added unit test

CHANGELOG.md

@@ -7,8 +7,7 @@ Tantivy 0.11.0
     - Better handling of whitespaces.
 - Closes #498 - add support for Elastic-style unbounded range queries for alphanumeric types eg. "title:>hello", "weight:>=70.5", "height:<200" (@petr-tik)
 - API change around `Box<BoxableTokenizer>`. See detail in #629
-- Avoid rebuilding Regex automaton whenever a regex query is reused. #639 (@brainlock)
-- Add footer with some metadata to index files. #605 (@fdb-hiroshima)
+- Avoid rebuilding Regex automaton whenever a regex query is reused. #630 (@brainlock)
  
 ## How to update?
 
@@ -16,12 +15,6 @@ Tantivy 0.11.0
 - Regex are now compiled when the `RegexQuery` instance is built. As a result, it can now return
 an error and handling the `Result` is required.
 
-
-Tantivy 0.10.2
-=====================
-
-- Closes #656. Solving memory leak.
-
 Tantivy 0.10.1
 =====================
 

Cargo.toml

@@ -15,13 +15,13 @@ edition = "2018"
 [dependencies]
 base64 = "0.10.0"
 byteorder = "1.0"
-crc32fast = "1.2.0"
 once_cell = "1.0"
 regex ={version = "1.3.0", default-features = false, features = ["std"]}
 tantivy-fst = "0.1"
 memmap = {version = "0.7", optional=true}
 lz4 = {version="1.20", optional=true}
 snap = {version="0.2"}
+derive_builder = "0.7"
 atomicwrites = {version="0.2.2", optional=true}
 tempfile = "3.0"
 log = "0.4"
@@ -31,7 +31,7 @@ serde_json = "1.0"
 num_cpus = "1.2"
 fs2={version="0.4", optional=true}
 itertools = "0.8"
-levenshtein_automata = {version="0.1", features=["fst_automaton"]}
+levenshtein_automata = "0.1"
 notify = {version="4", optional=true}
 bit-set = "0.5"
 uuid = { version = "0.7.2", features = ["v4", "serde"] }
@@ -82,7 +82,7 @@ unstable = [] # useful for benches.
 wasm-bindgen = ["uuid/wasm-bindgen"]
 
 [workspace]
-members = ["query-grammar"]
+members = ["query-grammar", "incremental-search"]
 
 [badges]
 travis-ci = { repository = "tantivy-search/tantivy" }

Makefile

@@ -1,3 +1,3 @@
 test:
 	echo "Run test only... No examples."
-	cargo test --tests --lib
+	cargo test --all --tests --lib

incremental-search/Cargo.toml

@@ -0,0 +1,10 @@
+[package]
+name = "incremental-search"
+version = "0.11.0"
+authors = ["Paul Masurel <paul.masurel@gmail.com>"]
+edition = "2018"
+
+
+[dependencies]
+derive_builder = "0.7"
+tantivy = {path = ".."}

incremental-search/src/bitset.rs

@@ -0,0 +1,395 @@
+use std::fmt;
+use std::u64;
+
+#[derive(Clone, Copy, Eq, PartialEq)]
+pub(crate) struct TinySet(u64);
+
+impl fmt::Debug for TinySet {
+    fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
+        self.into_iter().collect::<Vec<u32>>().fmt(f)
+    }
+}
+
+pub struct TinySetIterator(TinySet);
+impl Iterator for TinySetIterator {
+    type Item = u32;
+
+    fn next(&mut self) -> Option<Self::Item> {
+        self.0.pop_lowest()
+    }
+}
+
+impl IntoIterator for TinySet {
+    type Item = u32;
+    type IntoIter = TinySetIterator;
+    fn into_iter(self) -> Self::IntoIter {
+        TinySetIterator(self)
+    }
+}
+
+impl TinySet {
+    /// Returns an empty `TinySet`.
+    pub fn empty() -> TinySet {
+        TinySet(0u64)
+    }
+
+    /// Returns the complement of the set in `[0, 64[`.
+    fn complement(self) -> TinySet {
+        TinySet(!self.0)
+    }
+
+    /// Returns true iff the `TinySet` contains the element `el`.
+    pub fn contains(self, el: u32) -> bool {
+        !self.intersect(TinySet::singleton(el)).is_empty()
+    }
+
+    /// Returns the intersection of `self` and `other`
+    pub fn intersect(self, other: TinySet) -> TinySet {
+        TinySet(self.0 & other.0)
+    }
+
+    /// Creates a new `TinySet` containing only one element
+    /// within `[0; 64[`
+    #[inline(always)]
+    pub fn singleton(el: u32) -> TinySet {
+        TinySet(1u64 << u64::from(el))
+    }
+
+    /// Insert a new element within [0..64[
+    #[inline(always)]
+    pub fn insert(self, el: u32) -> TinySet {
+        self.union(TinySet::singleton(el))
+    }
+
+    /// Insert a new element within [0..64[
+    #[inline(always)]
+    pub fn insert_mut(&mut self, el: u32) -> bool {
+        let old = *self;
+        *self = old.insert(el);
+        old != *self
+    }
+
+    /// Returns the union of two tinysets
+    #[inline(always)]
+    pub fn union(self, other: TinySet) -> TinySet {
+        TinySet(self.0 | other.0)
+    }
+
+    /// Returns true iff the `TinySet` is empty.
+    #[inline(always)]
+    pub fn is_empty(self) -> bool {
+        self.0 == 0u64
+    }
+
+    /// Returns the lowest element in the `TinySet`
+    /// and removes it.
+    #[inline(always)]
+    pub fn pop_lowest(&mut self) -> Option<u32> {
+        if self.is_empty() {
+            None
+        } else {
+            let lowest = self.0.trailing_zeros() as u32;
+            self.0 ^= TinySet::singleton(lowest).0;
+            Some(lowest)
+        }
+    }
+
+    /// Returns a `TinySet` than contains all values up
+    /// to limit excluded.
+    ///
+    /// The limit is assumed to be strictly lower than 64.
+    pub fn range_lower(upper_bound: u32) -> TinySet {
+        TinySet((1u64 << u64::from(upper_bound % 64u32)) - 1u64)
+    }
+
+    /// Returns a `TinySet` that contains all values greater
+    /// or equal to the given limit, included. (and up to 63)
+    ///
+    /// The limit is assumed to be strictly lower than 64.
+    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, //< Technically it should be u32, but we
+    // count multiple inserts.
+    // `usize` guards us from overflow.
+    max_value: u32,
+}
+
+fn num_buckets(max_val: u32) -> u32 {
+    (max_val + 63u32) / 64u32
+}
+
+impl BitSet {
+    /// Create a new `BitSet` that may contain elements
+    /// within `[0, max_val[`.
+    pub fn with_max_value(max_value: u32) -> BitSet {
+        let num_buckets = num_buckets(max_value);
+        let tinybisets = vec![TinySet::empty(); num_buckets as usize].into_boxed_slice();
+        BitSet {
+            tinysets: tinybisets,
+            len: 0,
+            max_value,
+        }
+    }
+
+    /// Removes all elements from the `BitSet`.
+    pub fn clear(&mut self) {
+        for tinyset in self.tinysets.iter_mut() {
+            *tinyset = TinySet::empty();
+        }
+    }
+
+    /// Returns the number of elements in the `BitSet`.
+    pub fn len(&self) -> usize {
+        self.len
+    }
+
+    /// Inserts an element in the `BitSet`
+    pub fn insert(&mut self, el: u32) {
+        // we do not check saturated els.
+        let higher = el / 64u32;
+        let lower = el % 64u32;
+        self.len += if self.tinysets[higher as usize].insert_mut(lower) {
+            1
+        } else {
+            0
+        };
+    }
+
+    /// Returns true iff the elements is in the `BitSet`.
+    pub fn contains(&self, el: u32) -> bool {
+        self.tinyset(el / 64u32).contains(el % 64)
+    }
+
+    /// Returns the first non-empty `TinySet` associated to a bucket lower
+    /// or greater than bucket.
+    ///
+    /// Reminder: the tiny set with the bucket `bucket`, represents the
+    /// elements from `bucket * 64` to `(bucket+1) * 64`.
+    pub(crate) fn first_non_empty_bucket(&self, bucket: u32) -> Option<u32> {
+        self.tinysets[bucket as usize..]
+            .iter()
+            .cloned()
+            .position(|tinyset| !tinyset.is_empty())
+            .map(|delta_bucket| bucket + delta_bucket as u32)
+    }
+
+    pub fn max_value(&self) -> u32 {
+        self.max_value
+    }
+
+    /// Returns the tiny bitset representing the
+    /// the set restricted to the number range from
+    /// `bucket * 64` to `(bucket + 1) * 64`.
+    pub(crate) fn tinyset(&self, bucket: u32) -> TinySet {
+        self.tinysets[bucket as usize]
+    }
+}
+
+#[cfg(test)]
+mod tests {
+
+    use super::BitSet;
+    use super::TinySet;
+    use crate::docset::DocSet;
+    use crate::query::BitSetDocSet;
+    use crate::tests;
+    use crate::tests::generate_nonunique_unsorted;
+    use std::collections::BTreeSet;
+    use std::collections::HashSet;
+
+    #[test]
+    fn test_tiny_set() {
+        assert!(TinySet::empty().is_empty());
+        {
+            let mut u = TinySet::empty().insert(1u32);
+            assert_eq!(u.pop_lowest(), Some(1u32));
+            assert!(u.pop_lowest().is_none())
+        }
+        {
+            let mut u = TinySet::empty().insert(1u32).insert(1u32);
+            assert_eq!(u.pop_lowest(), Some(1u32));
+            assert!(u.pop_lowest().is_none())
+        }
+        {
+            let mut u = TinySet::empty().insert(2u32);
+            assert_eq!(u.pop_lowest(), Some(2u32));
+            u.insert_mut(1u32);
+            assert_eq!(u.pop_lowest(), Some(1u32));
+            assert!(u.pop_lowest().is_none());
+        }
+        {
+            let mut u = TinySet::empty().insert(63u32);
+            assert_eq!(u.pop_lowest(), Some(63u32));
+            assert!(u.pop_lowest().is_none());
+        }
+    }
+
+    #[test]
+    fn test_bitset() {
+        let test_against_hashset = |els: &[u32], max_value: u32| {
+            let mut hashset: HashSet<u32> = HashSet::new();
+            let mut bitset = BitSet::with_max_value(max_value);
+            for &el in els {
+                assert!(el < max_value);
+                hashset.insert(el);
+                bitset.insert(el);
+            }
+            for el in 0..max_value {
+                assert_eq!(hashset.contains(&el), bitset.contains(el));
+            }
+            assert_eq!(bitset.max_value(), max_value);
+        };
+
+        test_against_hashset(&[], 0);
+        test_against_hashset(&[], 1);
+        test_against_hashset(&[0u32], 1);
+        test_against_hashset(&[0u32], 100);
+        test_against_hashset(&[1u32, 2u32], 4);
+        test_against_hashset(&[99u32], 100);
+        test_against_hashset(&[63u32], 64);
+        test_against_hashset(&[62u32, 63u32], 64);
+    }
+
+    #[test]
+    fn test_bitset_large() {
+        let arr = generate_nonunique_unsorted(100_000, 5_000);
+        let mut btreeset: BTreeSet<u32> = BTreeSet::new();
+        let mut bitset = BitSet::with_max_value(100_000);
+        for el in arr {
+            btreeset.insert(el);
+            bitset.insert(el);
+        }
+        for i in 0..100_000 {
+            assert_eq!(btreeset.contains(&i), bitset.contains(i));
+        }
+        assert_eq!(btreeset.len(), bitset.len());
+        let mut bitset_docset = BitSetDocSet::from(bitset);
+        for el in btreeset.into_iter() {
+            bitset_docset.advance();
+            assert_eq!(bitset_docset.doc(), el);
+        }
+        assert!(!bitset_docset.advance());
+    }
+
+    #[test]
+    fn test_bitset_num_buckets() {
+        use super::num_buckets;
+        assert_eq!(num_buckets(0u32), 0);
+        assert_eq!(num_buckets(1u32), 1);
+        assert_eq!(num_buckets(64u32), 1);
+        assert_eq!(num_buckets(65u32), 2);
+        assert_eq!(num_buckets(128u32), 2);
+        assert_eq!(num_buckets(129u32), 3);
+    }
+
+    #[test]
+    fn test_tinyset_range() {
+        assert_eq!(
+            TinySet::range_lower(3).into_iter().collect::<Vec<u32>>(),
+            [0, 1, 2]
+        );
+        assert!(TinySet::range_lower(0).is_empty());
+        assert_eq!(
+            TinySet::range_lower(63).into_iter().collect::<Vec<u32>>(),
+            (0u32..63u32).collect::<Vec<_>>()
+        );
+        assert_eq!(
+            TinySet::range_lower(1).into_iter().collect::<Vec<u32>>(),
+            [0]
+        );
+        assert_eq!(
+            TinySet::range_lower(2).into_iter().collect::<Vec<u32>>(),
+            [0, 1]
+        );
+        assert_eq!(
+            TinySet::range_greater_or_equal(3)
+                .into_iter()
+                .collect::<Vec<u32>>(),
+            (3u32..64u32).collect::<Vec<_>>()
+        );
+    }
+
+    #[test]
+    fn test_bitset_len() {
+        let mut bitset = BitSet::with_max_value(1_000);
+        assert_eq!(bitset.len(), 0);
+        bitset.insert(3u32);
+        assert_eq!(bitset.len(), 1);
+        bitset.insert(103u32);
+        assert_eq!(bitset.len(), 2);
+        bitset.insert(3u32);
+        assert_eq!(bitset.len(), 2);
+        bitset.insert(103u32);
+        assert_eq!(bitset.len(), 2);
+        bitset.insert(104u32);
+        assert_eq!(bitset.len(), 3);
+    }
+
+    #[test]
+    fn test_bitset_clear() {
+        let mut bitset = BitSet::with_max_value(1_000);
+        let els = tests::sample(1_000, 0.01f64);
+        for &el in &els {
+            bitset.insert(el);
+        }
+        assert!(els.iter().all(|el| bitset.contains(*el)));
+        bitset.clear();
+        for el in 0u32..1000u32 {
+            assert!(!bitset.contains(el));
+        }
+    }
+}
+
+#[cfg(all(test, feature = "unstable"))]
+mod bench {
+
+    use super::BitSet;
+    use super::TinySet;
+    use test;
+
+    #[bench]
+    fn bench_tinyset_pop(b: &mut test::Bencher) {
+        b.iter(|| {
+            let mut tinyset = TinySet::singleton(test::black_box(31u32));
+            tinyset.pop_lowest();
+            tinyset.pop_lowest();
+            tinyset.pop_lowest();
+            tinyset.pop_lowest();
+            tinyset.pop_lowest();
+            tinyset.pop_lowest();
+        });
+    }
+
+    #[bench]
+    fn bench_tinyset_sum(b: &mut test::Bencher) {
+        let tiny_set = TinySet::empty().insert(10u32).insert(14u32).insert(21u32);
+        b.iter(|| {
+            assert_eq!(test::black_box(tiny_set).into_iter().sum::<u32>(), 45u32);
+        });
+    }
+
+    #[bench]
+    fn bench_tinyarr_sum(b: &mut test::Bencher) {
+        let v = [10u32, 14u32, 21u32];
+        b.iter(|| test::black_box(v).iter().cloned().sum::<u32>());
+    }
+
+    #[bench]
+    fn bench_bitset_initialize(b: &mut test::Bencher) {
+        b.iter(|| BitSet::with_max_value(1_000_000));
+    }
+}

incremental-search/src/lib.rs

@@ -0,0 +1,266 @@
+use tantivy::query::{BooleanQuery, FuzzyTermQuery, EmptyQuery};
+use derive_builder::Builder;
+use std::str::FromStr;
+use tantivy::query::{FuzzyConfiguration, FuzzyConfigurationBuilder, Query, Occur};
+use tantivy::schema::Field;
+use tantivy::{Searcher, TantivyError, DocAddress, Term, Document};
+use tantivy::collector::TopDocs;
+use std::ops::Deref;
+
+
+#[derive(Debug)]
+pub struct IncrementalSearchQuery {
+    pub terms: Vec<String>,
+    pub last_is_prefix: bool,
+}
+
+impl IncrementalSearchQuery {
+    pub fn fuzzy_configurations(&self) -> Vec<FuzzyConfigurations> {
+        if self.terms.is_empty() {
+            return Vec::default();
+        }
+        let single_term_confs: Vec<FuzzyConfigurationBuilder> = (0u8..3u8)
+            .map(|d: u8| {
+                let mut builder = FuzzyConfigurationBuilder::default();
+                builder.distance(d).transposition_cost_one(true);
+                builder
+            })
+            .collect();
+        let mut configurations: Vec<Vec<FuzzyConfigurationBuilder>> = single_term_confs
+            .iter()
+            .map(|conf| vec![conf.clone()])
+            .collect();
+        let mut new_configurations = Vec::new();
+        for _ in 1..self.terms.len() {
+            new_configurations.clear();
+            for single_term_conf in &single_term_confs {
+                for configuration in &configurations {
+                    let mut new_configuration: Vec<FuzzyConfigurationBuilder> = configuration.clone();
+                    new_configuration.push(single_term_conf.clone());
+                    new_configurations.push(new_configuration);
+                }
+            }
+            std::mem::swap(&mut configurations, &mut new_configurations);
+        }
+        if self.last_is_prefix {
+            for configuration in &mut configurations {
+                if let Some(last_conf) = configuration.last_mut() {
+                    last_conf.prefix(true);
+                }
+            }
+        }
+        let mut fuzzy_configurations: Vec<FuzzyConfigurations> = configurations
+            .into_iter()
+            .map(FuzzyConfigurations::from)
+            .collect();
+        fuzzy_configurations.sort_by(|left, right| left.cost.partial_cmp(&right.cost).unwrap());
+        fuzzy_configurations
+    }
+
+    fn search_query(&self, fields: &[Field], configurations: FuzzyConfigurations) -> Box<dyn Query> {
+        if self.terms.is_empty() {
+            Box::new(EmptyQuery)
+        } else if self.terms.len() == 1 {
+            build_query_for_fields(fields, &self.terms[0], &configurations.configurations[0])
+        } else {
+            Box::new(BooleanQuery::from(self.terms.iter()
+                .zip(configurations.configurations.iter())
+                .map(|(term, configuration)|
+                         (Occur::Must, build_query_for_fields(fields, &term, &configuration))
+                )
+                .collect::<Vec<_>>()))
+        }
+    }
+}
+
+#[derive(Debug)]
+pub struct FuzzyConfigurations {
+    configurations: Vec<FuzzyConfiguration>,
+    cost: f64,
+}
+
+
+fn compute_cost(fuzzy_confs: &[FuzzyConfiguration]) -> f64 {
+    fuzzy_confs
+        .iter()
+        .map(|fuzzy_conf| {
+            let weight = if fuzzy_conf.prefix { 30f64 } else { 5f64 };
+            weight * f64::from(fuzzy_conf.distance)
+        })
+        .sum()
+}
+
+impl From<Vec<FuzzyConfigurationBuilder>> for FuzzyConfigurations {
+    fn from(fuzzy_conf_builder: Vec<FuzzyConfigurationBuilder>) -> FuzzyConfigurations {
+        let configurations = fuzzy_conf_builder
+            .into_iter()
+            .map(|conf| conf.build().unwrap())
+            .collect::<Vec<FuzzyConfiguration>>();
+        let cost = compute_cost(&configurations);
+        FuzzyConfigurations {
+            configurations,
+            cost,
+        }
+    }
+}
+
+#[derive(Debug)]
+pub struct ParseIncrementalQueryError;
+
+impl Into<TantivyError> for ParseIncrementalQueryError {
+    fn into(self) -> TantivyError {
+        TantivyError::InvalidArgument(format!("Invalid query: {:?}", self))
+    }
+}
+
+impl FromStr for IncrementalSearchQuery {
+    type Err = ParseIncrementalQueryError;
+
+    fn from_str(query_str: &str) -> Result<Self, Self::Err> {
+        let terms: Vec<String> = query_str
+            .split_whitespace()
+            .map(ToString::to_string)
+            .collect();
+        Ok(IncrementalSearchQuery {
+            terms,
+            last_is_prefix: query_str
+                .chars()
+                .last()
+                .map(|c| !c.is_whitespace())
+                .unwrap_or(false),
+        })
+    }
+}
+
+fn build_query_for_fields(fields: &[Field], term_text: &str, conf: &FuzzyConfiguration) -> Box<dyn Query> {
+    assert!(fields.len() > 0);
+    if fields.len() > 1 {
+        let term_queries: Vec<(Occur, Box<dyn Query>)> = fields
+            .iter()
+            .map(|&field| {
+                let term = Term::from_field_text(field, term_text);
+                let query = FuzzyTermQuery::new_from_configuration(term, conf.clone());
+                let boxed_query: Box<dyn Query> = Box::new(query);
+                (Occur::Must, boxed_query)
+            })
+            .collect();
+        Box::new(BooleanQuery::from(term_queries))
+    } else {
+        let term = Term::from_field_text(fields[0], term_text);
+        Box::new( FuzzyTermQuery::new_from_configuration(term, conf.clone()))
+    }
+
+}
+
+pub struct IncrementalSearchResult {
+    pub docs: Vec<Document>
+}
+
+#[derive(Builder, Default)]
+pub struct IncrementalSearch {
+    nhits: usize,
+    #[builder(default)]
+    search_fields: Vec<Field>,
+    #[builder(default)]
+    return_fields: Vec<Field>,
+}
+
+impl IncrementalSearch {
+
+    pub fn search<S: Deref<Target=Searcher>>(
+        &self,
+        query: &str,
+        searcher: &S,
+    ) -> tantivy::Result<IncrementalSearchResult> {
+        let searcher = searcher.deref();
+        let inc_search_query: IncrementalSearchQuery =
+            FromStr::from_str(query).map_err(Into::<TantivyError>::into)?;
+
+        let mut results: Vec<DocAddress> = Vec::default();
+        let mut remaining = self.nhits;
+        for fuzzy_conf in inc_search_query.fuzzy_configurations() {
+            if remaining == 0 {
+                break;
+            }
+            let query = inc_search_query.search_query(&self.search_fields[..], fuzzy_conf);
+            let new_docs = searcher.search(query.as_ref(), &TopDocs::with_limit(remaining))?;
+            // TODO(pmasurel) remove already added docs.
+            results.extend(new_docs.into_iter()
+                .map(|(_, doc_address)| doc_address));
+            remaining = self.nhits - results.len();
+            if remaining == 0 {
+                break;
+            }
+        }
+        let docs: Vec<Document> = results.into_iter()
+            .map(|doc_address: DocAddress| searcher.doc(doc_address))
+            .collect::<tantivy::Result<_>>()?;
+        Ok(IncrementalSearchResult {
+            docs
+        })
+    }
+}
+#[cfg(test)]
+mod tests {
+    use tantivy::doc;
+    use crate::{IncrementalSearch, IncrementalSearchBuilder, IncrementalSearchQuery};
+    use std::str::FromStr;
+    use tantivy::schema::{SchemaBuilder, TEXT, STORED};
+    use tantivy::Index;
+
+    #[test]
+    fn test_incremental_search() {
+        let incremental_search = IncrementalSearchBuilder::default()
+            .nhits(10)
+            .build()
+            .unwrap();
+    }
+
+    #[test]
+    fn test_incremental_search_query_parse_empty() {
+        let query = IncrementalSearchQuery::from_str("").unwrap();
+        assert_eq!(query.terms, Vec::<String>::new());
+        assert!(!query.last_is_prefix);
+    }
+
+    #[test]
+    fn test_incremental_search_query_parse_trailing_whitespace() {
+        let query = IncrementalSearchQuery::from_str("hello happy tax pa ").unwrap();
+        assert_eq!(query.terms, vec!["hello", "happy", "tax", "pa"]);
+        assert!(!query.last_is_prefix);
+    }
+
+    #[test]
+    fn test_incremental_search_query_parse_unicode_whitespace() {
+        let query = IncrementalSearchQuery::from_str("hello  happy tax pa ").unwrap();
+        assert_eq!(query.terms, vec!["hello", "happy", "tax", "pa"]);
+        assert!(!query.last_is_prefix);
+    }
+
+    #[test]
+    fn test_incremental_search_query_parse() {
+        let query = IncrementalSearchQuery::from_str("hello happy tax pa").unwrap();
+        assert_eq!(query.terms, vec!["hello", "happy", "tax", "pa"]);
+        assert!(query.last_is_prefix);
+    }
+
+    #[test]
+    fn test_blop() {
+        let mut schema_builder = SchemaBuilder::new();
+        let body = schema_builder.add_text_field("body", TEXT | STORED);
+        let schema = schema_builder.build();
+        let index = Index::create_in_ram(schema);
+        let mut index_writer = index.writer_with_num_threads(1, 30_000_000).unwrap();
+        index_writer.add_document(doc!(body=> "hello happy tax payer"));
+        index_writer.commit().unwrap();
+        let reader = index.reader().unwrap();
+        let searcher = reader.searcher();
+        let incremental_search: IncrementalSearch = IncrementalSearchBuilder::default()
+            .nhits(1)
+            .search_fields(vec![body])
+            .build()
+            .unwrap();
+        let top_docs = incremental_search.search("hello hapy t", &searcher).unwrap();
+        assert_eq!(top_docs.docs.len(), 1);
+    }
+}

query-grammar/src/occur.rs

@@ -1,6 +1,3 @@
-use std::fmt;
-use std::fmt::Write;
-
 /// Defines whether a term in a query must be present,
 /// should be present or must not be present.
 #[derive(Debug, Clone, Hash, Copy, Eq, PartialEq)]
@@ -21,7 +18,7 @@ impl Occur {
     /// - `Should` => '?',
     /// - `Must` => '+'
     /// - `Not` => '-'
-    fn to_char(self) -> char {
+    pub fn to_char(self) -> char {
         match self {
             Occur::Should => '?',
             Occur::Must => '+',
@@ -50,9 +47,3 @@ impl Occur {
         }
     }
 }
-
-impl fmt::Display for Occur {
-    fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
-        f.write_char(self.to_char())
-    }
-}

query-grammar/src/user_input_ast.rs

@@ -151,7 +151,7 @@ impl fmt::Debug for UserInputAST {
                 Ok(())
             }
             UserInputAST::Unary(ref occur, ref subquery) => {
-                write!(formatter, "{}({:?})", occur, subquery)
+                write!(formatter, "{}({:?})", occur.to_char(), subquery)
             }
             UserInputAST::Leaf(ref subquery) => write!(formatter, "{:?}", subquery),
         }

src/collector/count_collector.rs

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

src/collector/facet_collector.rs

@@ -599,18 +599,19 @@ mod tests {
             );
         }
     }
+
 }
 
 #[cfg(all(test, feature = "unstable"))]
 mod bench {
 
-    use crate::collector::FacetCollector;
-    use crate::query::AllQuery;
-    use crate::schema::{Facet, Schema};
-    use crate::Index;
-    use rand::seq::SliceRandom;
-    use rand::thread_rng;
+    use collector::FacetCollector;
+    use query::AllQuery;
+    use rand::{thread_rng, Rng};
+    use schema::Facet;
+    use schema::Schema;
     use test::Bencher;
+    use Index;
 
     #[bench]
     fn bench_facet_collector(b: &mut Bencher) {
@@ -627,7 +628,7 @@ mod bench {
             }
         }
         // 40425 docs
-        docs[..].shuffle(&mut thread_rng());
+        thread_rng().shuffle(&mut docs[..]);
 
         let mut index_writer = index.writer_with_num_threads(1, 3_000_000).unwrap();
         for doc in docs {
@@ -636,7 +637,7 @@ mod bench {
         index_writer.commit().unwrap();
         let reader = index.reader().unwrap();
         b.iter(|| {
-            let searcher = reader.searcher();
+            let searcher = index.searcher();
             let facet_collector = FacetCollector::for_field(facet_field);
             searcher.search(&AllQuery, &facet_collector).unwrap();
         });

src/collector/top_score_collector.rs

@@ -592,4 +592,5 @@ mod tests {
         let query = query_parser.parse_query(query).unwrap();
         (index, query)
     }
+
 }

src/common/composite_file.rs

@@ -2,7 +2,7 @@ use crate::common::BinarySerializable;
 use crate::common::CountingWriter;
 use crate::common::VInt;
 use crate::directory::ReadOnlySource;
-use crate::directory::{TerminatingWrite, WritePtr};
+use crate::directory::WritePtr;
 use crate::schema::Field;
 use crate::space_usage::FieldUsage;
 use crate::space_usage::PerFieldSpaceUsage;
@@ -42,7 +42,7 @@ pub struct CompositeWrite<W = WritePtr> {
     offsets: HashMap<FileAddr, u64>,
 }
 
-impl<W: TerminatingWrite + Write> CompositeWrite<W> {
+impl<W: Write> CompositeWrite<W> {
     /// Crate a new API writer that writes a composite file
     /// in a given write.
     pub fn wrap(w: W) -> CompositeWrite<W> {
@@ -91,7 +91,8 @@ impl<W: TerminatingWrite + Write> CompositeWrite<W> {
 
         let footer_len = (self.write.written_bytes() - footer_offset) as u32;
         footer_len.serialize(&mut self.write)?;
-        self.write.terminate()
+        self.write.flush()?;
+        Ok(())
     }
 }
 
@@ -230,4 +231,5 @@ mod test {
             }
         }
     }
+
 }

src/common/counting_writer.rs

@@ -1,5 +1,3 @@
-use crate::directory::AntiCallToken;
-use crate::directory::TerminatingWrite;
 use std::io;
 use std::io::Write;
 
@@ -44,13 +42,6 @@ impl<W: Write> Write for CountingWriter<W> {
     }
 }
 
-impl<W: TerminatingWrite> TerminatingWrite for CountingWriter<W> {
-    fn terminate_ref(&mut self, token: AntiCallToken) -> io::Result<()> {
-        self.flush()?;
-        self.underlying.terminate_ref(token)
-    }
-}
-
 #[cfg(test)]
 mod test {
 

src/common/serialize.rs

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

src/core/index.rs

@@ -26,10 +26,9 @@ use crate::IndexWriter;
 use crate::Result;
 use num_cpus;
 use std::borrow::BorrowMut;
-use std::collections::HashSet;
 use std::fmt;
 #[cfg(feature = "mmap")]
-use std::path::{Path, PathBuf};
+use std::path::Path;
 use std::sync::Arc;
 
 fn load_metas(directory: &dyn Directory, inventory: &SegmentMetaInventory) -> Result<IndexMeta> {
@@ -369,11 +368,6 @@ impl Index {
             .map(SegmentMeta::id)
             .collect())
     }
-
-    /// Returns the set of corrupted files
-    pub fn validate_checksum(&self) -> Result<HashSet<PathBuf>> {
-        self.directory.list_damaged().map_err(Into::into)
-    }
 }
 
 impl fmt::Debug for Index {
@@ -601,4 +595,5 @@ mod tests {
         assert_eq!(searcher.num_docs(), 8_000);
         assert!(mem_right_after_merge_finished < mem_right_after_commit);
     }
+
 }

src/directory/directory.rs

@@ -118,8 +118,6 @@ pub trait Directory: DirectoryClone + fmt::Debug + Send + Sync + 'static {
     ///
     /// Specifically, subsequent writes or flushes should
     /// have no effect on the returned `ReadOnlySource` object.
-    ///
-    /// You should only use this to read files create with [`open_write`]
     fn open_read(&self, path: &Path) -> result::Result<ReadOnlySource, OpenReadError>;
 
     /// Removes a file
@@ -159,8 +157,6 @@ pub trait Directory: DirectoryClone + fmt::Debug + Send + Sync + 'static {
     /// atomic_write.
     ///
     /// This should only be used for small files.
-    ///
-    /// You should only use this to read files create with [`atomic_write`]
     fn atomic_read(&self, path: &Path) -> Result<Vec<u8>, OpenReadError>;
 
     /// Atomically replace the content of a file with data.

src/directory/footer.rs

@@ -1,213 +0,0 @@
-use crate::directory::read_only_source::ReadOnlySource;
-use crate::directory::{AntiCallToken, TerminatingWrite};
-use byteorder::{ByteOrder, LittleEndian};
-use crc32fast::Hasher;
-use std::io;
-use std::io::Write;
-
-const COMMON_FOOTER_SIZE: usize = 4 * 5;
-
-#[derive(Debug, Clone, PartialEq)]
-pub struct Footer {
-    pub tantivy_version: (u32, u32, u32),
-    pub meta: String,
-    pub versioned_footer: VersionedFooter,
-}
-
-impl Footer {
-    pub fn new(versioned_footer: VersionedFooter) -> Self {
-        let tantivy_version = (
-            env!("CARGO_PKG_VERSION_MAJOR").parse().unwrap(),
-            env!("CARGO_PKG_VERSION_MINOR").parse().unwrap(),
-            env!("CARGO_PKG_VERSION_PATCH").parse().unwrap(),
-        );
-        Footer {
-            tantivy_version,
-            meta: format!(
-                "tantivy {}.{}.{}, index v{}",
-                tantivy_version.0,
-                tantivy_version.1,
-                tantivy_version.2,
-                versioned_footer.version()
-            ),
-            versioned_footer,
-        }
-    }
-
-    pub fn to_bytes(&self) -> Vec<u8> {
-        let mut res = self.versioned_footer.to_bytes();
-        res.extend_from_slice(self.meta.as_bytes());
-        let len = res.len();
-        res.resize(len + COMMON_FOOTER_SIZE, 0);
-        let mut common_footer = &mut res[len..];
-        LittleEndian::write_u32(&mut common_footer, self.meta.len() as u32);
-        LittleEndian::write_u32(&mut common_footer[4..], self.tantivy_version.0);
-        LittleEndian::write_u32(&mut common_footer[8..], self.tantivy_version.1);
-        LittleEndian::write_u32(&mut common_footer[12..], self.tantivy_version.2);
-        LittleEndian::write_u32(&mut common_footer[16..], (len + COMMON_FOOTER_SIZE) as u32);
-        res
-    }
-
-    pub fn from_bytes(data: &[u8]) -> Result<Self, io::Error> {
-        let len = data.len();
-        if len < COMMON_FOOTER_SIZE + 4 {
-            // 4 bytes for index version, stored in versioned footer
-            return Err(io::Error::new(
-                    io::ErrorKind::UnexpectedEof,
-                    format!("File corrupted. The footer len must be over 24, while the entire file len is {}", len)
-                    )
-            );
-        }
-
-        let size = LittleEndian::read_u32(&data[len - 4..]) as usize;
-        if len < size as usize {
-            return Err(io::Error::new(
-                io::ErrorKind::UnexpectedEof,
-                format!(
-                    "File corrupted. The footer len is {}, while the entire file len is {}",
-                    size, len
-                ),
-            ));
-        }
-        let footer = &data[len - size as usize..];
-        let meta_len = LittleEndian::read_u32(&footer[size - 20..]) as usize;
-        let tantivy_major = LittleEndian::read_u32(&footer[size - 16..]);
-        let tantivy_minor = LittleEndian::read_u32(&footer[size - 12..]);
-        let tantivy_patch = LittleEndian::read_u32(&footer[size - 8..]);
-        Ok(Footer {
-            tantivy_version: (tantivy_major, tantivy_minor, tantivy_patch),
-            meta: String::from_utf8_lossy(&footer[size - meta_len - 20..size - 20]).into_owned(),
-            versioned_footer: VersionedFooter::from_bytes(&footer[..size - meta_len - 20])?,
-        })
-    }
-
-    pub fn extract_footer(source: ReadOnlySource) -> Result<(Footer, ReadOnlySource), io::Error> {
-        let footer = Footer::from_bytes(source.as_slice())?;
-        let reader = source.slice_to(source.as_slice().len() - footer.size());
-        Ok((footer, reader))
-    }
-
-    pub fn size(&self) -> usize {
-        self.versioned_footer.size() as usize + self.meta.len() + 20
-    }
-}
-
-#[derive(Debug, Clone, PartialEq)]
-pub enum VersionedFooter {
-    UnknownVersion { version: u32, size: u32 },
-    V0(u32), // crc
-}
-
-impl VersionedFooter {
-    pub fn to_bytes(&self) -> Vec<u8> {
-        match self {
-            VersionedFooter::V0(crc) => {
-                let mut res = vec![0; 8];
-                LittleEndian::write_u32(&mut res, 0);
-                LittleEndian::write_u32(&mut res[4..], *crc);
-                res
-            }
-            VersionedFooter::UnknownVersion { .. } => {
-                panic!("Unsupported index should never get serialized");
-            }
-        }
-    }
-
-    pub fn from_bytes(footer: &[u8]) -> Result<Self, io::Error> {
-        assert!(footer.len() >= 4);
-        let version = LittleEndian::read_u32(footer);
-        match version {
-            0 => {
-                if footer.len() == 8 {
-                    Ok(VersionedFooter::V0(LittleEndian::read_u32(&footer[4..])))
-                } else {
-                    Err(io::Error::new(
-                        io::ErrorKind::UnexpectedEof,
-                        format!(
-                            "File corrupted. The versioned footer len is {}, while it should be 8",
-                            footer.len()
-                        ),
-                    ))
-                }
-            }
-            version => Ok(VersionedFooter::UnknownVersion {
-                version,
-                size: footer.len() as u32,
-            }),
-        }
-    }
-
-    pub fn size(&self) -> u32 {
-        match self {
-            VersionedFooter::V0(_) => 8,
-            VersionedFooter::UnknownVersion { size, .. } => *size,
-        }
-    }
-
-    pub fn version(&self) -> u32 {
-        match self {
-            VersionedFooter::V0(_) => 0,
-            VersionedFooter::UnknownVersion { version, .. } => *version,
-        }
-    }
-
-    pub fn crc(&self) -> Option<u32> {
-        match self {
-            VersionedFooter::V0(crc) => Some(*crc),
-            VersionedFooter::UnknownVersion { .. } => None,
-        }
-    }
-}
-
-pub(crate) struct FooterProxy<W: TerminatingWrite> {
-    /// always Some except after terminate call
-    hasher: Option<Hasher>,
-    /// always Some except after terminate call
-    writer: Option<W>,
-}
-
-impl<W: TerminatingWrite> FooterProxy<W> {
-    pub fn new(writer: W) -> Self {
-        FooterProxy {
-            hasher: Some(Hasher::new()),
-            writer: Some(writer),
-        }
-    }
-}
-
-impl<W: TerminatingWrite> Write for FooterProxy<W> {
-    fn write(&mut self, buf: &[u8]) -> io::Result<usize> {
-        let count = self.writer.as_mut().unwrap().write(buf)?;
-        self.hasher.as_mut().unwrap().update(&buf[..count]);
-        Ok(count)
-    }
-
-    fn flush(&mut self) -> io::Result<()> {
-        self.writer.as_mut().unwrap().flush()
-    }
-}
-
-impl<W: TerminatingWrite> TerminatingWrite for FooterProxy<W> {
-    fn terminate_ref(&mut self, _: AntiCallToken) -> io::Result<()> {
-        let crc = self.hasher.take().unwrap().finalize();
-
-        let footer = Footer::new(VersionedFooter::V0(crc)).to_bytes();
-        let mut writer = self.writer.take().unwrap();
-        writer.write_all(&footer)?;
-        writer.terminate()
-    }
-}
-
-#[cfg(test)]
-mod tests {
-    use crate::directory::footer::{Footer, VersionedFooter};
-
-    #[test]
-    fn test_serialize_deserialize_footer() {
-        let crc = 123456;
-        let footer = Footer::new(VersionedFooter::V0(crc));
-        let footer_bytes = footer.to_bytes();
-
-        assert_eq!(Footer::from_bytes(&footer_bytes).unwrap(), footer);
-    }
-}

src/directory/managed_directory.rs

@@ -1,6 +1,5 @@
 use crate::core::MANAGED_FILEPATH;
 use crate::directory::error::{DeleteError, IOError, LockError, OpenReadError, OpenWriteError};
-use crate::directory::footer::{Footer, FooterProxy};
 use crate::directory::DirectoryLock;
 use crate::directory::Lock;
 use crate::directory::META_LOCK;
@@ -9,7 +8,6 @@ use crate::directory::{WatchCallback, WatchHandle};
 use crate::error::DataCorruption;
 use crate::Directory;
 use crate::Result;
-use crc32fast::Hasher;
 use serde_json;
 use std::collections::HashSet;
 use std::io;
@@ -209,59 +207,17 @@ impl ManagedDirectory {
         }
         Ok(())
     }
-
-    /// Verify checksum of a managed file
-    pub fn validate_checksum(&self, path: &Path) -> result::Result<bool, OpenReadError> {
-        let reader = self.directory.open_read(path)?;
-        let (footer, data) = Footer::extract_footer(reader)
-            .map_err(|err| IOError::with_path(path.to_path_buf(), err))?;
-        let mut hasher = Hasher::new();
-        hasher.update(data.as_slice());
-        let crc = hasher.finalize();
-        Ok(footer
-            .versioned_footer
-            .crc()
-            .map(|v| v == crc)
-            .unwrap_or(false))
-    }
-
-    /// List files for which checksum does not match content
-    pub fn list_damaged(&self) -> result::Result<HashSet<PathBuf>, OpenReadError> {
-        let mut hashset = HashSet::new();
-        let managed_paths = self
-            .meta_informations
-            .read()
-            .expect("Managed directory rlock poisoned in list damaged.")
-            .managed_paths
-            .clone();
-
-        for path in managed_paths.into_iter() {
-            if !self.validate_checksum(&path)? {
-                hashset.insert(path);
-            }
-        }
-        Ok(hashset)
-    }
 }
 
 impl Directory for ManagedDirectory {
     fn open_read(&self, path: &Path) -> result::Result<ReadOnlySource, OpenReadError> {
-        let read_only_source = self.directory.open_read(path)?;
-        let (_footer, reader) = Footer::extract_footer(read_only_source)
-            .map_err(|err| IOError::with_path(path.to_path_buf(), err))?;
-        Ok(reader)
+        self.directory.open_read(path)
     }
 
     fn open_write(&mut self, path: &Path) -> result::Result<WritePtr, OpenWriteError> {
         self.register_file_as_managed(path)
             .map_err(|e| IOError::with_path(path.to_owned(), e))?;
-        Ok(io::BufWriter::new(Box::new(FooterProxy::new(
-            self.directory
-                .open_write(path)?
-                .into_inner()
-                .map_err(|_| ())
-                .expect("buffer should be empty"),
-        ))))
+        self.directory.open_write(path)
     }
 
     fn atomic_write(&mut self, path: &Path, data: &[u8]) -> io::Result<()> {
@@ -303,9 +259,8 @@ impl Clone for ManagedDirectory {
 #[cfg(test)]
 mod tests_mmap_specific {
 
-    use crate::directory::{Directory, ManagedDirectory, MmapDirectory, TerminatingWrite};
+    use crate::directory::{Directory, ManagedDirectory, MmapDirectory};
     use std::collections::HashSet;
-    use std::fs::OpenOptions;
     use std::io::Write;
     use std::path::{Path, PathBuf};
     use tempfile::TempDir;
@@ -320,8 +275,8 @@ mod tests_mmap_specific {
         {
             let mmap_directory = MmapDirectory::open(&tempdir_path).unwrap();
             let mut managed_directory = ManagedDirectory::wrap(mmap_directory).unwrap();
-            let write_file = managed_directory.open_write(test_path1).unwrap();
-            write_file.terminate().unwrap();
+            let mut write_file = managed_directory.open_write(test_path1).unwrap();
+            write_file.flush().unwrap();
             managed_directory
                 .atomic_write(test_path2, &[0u8, 1u8])
                 .unwrap();
@@ -355,9 +310,9 @@ mod tests_mmap_specific {
 
         let mmap_directory = MmapDirectory::open(&tempdir_path).unwrap();
         let mut managed_directory = ManagedDirectory::wrap(mmap_directory).unwrap();
-        let mut write = managed_directory.open_write(test_path1).unwrap();
-        write.write_all(&[0u8, 1u8]).unwrap();
-        write.terminate().unwrap();
+        managed_directory
+            .atomic_write(test_path1, &vec![0u8, 1u8])
+            .unwrap();
         assert!(managed_directory.exists(test_path1));
 
         let _mmap_read = managed_directory.open_read(test_path1).unwrap();
@@ -376,38 +331,4 @@ mod tests_mmap_specific {
         }
     }
 
-    #[test]
-    fn test_checksum() {
-        let test_path1: &'static Path = Path::new("some_path_for_test");
-        let test_path2: &'static Path = Path::new("other_test_path");
-
-        let tempdir = TempDir::new().unwrap();
-        let tempdir_path = PathBuf::from(tempdir.path());
-
-        let mmap_directory = MmapDirectory::open(&tempdir_path).unwrap();
-        let mut managed_directory = ManagedDirectory::wrap(mmap_directory).unwrap();
-        let mut write = managed_directory.open_write(test_path1).unwrap();
-        write.write_all(&[0u8, 1u8]).unwrap();
-        write.terminate().unwrap();
-
-        let mut write = managed_directory.open_write(test_path2).unwrap();
-        write.write_all(&[3u8, 4u8, 5u8]).unwrap();
-        write.terminate().unwrap();
-
-        assert!(managed_directory.list_damaged().unwrap().is_empty());
-
-        let mut corrupted_path = tempdir_path.clone();
-        corrupted_path.push(test_path2);
-        let mut file = OpenOptions::new()
-            .write(true)
-            .open(&corrupted_path)
-            .unwrap();
-        file.write_all(&[255u8]).unwrap();
-        file.flush().unwrap();
-        drop(file);
-
-        let damaged = managed_directory.list_damaged().unwrap();
-        assert_eq!(damaged.len(), 1);
-        assert!(damaged.contains(test_path2));
-    }
 }

src/directory/mmap_directory.rs

@@ -11,7 +11,6 @@ use crate::directory::error::{
     DeleteError, IOError, OpenDirectoryError, OpenReadError, OpenWriteError,
 };
 use crate::directory::read_only_source::BoxedData;
-use crate::directory::AntiCallToken;
 use crate::directory::Directory;
 use crate::directory::DirectoryLock;
 use crate::directory::Lock;
@@ -19,7 +18,7 @@ use crate::directory::ReadOnlySource;
 use crate::directory::WatchCallback;
 use crate::directory::WatchCallbackList;
 use crate::directory::WatchHandle;
-use crate::directory::{TerminatingWrite, WritePtr};
+use crate::directory::WritePtr;
 use atomicwrites;
 use memmap::Mmap;
 use std::collections::HashMap;
@@ -142,28 +141,42 @@ impl MmapCache {
     }
 }
 
-struct WatcherWrapper {
+struct InnerWatcherWrapper {
     _watcher: Mutex<notify::RecommendedWatcher>,
-    watcher_router: Arc<WatchCallbackList>,
+    watcher_router: WatchCallbackList,
+}
+
+impl InnerWatcherWrapper {
+    pub fn new(path: &Path) -> Result<(Self, Receiver<notify::RawEvent>), notify::Error> {
+        let (tx, watcher_recv): (Sender<RawEvent>, Receiver<RawEvent>) = channel();
+        // We need to initialize the
+        let mut watcher = notify::raw_watcher(tx)?;
+        watcher.watch(path, RecursiveMode::Recursive)?;
+        let inner = InnerWatcherWrapper {
+            _watcher: Mutex::new(watcher),
+            watcher_router: Default::default(),
+        };
+        Ok((inner, watcher_recv))
+    }
+}
+
+#[derive(Clone)]
+struct WatcherWrapper {
+    inner: Arc<InnerWatcherWrapper>,
 }
 
 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)
-            .and_then(|mut watcher| {
-                watcher.watch(path, RecursiveMode::Recursive)?;
-                Ok(watcher)
-            })
-            .map_err(|err| match err {
-                notify::Error::PathNotFound => OpenDirectoryError::DoesNotExist(path.to_owned()),
-                _ => {
-                    panic!("Unknown error while starting watching directory {:?}", path);
-                }
-            })?;
-        let watcher_router: Arc<WatchCallbackList> = Default::default();
-        let watcher_router_clone = watcher_router.clone();
+        let (inner, watcher_recv) = InnerWatcherWrapper::new(path).map_err(|err| match err {
+            notify::Error::PathNotFound => OpenDirectoryError::DoesNotExist(path.to_owned()),
+            _ => {
+                panic!("Unknown error while starting watching directory {:?}", path);
+            }
+        })?;
+        let watcher_wrapper = WatcherWrapper {
+            inner: Arc::new(inner),
+        };
+        let watcher_wrapper_clone = watcher_wrapper.clone();
         thread::Builder::new()
             .name("meta-file-watch-thread".to_string())
             .spawn(move || {
@@ -174,7 +187,7 @@ impl WatcherWrapper {
                             // We might want to be more accurate than this at one point.
                             if let Some(filename) = changed_path.file_name() {
                                 if filename == *META_FILEPATH {
-                                    watcher_router_clone.broadcast();
+                                    watcher_wrapper_clone.inner.watcher_router.broadcast();
                                 }
                             }
                         }
@@ -187,15 +200,13 @@ impl WatcherWrapper {
                         }
                     }
                 }
-            })?;
-        Ok(WatcherWrapper {
-            _watcher: Mutex::new(watcher),
-            watcher_router,
-        })
+            })
+            .expect("Failed to spawn thread to watch meta.json");
+        Ok(watcher_wrapper)
     }
 
     pub fn watch(&mut self, watch_callback: WatchCallback) -> WatchHandle {
-        self.watcher_router.subscribe(watch_callback)
+        self.inner.watcher_router.subscribe(watch_callback)
     }
 }
 
@@ -254,7 +265,7 @@ impl MmapDirectoryInner {
             }
         }
         if let Some(watch_wrapper) = self.watcher.write().unwrap().as_mut() {
-            Ok(watch_wrapper.watch(watch_callback))
+            return Ok(watch_wrapper.watch(watch_callback));
         } else {
             unreachable!("At this point, watch wrapper is supposed to be initialized");
         }
@@ -401,12 +412,6 @@ impl Seek for SafeFileWriter {
     }
 }
 
-impl TerminatingWrite for SafeFileWriter {
-    fn terminate_ref(&mut self, _: AntiCallToken) -> io::Result<()> {
-        self.flush()
-    }
-}
-
 impl Directory for MmapDirectory {
     fn open_read(&self, path: &Path) -> result::Result<ReadOnlySource, OpenReadError> {
         debug!("Open Read {:?}", path);

src/directory/mod.rs

@@ -9,7 +9,6 @@ mod mmap_directory;
 
 mod directory;
 mod directory_lock;
-mod footer;
 mod managed_directory;
 mod ram_directory;
 mod read_only_source;
@@ -25,49 +24,18 @@ pub use self::ram_directory::RAMDirectory;
 pub use self::read_only_source::ReadOnlySource;
 pub(crate) use self::watch_event_router::WatchCallbackList;
 pub use self::watch_event_router::{WatchCallback, WatchHandle};
-use std::io::{self, BufWriter, Write};
+use std::io::{BufWriter, Write};
 
 #[cfg(feature = "mmap")]
 pub use self::mmap_directory::MmapDirectory;
 
 pub use self::managed_directory::ManagedDirectory;
 
-/// Struct used to prevent from calling [`terminate_ref`](trait.TerminatingWrite#method.terminate_ref) directly
-pub struct AntiCallToken(());
-
-/// Trait used to indicate when no more write need to be done on a writer
-pub trait TerminatingWrite: Write {
-    /// Indicate that the writer will no longer be used. Internally call terminate_ref.
-    fn terminate(mut self) -> io::Result<()>
-    where
-        Self: Sized,
-    {
-        self.terminate_ref(AntiCallToken(()))
-    }
-
-    /// You should implement this function to define custom behavior.
-    /// This function should flush any buffer it may hold.
-    fn terminate_ref(&mut self, _: AntiCallToken) -> io::Result<()>;
-}
-
-impl<W: TerminatingWrite + ?Sized> TerminatingWrite for Box<W> {
-    fn terminate_ref(&mut self, token: AntiCallToken) -> io::Result<()> {
-        self.as_mut().terminate_ref(token)
-    }
-}
-
-impl<W: TerminatingWrite> TerminatingWrite for BufWriter<W> {
-    fn terminate_ref(&mut self, a: AntiCallToken) -> io::Result<()> {
-        self.flush()?;
-        self.get_mut().terminate_ref(a)
-    }
-}
-
 /// Write object for Directory.
 ///
 /// `WritePtr` are required to implement both Write
 /// and Seek.
-pub type WritePtr = BufWriter<Box<dyn TerminatingWrite>>;
+pub type WritePtr = BufWriter<Box<dyn Write>>;
 
 #[cfg(test)]
 mod tests;

src/directory/ram_directory.rs

@@ -1,9 +1,8 @@
 use crate::core::META_FILEPATH;
 use crate::directory::error::{DeleteError, OpenReadError, OpenWriteError};
-use crate::directory::AntiCallToken;
 use crate::directory::WatchCallbackList;
+use crate::directory::WritePtr;
 use crate::directory::{Directory, ReadOnlySource, WatchCallback, WatchHandle};
-use crate::directory::{TerminatingWrite, WritePtr};
 use fail::fail_point;
 use std::collections::HashMap;
 use std::fmt;
@@ -72,12 +71,6 @@ impl Write for VecWriter {
     }
 }
 
-impl TerminatingWrite for VecWriter {
-    fn terminate_ref(&mut self, _: AntiCallToken) -> io::Result<()> {
-        self.flush()
-    }
-}
-
 #[derive(Default)]
 struct InnerDirectory {
     fs: HashMap<PathBuf, ReadOnlySource>,

src/directory/tests.rs

@@ -127,7 +127,7 @@ fn test_watch(directory: &mut dyn Directory) {
         assert!(directory
             .atomic_write(Path::new("meta.json"), b"random_test_data_2")
             .is_ok());
-        for _ in 0..1_000 {
+        for _ in 0..100 {
             if counter.load(Ordering::SeqCst) > i {
                 break;
             }

src/directory/watch_event_router.rs

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

src/fastfield/mod.rs

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

src/fastfield/multivalued/writer.rs

@@ -5,8 +5,8 @@ use crate::postings::UnorderedTermId;
 use crate::schema::{Document, Field};
 use crate::termdict::TermOrdinal;
 use crate::DocId;
-use fnv::FnvHashMap;
 use itertools::Itertools;
+use std::collections::HashMap;
 use std::io;
 
 /// Writer for multi-valued (as in, more than one value per document)
@@ -102,7 +102,7 @@ impl MultiValueIntFastFieldWriter {
     pub fn serialize(
         &self,
         serializer: &mut FastFieldSerializer,
-        mapping_opt: Option<&FnvHashMap<UnorderedTermId, TermOrdinal>>,
+        mapping_opt: Option<&HashMap<UnorderedTermId, TermOrdinal>>,
     ) -> io::Result<()> {
         {
             // writing the offset index

src/fastfield/writer.rs

@@ -6,7 +6,6 @@ use crate::fastfield::{BytesFastFieldWriter, FastFieldSerializer};
 use crate::postings::UnorderedTermId;
 use crate::schema::{Cardinality, Document, Field, FieldType, Schema};
 use crate::termdict::TermOrdinal;
-use fnv::FnvHashMap;
 use std::collections::HashMap;
 use std::io;
 
@@ -117,7 +116,7 @@ impl FastFieldsWriter {
     pub fn serialize(
         &self,
         serializer: &mut FastFieldSerializer,
-        mapping: &HashMap<Field, FnvHashMap<UnorderedTermId, TermOrdinal>>,
+        mapping: &HashMap<Field, HashMap<UnorderedTermId, TermOrdinal>>,
     ) -> io::Result<()> {
         for field_writer in &self.single_value_writers {
             field_writer.serialize(serializer)?;

src/indexer/index_writer.rs

@@ -8,7 +8,6 @@ use crate::core::SegmentId;
 use crate::core::SegmentMeta;
 use crate::core::SegmentReader;
 use crate::directory::DirectoryLock;
-use crate::directory::TerminatingWrite;
 use crate::docset::DocSet;
 use crate::error::TantivyError;
 use crate::fastfield::write_delete_bitset;
@@ -169,7 +168,6 @@ pub(crate) fn advance_deletes(
             segment = segment.with_delete_meta(num_deleted_docs as u32, target_opstamp);
             let mut delete_file = segment.open_write(SegmentComponent::DELETE)?;
             write_delete_bitset(&delete_bitset, &mut delete_file)?;
-            delete_file.terminate()?;
         }
     }
     segment_entry.set_meta(segment.meta().clone());
@@ -1179,4 +1177,5 @@ mod tests {
         assert!(clear_again.is_ok());
         assert!(commit_again.is_ok());
     }
+
 }

src/indexer/segment_register.rs

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

src/indexer/segment_writer.rs

@@ -296,4 +296,5 @@ mod tests {
         assert_eq!(initial_table_size(10_000_000).unwrap(), 17);
         assert_eq!(initial_table_size(1_000_000_000).unwrap(), 19);
     }
+
 }

src/postings/compression/mod.rs

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

src/postings/mod.rs

@@ -622,23 +622,23 @@ pub mod tests {
             assert!(!postings_unopt.advance());
         }
     }
+
 }
 
 #[cfg(all(test, feature = "unstable"))]
 mod bench {
 
     use super::tests::*;
-    use crate::docset::SkipResult;
-    use crate::query::Intersection;
-    use crate::schema::IndexRecordOption;
-    use crate::tests;
-    use crate::DocSet;
+    use docset::SkipResult;
+    use query::Intersection;
+    use schema::IndexRecordOption;
     use test::{self, Bencher};
+    use tests;
+    use DocSet;
 
     #[bench]
     fn bench_segment_postings(b: &mut Bencher) {
-        let reader = INDEX.reader().unwrap();
-        let searcher = reader.searcher();
+        let searcher = INDEX.searcher();
         let segment_reader = searcher.segment_reader(0);
 
         b.iter(|| {
@@ -652,8 +652,7 @@ mod bench {
 
     #[bench]
     fn bench_segment_intersection(b: &mut Bencher) {
-        let reader = INDEX.reader().unwrap();
-        let searcher = reader.searcher();
+        let searcher = INDEX.searcher();
         let segment_reader = searcher.segment_reader(0);
         b.iter(|| {
             let segment_postings_a = segment_reader
@@ -683,8 +682,7 @@ mod bench {
     }
 
     fn bench_skip_next(p: f64, b: &mut Bencher) {
-        let reader = INDEX.reader().unwrap();
-        let searcher = reader.searcher();
+        let searcher = INDEX.searcher();
         let segment_reader = searcher.segment_reader(0);
         let docs = tests::sample(segment_reader.num_docs(), p);
 
@@ -739,8 +737,7 @@ mod bench {
 
     #[bench]
     fn bench_iterate_segment_postings(b: &mut Bencher) {
-        let reader = INDEX.reader().unwrap();
-        let searcher = reader.searcher();
+        let searcher = INDEX.searcher();
         let segment_reader = searcher.segment_reader(0);
         b.iter(|| {
             let n: u32 = test::black_box(17);

src/postings/postings_writer.rs

@@ -12,7 +12,6 @@ use crate::tokenizer::TokenStream;
 use crate::tokenizer::{Token, MAX_TOKEN_LEN};
 use crate::DocId;
 use crate::Result;
-use fnv::FnvHashMap;
 use std::collections::HashMap;
 use std::io;
 use std::marker::PhantomData;
@@ -128,12 +127,12 @@ impl MultiFieldPostingsWriter {
     pub fn serialize(
         &self,
         serializer: &mut InvertedIndexSerializer,
-    ) -> Result<HashMap<Field, FnvHashMap<UnorderedTermId, TermOrdinal>>> {
+    ) -> Result<HashMap<Field, HashMap<UnorderedTermId, TermOrdinal>>> {
         let mut term_offsets: Vec<(&[u8], Addr, UnorderedTermId)> =
             self.term_index.iter().collect();
         term_offsets.sort_unstable_by_key(|&(k, _, _)| k);
 
-        let mut unordered_term_mappings: HashMap<Field, FnvHashMap<UnorderedTermId, TermOrdinal>> =
+        let mut unordered_term_mappings: HashMap<Field, HashMap<UnorderedTermId, TermOrdinal>> =
             HashMap::new();
 
         let field_offsets = make_field_partition(&term_offsets);
@@ -148,7 +147,7 @@ impl MultiFieldPostingsWriter {
                     let unordered_term_ids = term_offsets[start..stop]
                         .iter()
                         .map(|&(_, _, bucket)| bucket);
-                    let mapping: FnvHashMap<UnorderedTermId, TermOrdinal> = unordered_term_ids
+                    let mapping: HashMap<UnorderedTermId, TermOrdinal> = unordered_term_ids
                         .enumerate()
                         .map(|(term_ord, unord_term_id)| {
                             (unord_term_id as UnorderedTermId, term_ord as TermOrdinal)

src/postings/serializer.rs

@@ -141,7 +141,10 @@ impl<'a> FieldSerializer<'a> {
             FieldType::Str(ref text_options) => {
                 if let Some(text_indexing_options) = text_options.get_indexing_options() {
                     let index_option = text_indexing_options.index_option();
-                    (index_option.has_freq(), index_option.has_positions())
+                    (
+                        index_option.is_termfreq_enabled(),
+                        index_option.is_position_enabled(),
+                    )
                 } else {
                     (false, false)
                 }

src/postings/stacker/expull.rs

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

src/query/all_query.rs

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

src/query/bitset/mod.rs

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

src/query/bm25.rs

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

src/query/boolean_query/mod.rs

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

src/query/exclude.rs

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

src/query/fuzzy_query.rs

@@ -1,12 +1,14 @@
 use crate::error::TantivyError::InvalidArgument;
 use crate::query::{AutomatonWeight, Query, Weight};
 use crate::schema::Term;
+use crate::termdict::WrappedDFA;
 use crate::Result;
 use crate::Searcher;
-use levenshtein_automata::{LevenshteinAutomatonBuilder, DFA};
+use levenshtein_automata::{Distance, LevenshteinAutomatonBuilder, DFA};
 use once_cell::sync::Lazy;
 use std::collections::HashMap;
 use std::ops::Range;
+use derive_builder::Builder;
 
 /// 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
@@ -24,6 +26,38 @@ static LEV_BUILDER: Lazy<HashMap<(u8, bool), LevenshteinAutomatonBuilder>> = Laz
     lev_builder_cache
 });
 
+
+#[derive(Builder, Default, Clone, Debug)]
+pub struct FuzzyConfiguration {
+    /// How many changes are we going to allow
+    pub distance: u8,
+    /// Should a transposition cost 1 or 2?
+    #[builder(default)]
+    pub transposition_cost_one: bool,
+    #[builder(default)]
+    pub prefix: bool,
+    /// If true, only the term with a levenshtein of exactly `distance` will match.
+    /// If false, terms at a distance `<=` to `distance` will match.
+    #[builder(default)]
+    pub exact_distance: bool,
+}
+
+fn build_dfa(fuzzy_configuration: &FuzzyConfiguration, term_text: &str) -> Result<DFA> {
+    let automaton_builder = LEV_BUILDER
+        .get(&(fuzzy_configuration.distance, fuzzy_configuration.transposition_cost_one))
+        .ok_or_else(|| {
+            InvalidArgument(format!(
+                "Levenshtein distance of {} is not allowed. Choose a value in the {:?} range",
+                fuzzy_configuration.distance, VALID_LEVENSHTEIN_DISTANCE_RANGE
+            ))
+        })?;
+    if fuzzy_configuration.prefix {
+        Ok(automaton_builder.build_prefix_dfa(term_text))
+    } else {
+        Ok(automaton_builder.build_dfa(term_text))
+    }
+}
+
 /// A Fuzzy Query matches all of the documents
 /// containing a specific term that is within
 /// Levenshtein distance
@@ -41,32 +75,19 @@ static LEV_BUILDER: Lazy<HashMap<(u8, bool), LevenshteinAutomatonBuilder>> = Laz
 ///     let index = Index::create_in_ram(schema);
 ///     {
 ///         let mut index_writer = index.writer(3_000_000)?;
-///         index_writer.add_document(doc!(
-///             title => "The Name of the Wind",
-///         ));
-///         index_writer.add_document(doc!(
-///             title => "The Diary of Muadib",
-///         ));
-///         index_writer.add_document(doc!(
-///             title => "A Dairy Cow",
-///         ));
-///         index_writer.add_document(doc!(
-///             title => "The Diary of a Young Girl",
-///         ));
+///         index_writer.add_document(doc!(title => "The Name of the Wind"));
+///         index_writer.add_document(doc!(title => "The Diary of Muadib"));
+///         index_writer.add_document(doc!(title => "A Dairy Cow"));
+///         index_writer.add_document(doc!(title => "The Diary of a Young Girl"));
 ///         index_writer.commit().unwrap();
 ///     }
 ///     let reader = index.reader()?;
 ///     let searcher = reader.searcher();
-///
-///     {
-///
-///         let term = Term::from_field_text(title, "Diary");
-///         let query = FuzzyTermQuery::new(term, 1, true);
-///         let (top_docs, count) = searcher.search(&query, &(TopDocs::with_limit(2), Count)).unwrap();
-///         assert_eq!(count, 2);
-///         assert_eq!(top_docs.len(), 2);
-///     }
-///
+///     let term = Term::from_field_text(title, "Diary");
+///     let query = FuzzyTermQuery::new(term, 1, true);
+///     let (top_docs, count) = searcher.search(&query, &(TopDocs::with_limit(2), Count)).unwrap();
+///     assert_eq!(count, 2);
+///     assert_eq!(top_docs.len(), 2);
 ///     Ok(())
 /// }
 /// ```
@@ -74,54 +95,58 @@ static LEV_BUILDER: Lazy<HashMap<(u8, bool), LevenshteinAutomatonBuilder>> = Laz
 pub struct FuzzyTermQuery {
     /// What term are we searching
     term: Term,
-    /// How many changes are we going to allow
-    distance: u8,
-    /// Should a transposition cost 1 or 2?
-    transposition_cost_one: bool,
-    ///
-    prefix: bool,
+    configuration: FuzzyConfiguration
 }
 
 impl FuzzyTermQuery {
+    pub fn new_from_configuration(term: Term, configuration: FuzzyConfiguration) -> FuzzyTermQuery {
+        FuzzyTermQuery {
+            term,
+            configuration
+        }
+    }
+
     /// Creates a new Fuzzy Query
     pub fn new(term: Term, distance: u8, transposition_cost_one: bool) -> FuzzyTermQuery {
         FuzzyTermQuery {
             term,
-            distance,
-            transposition_cost_one,
-            prefix: false,
-        }
-    }
-
-    /// 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,
-            distance,
-            transposition_cost_one,
-            prefix: true,
-        }
-    }
-
-    fn specialized_weight(&self) -> 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 = automaton_builder.build_dfa(self.term.text());
-                Ok(AutomatonWeight::new(self.term.field(), automaton))
+            configuration: FuzzyConfiguration {
+                distance,
+                transposition_cost_one,
+                prefix: false,
+                exact_distance: false
             }
-            None => Err(InvalidArgument(format!(
-                "Levenshtein distance of {} is not allowed. Choose a value in the {:?} range",
-                self.distance, VALID_LEVENSHTEIN_DISTANCE_RANGE
-            ))),
         }
     }
 }
 
 impl Query for FuzzyTermQuery {
     fn weight(&self, _searcher: &Searcher, _scoring_enabled: bool) -> Result<Box<dyn Weight>> {
-        Ok(Box::new(self.specialized_weight()?))
+        let dfa = build_dfa(&self.configuration, self.term.text())?;
+        // TODO optimize for distance = 0 and possibly prefix
+        if self.configuration.exact_distance {
+            let target_distance = self.configuration.distance;
+            let wrapped_dfa = WrappedDFA {
+                dfa,
+                condition: move |distance: Distance| distance == Distance::Exact(target_distance),
+            };
+            Ok(Box::new(AutomatonWeight::new(
+                self.term.field(),
+                wrapped_dfa,
+            )))
+        } else {
+            let wrapped_dfa = WrappedDFA {
+                dfa,
+                condition: move |distance: Distance| match distance {
+                    Distance::Exact(_) => true,
+                    Distance::AtLeast(_) => false,
+                },
+            };
+            Ok(Box::new(AutomatonWeight::new(
+                self.term.field(),
+                wrapped_dfa,
+            )))
+        }
     }
 }
 
@@ -134,6 +159,7 @@ mod test {
     use crate::tests::assert_nearly_equals;
     use crate::Index;
     use crate::Term;
+    use super::FuzzyConfigurationBuilder;
 
     #[test]
     pub fn test_fuzzy_term() {
@@ -155,7 +181,6 @@ mod test {
         let searcher = reader.searcher();
         {
             let term = Term::from_field_text(country_field, "japon");
-
             let fuzzy_query = FuzzyTermQuery::new(term, 1, true);
             let top_docs = searcher
                 .search(&fuzzy_query, &TopDocs::with_limit(2))
@@ -164,5 +189,73 @@ mod test {
             let (score, _) = top_docs[0];
             assert_nearly_equals(1f32, score);
         }
+        {
+            let term = Term::from_field_text(country_field, "japon");
+            let fuzzy_conf = FuzzyConfigurationBuilder::default()
+                .distance(2)
+                .exact_distance(true)
+                .build()
+                .unwrap();
+            let fuzzy_query = FuzzyTermQuery::new_from_configuration(term, fuzzy_conf);
+            let top_docs = searcher
+                .search(&fuzzy_query, &TopDocs::with_limit(2))
+                .unwrap();
+            assert!(top_docs.is_empty());
+        }
+        {
+            let term = Term::from_field_text(country_field, "japon");
+            let fuzzy_conf = FuzzyConfigurationBuilder::default()
+                .distance(1)
+                .exact_distance(true)
+                .build()
+                .unwrap();
+            let fuzzy_query = FuzzyTermQuery::new_from_configuration(term, fuzzy_conf);
+            let top_docs = searcher
+                .search(&fuzzy_query, &TopDocs::with_limit(2))
+                .unwrap();
+            assert_eq!(top_docs.len(), 1);
+        }
+        {
+            let term = Term::from_field_text(country_field, "jpp");
+            let fuzzy_conf = FuzzyConfigurationBuilder::default()
+                .distance(1)
+                .prefix(true)
+                .build()
+                .unwrap();
+            let fuzzy_query = FuzzyTermQuery::new_from_configuration(term, fuzzy_conf);
+            let top_docs = searcher
+                .search(&fuzzy_query, &TopDocs::with_limit(2))
+                .unwrap();
+            assert_eq!(top_docs.len(), 1);
+        }
+        {
+            let term = Term::from_field_text(country_field, "jpaan");
+            let fuzzy_conf = FuzzyConfigurationBuilder::default()
+                .distance(1)
+                .exact_distance(true)
+                .transposition_cost_one(true)
+                .build()
+                .unwrap();
+            let fuzzy_query = FuzzyTermQuery::new_from_configuration(term, fuzzy_conf);
+            let top_docs = searcher
+                .search(&fuzzy_query, &TopDocs::with_limit(2))
+                .unwrap();
+            assert_eq!(top_docs.len(), 1);
+        }
+        {
+            let term = Term::from_field_text(country_field, "jpaan");
+            let fuzzy_conf = FuzzyConfigurationBuilder::default()
+                .distance(2)
+                .exact_distance(true)
+                .transposition_cost_one(false)
+                .build()
+                .unwrap();
+            let fuzzy_query = FuzzyTermQuery::new_from_configuration(term, fuzzy_conf);
+            let top_docs = searcher
+                .search(&fuzzy_query, &TopDocs::with_limit(2))
+                .unwrap();
+            assert_eq!(top_docs.len(), 1);
+        }
     }
+
 }

src/query/intersection.rs

@@ -45,7 +45,7 @@ pub fn intersect_scorers(mut scorers: Vec<Box<dyn Scorer>>) -> Box<dyn Scorer> {
     })
 }
 
-/// Creates a `DocSet` that iterate through the intersection of two or more `DocSet`s.
+/// Creates a `DocSet` that iterator through the intersection of two `DocSet`s.
 pub struct Intersection<TDocSet: DocSet, TOtherDocSet: DocSet = Box<dyn Scorer>> {
     left: TDocSet,
     right: TDocSet,

src/query/intersection_two.rs

@@ -5,7 +5,7 @@ use Score;
 use SkipResult;
 
 
-/// Creates a `DocSet` that iterate through the intersection of two `DocSet`s.
+/// Creates a `DocSet` that iterator through the intersection of two `DocSet`s.
 pub struct IntersectionTwoTerms<TDocSet> {
     left: TDocSet,
     right: TDocSet

src/query/mod.rs

@@ -40,7 +40,7 @@ pub use self::boolean_query::BooleanQuery;
 pub use self::empty_query::{EmptyQuery, EmptyScorer, EmptyWeight};
 pub use self::exclude::Exclude;
 pub use self::explanation::Explanation;
-pub use self::fuzzy_query::FuzzyTermQuery;
+pub use self::fuzzy_query::{FuzzyTermQuery, FuzzyConfiguration, FuzzyConfigurationBuilder};
 pub use self::intersection::intersect_scorers;
 pub use self::phrase_query::PhraseQuery;
 pub use self::query::Query;

src/query/range_query.rs

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

src/query/reqopt_scorer.rs

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

src/query/union.rs

@@ -28,7 +28,7 @@ where
     }
 }
 
-/// Creates a `DocSet` that iterate through the union of two or more `DocSet`s.
+/// Creates a `DocSet` that iterator through the intersection of two `DocSet`s.
 pub struct Union<TScorer, TScoreCombiner = DoNothingCombiner> {
     docsets: Vec<TScorer>,
     bitsets: Box<[TinySet; HORIZON_NUM_TINYBITSETS]>,
@@ -409,17 +409,20 @@ mod tests {
             vec![1, 2, 3, 7, 8, 9, 99, 100, 101, 500, 20000],
         );
     }
+
 }
 
 #[cfg(all(test, feature = "unstable"))]
 mod bench {
 
-    use crate::query::score_combiner::DoNothingCombiner;
-    use crate::query::{ConstScorer, Union, VecDocSet};
-    use crate::tests;
-    use crate::DocId;
-    use crate::DocSet;
+    use query::score_combiner::DoNothingCombiner;
+    use query::ConstScorer;
+    use query::Union;
+    use query::VecDocSet;
     use test::Bencher;
+    use tests;
+    use DocId;
+    use DocSet;
 
     #[bench]
     fn bench_union_3_high(bench: &mut Bencher) {

src/query/vec_docset.rs

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

src/schema/document.rs

@@ -178,4 +178,5 @@ mod tests {
         doc.add_text(text_field, "My title");
         assert_eq!(doc.field_values().len(), 1);
     }
+
 }

src/schema/index_record_option.rs

@@ -29,6 +29,22 @@ pub enum IndexRecordOption {
 }
 
 impl IndexRecordOption {
+    /// Returns true iff the term frequency will be encoded.
+    pub fn is_termfreq_enabled(self) -> bool {
+        match self {
+            IndexRecordOption::WithFreqsAndPositions | IndexRecordOption::WithFreqs => true,
+            _ => false,
+        }
+    }
+
+    /// Returns true iff the term positions within the document are stored as well.
+    pub fn is_position_enabled(self) -> bool {
+        match self {
+            IndexRecordOption::WithFreqsAndPositions => true,
+            _ => false,
+        }
+    }
+
     /// Returns true iff this option includes encoding
     /// term frequencies.
     pub fn has_freq(self) -> bool {

src/schema/mod.rs

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

src/schema/term.rs

@@ -22,10 +22,10 @@ impl Term {
     /// Builds a term given a field, and a i64-value
     ///
     /// Assuming the term has a field id of 1, and a i64 value of 3234,
-    /// the Term will have 12 bytes.
+    /// the Term will have 8 bytes.
     ///
     /// The first four byte are dedicated to storing the field id as a u64.
-    /// The 8 following bytes are encoding the u64 value.
+    /// The 4 following bytes are encoding the u64 value.
     pub fn from_field_i64(field: Field, val: i64) -> Term {
         let val_u64: u64 = common::i64_to_u64(val);
         Term::from_field_u64(field, val_u64)
@@ -33,11 +33,11 @@ impl Term {
 
     /// Builds a term given a field, and a f64-value
     ///
-    /// Assuming the term has a field id of 1, and a f64 value of 1.5,
-    /// the Term will have 12 bytes.
+    /// Assuming the term has a field id of 1, and a u64 value of 3234,
+    /// the Term will have 8 bytes. <= this is wrong
     ///
     /// The first four byte are dedicated to storing the field id as a u64.
-    /// The 8 following bytes are encoding the f64 as a u64 value.
+    /// The 4 following bytes are encoding the u64 value.
     pub fn from_field_f64(field: Field, val: f64) -> Term {
         let val_u64: u64 = common::f64_to_u64(val);
         Term::from_field_u64(field, val_u64)
@@ -46,10 +46,10 @@ impl Term {
     /// Builds a term given a field, and a DateTime value
     ///
     /// Assuming the term has a field id of 1, and a timestamp i64 value of 3234,
-    /// the Term will have 12 bytes.
+    /// the Term will have 8 bytes.
     ///
     /// The first four byte are dedicated to storing the field id as a u64.
-    /// The 8 following bytes are encoding the DateTime as i64 timestamp value.
+    /// The 4 following bytes are encoding the DateTime as i64 timestamp value.
     pub fn from_field_date(field: Field, val: &DateTime) -> Term {
         let val_timestamp = val.timestamp();
         Term::from_field_i64(field, val_timestamp)
@@ -82,10 +82,10 @@ impl Term {
     /// Builds a term given a field, and a u64-value
     ///
     /// Assuming the term has a field id of 1, and a u64 value of 3234,
-    /// the Term will have 12 bytes.
+    /// the Term will have 8 bytes.
     ///
     /// The first four byte are dedicated to storing the field id as a u64.
-    /// The 8 following bytes are encoding the u64 value.
+    /// The 4 following bytes are encoding the u64 value.
     pub fn from_field_u64(field: Field, val: u64) -> Term {
         let mut term = Term(vec![0u8; INT_TERM_LEN]);
         term.set_field(field);
@@ -182,7 +182,7 @@ where
     ///
     /// # Panics
     /// ... or returns an invalid value
-    /// if the term is not a `f64` field.
+    /// if the term is not a `i64` field.
     pub fn get_f64(&self) -> f64 {
         common::u64_to_f64(BigEndian::read_u64(&self.0.as_ref()[4..]))
     }

src/store/mod.rs

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

src/store/skiplist/mod.rs

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

src/store/writer.rs

@@ -3,7 +3,6 @@ use super::skiplist::SkipListBuilder;
 use super::StoreReader;
 use crate::common::CountingWriter;
 use crate::common::{BinarySerializable, VInt};
-use crate::directory::TerminatingWrite;
 use crate::directory::WritePtr;
 use crate::schema::Document;
 use crate::DocId;
@@ -110,6 +109,6 @@ impl StoreWriter {
         self.offset_index_writer.write(&mut self.writer)?;
         header_offset.serialize(&mut self.writer)?;
         self.doc.serialize(&mut self.writer)?;
-        self.writer.terminate()
+        self.writer.flush()
     }
 }

src/termdict/mod.rs

@@ -31,14 +31,43 @@ mod termdict;
 pub use self::merger::TermMerger;
 pub use self::streamer::{TermStreamer, TermStreamerBuilder};
 pub use self::termdict::{TermDictionary, TermDictionaryBuilder};
+use levenshtein_automata::{Distance, DFA, SINK_STATE};
+use tantivy_fst::Automaton;
+
+pub(crate) struct WrappedDFA<Cond> {
+    pub dfa: DFA,
+    pub condition: Cond,
+}
+
+impl<Cond: Fn(Distance) -> bool> Automaton for WrappedDFA<Cond> {
+    type State = u32;
+
+    fn start(&self) -> Self::State {
+        self.dfa.initial_state()
+    }
+
+    fn is_match(&self, state: &Self::State) -> bool {
+        let distance = self.dfa.distance(*state);
+        (self.condition)(distance)
+    }
+
+    fn can_match(&self, state: &Self::State) -> bool {
+        *state != SINK_STATE
+    }
+
+    fn accept(&self, state: &Self::State, byte: u8) -> Self::State {
+        self.dfa.transition(*state, byte)
+    }
+}
 
 #[cfg(test)]
 mod tests {
-    use super::{TermDictionary, TermDictionaryBuilder, TermStreamer};
+    use super::{TermDictionary, TermDictionaryBuilder, TermStreamer, WrappedDFA};
     use crate::core::Index;
     use crate::directory::{Directory, RAMDirectory, ReadOnlySource};
     use crate::postings::TermInfo;
     use crate::schema::{Document, FieldType, Schema, TEXT};
+    use levenshtein_automata::Distance;
     use std::path::PathBuf;
     use std::str;
 
@@ -423,9 +452,14 @@ mod tests {
 
         // We can now build an entire dfa.
         let lev_automaton_builder = LevenshteinAutomatonBuilder::new(2, true);
-        let automaton = lev_automaton_builder.build_dfa("Spaen");
-
-        let mut range = term_dict.search(automaton).into_stream();
+        let wrapped_dfa = WrappedDFA {
+            dfa: lev_automaton_builder.build_dfa("Spaen"),
+            condition: |distance| match distance {
+                Distance::Exact(_) => true,
+                Distance::AtLeast(_) => false,
+            },
+        };
+        let mut range = term_dict.search(wrapped_dfa).into_stream();
 
         // get the first finding
         assert!(range.advance());

src/termdict/term_info_store.rs

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

src/tokenizer/lower_caser.rs

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

src/tokenizer/mod.rs

@@ -281,4 +281,5 @@ pub mod tests {
             assert!(tokens.is_empty());
         }
     }
+
 }

src/tokenizer/ngram_tokenizer.rs

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

src/tokenizer/token_stream_chain.rs

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

tests/failpoints/mod.rs

@@ -1,7 +1,7 @@
 use fail;
 use std::io::Write;
 use std::path::Path;
-use tantivy::directory::{Directory, ManagedDirectory, RAMDirectory, TerminatingWrite};
+use tantivy::directory::{Directory, ManagedDirectory, RAMDirectory};
 use tantivy::doc;
 use tantivy::schema::{Schema, TEXT};
 use tantivy::{Index, Term};
@@ -17,7 +17,7 @@ fn test_failpoints_managed_directory_gc_if_delete_fails() {
     managed_directory
         .open_write(test_path)
         .unwrap()
-        .terminate()
+        .flush()
         .unwrap();
     assert!(managed_directory.exists(test_path));
     // triggering gc and setting the delete operation to fail.