};
-use crate::{DocId, DocSet, Searcher, SegmentReader, Term};
-use std::collections::BTreeSet;
-use std::fmt;
-
-/// `BoostQuery` is a wrapper over a query used to boost its score.
-///
-/// The document set matched by the `BoostQuery` is strictly the same as the underlying query.
-/// The score of each document, is the score of the underlying query multiplied by the `boost`
-/// factor.
-pub struct BoostQuery {
-    query: Box<dyn Query>,
-    boost: f32,
-}
-
-impl BoostQuery {
-    /// Builds a boost query.
-    pub fn new(query: Box<dyn Query>, boost: f32) -> BoostQuery {
-        BoostQuery { query, boost }
-    }
-}
-
-impl Clone for BoostQuery {
-    fn clone(&self) -> Self {
-        BoostQuery {
-            query: self.query.box_clone(),
-            boost: self.boost,
-        }
-    }
-}
-
-impl fmt::Debug for BoostQuery {
-    fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
-        write!(f, "Boost(query={:?}, boost={})", self.query, self.boost)
-    }
-}
-
-impl Query for BoostQuery {
-    fn weight(&self, searcher: &Searcher, scoring_enabled: bool) -> crate::Result<Box<dyn Weight>> {
-        let weight_without_boost = self.query.weight(searcher, scoring_enabled)?;
-        let boosted_weight = if scoring_enabled {
-            Box::new(BoostWeight::new(weight_without_boost, self.boost))
-        } else {
-            weight_without_boost
-        };
-        Ok(boosted_weight)
-    }
-
-    fn query_terms(&self, term_set: &mut BTreeSet<Term>) {
-        self.query.query_terms(term_set)
-    }
-}
-
-pub(crate) struct BoostWeight {
-    weight: Box<dyn Weight>,
-    boost: f32,
-}
-
-impl BoostWeight {
-    pub fn new(weight: Box<dyn Weight>, boost: f32) -> Self {
-        BoostWeight { weight, boost }
-    }
-}
-
-impl Weight for BoostWeight {
-    fn scorer(&self, reader: &SegmentReader, boost: f32) -> crate::Result<Box<dyn Scorer>> {
-        self.weight.scorer(reader, boost * self.boost)
-    }
-
-    fn explain(&self, reader: &SegmentReader, doc: u32) -> crate::Result<Explanation> {
-        let mut scorer = self.scorer(reader, 1.0f32)?;
-        if scorer.seek(doc) != doc {
-            return Err(does_not_match(doc));
-        }
-        let mut explanation =
-            Explanation::new(format!("Boost x{} of ...", self.boost), scorer.score());
-        let underlying_explanation = self.weight.explain(reader, doc)?;
-        explanation.add_detail(underlying_explanation);
-        Ok(explanation)
-    }
-
-    fn count(&self, reader: &SegmentReader) -> crate::Result<u32> {
-        self.weight.count(reader)
-    }
-}
-
-pub(crate) struct BoostScorer<S: Scorer> {
-    underlying: S,
-    boost: f32,
-}
-
-impl<S: Scorer> BoostScorer<S> {
-    pub fn new(underlying: S, boost: f32) -> BoostScorer<S> {
-        BoostScorer { underlying, boost }
-    }
-}
-
-impl<S: Scorer> DocSet for BoostScorer<S> {
-    fn advance(&mut self) -> DocId {
-        self.underlying.advance()
-    }
-
-    fn seek(&mut self, target: DocId) -> DocId {
-        self.underlying.seek(target)
-    }
-
-    fn fill_buffer(&mut self, buffer: &mut [DocId]) -> usize {
-        self.underlying.fill_buffer(buffer)
-    }
-
-    fn doc(&self) -> u32 {
-        self.underlying.doc()
-    }
-
-    fn size_hint(&self) -> u32 {
-        self.underlying.size_hint()
-    }
-
-    fn count(&mut self, delete_bitset: &DeleteBitSet) -> u32 {
-        self.underlying.count(delete_bitset)
-    }
-
-    fn count_including_deleted(&mut self) -> u32 {
-        self.underlying.count_including_deleted()
-    }
-}
-
-impl<S: Scorer> Scorer for BoostScorer<S> {
-    fn score(&mut self) -> f32 {
-        self.underlying.score() * self.boost
-    }
-}
-
-#[cfg(test)]
-mod tests {
-    use super::BoostQuery;
-    use crate::query::{AllQuery, Query};
-    use crate::schema::Schema;
-    use crate::{DocAddress, Document, Index};
-
-    #[test]
-    fn test_boost_query_explain() {
-        let schema = Schema::builder().build();
-        let index = Index::create_in_ram(schema);
-        let mut index_writer = index.writer_with_num_threads(1, 3_000_000).unwrap();
-        index_writer.add_document(Document::new());
-        assert!(index_writer.commit().is_ok());
-        let reader = index.reader().unwrap();
-        let searcher = reader.searcher();
-        let query = BoostQuery::new(Box::new(AllQuery), 0.2);
-        let explanation = query.explain(&searcher, DocAddress(0, 0u32)).unwrap();
-        assert_eq!(
-            explanation.to_pretty_json(),
-            "{\n  \"value\": 0.2,\n  \"description\": \"Boost x0.2 of ...\",\n  \"details\": [\n    {\n      \"value\": 1.0,\n      \"description\": \"AllQuery\"\n    }\n  ]\n}"
-        )
-    }
-}

src/query/empty_query.rs

@@ -1,5 +1,4 @@
 use super::Scorer;
-use crate::docset::TERMINATED;
 use crate::query::explanation::does_not_match;
 use crate::query::Weight;
 use crate::query::{Explanation, Query};
@@ -34,7 +33,7 @@ impl Query for EmptyQuery {
 /// It is useful for tests and handling edge cases.
 pub struct EmptyWeight;
 impl Weight for EmptyWeight {
-    fn scorer(&self, _reader: &SegmentReader, _boost: f32) -> crate::Result<Box<dyn Scorer>> {
+    fn scorer(&self, _reader: &SegmentReader) -> crate::Result<Box<dyn Scorer>> {
         Ok(Box::new(EmptyScorer))
     }
 
@@ -49,12 +48,15 @@ impl Weight for EmptyWeight {
 pub struct EmptyScorer;
 
 impl DocSet for EmptyScorer {
-    fn advance(&mut self) -> DocId {
-        TERMINATED
+    fn advance(&mut self) -> bool {
+        false
     }
 
     fn doc(&self) -> DocId {
-        TERMINATED
+        panic!(
+            "You may not call .doc() on a scorer \
+             where the last call to advance() did not return true."
+        );
     }
 
     fn size_hint(&self) -> u32 {
@@ -70,15 +72,18 @@ impl Scorer for EmptyScorer {
 
 #[cfg(test)]
 mod tests {
-    use crate::docset::TERMINATED;
     use crate::query::EmptyScorer;
     use crate::DocSet;
 
     #[test]
     fn test_empty_scorer() {
         let mut empty_scorer = EmptyScorer;
-        assert_eq!(empty_scorer.doc(), TERMINATED);
-        assert_eq!(empty_scorer.advance(), TERMINATED);
-        assert_eq!(empty_scorer.doc(), TERMINATED);
+        assert!(!empty_scorer.advance());
+    }
+
+    #[test]
+    #[should_panic]
+    fn test_empty_scorer_panic_on_doc_call() {
+        EmptyScorer.doc();
     }
 }

src/query/exclude.rs

@@ -1,37 +1,41 @@
-use crate::docset::{DocSet, TERMINATED};
+use crate::docset::{DocSet, SkipResult};
 use crate::query::Scorer;
 use crate::DocId;
 use crate::Score;
 
+#[derive(Clone, Copy, Debug)]
+enum State {
+    ExcludeOne(DocId),
+    Finished,
+}
+
 /// Filters a given `DocSet` by removing the docs from a given `DocSet`.
 ///
 /// The excluding docset has no impact on scoring.
 pub struct Exclude<TDocSet, TDocSetExclude> {
     underlying_docset: TDocSet,
     excluding_docset: TDocSetExclude,
+    excluding_state: State,
 }
 
 impl<TDocSet, TDocSetExclude> Exclude<TDocSet, TDocSetExclude>
 where
-    TDocSet: DocSet,
     TDocSetExclude: DocSet,
 {
     /// Creates a new `ExcludeScorer`
     pub fn new(
-        mut underlying_docset: TDocSet,
+        underlying_docset: TDocSet,
         mut excluding_docset: TDocSetExclude,
     ) -> Exclude<TDocSet, TDocSetExclude> {
-        while underlying_docset.doc() != TERMINATED {
-            let target = underlying_docset.doc();
-            if excluding_docset.seek(target) != target {
-                // this document is not excluded.
-                break;
-            }
-            underlying_docset.advance();
-        }
+        let state = if excluding_docset.advance() {
+            State::ExcludeOne(excluding_docset.doc())
+        } else {
+            State::Finished
+        };
         Exclude {
             underlying_docset,
             excluding_docset,
+            excluding_state: state,
         }
     }
 }
@@ -47,7 +51,28 @@ where
     /// increasing `doc`.
     fn accept(&mut self) -> bool {
         let doc = self.underlying_docset.doc();
-        self.excluding_docset.seek(doc) != doc
+        match self.excluding_state {
+            State::ExcludeOne(excluded_doc) => {
+                if doc == excluded_doc {
+                    return false;
+                }
+                if excluded_doc > doc {
+                    return true;
+                }
+                match self.excluding_docset.skip_next(doc) {
+                    SkipResult::OverStep => {
+                        self.excluding_state = State::ExcludeOne(self.excluding_docset.doc());
+                        true
+                    }
+                    SkipResult::End => {
+                        self.excluding_state = State::Finished;
+                        true
+                    }
+                    SkipResult::Reached => false,
+                }
+            }
+            State::Finished => true,
+        }
     }
 }
 
@@ -56,24 +81,27 @@ where
     TDocSet: DocSet,
     TDocSetExclude: DocSet,
 {
-    fn advance(&mut self) -> DocId {
-        while self.underlying_docset.advance() != TERMINATED {
+    fn advance(&mut self) -> bool {
+        while self.underlying_docset.advance() {
             if self.accept() {
-                return self.doc();
+                return true;
             }
         }
-        TERMINATED
+        false
     }
 
-    fn seek(&mut self, target: DocId) -> DocId {
-        let underlying_seek_result = self.underlying_docset.seek(target);
-        if underlying_seek_result == TERMINATED {
-            return TERMINATED;
+    fn skip_next(&mut self, target: DocId) -> SkipResult {
+        let underlying_skip_result = self.underlying_docset.skip_next(target);
+        if underlying_skip_result == SkipResult::End {
+            return SkipResult::End;
         }
         if self.accept() {
-            return underlying_seek_result;
+            underlying_skip_result
+        } else if self.advance() {
+            SkipResult::OverStep
+        } else {
+            SkipResult::End
         }
-        self.advance()
     }
 
     fn doc(&self) -> DocId {
@@ -113,9 +141,8 @@ mod tests {
             VecDocSet::from(vec![1, 2, 3, 10, 16, 24]),
         );
         let mut els = vec![];
-        while exclude_scorer.doc() != TERMINATED {
+        while exclude_scorer.advance() {
             els.push(exclude_scorer.doc());
-            exclude_scorer.advance();
         }
         assert_eq!(els, vec![5, 8, 15]);
     }

src/query/explanation.rs

@@ -1,5 +1,4 @@
 use crate::{DocId, TantivyError};
-use serde::Serialize;
 
 pub(crate) fn does_not_match(doc: DocId) -> TantivyError {
     TantivyError::InvalidArgument(format!("Document #({}) does not match", doc))

src/query/fuzzy_query.rs

@@ -2,40 +2,14 @@ use crate::query::{AutomatonWeight, Query, Weight};
 use crate::schema::Term;
 use crate::Searcher;
 use crate::TantivyError::InvalidArgument;
-use levenshtein_automata::{Distance, LevenshteinAutomatonBuilder, DFA};
+use levenshtein_automata::{LevenshteinAutomatonBuilder, DFA};
 use once_cell::sync::Lazy;
 use std::collections::HashMap;
 use std::ops::Range;
-use tantivy_fst::Automaton;
-
-pub(crate) struct DFAWrapper(pub DFA);
-
-impl Automaton for DFAWrapper {
-    type State = u32;
-
-    fn start(&self) -> Self::State {
-        self.0.initial_state()
-    }
-
-    fn is_match(&self, state: &Self::State) -> bool {
-        match self.0.distance(*state) {
-            Distance::Exact(_) => true,
-            Distance::AtLeast(_) => false,
-        }
-    }
-
-    fn can_match(&self, state: &u32) -> bool {
-        *state != levenshtein_automata::SINK_STATE
-    }
-
-    fn accept(&self, state: &Self::State, byte: u8) -> Self::State {
-        self.0.transition(*state, byte)
-    }
-}
 
 /// A range of Levenshtein distances that we will build DFAs for our terms
 /// The computation is exponential, so best keep it to low single digits
-const VALID_LEVENSHTEIN_DISTANCE_RANGE: Range<u8> = 0..3;
+const VALID_LEVENSHTEIN_DISTANCE_RANGE: Range<u8> = (0..3);
 
 static LEV_BUILDER: Lazy<HashMap<(u8, bool), LevenshteinAutomatonBuilder>> = Lazy::new(|| {
     let mut lev_builder_cache = HashMap::new();
@@ -117,7 +91,7 @@ impl FuzzyTermQuery {
         }
     }
 
-    /// Creates a new Fuzzy Query of the Term prefix
+    /// Creates a new Fuzzy Query that treats transpositions as cost one rather than two
     pub fn new_prefix(term: Term, distance: u8, transposition_cost_one: bool) -> FuzzyTermQuery {
         FuzzyTermQuery {
             term,
@@ -127,20 +101,13 @@ impl FuzzyTermQuery {
         }
     }
 
-    fn specialized_weight(&self) -> crate::Result<AutomatonWeight<DFAWrapper>> {
+    fn specialized_weight(&self) -> crate::Result<AutomatonWeight<DFA>> {
         // LEV_BUILDER is a HashMap, whose `get` method returns an Option
         match LEV_BUILDER.get(&(self.distance, false)) {
             // Unwrap the option and build the Ok(AutomatonWeight)
             Some(automaton_builder) => {
-                let automaton = if self.prefix {
-                    automaton_builder.build_prefix_dfa(self.term.text())
-                } else {
-                    automaton_builder.build_dfa(self.term.text())
-                };
-                Ok(AutomatonWeight::new(
-                    self.term.field(),
-                    DFAWrapper(automaton),
-                ))
+                let automaton = automaton_builder.build_dfa(self.term.text());
+                Ok(AutomatonWeight::new(self.term.field(), automaton))
             }
             None => Err(InvalidArgument(format!(
                 "Levenshtein distance of {} is not allowed. Choose a value in the {:?} range",
@@ -188,8 +155,6 @@ mod test {
         }
         let reader = index.reader().unwrap();
         let searcher = reader.searcher();
-
-        // passes because Levenshtein distance is 1 (substitute 'o' with 'a')
         {
             let term = Term::from_field_text(country_field, "japon");
 
@@ -201,29 +166,5 @@ mod test {
             let (score, _) = top_docs[0];
             assert_nearly_equals(1f32, score);
         }
-
-        // fails because non-prefix Levenshtein distance is more than 1 (add 'a' and 'n')
-        {
-            let term = Term::from_field_text(country_field, "jap");
-
-            let fuzzy_query = FuzzyTermQuery::new(term, 1, true);
-            let top_docs = searcher
-                .search(&fuzzy_query, &TopDocs::with_limit(2))
-                .unwrap();
-            assert_eq!(top_docs.len(), 0, "Expected no document");
-        }
-
-        // passes because prefix Levenshtein distance is 0
-        {
-            let term = Term::from_field_text(country_field, "jap");
-
-            let fuzzy_query = FuzzyTermQuery::new_prefix(term, 1, true);
-            let top_docs = searcher
-                .search(&fuzzy_query, &TopDocs::with_limit(2))
-                .unwrap();
-            assert_eq!(top_docs.len(), 1, "Expected only 1 document");
-            let (score, _) = top_docs[0];
-            assert_nearly_equals(1f32, score);
-        }
     }
 }

src/query/intersection.rs

@@ -1,4 +1,4 @@
-use crate::docset::{DocSet, TERMINATED};
+use crate::docset::{DocSet, SkipResult};
 use crate::query::term_query::TermScorer;
 use crate::query::EmptyScorer;
 use crate::query::Scorer;
@@ -20,14 +20,12 @@ pub fn intersect_scorers(mut scorers: Vec<Box<dyn Scorer>>) -> Box<dyn Scorer> {
     if scorers.len() == 1 {
         return scorers.pop().unwrap();
     }
-    scorers.sort_by_key(|scorer| scorer.size_hint());
-    let doc = go_to_first_doc(&mut scorers[..]);
-    if doc == TERMINATED {
-        return Box::new(EmptyScorer);
-    }
     // We know that we have at least 2 elements.
-    let left = scorers.remove(0);
-    let right = scorers.remove(0);
+    let num_docsets = scorers.len();
+    scorers.sort_by(|left, right| right.size_hint().cmp(&left.size_hint()));
+    let left = scorers.pop().unwrap();
+    let right = scorers.pop().unwrap();
+    scorers.reverse();
     let all_term_scorers = [&left, &right]
         .iter()
         .all(|&scorer| scorer.is::<TermScorer>());
@@ -36,12 +34,14 @@ pub fn intersect_scorers(mut scorers: Vec<Box<dyn Scorer>>) -> Box<dyn Scorer> {
             left: *(left.downcast::<TermScorer>().map_err(|_| ()).unwrap()),
             right: *(right.downcast::<TermScorer>().map_err(|_| ()).unwrap()),
             others: scorers,
+            num_docsets,
         });
     }
     Box::new(Intersection {
         left,
         right,
         others: scorers,
+        num_docsets,
     })
 }
 
@@ -50,34 +50,22 @@ pub struct Intersection<TDocSet: DocSet, TOtherDocSet: DocSet = Box<dyn Scorer>>
     left: TDocSet,
     right: TDocSet,
     others: Vec<TOtherDocSet>,
-}
-
-fn go_to_first_doc<TDocSet: DocSet>(docsets: &mut [TDocSet]) -> DocId {
-    let mut candidate = 0;
-    'outer: loop {
-        for docset in docsets.iter_mut() {
-            let seek_doc = docset.seek(candidate);
-            if seek_doc > candidate {
-                candidate = docset.doc();
-                continue 'outer;
-            }
-        }
-        return candidate;
-    }
+    num_docsets: usize,
 }
 
 impl<TDocSet: DocSet> Intersection<TDocSet, TDocSet> {
     pub(crate) fn new(mut docsets: Vec<TDocSet>) -> Intersection<TDocSet, TDocSet> {
         let num_docsets = docsets.len();
         assert!(num_docsets >= 2);
-        docsets.sort_by_key(|docset| docset.size_hint());
-        go_to_first_doc(&mut docsets);
-        let left = docsets.remove(0);
-        let right = docsets.remove(0);
+        docsets.sort_by(|left, right| right.size_hint().cmp(&left.size_hint()));
+        let left = docsets.pop().unwrap();
+        let right = docsets.pop().unwrap();
+        docsets.reverse();
         Intersection {
             left,
             right,
             others: docsets,
+            num_docsets,
         }
     }
 }
@@ -92,49 +80,128 @@ impl<TDocSet: DocSet> Intersection<TDocSet, TDocSet> {
     }
 }
 
+impl<TDocSet: DocSet, TOtherDocSet: DocSet> Intersection<TDocSet, TOtherDocSet> {
+    pub(crate) fn docset_mut(&mut self, ord: usize) -> &mut dyn DocSet {
+        match ord {
+            0 => &mut self.left,
+            1 => &mut self.right,
+            n => &mut self.others[n - 2],
+        }
+    }
+}
+
 impl<TDocSet: DocSet, TOtherDocSet: DocSet> DocSet for Intersection<TDocSet, TOtherDocSet> {
-    fn advance(&mut self) -> DocId {
+    fn advance(&mut self) -> bool {
         let (left, right) = (&mut self.left, &mut self.right);
-        let mut candidate = left.advance();
+
+        if !left.advance() {
+            return false;
+        }
+
+        let mut candidate = left.doc();
+        let mut other_candidate_ord: usize = usize::max_value();
 
         'outer: loop {
             // In the first part we look for a document in the intersection
             // of the two rarest `DocSet` in the intersection.
-
             loop {
-                let right_doc = right.seek(candidate);
-                candidate = left.seek(right_doc);
-                if candidate == right_doc {
-                    break;
+                match right.skip_next(candidate) {
+                    SkipResult::Reached => {
+                        break;
+                    }
+                    SkipResult::OverStep => {
+                        candidate = right.doc();
+                        other_candidate_ord = usize::max_value();
+                    }
+                    SkipResult::End => {
+                        return false;
+                    }
+                }
+                match left.skip_next(candidate) {
+                    SkipResult::Reached => {
+                        break;
+                    }
+                    SkipResult::OverStep => {
+                        candidate = left.doc();
+                        other_candidate_ord = usize::max_value();
+                    }
+                    SkipResult::End => {
+                        return false;
+                    }
                 }
             }
-
-            debug_assert_eq!(left.doc(), right.doc());
             // test the remaining scorers;
-            for docset in self.others.iter_mut() {
+            for (ord, docset) in self.others.iter_mut().enumerate() {
+                if ord == other_candidate_ord {
+                    continue;
+                }
                 // `candidate_ord` is already at the
                 // right position.
                 //
                 // Calling `skip_next` would advance this docset
                 // and miss it.
-                let seek_doc = docset.seek(candidate);
-                if seek_doc > candidate {
-                    candidate = left.seek(seek_doc);
-                    continue 'outer;
+                match docset.skip_next(candidate) {
+                    SkipResult::Reached => {}
+                    SkipResult::OverStep => {
+                        // this is not in the intersection,
+                        // let's update our candidate.
+                        candidate = docset.doc();
+                        match left.skip_next(candidate) {
+                            SkipResult::Reached => {
+                                other_candidate_ord = ord;
+                            }
+                            SkipResult::OverStep => {
+                                candidate = left.doc();
+                                other_candidate_ord = usize::max_value();
+                            }
+                            SkipResult::End => {
+                                return false;
+                            }
+                        }
+                        continue 'outer;
+                    }
+                    SkipResult::End => {
+                        return false;
+                    }
                 }
             }
-
-            return candidate;
+            return true;
         }
     }
 
-    fn seek(&mut self, target: DocId) -> DocId {
-        self.left.seek(target);
-        let mut docsets: Vec<&mut dyn DocSet> = vec![&mut self.left, &mut self.right];
-        for docset in &mut self.others {
-            docsets.push(docset);
+    fn skip_next(&mut self, target: DocId) -> SkipResult {
+        // We optimize skipping by skipping every single member
+        // of the intersection to target.
+        let mut current_target: DocId = target;
+        let mut current_ord = self.num_docsets;
+
+        'outer: loop {
+            for ord in 0..self.num_docsets {
+                let docset = self.docset_mut(ord);
+                if ord == current_ord {
+                    continue;
+                }
+                match docset.skip_next(current_target) {
+                    SkipResult::End => {
+                        return SkipResult::End;
+                    }
+                    SkipResult::OverStep => {
+                        // update the target
+                        // for the remaining members of the intersection.
+                        current_target = docset.doc();
+                        current_ord = ord;
+                        continue 'outer;
+                    }
+                    SkipResult::Reached => {}
+                }
+            }
+            if target == current_target {
+                return SkipResult::Reached;
+            } else {
+                assert!(current_target > target);
+                return SkipResult::OverStep;
+            }
         }
-        go_to_first_doc(&mut docsets[..])
     }
 
     fn doc(&self) -> DocId {
@@ -161,7 +228,7 @@ where
 #[cfg(test)]
 mod tests {
     use super::Intersection;
-    use crate::docset::{DocSet, TERMINATED};
+    use crate::docset::{DocSet, SkipResult};
     use crate::postings::tests::test_skip_against_unoptimized;
     use crate::query::VecDocSet;
 
@@ -171,18 +238,20 @@ mod tests {
             let left = VecDocSet::from(vec![1, 3, 9]);
             let right = VecDocSet::from(vec![3, 4, 9, 18]);
             let mut intersection = Intersection::new(vec![left, right]);
+            assert!(intersection.advance());
             assert_eq!(intersection.doc(), 3);
-            assert_eq!(intersection.advance(), 9);
+            assert!(intersection.advance());
             assert_eq!(intersection.doc(), 9);
-            assert_eq!(intersection.advance(), TERMINATED);
+            assert!(!intersection.advance());
         }
         {
             let a = VecDocSet::from(vec![1, 3, 9]);
             let b = VecDocSet::from(vec![3, 4, 9, 18]);
             let c = VecDocSet::from(vec![1, 5, 9, 111]);
             let mut intersection = Intersection::new(vec![a, b, c]);
+            assert!(intersection.advance());
             assert_eq!(intersection.doc(), 9);
-            assert_eq!(intersection.advance(), TERMINATED);
+            assert!(!intersection.advance());
         }
     }
 
@@ -191,8 +260,8 @@ mod tests {
         let left = VecDocSet::from(vec![0]);
         let right = VecDocSet::from(vec![0]);
         let mut intersection = Intersection::new(vec![left, right]);
+        assert!(intersection.advance());
         assert_eq!(intersection.doc(), 0);
-        assert_eq!(intersection.advance(), TERMINATED);
     }
 
     #[test]
@@ -200,7 +269,7 @@ mod tests {
         let left = VecDocSet::from(vec![0, 1, 2, 4]);
         let right = VecDocSet::from(vec![2, 5]);
         let mut intersection = Intersection::new(vec![left, right]);
-        assert_eq!(intersection.seek(2), 2);
+        assert_eq!(intersection.skip_next(2), SkipResult::Reached);
         assert_eq!(intersection.doc(), 2);
     }
 
@@ -243,7 +312,7 @@ mod tests {
         let a = VecDocSet::from(vec![1, 3]);
         let b = VecDocSet::from(vec![1, 4]);
         let c = VecDocSet::from(vec![3, 9]);
-        let intersection = Intersection::new(vec![a, b, c]);
-        assert_eq!(intersection.doc(), TERMINATED);
+        let mut intersection = Intersection::new(vec![a, b, c]);
+        assert!(!intersection.advance());
     }
 }

src/query/mod.rs

@@ -1,11 +1,12 @@
-/*! Query Module */
+/*!
+Query
+*/
 
 mod all_query;
 mod automaton_weight;
 mod bitset;
 mod bm25;
 mod boolean_query;
-mod boost_query;
 mod empty_query;
 mod exclude;
 mod explanation;
@@ -36,12 +37,9 @@ pub use self::all_query::{AllQuery, AllScorer, AllWeight};
 pub use self::automaton_weight::AutomatonWeight;
 pub use self::bitset::BitSetDocSet;
 pub use self::boolean_query::BooleanQuery;
-pub use self::boost_query::BoostQuery;
 pub use self::empty_query::{EmptyQuery, EmptyScorer, EmptyWeight};
 pub use self::exclude::Exclude;
 pub use self::explanation::Explanation;
-#[cfg(test)]
-pub(crate) use self::fuzzy_query::DFAWrapper;
 pub use self::fuzzy_query::FuzzyTermQuery;
 pub use self::intersection::intersect_scorers;
 pub use self::phrase_query::PhraseQuery;

src/query/phrase_query/mod.rs

@@ -7,17 +7,18 @@ pub use self::phrase_scorer::PhraseScorer;
 pub use self::phrase_weight::PhraseWeight;
 
 #[cfg(test)]
-pub mod tests {
+mod tests {
 
     use super::*;
     use crate::collector::tests::{TEST_COLLECTOR_WITHOUT_SCORE, TEST_COLLECTOR_WITH_SCORE};
     use crate::core::Index;
+    use crate::error::TantivyError;
     use crate::schema::{Schema, Term, TEXT};
     use crate::tests::assert_nearly_equals;
-    use crate::DocAddress;
     use crate::DocId;
+    use crate::{DocAddress, DocSet};
 
-    pub fn create_index(texts: &[&'static str]) -> Index {
+    fn create_index(texts: &[&'static str]) -> Index {
         let mut schema_builder = Schema::builder();
         let text_field = schema_builder.add_text_field("text", TEXT);
         let schema = schema_builder.build();
@@ -60,8 +61,8 @@ pub mod tests {
                 .map(|docaddr| docaddr.1)
                 .collect::<Vec<_>>()
         };
-        assert_eq!(test_query(vec!["a", "b"]), vec![1, 2, 3, 4]);
         assert_eq!(test_query(vec!["a", "b", "c"]), vec![2, 4]);
+        assert_eq!(test_query(vec!["a", "b"]), vec![1, 2, 3, 4]);
         assert_eq!(test_query(vec!["b", "b"]), vec![0, 1]);
         assert!(test_query(vec!["g", "ewrwer"]).is_empty());
         assert!(test_query(vec!["g", "a"]).is_empty());
@@ -101,6 +102,30 @@ pub mod tests {
         assert!(test_query(vec!["g", "a"]).is_empty());
     }
 
+    #[test]
+    pub fn test_phrase_count() {
+        let index = create_index(&["a c", "a a b d a b c", " a b"]);
+        let schema = index.schema();
+        let text_field = schema.get_field("text").unwrap();
+        let searcher = index.reader().unwrap().searcher();
+        let phrase_query = PhraseQuery::new(vec![
+            Term::from_field_text(text_field, "a"),
+            Term::from_field_text(text_field, "b"),
+        ]);
+        let phrase_weight = phrase_query.phrase_weight(&searcher, true).unwrap();
+        let mut phrase_scorer = phrase_weight
+            .phrase_scorer(searcher.segment_reader(0u32))
+            .unwrap()
+            .unwrap();
+        assert!(phrase_scorer.advance());
+        assert_eq!(phrase_scorer.doc(), 1);
+        assert_eq!(phrase_scorer.phrase_count(), 2);
+        assert!(phrase_scorer.advance());
+        assert_eq!(phrase_scorer.doc(), 2);
+        assert_eq!(phrase_scorer.phrase_count(), 1);
+        assert!(!phrase_scorer.advance());
+    }
+
     #[test]
     pub fn test_phrase_query_no_positions() {
         let mut schema_builder = Schema::builder();
@@ -126,16 +151,21 @@ pub mod tests {
             Term::from_field_text(text_field, "a"),
             Term::from_field_text(text_field, "b"),
         ]);
-
-        let search_result = searcher
+        match searcher
             .search(&phrase_query, &TEST_COLLECTOR_WITH_SCORE)
-            .map(|_| ());
-        assert!(matches!(
-            search_result,
-            Err(crate::TantivyError::SchemaError(msg))
-            if msg == "Applied phrase query on field \"text\", which does not have positions \
-            indexed"
-        ));
+            .map(|_| ())
+            .unwrap_err()
+        {
+            TantivyError::SchemaError(ref msg) => {
+                assert_eq!(
+                    "Applied phrase query on field \"text\", which does not have positions indexed",
+                    msg.as_str()
+                );
+            }
+            _ => {
+                panic!("Should have returned an error");
+            }
+        }
     }
 
     #[test]

src/query/phrase_query/phrase_scorer.rs

@@ -1,4 +1,4 @@
-use crate::docset::{DocSet, TERMINATED};
+use crate::docset::{DocSet, SkipResult};
 use crate::fieldnorm::FieldNormReader;
 use crate::postings::Postings;
 use crate::query::bm25::BM25Weight;
@@ -25,12 +25,12 @@ impl<TPostings: Postings> PostingsWithOffset<TPostings> {
 }
 
 impl<TPostings: Postings> DocSet for PostingsWithOffset<TPostings> {
-    fn advance(&mut self) -> DocId {
+    fn advance(&mut self) -> bool {
         self.postings.advance()
     }
 
-    fn seek(&mut self, target: DocId) -> DocId {
-        self.postings.seek(target)
+    fn skip_next(&mut self, target: DocId) -> SkipResult {
+        self.postings.skip_next(target)
     }
 
     fn doc(&self) -> DocId {
@@ -149,7 +149,7 @@ impl<TPostings: Postings> PhraseScorer<TPostings> {
                 PostingsWithOffset::new(postings, (max_offset - offset) as u32)
             })
             .collect::<Vec<_>>();
-        let mut scorer = PhraseScorer {
+        PhraseScorer {
             intersection_docset: Intersection::new(postings_with_offsets),
             num_terms: num_docsets,
             left: Vec::with_capacity(100),
@@ -158,11 +158,7 @@ impl<TPostings: Postings> PhraseScorer<TPostings> {
             similarity_weight,
             fieldnorm_reader,
             score_needed,
-        };
-        if scorer.doc() != TERMINATED && !scorer.phrase_match() {
-            scorer.advance();
         }
-        scorer
     }
 
     pub fn phrase_count(&self) -> u32 {
@@ -229,21 +225,31 @@ impl<TPostings: Postings> PhraseScorer<TPostings> {
 }
 
 impl<TPostings: Postings> DocSet for PhraseScorer<TPostings> {
-    fn advance(&mut self) -> DocId {
-        loop {
-            let doc = self.intersection_docset.advance();
-            if doc == TERMINATED || self.phrase_match() {
-                return doc;
+    fn advance(&mut self) -> bool {
+        while self.intersection_docset.advance() {
+            if self.phrase_match() {
+                return true;
             }
         }
+        false
     }
 
-    fn seek(&mut self, target: DocId) -> DocId {
-        let doc = self.intersection_docset.seek(target);
-        if doc == TERMINATED || self.phrase_match() {
-            return doc;
+    fn skip_next(&mut self, target: DocId) -> SkipResult {
+        if self.intersection_docset.skip_next(target) == SkipResult::End {
+            return SkipResult::End;
+        }
+        if self.phrase_match() {
+            if self.doc() == target {
+                return SkipResult::Reached;
+            } else {
+                return SkipResult::OverStep;
+            }
+        }
+        if self.advance() {
+            SkipResult::OverStep
+        } else {
+            SkipResult::End
         }
-        self.advance()
     }
 
     fn doc(&self) -> DocId {

src/query/phrase_query/phrase_weight.rs

@@ -9,8 +9,8 @@ use crate::query::Weight;
 use crate::query::{EmptyScorer, Explanation};
 use crate::schema::IndexRecordOption;
 use crate::schema::Term;
-use crate::Result;
 use crate::{DocId, DocSet};
+use crate::{Result, SkipResult};
 
 pub struct PhraseWeight {
     phrase_terms: Vec<(usize, Term)>,
@@ -37,12 +37,11 @@ impl PhraseWeight {
         reader.get_fieldnorms_reader(field)
     }
 
-    fn phrase_scorer(
+    pub fn phrase_scorer(
         &self,
         reader: &SegmentReader,
-        boost: f32,
     ) -> Result<Option<PhraseScorer<SegmentPostings>>> {
-        let similarity_weight = self.similarity_weight.boost_by(boost);
+        let similarity_weight = self.similarity_weight.clone();
         let fieldnorm_reader = self.fieldnorm_reader(reader);
         if reader.has_deletes() {
             let mut term_postings_list = Vec::new();
@@ -85,8 +84,8 @@ impl PhraseWeight {
 }
 
 impl Weight for PhraseWeight {
-    fn scorer(&self, reader: &SegmentReader, boost: f32) -> Result<Box<dyn Scorer>> {
-        if let Some(scorer) = self.phrase_scorer(reader, boost)? {
+    fn scorer(&self, reader: &SegmentReader) -> Result<Box<dyn Scorer>> {
+        if let Some(scorer) = self.phrase_scorer(reader)? {
             Ok(Box::new(scorer))
         } else {
             Ok(Box::new(EmptyScorer))
@@ -94,12 +93,12 @@ impl Weight for PhraseWeight {
     }
 
     fn explain(&self, reader: &SegmentReader, doc: DocId) -> Result<Explanation> {
-        let scorer_opt = self.phrase_scorer(reader, 1.0f32)?;
+        let scorer_opt = self.phrase_scorer(reader)?;
         if scorer_opt.is_none() {
             return Err(does_not_match(doc));
         }
         let mut scorer = scorer_opt.unwrap();
-        if scorer.seek(doc) != doc {
+        if scorer.skip_next(doc) != SkipResult::Reached {
             return Err(does_not_match(doc));
         }
         let fieldnorm_reader = self.fieldnorm_reader(reader);
@@ -110,34 +109,3 @@ impl Weight for PhraseWeight {
         Ok(explanation)
     }
 }
-
-#[cfg(test)]
-mod tests {
-    use super::super::tests::create_index;
-    use crate::docset::TERMINATED;
-    use crate::query::PhraseQuery;
-    use crate::{DocSet, Term};
-
-    #[test]
-    pub fn test_phrase_count() {
-        let index = create_index(&["a c", "a a b d a b c", " a b"]);
-        let schema = index.schema();
-        let text_field = schema.get_field("text").unwrap();
-        let searcher = index.reader().unwrap().searcher();
-        let phrase_query = PhraseQuery::new(vec![
-            Term::from_field_text(text_field, "a"),
-            Term::from_field_text(text_field, "b"),
-        ]);
-        let phrase_weight = phrase_query.phrase_weight(&searcher, true).unwrap();
-        let mut phrase_scorer = phrase_weight
-            .phrase_scorer(searcher.segment_reader(0u32), 1.0f32)
-            .unwrap()
-            .unwrap();
-        assert_eq!(phrase_scorer.doc(), 1);
-        assert_eq!(phrase_scorer.phrase_count(), 2);
-        assert_eq!(phrase_scorer.advance(), 2);
-        assert_eq!(phrase_scorer.doc(), 2);
-        assert_eq!(phrase_scorer.phrase_count(), 1);
-        assert_eq!(phrase_scorer.advance(), TERMINATED);
-    }
-}

src/query/query_parser/logical_ast.rs

@@ -21,17 +21,6 @@ pub enum LogicalLiteral {
 pub enum LogicalAST {
     Clause(Vec<(Occur, LogicalAST)>),
     Leaf(Box<LogicalLiteral>),
-    Boost(Box<LogicalAST>, f32),
-}
-
-impl LogicalAST {
-    pub fn boost(self, boost: f32) -> LogicalAST {
-        if (boost - 1.0f32).abs() < std::f32::EPSILON {
-            self
-        } else {
-            LogicalAST::Boost(Box::new(self), boost)
-        }
-    }
 }
 
 fn occur_letter(occur: Occur) -> &'static str {
@@ -58,7 +47,6 @@ impl fmt::Debug for LogicalAST {
                 }
                 Ok(())
             }
-            LogicalAST::Boost(ref ast, boost) => write!(formatter, "{:?}^{}", ast, boost),
             LogicalAST::Leaf(ref literal) => write!(formatter, "{:?}", literal),
         }
     }

src/query/query_parser/query_parser.rs

@@ -1,5 +1,6 @@
 use super::logical_ast::*;
 use crate::core::Index;
+use crate::query::AllQuery;
 use crate::query::BooleanQuery;
 use crate::query::EmptyQuery;
 use crate::query::Occur;
@@ -7,13 +8,11 @@ use crate::query::PhraseQuery;
 use crate::query::Query;
 use crate::query::RangeQuery;
 use crate::query::TermQuery;
-use crate::query::{AllQuery, BoostQuery};
 use crate::schema::{Facet, IndexRecordOption};
 use crate::schema::{Field, Schema};
 use crate::schema::{FieldType, Term};
 use crate::tokenizer::TokenizerManager;
 use std::borrow::Cow;
-use std::collections::HashMap;
 use std::num::{ParseFloatError, ParseIntError};
 use std::ops::Bound;
 use std::str::FromStr;
@@ -113,9 +112,8 @@ fn trim_ast(logical_ast: LogicalAST) -> Option<LogicalAST> {
 /// The language covered by the current parser is extremely simple.
 ///
 /// * simple terms: "e.g.: `Barack Obama` are simply tokenized using
-///   tantivy's [`SimpleTokenizer`](tantivy::tokenizer::SimpleTokenizer), hence
-///   becoming `["barack", "obama"]`. The terms are then searched within
-///   the default terms of the query parser.
+///   tantivy's `StandardTokenizer`, hence becoming `["barack", "obama"]`.
+///   The terms are then searched within the default terms of the query parser.
 ///
 ///   e.g. If `body` and `title` are default fields, our example terms are
 ///   `["title:barack", "body:barack", "title:obama", "body:obama"]`.
@@ -146,6 +144,7 @@ fn trim_ast(logical_ast: LogicalAST) -> Option<LogicalAST> {
 ///
 /// * must terms: By prepending a term by a `+`, a term can be made required for the search.
 ///
+///
 /// * phrase terms: Quoted terms become phrase searches on fields that have positions indexed.
 ///   e.g., `title:"Barack Obama"` will only find documents that have "barack" immediately followed
 ///   by "obama".
@@ -159,30 +158,12 @@ fn trim_ast(logical_ast: LogicalAST) -> Option<LogicalAST> {
 ///
 /// *  all docs query: A plain `*` will match all documents in the index.
 ///
-/// Parts of the queries can be boosted by appending `^boostfactor`.
-/// For instance, `"SRE"^2.0 OR devops^0.4` will boost documents containing `SRE` instead of
-/// devops. Negative boosts are not allowed.
-///
-/// It is also possible to define a boost for a some specific field, at the query parser level.
-/// (See [`set_boost(...)`](#method.set_field_boost) ). Typically you may want to boost a title
-/// field.
 #[derive(Clone)]
 pub struct QueryParser {
     schema: Schema,
     default_fields: Vec<Field>,
     conjunction_by_default: bool,
     tokenizer_manager: TokenizerManager,
-    boost: HashMap<Field, f32>,
-}
-
-fn all_negative(ast: &LogicalAST) -> bool {
-    match ast {
-        LogicalAST::Leaf(_) => false,
-        LogicalAST::Boost(ref child_ast, _) => all_negative(&*child_ast),
-        LogicalAST::Clause(children) => children
-            .iter()
-            .all(|(ref occur, child)| (*occur == Occur::MustNot) || all_negative(child)),
-    }
 }
 
 impl QueryParser {
@@ -200,7 +181,6 @@ impl QueryParser {
             default_fields,
             tokenizer_manager,
             conjunction_by_default: false,
-            boost: Default::default(),
         }
     }
 
@@ -221,17 +201,6 @@ impl QueryParser {
         self.conjunction_by_default = true;
     }
 
-    /// Sets a boost for a specific field.
-    ///
-    /// The parse query will automatically boost this field.
-    ///
-    /// If the query defines a query boost through the query language (e.g: `country:France^3.0`),
-    /// the two boosts (the one defined in the query, and the one defined in the `QueryParser`)
-    /// are multiplied together.
-    pub fn set_field_boost(&mut self, field: Field, boost: f32) {
-        self.boost.insert(field, boost);
-    }
-
     /// Parse a query
     ///
     /// Note that `parse_query` returns an error if the input
@@ -264,13 +233,8 @@ impl QueryParser {
         &self,
         user_input_ast: UserInputAST,
     ) -> Result<LogicalAST, QueryParserError> {
-        let ast = self.compute_logical_ast_with_occur(user_input_ast)?;
-        if let LogicalAST::Clause(children) = &ast {
-            if children.is_empty() {
-                return Ok(ast);
-            }
-        }
-        if all_negative(&ast) {
+        let (occur, ast) = self.compute_logical_ast_with_occur(user_input_ast)?;
+        if occur == Occur::MustNot {
             return Err(QueryParserError::AllButQueryForbidden);
         }
         Ok(ast)
@@ -426,30 +390,30 @@ impl QueryParser {
     fn compute_logical_ast_with_occur(
         &self,
         user_input_ast: UserInputAST,
-    ) -> Result<LogicalAST, QueryParserError> {
+    ) -> Result<(Occur, LogicalAST), QueryParserError> {
         match user_input_ast {
             UserInputAST::Clause(sub_queries) => {
                 let default_occur = self.default_occur();
                 let mut logical_sub_queries: Vec<(Occur, LogicalAST)> = Vec::new();
-                for (occur_opt, sub_ast) in sub_queries {
-                    let sub_ast = self.compute_logical_ast_with_occur(sub_ast)?;
-                    let occur = occur_opt.unwrap_or(default_occur);
-                    logical_sub_queries.push((occur, sub_ast));
+                for sub_query in sub_queries {
+                    let (occur, sub_ast) = self.compute_logical_ast_with_occur(sub_query)?;
+                    let new_occur = Occur::compose(default_occur, occur);
+                    logical_sub_queries.push((new_occur, sub_ast));
                 }
-                Ok(LogicalAST::Clause(logical_sub_queries))
+                Ok((Occur::Should, LogicalAST::Clause(logical_sub_queries)))
             }
-            UserInputAST::Boost(ast, boost) => {
-                let ast = self.compute_logical_ast_with_occur(*ast)?;
-                Ok(ast.boost(boost))
+            UserInputAST::Unary(left_occur, subquery) => {
+                let (right_occur, logical_sub_queries) =
+                    self.compute_logical_ast_with_occur(*subquery)?;
+                Ok((Occur::compose(left_occur, right_occur), logical_sub_queries))
+            }
+            UserInputAST::Leaf(leaf) => {
+                let result_ast = self.compute_logical_ast_from_leaf(*leaf)?;
+                Ok((Occur::Should, result_ast))
             }
-            UserInputAST::Leaf(leaf) => self.compute_logical_ast_from_leaf(*leaf),
         }
     }
 
-    fn field_boost(&self, field: Field) -> f32 {
-        self.boost.get(&field).cloned().unwrap_or(1.0f32)
-    }
-
     fn compute_logical_ast_from_leaf(
         &self,
         leaf: UserInputLeaf,
@@ -475,9 +439,7 @@ impl QueryParser {
                 let mut asts: Vec<LogicalAST> = Vec::new();
                 for (field, phrase) in term_phrases {
                     if let Some(ast) = self.compute_logical_ast_for_leaf(field, &phrase)? {
-                        // Apply some field specific boost defined at the query parser level.
-                        let boost = self.field_boost(field);
-                        asts.push(LogicalAST::Leaf(Box::new(ast)).boost(boost));
+                        asts.push(LogicalAST::Leaf(Box::new(ast)));
                     }
                 }
                 let result_ast: LogicalAST = if asts.len() == 1 {
@@ -497,16 +459,14 @@ impl QueryParser {
                 let mut clauses = fields
                     .iter()
                     .map(|&field| {
-                        let boost = self.field_boost(field);
                         let field_entry = self.schema.get_field_entry(field);
                         let value_type = field_entry.field_type().value_type();
-                        let logical_ast = LogicalAST::Leaf(Box::new(LogicalLiteral::Range {
+                        Ok(LogicalAST::Leaf(Box::new(LogicalLiteral::Range {
                             field,
                             value_type,
                             lower: self.resolve_bound(field, &lower)?,
                             upper: self.resolve_bound(field, &upper)?,
-                        }));
-                        Ok(logical_ast.boost(boost))
+                        })))
                     })
                     .collect::<Result<Vec<_>, QueryParserError>>()?;
                 let result_ast = if clauses.len() == 1 {
@@ -559,11 +519,6 @@ fn convert_to_query(logical_ast: LogicalAST) -> Box<dyn Query> {
         Some(LogicalAST::Leaf(trimmed_logical_literal)) => {
             convert_literal_to_query(*trimmed_logical_literal)
         }
-        Some(LogicalAST::Boost(ast, boost)) => {
-            let query = convert_to_query(*ast);
-            let boosted_query = BoostQuery::new(query, boost);
-            Box::new(boosted_query)
-        }
         None => Box::new(EmptyQuery),
     }
 }
@@ -583,7 +538,7 @@ mod test {
     use crate::Index;
     use matches::assert_matches;
 
-    fn make_schema() -> Schema {
+    fn make_query_parser() -> QueryParser {
         let mut schema_builder = Schema::builder();
         let text_field_indexing = TextFieldIndexing::default()
             .set_tokenizer("en_with_stop_words")
@@ -591,8 +546,8 @@ mod test {
         let text_options = TextOptions::default()
             .set_indexing_options(text_field_indexing)
             .set_stored();
-        schema_builder.add_text_field("title", TEXT);
-        schema_builder.add_text_field("text", TEXT);
+        let title = schema_builder.add_text_field("title", TEXT);
+        let text = schema_builder.add_text_field("text", TEXT);
         schema_builder.add_i64_field("signed", INDEXED);
         schema_builder.add_u64_field("unsigned", INDEXED);
         schema_builder.add_text_field("notindexed_text", STORED);
@@ -603,15 +558,8 @@ mod test {
         schema_builder.add_date_field("date", INDEXED);
         schema_builder.add_f64_field("float", INDEXED);
         schema_builder.add_facet_field("facet");
-        schema_builder.build()
-    }
-
-    fn make_query_parser() -> QueryParser {
-        let schema = make_schema();
-        let default_fields: Vec<Field> = vec!["title", "text"]
-            .into_iter()
-            .flat_map(|field_name| schema.get_field(field_name))
-            .collect();
+        let schema = schema_builder.build();
+        let default_fields = vec![title, text];
         let tokenizer_manager = TokenizerManager::default();
         tokenizer_manager.register(
             "en_with_stop_words",
@@ -653,45 +601,6 @@ mod test {
         );
     }
 
-    #[test]
-    pub fn test_parse_query_with_boost() {
-        let mut query_parser = make_query_parser();
-        let schema = make_schema();
-        let text_field = schema.get_field("text").unwrap();
-        query_parser.set_field_boost(text_field, 2.0f32);
-        let query = query_parser.parse_query("text:hello").unwrap();
-        assert_eq!(
-            format!("{:?}", query),
-            "Boost(query=TermQuery(Term(field=1,bytes=[104, 101, 108, 108, 111])), boost=2)"
-        );
-    }
-
-    #[test]
-    pub fn test_parse_query_range_with_boost() {
-        let mut query_parser = make_query_parser();
-        let schema = make_schema();
-        let title_field = schema.get_field("title").unwrap();
-        query_parser.set_field_boost(title_field, 2.0f32);
-        let query = query_parser.parse_query("title:[A TO B]").unwrap();
-        assert_eq!(
-            format!("{:?}", query),
-            "Boost(query=RangeQuery { field: Field(0), value_type: Str, left_bound: Included([97]), right_bound: Included([98]) }, boost=2)"
-        );
-    }
-
-    #[test]
-    pub fn test_parse_query_with_default_boost_and_custom_boost() {
-        let mut query_parser = make_query_parser();
-        let schema = make_schema();
-        let text_field = schema.get_field("text").unwrap();
-        query_parser.set_field_boost(text_field, 2.0f32);
-        let query = query_parser.parse_query("text:hello^2").unwrap();
-        assert_eq!(
-            format!("{:?}", query),
-            "Boost(query=Boost(query=TermQuery(Term(field=1,bytes=[104, 101, 108, 108, 111])), boost=2), boost=2)"
-        );
-    }
-
     #[test]
     pub fn test_parse_nonindexed_field_yields_error() {
         let query_parser = make_query_parser();
@@ -790,20 +699,6 @@ mod test {
         );
     }
 
-    #[test]
-    fn test_parse_query_to_ast_ab_c() {
-        test_parse_query_to_logical_ast_helper(
-            "(+title:a +title:b) title:c",
-            "((+Term(field=0,bytes=[97]) +Term(field=0,bytes=[98])) Term(field=0,bytes=[99]))",
-            false,
-        );
-        test_parse_query_to_logical_ast_helper(
-            "(+title:a +title:b) title:c",
-            "(+(+Term(field=0,bytes=[97]) +Term(field=0,bytes=[98])) +Term(field=0,bytes=[99]))",
-            true,
-        );
-    }
-
     #[test]
     pub fn test_parse_query_to_ast_single_term() {
         test_parse_query_to_logical_ast_helper(
@@ -823,13 +718,11 @@ mod test {
              Term(field=1,bytes=[116, 105, 116, 105])))",
             false,
         );
-    }
-
-    #[test]
-    fn test_single_negative_term() {
-        assert_matches!(
-            parse_query_to_logical_ast("-title:toto", false),
-            Err(QueryParserError::AllButQueryForbidden)
+        assert_eq!(
+            parse_query_to_logical_ast("-title:toto", false)
+                .err()
+                .unwrap(),
+            QueryParserError::AllButQueryForbidden
         );
     }
 
@@ -989,18 +882,6 @@ mod test {
         assert!(query_parser.parse_query("with_stop_words:the").is_ok());
     }
 
-    #[test]
-    pub fn test_parse_query_single_negative_term_through_error() {
-        assert_matches!(
-            parse_query_to_logical_ast("-title:toto", true),
-            Err(QueryParserError::AllButQueryForbidden)
-        );
-        assert_matches!(
-            parse_query_to_logical_ast("-title:toto", false),
-            Err(QueryParserError::AllButQueryForbidden)
-        );
-    }
-
     #[test]
     pub fn test_parse_query_to_ast_conjunction() {
         test_parse_query_to_logical_ast_helper(
@@ -1020,6 +901,12 @@ mod test {
              Term(field=1,bytes=[116, 105, 116, 105])))",
             true,
         );
+        assert_eq!(
+            parse_query_to_logical_ast("-title:toto", true)
+                .err()
+                .unwrap(),
+            QueryParserError::AllButQueryForbidden
+        );
         test_parse_query_to_logical_ast_helper(
             "title:a b",
             "(+Term(field=0,bytes=[97]) \
@@ -1043,26 +930,4 @@ mod test {
             false
         );
     }
-
-    #[test]
-    fn test_and_default_regardless_of_default_conjunctive() {
-        for &default_conjunction in &[false, true] {
-            test_parse_query_to_logical_ast_helper(
-                "title:a AND title:b",
-                "(+Term(field=0,bytes=[97]) +Term(field=0,bytes=[98]))",
-                default_conjunction,
-            );
-        }
-    }
-
-    #[test]
-    fn test_or_default_conjunctive() {
-        for &default_conjunction in &[false, true] {
-            test_parse_query_to_logical_ast_helper(
-                "title:a OR title:b",
-                "(Term(field=0,bytes=[97]) Term(field=0,bytes=[98]))",
-                default_conjunction,
-            );
-        }
-    }
 }

src/query/range_query.rs

@@ -10,7 +10,7 @@ use crate::schema::Type;
 use crate::schema::{Field, IndexRecordOption, Term};
 use crate::termdict::{TermDictionary, TermStreamer};
 use crate::DocId;
-use crate::Result;
+use crate::{Result, SkipResult};
 use std::collections::Bound;
 use std::ops::Range;
 
@@ -289,7 +289,7 @@ impl RangeWeight {
 }
 
 impl Weight for RangeWeight {
-    fn scorer(&self, reader: &SegmentReader, boost: f32) -> Result<Box<dyn Scorer>> {
+    fn scorer(&self, reader: &SegmentReader) -> Result<Box<dyn Scorer>> {
         let max_doc = reader.max_doc();
         let mut doc_bitset = BitSet::with_max_value(max_doc);
 
@@ -307,12 +307,12 @@ impl Weight for RangeWeight {
             }
         }
         let doc_bitset = BitSetDocSet::from(doc_bitset);
-        Ok(Box::new(ConstScorer::new(doc_bitset, boost)))
+        Ok(Box::new(ConstScorer::new(doc_bitset)))
     }
 
     fn explain(&self, reader: &SegmentReader, doc: DocId) -> Result<Explanation> {
-        let mut scorer = self.scorer(reader, 1.0f32)?;
-        if scorer.seek(doc) != doc {
+        let mut scorer = self.scorer(reader)?;
+        if scorer.skip_next(doc) != SkipResult::Reached {
             return Err(does_not_match(doc));
         }
         Ok(Explanation::new("RangeQuery", 1.0f32))

src/query/reqopt_scorer.rs

@@ -1,8 +1,9 @@
-use crate::docset::DocSet;
+use crate::docset::{DocSet, SkipResult};
 use crate::query::score_combiner::ScoreCombiner;
 use crate::query::Scorer;
 use crate::DocId;
 use crate::Score;
+use std::cmp::Ordering;
 use std::marker::PhantomData;
 
 /// Given a required scorer and an optional scorer
@@ -16,6 +17,7 @@ pub struct RequiredOptionalScorer<TReqScorer, TOptScorer, TScoreCombiner> {
     req_scorer: TReqScorer,
     opt_scorer: TOptScorer,
     score_cache: Option<Score>,
+    opt_finished: bool,
     _phantom: PhantomData<TScoreCombiner>,
 }
 
@@ -27,12 +29,14 @@ where
     /// Creates a new `RequiredOptionalScorer`.
     pub fn new(
         req_scorer: TReqScorer,
-        opt_scorer: TOptScorer,
+        mut opt_scorer: TOptScorer,
     ) -> RequiredOptionalScorer<TReqScorer, TOptScorer, TScoreCombiner> {
+        let opt_finished = !opt_scorer.advance();
         RequiredOptionalScorer {
             req_scorer,
             opt_scorer,
             score_cache: None,
+            opt_finished,
             _phantom: PhantomData,
         }
     }
@@ -44,7 +48,7 @@ where
     TReqScorer: DocSet,
     TOptScorer: DocSet,
 {
-    fn advance(&mut self) -> DocId {
+    fn advance(&mut self) -> bool {
         self.score_cache = None;
         self.req_scorer.advance()
     }
@@ -72,8 +76,22 @@ where
         let doc = self.doc();
         let mut score_combiner = TScoreCombiner::default();
         score_combiner.update(&mut self.req_scorer);
-        if self.opt_scorer.seek(doc) == doc {
-            score_combiner.update(&mut self.opt_scorer);
+        if !self.opt_finished {
+            match self.opt_scorer.doc().cmp(&doc) {
+                Ordering::Greater => {}
+                Ordering::Equal => {
+                    score_combiner.update(&mut self.opt_scorer);
+                }
+                Ordering::Less => match self.opt_scorer.skip_next(doc) {
+                    SkipResult::Reached => {
+                        score_combiner.update(&mut self.opt_scorer);
+                    }
+                    SkipResult::End => {
+                        self.opt_finished = true;
+                    }
+                    SkipResult::OverStep => {}
+                },
+            }
         }
         let score = score_combiner.score();
         self.score_cache = Some(score);
@@ -84,7 +102,7 @@ where
 #[cfg(test)]
 mod tests {
     use super::RequiredOptionalScorer;
-    use crate::docset::{DocSet, TERMINATED};
+    use crate::docset::DocSet;
     use crate::postings::tests::test_skip_against_unoptimized;
     use crate::query::score_combiner::{DoNothingCombiner, SumCombiner};
     use crate::query::ConstScorer;
@@ -97,11 +115,13 @@ mod tests {
         let req = vec![1, 3, 7];
         let mut reqoptscorer: RequiredOptionalScorer<_, _, SumCombiner> =
             RequiredOptionalScorer::new(
-                ConstScorer::from(VecDocSet::from(req.clone())),
-                ConstScorer::from(VecDocSet::from(vec![])),
+                ConstScorer::new(VecDocSet::from(req.clone())),
+                ConstScorer::new(VecDocSet::from(vec![])),
             );
         let mut docs = vec![];
-        reqoptscorer.for_each(&mut |doc, _| docs.push(doc));
+        while reqoptscorer.advance() {
+            docs.push(reqoptscorer.doc());
+        }
         assert_eq!(docs, req);
     }
 
@@ -109,49 +129,50 @@ mod tests {
     fn test_reqopt_scorer() {
         let mut reqoptscorer: RequiredOptionalScorer<_, _, SumCombiner> =
             RequiredOptionalScorer::new(
-                ConstScorer::new(VecDocSet::from(vec![1, 3, 7, 8, 9, 10, 13, 15]), 1.0f32),
-                ConstScorer::new(VecDocSet::from(vec![1, 2, 7, 11, 12, 15]), 1.0f32),
+                ConstScorer::new(VecDocSet::from(vec![1, 3, 7, 8, 9, 10, 13, 15])),
+                ConstScorer::new(VecDocSet::from(vec![1, 2, 7, 11, 12, 15])),
             );
         {
+            assert!(reqoptscorer.advance());
             assert_eq!(reqoptscorer.doc(), 1);
             assert_eq!(reqoptscorer.score(), 2f32);
         }
         {
-            assert_eq!(reqoptscorer.advance(), 3);
+            assert!(reqoptscorer.advance());
             assert_eq!(reqoptscorer.doc(), 3);
             assert_eq!(reqoptscorer.score(), 1f32);
         }
         {
-            assert_eq!(reqoptscorer.advance(), 7);
+            assert!(reqoptscorer.advance());
             assert_eq!(reqoptscorer.doc(), 7);
             assert_eq!(reqoptscorer.score(), 2f32);
         }
         {
-            assert_eq!(reqoptscorer.advance(), 8);
+            assert!(reqoptscorer.advance());
             assert_eq!(reqoptscorer.doc(), 8);
             assert_eq!(reqoptscorer.score(), 1f32);
         }
         {
-            assert_eq!(reqoptscorer.advance(), 9);
+            assert!(reqoptscorer.advance());
             assert_eq!(reqoptscorer.doc(), 9);
             assert_eq!(reqoptscorer.score(), 1f32);
         }
         {
-            assert_eq!(reqoptscorer.advance(), 10);
+            assert!(reqoptscorer.advance());
             assert_eq!(reqoptscorer.doc(), 10);
             assert_eq!(reqoptscorer.score(), 1f32);
         }
         {
-            assert_eq!(reqoptscorer.advance(), 13);
+            assert!(reqoptscorer.advance());
             assert_eq!(reqoptscorer.doc(), 13);
             assert_eq!(reqoptscorer.score(), 1f32);
         }
         {
-            assert_eq!(reqoptscorer.advance(), 15);
+            assert!(reqoptscorer.advance());
             assert_eq!(reqoptscorer.doc(), 15);
             assert_eq!(reqoptscorer.score(), 2f32);
         }
-        assert_eq!(reqoptscorer.advance(), TERMINATED);
+        assert!(!reqoptscorer.advance());
     }
 
     #[test]
@@ -162,8 +183,8 @@ mod tests {
         test_skip_against_unoptimized(
             || {
                 Box::new(RequiredOptionalScorer::<_, _, DoNothingCombiner>::new(
-                    ConstScorer::from(VecDocSet::from(req_docs.clone())),
-                    ConstScorer::from(VecDocSet::from(opt_docs.clone())),
+                    ConstScorer::new(VecDocSet::from(req_docs.clone())),
+                    ConstScorer::new(VecDocSet::from(opt_docs.clone())),
                 ))
             },
             skip_docs,

src/query/scorer.rs

@@ -1,4 +1,5 @@
-use crate::docset::{DocSet, TERMINATED};
+use crate::common::BitSet;
+use crate::docset::{DocSet, SkipResult};
 use crate::DocId;
 use crate::Score;
 use downcast_rs::impl_downcast;
@@ -16,35 +17,8 @@ pub trait Scorer: downcast_rs::Downcast + DocSet + 'static {
     /// Iterates through all of the document matched by the DocSet
     /// `DocSet` and push the scored documents to the collector.
     fn for_each(&mut self, callback: &mut dyn FnMut(DocId, Score)) {
-        let mut doc = self.doc();
-        while doc != TERMINATED {
-            callback(doc, self.score());
-            doc = self.advance();
-        }
-    }
-
-    /// Calls `callback` with all of the `(doc, score)` for which score
-    /// is exceeding a given threshold.
-    ///
-    /// This method is useful for the TopDocs collector.
-    /// For all docsets, the blanket implementation has the benefit
-    /// of prefiltering (doc, score) pairs, avoiding the
-    /// virtual dispatch cost.
-    ///
-    /// More importantly, it makes it possible for scorers to implement
-    /// important optimization (e.g. BlockWAND for union).
-    fn for_each_pruning(
-        &mut self,
-        mut threshold: f32,
-        callback: &mut dyn FnMut(DocId, Score) -> Score,
-    ) {
-        let mut doc = self.doc();
-        while doc != TERMINATED {
-            let score = self.score();
-            if score > threshold {
-                threshold = callback(doc, score);
-            }
-            doc = self.advance();
+        while self.advance() {
+            callback(self.doc(), self.score());
         }
     }
 }
@@ -75,24 +49,26 @@ pub struct ConstScorer<TDocSet: DocSet> {
 
 impl<TDocSet: DocSet> ConstScorer<TDocSet> {
     /// Creates a new `ConstScorer`.
-    pub fn new(docset: TDocSet, score: f32) -> ConstScorer<TDocSet> {
-        ConstScorer { docset, score }
+    pub fn new(docset: TDocSet) -> ConstScorer<TDocSet> {
+        ConstScorer {
+            docset,
+            score: 1f32,
+        }
     }
-}
 
-impl<TDocSet: DocSet> From<TDocSet> for ConstScorer<TDocSet> {
-    fn from(docset: TDocSet) -> Self {
-        ConstScorer::new(docset, 1.0f32)
+    /// Sets the constant score to a different value.
+    pub fn set_score(&mut self, score: Score) {
+        self.score = score;
     }
 }
 
 impl<TDocSet: DocSet> DocSet for ConstScorer<TDocSet> {
-    fn advance(&mut self) -> DocId {
+    fn advance(&mut self) -> bool {
         self.docset.advance()
     }
 
-    fn seek(&mut self, target: DocId) -> DocId {
-        self.docset.seek(target)
+    fn skip_next(&mut self, target: DocId) -> SkipResult {
+        self.docset.skip_next(target)
     }
 
     fn fill_buffer(&mut self, buffer: &mut [DocId]) -> usize {
@@ -106,10 +82,14 @@ impl<TDocSet: DocSet> DocSet for ConstScorer<TDocSet> {
     fn size_hint(&self) -> u32 {
         self.docset.size_hint()
     }
+
+    fn append_to_bitset(&mut self, bitset: &mut BitSet) {
+        self.docset.append_to_bitset(bitset);
+    }
 }
 
 impl<TDocSet: DocSet + 'static> Scorer for ConstScorer<TDocSet> {
     fn score(&mut self) -> Score {
-        self.score
+        1f32
     }
 }

src/query/term_query/mod.rs

@@ -26,8 +26,10 @@ mod tests {
         {
             // writing the segment
             let mut index_writer = index.writer_with_num_threads(1, 3_000_000).unwrap();
-            let doc = doc!(text_field => "a");
-            index_writer.add_document(doc);
+            {
+                let doc = doc!(text_field => "a");
+                index_writer.add_document(doc);
+            }
             assert!(index_writer.commit().is_ok());
         }
         let searcher = index.reader().unwrap().searcher();
@@ -37,7 +39,8 @@ mod tests {
         );
         let term_weight = term_query.weight(&searcher, true).unwrap();
         let segment_reader = searcher.segment_reader(0);
-        let mut term_scorer = term_weight.scorer(segment_reader, 1.0f32).unwrap();
+        let mut term_scorer = term_weight.scorer(segment_reader).unwrap();
+        assert!(term_scorer.advance());
         assert_eq!(term_scorer.doc(), 0);
         assert_eq!(term_scorer.score(), 0.28768212);
     }

src/query/term_query/term_scorer.rs

@@ -1,4 +1,4 @@
-use crate::docset::DocSet;
+use crate::docset::{DocSet, SkipResult};
 use crate::query::{Explanation, Scorer};
 use crate::DocId;
 use crate::Score;
@@ -45,12 +45,12 @@ impl TermScorer {
 }
 
 impl DocSet for TermScorer {
-    fn advance(&mut self) -> DocId {
+    fn advance(&mut self) -> bool {
         self.postings.advance()
     }
 
-    fn seek(&mut self, target: DocId) -> DocId {
-        self.postings.seek(target)
+    fn skip_next(&mut self, target: DocId) -> SkipResult {
+        self.postings.skip_next(target)
     }
 
     fn doc(&self) -> DocId {

src/query/term_query/term_weight.rs

@@ -8,8 +8,8 @@ use crate::query::Weight;
 use crate::query::{Explanation, Scorer};
 use crate::schema::IndexRecordOption;
 use crate::DocId;
-use crate::Result;
 use crate::Term;
+use crate::{Result, SkipResult};
 
 pub struct TermWeight {
     term: Term,
@@ -18,14 +18,14 @@ pub struct TermWeight {
 }
 
 impl Weight for TermWeight {
-    fn scorer(&self, reader: &SegmentReader, boost: f32) -> Result<Box<dyn Scorer>> {
-        let term_scorer = self.scorer_specialized(reader, boost)?;
+    fn scorer(&self, reader: &SegmentReader) -> Result<Box<dyn Scorer>> {
+        let term_scorer = self.scorer_specialized(reader)?;
         Ok(Box::new(term_scorer))
     }
 
     fn explain(&self, reader: &SegmentReader, doc: DocId) -> Result<Explanation> {
-        let mut scorer = self.scorer_specialized(reader, 1.0f32)?;
-        if scorer.seek(doc) != doc {
+        let mut scorer = self.scorer_specialized(reader)?;
+        if scorer.skip_next(doc) != SkipResult::Reached {
             return Err(does_not_match(doc));
         }
         Ok(scorer.explain())
@@ -33,7 +33,7 @@ impl Weight for TermWeight {
 
     fn count(&self, reader: &SegmentReader) -> Result<u32> {
         if let Some(delete_bitset) = reader.delete_bitset() {
-            Ok(self.scorer(reader, 1.0f32)?.count(delete_bitset))
+            Ok(self.scorer(reader)?.count(delete_bitset))
         } else {
             let field = self.term.field();
             Ok(reader
@@ -58,11 +58,11 @@ impl TermWeight {
         }
     }
 
-    fn scorer_specialized(&self, reader: &SegmentReader, boost: f32) -> Result<TermScorer> {
+    fn scorer_specialized(&self, reader: &SegmentReader) -> Result<TermScorer> {
         let field = self.term.field();
         let inverted_index = reader.inverted_index(field);
         let fieldnorm_reader = reader.get_fieldnorms_reader(field);
-        let similarity_weight = self.similarity_weight.boost_by(boost);
+        let similarity_weight = self.similarity_weight.clone();
         let postings_opt: Option<SegmentPostings> =
             inverted_index.read_postings(&self.term, self.index_record_option);
         if let Some(segment_postings) = postings_opt {

src/query/union.rs

@@ -1,9 +1,10 @@
 use crate::common::TinySet;
-use crate::docset::{DocSet, TERMINATED};
+use crate::docset::{DocSet, SkipResult};
 use crate::query::score_combiner::{DoNothingCombiner, ScoreCombiner};
 use crate::query::Scorer;
 use crate::DocId;
 use crate::Score;
+use std::cmp::Ordering;
 
 const HORIZON_NUM_TINYBITSETS: usize = 64;
 const HORIZON: u32 = 64u32 * HORIZON_NUM_TINYBITSETS as u32;
@@ -46,9 +47,17 @@ where
     fn from(docsets: Vec<TScorer>) -> Union<TScorer, TScoreCombiner> {
         let non_empty_docsets: Vec<TScorer> = docsets
             .into_iter()
-            .filter(|docset| docset.doc() != TERMINATED)
+            .flat_map(
+                |mut docset| {
+                    if docset.advance() {
+                        Some(docset)
+                    } else {
+                        None
+                    }
+                },
+            )
             .collect();
-        let mut union = Union {
+        Union {
             docsets: non_empty_docsets,
             bitsets: Box::new([TinySet::empty(); HORIZON_NUM_TINYBITSETS]),
             scores: Box::new([TScoreCombiner::default(); HORIZON as usize]),
@@ -56,13 +65,7 @@ where
             offset: 0,
             doc: 0,
             score: 0f32,
-        };
-        if union.refill() {
-            union.advance();
-        } else {
-            union.doc = TERMINATED;
         }
-        union
     }
 }
 
@@ -83,7 +86,7 @@ fn refill<TScorer: Scorer, TScoreCombiner: ScoreCombiner>(
             let delta = doc - min_doc;
             bitsets[(delta / 64) as usize].insert_mut(delta % 64u32);
             score_combiner[delta as usize].update(scorer);
-            if scorer.advance() == TERMINATED {
+            if !scorer.advance() {
                 // remove the docset, it has been entirely consumed.
                 return true;
             }
@@ -96,7 +99,6 @@ impl<TScorer: Scorer, TScoreCombiner: ScoreCombiner> Union<TScorer, TScoreCombin
         if let Some(min_doc) = self.docsets.iter().map(DocSet::doc).min() {
             self.offset = min_doc;
             self.cursor = 0;
-            self.doc = min_doc;
             refill(
                 &mut self.docsets,
                 &mut *self.bitsets,
@@ -131,23 +133,50 @@ where
     TScorer: Scorer,
     TScoreCombiner: ScoreCombiner,
 {
-    fn advance(&mut self) -> DocId {
+    fn advance(&mut self) -> bool {
         if self.advance_buffered() {
-            return self.doc;
+            return true;
         }
-        if !self.refill() {
-            self.doc = TERMINATED;
-            return TERMINATED;
+        if self.refill() {
+            self.advance();
+            true
+        } else {
+            false
         }
-        if !self.advance_buffered() {
-            return TERMINATED;
-        }
-        self.doc
     }
 
-    fn seek(&mut self, target: DocId) -> DocId {
-        if self.doc >= target {
-            return self.doc;
+    fn count_including_deleted(&mut self) -> u32 {
+        let mut count = self.bitsets[self.cursor..HORIZON_NUM_TINYBITSETS]
+            .iter()
+            .map(|bitset| bitset.len())
+            .sum::<u32>();
+        for bitset in self.bitsets.iter_mut() {
+            bitset.clear();
+        }
+        while self.refill() {
+            count += self.bitsets.iter().map(|bitset| bitset.len()).sum::<u32>();
+            for bitset in self.bitsets.iter_mut() {
+                bitset.clear();
+            }
+        }
+        self.cursor = HORIZON_NUM_TINYBITSETS;
+        count
+    }
+
+    // TODO implement `count` efficiently.
+
+    fn skip_next(&mut self, target: DocId) -> SkipResult {
+        if !self.advance() {
+            return SkipResult::End;
+        }
+        match self.doc.cmp(&target) {
+            Ordering::Equal => {
+                return SkipResult::Reached;
+            }
+            Ordering::Greater => {
+                return SkipResult::OverStep;
+            }
+            Ordering::Less => {}
         }
         let gap = target - self.offset;
         if gap < HORIZON {
@@ -165,11 +194,18 @@ where
 
             // Advancing until we reach the end of the bucket
             // or we reach a doc greater or equal to the target.
-            let mut doc = self.doc();
-            while doc < target {
-                doc = self.advance();
+            while self.advance() {
+                match self.doc().cmp(&target) {
+                    Ordering::Equal => {
+                        return SkipResult::Reached;
+                    }
+                    Ordering::Greater => {
+                        return SkipResult::OverStep;
+                    }
+                    Ordering::Less => {}
+                }
             }
-            doc
+            SkipResult::End
         } else {
             // clear the buffered info.
             for obsolete_tinyset in self.bitsets.iter_mut() {
@@ -183,41 +219,28 @@ where
             // advance all docsets to a doc >= to the target.
             #[cfg_attr(feature = "cargo-clippy", allow(clippy::clippy::collapsible_if))]
             unordered_drain_filter(&mut self.docsets, |docset| {
-                docset.seek(target) == TERMINATED
+                if docset.doc() < target {
+                    if docset.skip_next(target) == SkipResult::End {
+                        return true;
+                    }
+                }
+                false
             });
 
             // at this point all of the docsets
             // are positionned on a doc >= to the target.
-            if !self.refill() {
-                self.doc = TERMINATED;
-                return TERMINATED;
-            }
-            self.advance()
-        }
-    }
-
-    // TODO Also implement `count` with deletes efficiently.
-
-    fn count_including_deleted(&mut self) -> u32 {
-        if self.doc == TERMINATED {
-            return 0;
-        }
-        let mut count = self.bitsets[self.cursor..HORIZON_NUM_TINYBITSETS]
-            .iter()
-            .map(|bitset| bitset.len())
-            .sum::<u32>()
-            + 1;
-        for bitset in self.bitsets.iter_mut() {
-            bitset.clear();
-        }
-        while self.refill() {
-            count += self.bitsets.iter().map(|bitset| bitset.len()).sum::<u32>();
-            for bitset in self.bitsets.iter_mut() {
-                bitset.clear();
+            if self.refill() {
+                self.advance();
+                if self.doc() == target {
+                    SkipResult::Reached
+                } else {
+                    debug_assert!(self.doc() > target);
+                    SkipResult::OverStep
+                }
+            } else {
+                SkipResult::End
             }
         }
-        self.cursor = HORIZON_NUM_TINYBITSETS;
-        count
     }
 
     fn doc(&self) -> DocId {
@@ -225,11 +248,7 @@ where
     }
 
     fn size_hint(&self) -> u32 {
-        self.docsets
-            .iter()
-            .map(|docset| docset.size_hint())
-            .max()
-            .unwrap_or(0u32)
+        0u32
     }
 }
 
@@ -248,7 +267,7 @@ mod tests {
 
     use super::Union;
     use super::HORIZON;
-    use crate::docset::{DocSet, TERMINATED};
+    use crate::docset::{DocSet, SkipResult};
     use crate::postings::tests::test_skip_against_unoptimized;
     use crate::query::score_combiner::DoNothingCombiner;
     use crate::query::ConstScorer;
@@ -271,18 +290,18 @@ mod tests {
                 vals.iter()
                     .cloned()
                     .map(VecDocSet::from)
-                    .map(|docset| ConstScorer::new(docset, 1.0f32))
+                    .map(ConstScorer::new)
                     .collect::<Vec<ConstScorer<VecDocSet>>>(),
             )
         };
         let mut union: Union<_, DoNothingCombiner> = make_union();
         let mut count = 0;
-        while union.doc() != TERMINATED {
+        while union.advance() {
+            assert!(union_expected.advance());
             assert_eq!(union_expected.doc(), union.doc());
-            assert_eq!(union_expected.advance(), union.advance());
             count += 1;
         }
-        assert_eq!(union_expected.advance(), TERMINATED);
+        assert!(!union_expected.advance());
         assert_eq!(count, make_union().count_including_deleted());
     }
 
@@ -310,7 +329,9 @@ mod tests {
     fn test_aux_union_skip(docs_list: &[Vec<DocId>], skip_targets: Vec<DocId>) {
         let mut btree_set = BTreeSet::new();
         for docs in docs_list {
-            btree_set.extend(docs.iter().cloned());
+            for &doc in docs.iter() {
+                btree_set.insert(doc);
+            }
         }
         let docset_factory = || {
             let res: Box<dyn DocSet> = Box::new(Union::<_, DoNothingCombiner>::from(
@@ -318,17 +339,17 @@ mod tests {
                     .iter()
                     .map(|docs| docs.clone())
                     .map(VecDocSet::from)
-                    .map(|docset| ConstScorer::new(docset, 1.0f32))
+                    .map(ConstScorer::new)
                     .collect::<Vec<_>>(),
             ));
             res
         };
         let mut docset = docset_factory();
         for el in btree_set {
+            assert!(docset.advance());
             assert_eq!(el, docset.doc());
-            docset.advance();
         }
-        assert_eq!(docset.doc(), TERMINATED);
+        assert!(!docset.advance());
         test_skip_against_unoptimized(docset_factory, skip_targets);
     }
 
@@ -348,13 +369,13 @@ mod tests {
     #[test]
     fn test_union_skip_corner_case3() {
         let mut docset = Union::<_, DoNothingCombiner>::from(vec![
-            ConstScorer::from(VecDocSet::from(vec![0u32, 5u32])),
-            ConstScorer::from(VecDocSet::from(vec![1u32, 4u32])),
+            ConstScorer::new(VecDocSet::from(vec![0u32, 5u32])),
+            ConstScorer::new(VecDocSet::from(vec![1u32, 4u32])),
         ]);
+        assert!(docset.advance());
         assert_eq!(docset.doc(), 0u32);
-        assert_eq!(docset.seek(0u32), 0u32);
-        assert_eq!(docset.seek(0u32), 0u32);
-        assert_eq!(docset.doc(), 0u32)
+        assert_eq!(docset.skip_next(0u32), SkipResult::OverStep);
+        assert_eq!(docset.doc(), 1u32)
     }
 
     #[test]

src/query/vec_docset.rs

@@ -1,8 +1,9 @@
 #![allow(dead_code)]
 
 use crate::common::HasLen;
-use crate::docset::{DocSet, TERMINATED};
+use crate::docset::DocSet;
 use crate::DocId;
+use std::num::Wrapping;
 
 /// Simulate a `Postings` objects from a `VecPostings`.
 /// `VecPostings` only exist for testing purposes.
@@ -11,30 +12,26 @@ use crate::DocId;
 /// No positions are returned.
 pub struct VecDocSet {
     doc_ids: Vec<DocId>,
-    cursor: usize,
+    cursor: Wrapping<usize>,
 }
 
 impl From<Vec<DocId>> for VecDocSet {
     fn from(doc_ids: Vec<DocId>) -> VecDocSet {
-        VecDocSet { doc_ids, cursor: 0 }
+        VecDocSet {
+            doc_ids,
+            cursor: Wrapping(usize::max_value()),
+        }
     }
 }
 
 impl DocSet for VecDocSet {
-    fn advance(&mut self) -> DocId {
-        self.cursor += 1;
-        if self.cursor >= self.doc_ids.len() {
-            self.cursor = self.doc_ids.len();
-            return TERMINATED;
-        }
-        self.doc()
+    fn advance(&mut self) -> bool {
+        self.cursor += Wrapping(1);
+        self.doc_ids.len() > self.cursor.0
     }
 
     fn doc(&self) -> DocId {
-        if self.cursor == self.doc_ids.len() {
-            return TERMINATED;
-        }
-        self.doc_ids[self.cursor]
+        self.doc_ids[self.cursor.0]
     }
 
     fn size_hint(&self) -> u32 {
@@ -52,21 +49,22 @@ impl HasLen for VecDocSet {
 pub mod tests {
 
     use super::*;
-    use crate::docset::DocSet;
+    use crate::docset::{DocSet, SkipResult};
     use crate::DocId;
 
     #[test]
     pub fn test_vec_postings() {
         let doc_ids: Vec<DocId> = (0u32..1024u32).map(|e| e * 3).collect();
         let mut postings = VecDocSet::from(doc_ids);
+        assert!(postings.advance());
         assert_eq!(postings.doc(), 0u32);
-        assert_eq!(postings.advance(), 3u32);
+        assert!(postings.advance());
         assert_eq!(postings.doc(), 3u32);
-        assert_eq!(postings.seek(14u32), 15u32);
+        assert_eq!(postings.skip_next(14u32), SkipResult::OverStep);
         assert_eq!(postings.doc(), 15u32);
-        assert_eq!(postings.seek(300u32), 300u32);
+        assert_eq!(postings.skip_next(300u32), SkipResult::Reached);
         assert_eq!(postings.doc(), 300u32);
-        assert_eq!(postings.seek(6000u32), TERMINATED);
+        assert_eq!(postings.skip_next(6000u32), SkipResult::End);
     }
 
     #[test]

src/query/weight.rs

@@ -1,7 +1,7 @@
 use super::Scorer;
 use crate::core::SegmentReader;
 use crate::query::Explanation;
-use crate::DocId;
+use crate::{DocId, Result};
 
 /// A Weight is the specialization of a Query
 /// for a given set of segments.
@@ -9,18 +9,15 @@ use crate::DocId;
 /// See [`Query`](./trait.Query.html).
 pub trait Weight: Send + Sync + 'static {
     /// Returns the scorer for the given segment.
-    ///
-    /// `boost` is a multiplier to apply to the score.
-    ///
     /// See [`Query`](./trait.Query.html).
-    fn scorer(&self, reader: &SegmentReader, boost: f32) -> crate::Result<Box<dyn Scorer>>;
+    fn scorer(&self, reader: &SegmentReader) -> Result<Box<dyn Scorer>>;
 
     /// Returns an `Explanation` for the given document.
-    fn explain(&self, reader: &SegmentReader, doc: DocId) -> crate::Result<Explanation>;
+    fn explain(&self, reader: &SegmentReader, doc: DocId) -> Result<Explanation>;
 
     /// Returns the number documents within the given `SegmentReader`.
-    fn count(&self, reader: &SegmentReader) -> crate::Result<u32> {
-        let mut scorer = self.scorer(reader, 1.0f32)?;
+    fn count(&self, reader: &SegmentReader) -> Result<u32> {
+        let mut scorer = self.scorer(reader)?;
         if let Some(delete_bitset) = reader.delete_bitset() {
             Ok(scorer.count(delete_bitset))
         } else {

src/reader/mod.rs

@@ -9,7 +9,6 @@ use crate::directory::META_LOCK;
 use crate::Index;
 use crate::Searcher;
 use crate::SegmentReader;
-use std::convert::TryInto;
 use std::sync::Arc;
 
 /// Defines when a new version of the index should be reloaded.
@@ -61,6 +60,7 @@ impl IndexReaderBuilder {
     /// Building the reader is a non-trivial operation that requires
     /// to open different segment readers. It may take hundreds of milliseconds
     /// of time and it may return an error.
+    /// TODO(pmasurel) Use the `TryInto` trait once it is available in stable.
     pub fn try_into(self) -> crate::Result<IndexReader> {
         let inner_reader = InnerIndexReader {
             index: self.index,
@@ -113,14 +113,6 @@ impl IndexReaderBuilder {
     }
 }
 
-impl TryInto<IndexReader> for IndexReaderBuilder {
-    type Error = crate::TantivyError;
-
-    fn try_into(self) -> crate::Result<IndexReader> {
-        IndexReaderBuilder::try_into(self)
-    }
-}
-
 struct InnerIndexReader {
     num_searchers: usize,
     searcher_pool: Pool<Searcher>,

src/reader/pool.rs

@@ -68,9 +68,7 @@ impl<T> Pool<T> {
     /// After publish, all new `Searcher` acquired will be
     /// of the new generation.
     pub fn publish_new_generation(&self, items: Vec<T>) {
-        assert!(!items.is_empty());
         let next_generation = self.next_generation.fetch_add(1, Ordering::SeqCst) + 1;
-        let num_items = items.len();
         for item in items {
             let gen_item = GenerationItem {
                 item,
@@ -79,23 +77,6 @@ impl<T> Pool<T> {
             self.queue.push(gen_item);
         }
         self.advertise_generation(next_generation);
-        // Purge possible previous searchers.
-        //
-        // Assuming at this point no searcher is held more than duration T by the user,
-        // this guarantees that an obsolete searcher will not be uselessly held (and its associated
-        // mmap) for more than duration T.
-        //
-        // Proof: At this point, obsolete searcher that are held by the user will be held for less
-        // than T. When released, they will be dropped as their generation is detected obsolete.
-        //
-        // We still need to ensure that the searcher that are obsolete and in the pool get removed.
-        // The queue currently contains up to 2n searchers, in any random order.
-        //
-        // Half of them are obsoletes. By requesting `(n+1)` fresh searchers, we ensure that all
-        // searcher will be inspected.
-        for _ in 0..=num_items {
-            let _ = self.acquire();
-        }
     }
 
     /// At the exit of this method,

src/schema/document.rs

@@ -3,8 +3,8 @@ use crate::common::BinarySerializable;
 use crate::common::VInt;
 use crate::tokenizer::PreTokenizedString;
 use crate::DateTime;
+use itertools::Itertools;
 use std::io::{self, Read, Write};
-use std::mem;
 
 /// Tantivy's Document is the object that can
 /// be indexed and then searched for.
@@ -16,7 +16,7 @@ use std::mem;
 
 /// Documents are really just a list of couple `(field, value)`.
 /// In this list, one field may appear more than once.
-#[derive(Clone, Debug, serde::Serialize, serde::Deserialize, Default)]
+#[derive(Clone, Debug, Serialize, Deserialize, Default)]
 pub struct Document {
     field_values: Vec<FieldValue>,
 }
@@ -131,34 +131,12 @@ impl Document {
     pub fn get_sorted_field_values(&self) -> Vec<(Field, Vec<&FieldValue>)> {
         let mut field_values: Vec<&FieldValue> = self.field_values().iter().collect();
         field_values.sort_by_key(|field_value| field_value.field());
-
-        let mut grouped_field_values = vec![];
-
-        let mut current_field;
-        let mut current_group;
-
-        let mut field_values_it = field_values.into_iter();
-        if let Some(field_value) = field_values_it.next() {
-            current_field = field_value.field();
-            current_group = vec![field_value]
-        } else {
-            return grouped_field_values;
-        }
-
-        for field_value in field_values_it {
-            if field_value.field() == current_field {
-                current_group.push(field_value);
-            } else {
-                grouped_field_values.push((
-                    current_field,
-                    mem::replace(&mut current_group, vec![field_value]),
-                ));
-                current_field = field_value.field();
-            }
-        }
-
-        grouped_field_values.push((current_field, current_group));
-        grouped_field_values
+        field_values
+            .into_iter()
+            .group_by(|field_value| field_value.field())
+            .into_iter()
+            .map(|(key, group)| (key, group.collect()))
+            .collect::<Vec<(Field, Vec<&FieldValue>)>>()
     }
 
     /// Returns all of the `FieldValue`s associated the given field

src/schema/facet.rs

@@ -125,7 +125,7 @@ impl Facet {
 
     /// This function is the inverse of Facet::from(&str).
     pub fn to_path_string(&self) -> String {
-        format!("{}", self)
+        format!("{}", self.to_string())
     }
 }
 

src/schema/field.rs

@@ -5,20 +5,18 @@ use std::io::Write;
 
 /// `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, serde::Serialize, serde::Deserialize,
-)]
+#[derive(Copy, Clone, Debug, PartialEq, PartialOrd, Eq, Ord, Hash, Serialize, Deserialize)]
 pub struct Field(u32);
 
 impl Field {
     /// Create a new field object for the given FieldId.
-    pub const fn from_field_id(field_id: u32) -> Field {
+    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 const fn field_id(&self) -> u32 {
+    pub fn field_id(&self) -> u32 {
         self.0
     }
 }

src/schema/field_value.rs

@@ -1,10 +1,12 @@
 use crate::common::BinarySerializable;
 use crate::schema::Field;
 use crate::schema::Value;
-use std::io::{self, Read, Write};
+use std::io;
+use std::io::Read;
+use std::io::Write;
 
 /// `FieldValue` holds together a `Field` and its `Value`.
-#[derive(Debug, Clone, Ord, PartialEq, Eq, PartialOrd, serde::Serialize, serde::Deserialize)]
+#[derive(Debug, Clone, Ord, PartialEq, Eq, PartialOrd, Serialize, Deserialize)]
 pub struct FieldValue {
     field: Field,
     value: Value,

src/schema/index_record_option.rs

@@ -1,5 +1,3 @@
-use serde::{Deserialize, Serialize};
-
 /// `IndexRecordOption` describes an amount information associated
 /// to a given indexed field.
 ///

src/schema/int_options.rs

@@ -1,5 +1,4 @@
 use crate::schema::flags::{FastFlag, IndexedFlag, SchemaFlagList, StoredFlag};
-use serde::{Deserialize, Serialize};
 use std::ops::BitOr;
 
 /// Express whether a field is single-value or multi-valued.

src/schema/named_field_document.rs

@@ -1,5 +1,4 @@
 use crate::schema::Value;
-use serde::Serialize;
 use std::collections::BTreeMap;
 
 /// Internal representation of a document used for JSON

src/schema/text_options.rs

@@ -1,7 +1,6 @@
 use crate::schema::flags::SchemaFlagList;
 use crate::schema::flags::StoredFlag;
 use crate::schema::IndexRecordOption;
-use serde::{Deserialize, Serialize};
 use std::borrow::Cow;
 use std::ops::BitOr;
 
@@ -156,17 +155,30 @@ mod tests {
 
     #[test]
     fn test_field_options() {
-        let field_options = STORED | TEXT;
-        assert!(field_options.is_stored());
-        assert!(field_options.get_indexing_options().is_some());
-        let mut schema_builder = Schema::builder();
-        schema_builder.add_text_field("body", TEXT);
-        let schema = schema_builder.build();
-        let field = schema.get_field("body").unwrap();
-        let field_entry = schema.get_field_entry(field);
-        assert!(matches!(field_entry.field_type(),
-                &FieldType::Str(ref text_options)
-                if text_options.get_indexing_options().unwrap().tokenizer() == "default"));
+        {
+            let field_options = STORED | TEXT;
+            assert!(field_options.is_stored());
+            assert!(field_options.get_indexing_options().is_some());
+        }
+        {
+            let mut schema_builder = Schema::builder();
+            schema_builder.add_text_field("body", TEXT);
+            let schema = schema_builder.build();
+            let field = schema.get_field("body").unwrap();
+            let field_entry = schema.get_field_entry(field);
+            match field_entry.field_type() {
+                &FieldType::Str(ref text_options) => {
+                    assert!(text_options.get_indexing_options().is_some());
+                    assert_eq!(
+                        text_options.get_indexing_options().unwrap().tokenizer(),
+                        "default"
+                    );
+                }
+                _ => {
+                    panic!("");
+                }
+            }
+        }
     }
 
     #[test]

src/space_usage/mod.rs

@@ -11,7 +11,6 @@ under-count actual resultant space usage by up to 4095 bytes per file.
 
 use crate::schema::Field;
 use crate::SegmentComponent;
-use serde::{Deserialize, Serialize};
 use std::collections::HashMap;
 
 /// Indicates space usage in bytes

src/store/compression_snap.rs

@@ -1,3 +1,5 @@
+use snap;
+
 use std::io::{self, Read, Write};
 
 /// Name of the compression scheme used in the doc store.
@@ -7,7 +9,7 @@ pub const COMPRESSION: &str = "snappy";
 
 pub fn compress(uncompressed: &[u8], compressed: &mut Vec<u8>) -> io::Result<()> {
     compressed.clear();
-    let mut encoder = snap::write::FrameEncoder::new(compressed);
+    let mut encoder = snap::Writer::new(compressed);
     encoder.write_all(&uncompressed)?;
     encoder.flush()?;
     Ok(())
@@ -15,6 +17,6 @@ pub fn compress(uncompressed: &[u8], compressed: &mut Vec<u8>) -> io::Result<()>
 
 pub fn decompress(compressed: &[u8], decompressed: &mut Vec<u8>) -> io::Result<()> {
     decompressed.clear();
-    snap::read::FrameDecoder::new(compressed).read_to_end(decompressed)?;
+    snap::Reader::new(compressed).read_to_end(decompressed)?;
     Ok(())
 }

src/termdict/mod.rs

@@ -434,7 +434,6 @@ mod tests {
 
     #[test]
     fn test_automaton_search() {
-        use crate::query::DFAWrapper;
         use levenshtein_automata::LevenshteinAutomatonBuilder;
 
         const COUNTRIES: [&'static str; 7] = [
@@ -464,7 +463,7 @@ mod tests {
 
         // We can now build an entire dfa.
         let lev_automaton_builder = LevenshteinAutomatonBuilder::new(2, true);
-        let automaton = DFAWrapper(lev_automaton_builder.build_dfa("Spaen"));
+        let automaton = lev_automaton_builder.build_dfa("Spaen");
 
         let mut range = term_dict.search(automaton).into_stream();
 

src/termdict/termdict.rs

@@ -7,6 +7,7 @@ use crate::postings::TermInfo;
 use crate::termdict::TermOrdinal;
 use once_cell::sync::Lazy;
 use std::io::{self, Write};
+use tantivy_fst;
 use tantivy_fst::raw::Fst;
 use tantivy_fst::Automaton;
 

src/tokenizer/stemmer.rs

@@ -1,7 +1,6 @@
 use super::{Token, TokenFilter, TokenStream};
 use crate::tokenizer::BoxTokenStream;
 use rust_stemmers::{self, Algorithm};
-use serde::{Deserialize, Serialize};
 
 /// Available stemmer languages.
 #[derive(Debug, Serialize, Deserialize, Eq, PartialEq, Copy, Clone)]

src/tokenizer/tokenized_string.rs

@@ -1,5 +1,4 @@
 use crate::tokenizer::{BoxTokenStream, Token, TokenStream, TokenStreamChain};
-use serde::{Deserialize, Serialize};
 use std::cmp::Ordering;
 
 /// Struct representing pre-tokenized text

src/tokenizer/tokenizer.rs

@@ -1,5 +1,4 @@
 use crate::tokenizer::TokenStreamChain;
-use serde::{Deserialize, Serialize};
 /// The tokenizer module contains all of the tools used to process
 /// text in `tantivy`.
 use std::borrow::{Borrow, BorrowMut};