added unit test

* Remove mut in add_document and delete_term. Made stamper ordering relaxed.

* Made batch operations &mut self -> &self

* Added example

.travis.yml

@@ -61,6 +61,9 @@ before_script:
 script:
   - bash ci/script.sh
 
+after_success:
+  - cargo doc-upload
+
 before_deploy:
   - sh ci/before_deploy.sh
 

CHANGELOG.md

@@ -1,13 +1,39 @@
 Tantivy 0.10.0
-====================
+=====================
 
+*Tantivy 0.10.0 index format is compatible with the index format in 0.9.0.*
+
+- Added an ASCII folding filter (@drusellers)
+- Bugfix in `query.count` in presence of deletes (@pmasurel)
 
 Minor
 ---------
 - Small simplification of the code. 
-Calling .freq() or .doc() when .advance() has never 
+Calling .freq() or .doc() when .advance() has never been called
 on segment postings should panic from now on.
 - Tokens exceeding `u16::max_value() - 4` chars are discarded silently instead of panicking.
+- Fast fields are now preloaded when the `SegmentReader` is created.
+- `IndexMeta` is now public.  (@hntd187)
+- `IndexWriter` `add_document`, `delete_term`. `IndexWriter` is `Sync`, making it possible to use it with a `
+Arc<RwLock<IndexWriter>>`. `add_document` and `delete_term` can 
+only require a read lock. (@pmasurel)
+- Introducing `Opstamp` as an expressive type alias for `u64`. (@petr-tik)
+- Stamper now relies on `AtomicU64` on all platforms (@petr-tik)
+
+## How to update?
+
+Your existing indexes are usable as is, but you may need some 
+trivial updates.
+
+### Fast fields
+
+Fast fields used to be accessed directly from the `SegmentReader`.
+The API changed, you are now required to acquire your fast field reader via the
+`segment_reader.fast_fields()`, and use one of the typed method: 
+- `.u64()`, `.i64()` if your field is single-valued ;
+- `.u64s()`, `.i64s()` if your field is multi-valued ;
+- `.bytes()` if your field is bytes fast field.
+
 
 
 Tantivy 0.9.0

examples/custom_collector.rs

@@ -18,8 +18,8 @@ use tantivy::fastfield::FastFieldReader;
 use tantivy::query::QueryParser;
 use tantivy::schema::Field;
 use tantivy::schema::{Schema, FAST, INDEXED, TEXT};
-use tantivy::Index;
 use tantivy::SegmentReader;
+use tantivy::{Index, TantivyError};
 
 #[derive(Default)]
 struct Stats {
@@ -75,9 +75,18 @@ impl Collector for StatsCollector {
     fn for_segment(
         &self,
         _segment_local_id: u32,
-        segment: &SegmentReader,
+        segment_reader: &SegmentReader,
     ) -> tantivy::Result<StatsSegmentCollector> {
-        let fast_field_reader = segment.fast_field_reader(self.field)?;
+        let fast_field_reader = segment_reader
+            .fast_fields()
+            .u64(self.field)
+            .ok_or_else(|| {
+                let field_name = segment_reader.schema().get_field_name(self.field);
+                TantivyError::SchemaError(format!(
+                    "Field {:?} is not a u64 fast field.",
+                    field_name
+                ))
+            })?;
         Ok(StatsSegmentCollector {
             fast_field_reader,
             stats: Stats::default(),

examples/multiple_producer.rs

@@ -0,0 +1,107 @@
+// # Indexing from different threads.
+//
+// It is fairly common to have to index from different threads.
+// Tantivy forbids to create more than one `IndexWriter` at a time.
+//
+// This `IndexWriter` itself has its own multithreaded layer, so managing your own
+// indexing threads will not help. However, it can still be useful for some applications.
+//
+// For instance, if preparing documents to send to tantivy before indexing is the bottleneck of
+// your application, it is reasonable to have multiple threads.
+//
+// Another very common reason to want to index from multiple threads, is implementing a webserver
+// with CRUD capabilities. The server framework will most likely handle request from
+// different threads.
+//
+// The recommended way to address both of these use case is to wrap your `IndexWriter` into a
+// `Arc<RwLock<IndexWriter>>`.
+//
+// While this is counterintuitive, adding and deleting documents do not require mutability
+// over the `IndexWriter`, so several threads will be able to do this operation concurrently.
+//
+// The example below does not represent an actual real-life use case (who would spawn thread to
+// index a single document?), but aims at demonstrating the mechanism that makes indexing
+// from several threads possible.
+
+extern crate tempdir;
+
+// ---
+// Importing tantivy...
+#[macro_use]
+extern crate tantivy;
+use std::sync::{Arc, RwLock};
+use std::thread;
+use std::time::Duration;
+use tantivy::schema::{Schema, STORED, TEXT};
+use tantivy::Opstamp;
+use tantivy::{Index, IndexWriter};
+
+fn main() -> tantivy::Result<()> {
+    // # Defining the schema
+    let mut schema_builder = Schema::builder();
+    let title = schema_builder.add_text_field("title", TEXT | STORED);
+    let body = schema_builder.add_text_field("body", TEXT);
+    let schema = schema_builder.build();
+
+    let index = Index::create_in_ram(schema);
+    let index_writer: Arc<RwLock<IndexWriter>> = Arc::new(RwLock::new(index.writer(50_000_000)?));
+
+    // # First indexing thread.
+    let index_writer_clone_1 = index_writer.clone();
+    thread::spawn(move || {
+        // we index 100 times the document... for the sake of the example.
+        for i in 0..100 {
+            let opstamp = {
+                // A read lock is sufficient here.
+                let index_writer_rlock = index_writer_clone_1.read().unwrap();
+                index_writer_rlock.add_document(
+                    doc!(
+                        title => "Of Mice and Men",
+                        body => "A few miles south of Soledad, the Salinas River drops in close to the hillside \
+                        bank and runs deep and green. The water is warm too, for it has slipped twinkling \
+                        over the yellow sands in the sunlight before reaching the narrow pool. On one \
+                        side of the river the golden foothill slopes curve up to the strong and rocky \
+                        Gabilan Mountains, but on the valley side the water is lined with trees—willows \
+                        fresh and green with every spring, carrying in their lower leaf junctures the \
+                        debris of the winter’s flooding; and sycamores with mottled, white, recumbent \
+                        limbs and branches that arch over the pool"
+                    ))
+            };
+            println!("add doc {} from thread 1 - opstamp {}", i, opstamp);
+            thread::sleep(Duration::from_millis(20));
+        }
+    });
+
+    // # Second indexing thread.
+    let index_writer_clone_2 = index_writer.clone();
+    // For convenience, tantivy also comes with a macro to
+    // reduce the boilerplate above.
+    thread::spawn(move || {
+        // we index 100 times the document... for the sake of the example.
+        for i in 0..100 {
+            // A read lock is sufficient here.
+            let opstamp = {
+                let index_writer_rlock = index_writer_clone_2.read().unwrap();
+                index_writer_rlock.add_document(doc!(
+                    title => "Manufacturing consent",
+                    body => "Some great book description..."
+                ))
+            };
+            println!("add doc {} from thread 2 - opstamp {}", i, opstamp);
+            thread::sleep(Duration::from_millis(10));
+        }
+    });
+
+    // # In the main thread, we commit 10 times, once every 500ms.
+    for _ in 0..10 {
+        let opstamp: Opstamp = {
+            // Committing or rollbacking on the other hand requires write lock. This will block other threads.
+            let mut index_writer_wlock = index_writer.write().unwrap();
+            index_writer_wlock.commit().unwrap()
+        };
+        println!("committed with opstamp {}", opstamp);
+        thread::sleep(Duration::from_millis(500));
+    }
+
+    Ok(())
+}

src/collector/facet_collector.rs

@@ -17,6 +17,7 @@ use Result;
 use Score;
 use SegmentLocalId;
 use SegmentReader;
+use TantivyError;
 
 struct Hit<'a> {
     count: u64,
@@ -264,7 +265,10 @@ impl Collector for FacetCollector {
         _: SegmentLocalId,
         reader: &SegmentReader,
     ) -> Result<FacetSegmentCollector> {
-        let facet_reader = reader.facet_reader(self.field)?;
+        let field_name = reader.schema().get_field_name(self.field);
+        let facet_reader = reader.facet_reader(self.field).ok_or_else(|| {
+            TantivyError::SchemaError(format!("Field {:?} is not a facet field.", field_name))
+        })?;
 
         let mut collapse_mapping = Vec::new();
         let mut counts = Vec::new();

src/collector/multi_collector.rs

@@ -2,6 +2,7 @@ use super::Collector;
 use super::SegmentCollector;
 use collector::Fruit;
 use std::marker::PhantomData;
+use std::ops::Deref;
 use DocId;
 use Result;
 use Score;
@@ -199,7 +200,10 @@ impl<'a> Collector for MultiCollector<'a> {
     }
 
     fn requires_scoring(&self) -> bool {
-        self.collector_wrappers.iter().any(|c| c.requires_scoring())
+        self.collector_wrappers
+            .iter()
+            .map(Deref::deref)
+            .any(Collector::requires_scoring)
     }
 
     fn merge_fruits(&self, segments_multifruits: Vec<MultiFruit>) -> Result<MultiFruit> {

src/collector/tests.rs

@@ -114,11 +114,15 @@ impl Collector for FastFieldTestCollector {
     fn for_segment(
         &self,
         _: SegmentLocalId,
-        reader: &SegmentReader,
+        segment_reader: &SegmentReader,
     ) -> Result<FastFieldSegmentCollector> {
+        let reader = segment_reader
+            .fast_fields()
+            .u64(self.field)
+            .expect("Requested field is not a fast field.");
         Ok(FastFieldSegmentCollector {
             vals: Vec::new(),
-            reader: reader.fast_field_reader(self.field)?,
+            reader,
         })
     }
 
@@ -170,11 +174,14 @@ impl Collector for BytesFastFieldTestCollector {
     fn for_segment(
         &self,
         _segment_local_id: u32,
-        segment: &SegmentReader,
+        segment_reader: &SegmentReader,
     ) -> Result<BytesFastFieldSegmentCollector> {
         Ok(BytesFastFieldSegmentCollector {
             vals: Vec::new(),
-            reader: segment.bytes_fast_field_reader(self.field)?,
+            reader: segment_reader
+                .fast_fields()
+                .bytes(self.field)
+                .expect("Field is not a bytes fast field."),
         })
     }
 
@@ -191,7 +198,7 @@ impl SegmentCollector for BytesFastFieldSegmentCollector {
     type Fruit = Vec<u8>;
 
     fn collect(&mut self, doc: u32, _score: f32) {
-        let data = self.reader.get_val(doc);
+        let data = self.reader.get_bytes(doc);
         self.vals.extend(data);
     }
 

src/collector/top_collector.rs

@@ -98,11 +98,11 @@ where
             .collect())
     }
 
-    pub(crate) fn for_segment(
+    pub(crate) fn for_segment<F: PartialOrd>(
         &self,
         segment_id: SegmentLocalId,
         _: &SegmentReader,
-    ) -> Result<TopSegmentCollector<T>> {
+    ) -> Result<TopSegmentCollector<F>> {
         Ok(TopSegmentCollector::new(segment_id, self.limit))
     }
 }

src/collector/top_field_collector.rs

@@ -5,10 +5,12 @@ use collector::SegmentCollector;
 use fastfield::FastFieldReader;
 use fastfield::FastValue;
 use schema::Field;
+use std::marker::PhantomData;
 use DocAddress;
 use Result;
 use SegmentLocalId;
 use SegmentReader;
+use TantivyError;
 
 /// The Top Field Collector keeps track of the K documents
 /// sorted by a fast field in the index
@@ -106,8 +108,15 @@ impl<T: FastValue + PartialOrd + Send + Sync + 'static> Collector for TopDocsByF
         reader: &SegmentReader,
     ) -> Result<TopFieldSegmentCollector<T>> {
         let collector = self.collector.for_segment(segment_local_id, reader)?;
-        let reader = reader.fast_field_reader(self.field)?;
-        Ok(TopFieldSegmentCollector { collector, reader })
+        let reader = reader.fast_fields().u64(self.field).ok_or_else(|| {
+            let field_name = reader.schema().get_field_name(self.field);
+            TantivyError::SchemaError(format!("Failed to find fast field reader {:?}", field_name))
+        })?;
+        Ok(TopFieldSegmentCollector {
+            collector,
+            reader,
+            _type: PhantomData,
+        })
     }
 
     fn requires_scoring(&self) -> bool {
@@ -122,9 +131,10 @@ impl<T: FastValue + PartialOrd + Send + Sync + 'static> Collector for TopDocsByF
     }
 }
 
-pub struct TopFieldSegmentCollector<T: FastValue + PartialOrd> {
-    collector: TopSegmentCollector<T>,
-    reader: FastFieldReader<T>,
+pub struct TopFieldSegmentCollector<T> {
+    collector: TopSegmentCollector<u64>,
+    reader: FastFieldReader<u64>,
+    _type: PhantomData<T>,
 }
 
 impl<T: FastValue + PartialOrd + Send + Sync + 'static> SegmentCollector
@@ -138,7 +148,11 @@ impl<T: FastValue + PartialOrd + Send + Sync + 'static> SegmentCollector
     }
 
     fn harvest(self) -> Vec<(T, DocAddress)> {
-        self.collector.harvest()
+        self.collector
+            .harvest()
+            .into_iter()
+            .map(|(val, doc_address)| (T::from_u64(val), doc_address))
+            .collect()
     }
 }
 
@@ -235,7 +249,7 @@ mod tests {
                 .for_segment(0, segment)
                 .map(|_| ())
                 .unwrap_err(),
-            TantivyError::FastFieldError(_)
+            TantivyError::SchemaError(_)
         );
     }
 

src/core/index.rs

@@ -340,7 +340,7 @@ impl Index {
         Ok(self
             .searchable_segment_metas()?
             .iter()
-            .map(|segment_meta| segment_meta.id())
+            .map(SegmentMeta::id)
             .collect())
     }
 }

src/core/index_meta.rs

@@ -2,6 +2,7 @@ use core::SegmentMeta;
 use schema::Schema;
 use serde_json;
 use std::fmt;
+use Opstamp;
 
 /// Meta information about the `Index`.
 ///
@@ -15,7 +16,7 @@ use std::fmt;
 pub struct IndexMeta {
     pub segments: Vec<SegmentMeta>,
     pub schema: Schema,
-    pub opstamp: u64,
+    pub opstamp: Opstamp,
     #[serde(skip_serializing_if = "Option::is_none")]
     pub payload: Option<String>,
 }

src/core/searcher.rs

@@ -59,7 +59,7 @@ impl Searcher {
     ) -> Searcher {
         let store_readers = segment_readers
             .iter()
-            .map(|segment_reader| segment_reader.get_store_reader())
+            .map(SegmentReader::get_store_reader)
             .collect();
         Searcher {
             schema,
@@ -218,7 +218,7 @@ impl fmt::Debug for Searcher {
         let segment_ids = self
             .segment_readers
             .iter()
-            .map(|segment_reader| segment_reader.segment_id())
+            .map(SegmentReader::segment_id)
             .collect::<Vec<_>>();
         write!(f, "Searcher({:?})", segment_ids)
     }

src/core/segment.rs

@@ -10,6 +10,7 @@ use schema::Schema;
 use std::fmt;
 use std::path::PathBuf;
 use std::result;
+use Opstamp;
 use Result;
 
 /// A segment is a piece of the index.
@@ -50,7 +51,7 @@ impl Segment {
     }
 
     #[doc(hidden)]
-    pub fn with_delete_meta(self, num_deleted_docs: u32, opstamp: u64) -> Segment {
+    pub fn with_delete_meta(self, num_deleted_docs: u32, opstamp: Opstamp) -> Segment {
         Segment {
             index: self.index,
             meta: self.meta.with_delete_meta(num_deleted_docs, opstamp),

src/core/segment_meta.rs

@@ -5,6 +5,7 @@ use serde;
 use std::collections::HashSet;
 use std::fmt;
 use std::path::PathBuf;
+use Opstamp;
 
 lazy_static! {
     static ref INVENTORY: Inventory<InnerSegmentMeta> = { Inventory::new() };
@@ -13,7 +14,7 @@ lazy_static! {
 #[derive(Clone, Debug, Serialize, Deserialize)]
 struct DeleteMeta {
     num_deleted_docs: u32,
-    opstamp: u64,
+    opstamp: Opstamp,
 }
 
 /// `SegmentMeta` contains simple meta information about a segment.
@@ -136,9 +137,9 @@ impl SegmentMeta {
         self.max_doc() - self.num_deleted_docs()
     }
 
-    /// Returns the opstamp of the last delete operation
+    /// Returns the `Opstamp` of the last delete operation
     /// taken in account in this segment.
-    pub fn delete_opstamp(&self) -> Option<u64> {
+    pub fn delete_opstamp(&self) -> Option<Opstamp> {
         self.tracked
             .deletes
             .as_ref()
@@ -152,7 +153,7 @@ impl SegmentMeta {
     }
 
     #[doc(hidden)]
-    pub fn with_delete_meta(self, num_deleted_docs: u32, opstamp: u64) -> SegmentMeta {
+    pub fn with_delete_meta(self, num_deleted_docs: u32, opstamp: Opstamp) -> SegmentMeta {
         let delete_meta = DeleteMeta {
             num_deleted_docs,
             opstamp,

src/core/segment_reader.rs

@@ -5,14 +5,10 @@ use core::Segment;
 use core::SegmentComponent;
 use core::SegmentId;
 use directory::ReadOnlySource;
-use error::TantivyError;
 use fastfield::DeleteBitSet;
 use fastfield::FacetReader;
-use fastfield::FastFieldReader;
-use fastfield::{self, FastFieldNotAvailableError};
-use fastfield::{BytesFastFieldReader, FastValue, MultiValueIntFastFieldReader};
+use fastfield::FastFieldReaders;
 use fieldnorm::FieldNormReader;
-use schema::Cardinality;
 use schema::Field;
 use schema::FieldType;
 use schema::Schema;
@@ -51,7 +47,7 @@ pub struct SegmentReader {
     postings_composite: CompositeFile,
     positions_composite: CompositeFile,
     positions_idx_composite: CompositeFile,
-    fast_fields_composite: CompositeFile,
+    fast_fields_readers: Arc<FastFieldReaders>,
     fieldnorms_composite: CompositeFile,
 
     store_source: ReadOnlySource,
@@ -105,93 +101,21 @@ impl SegmentReader {
     ///
     /// # Panics
     /// May panic if the index is corrupted.
-    pub fn fast_field_reader<Item: FastValue>(
-        &self,
-        field: Field,
-    ) -> fastfield::Result<FastFieldReader<Item>> {
-        let field_entry = self.schema.get_field_entry(field);
-        if Item::fast_field_cardinality(field_entry.field_type()) == Some(Cardinality::SingleValue)
-        {
-            self.fast_fields_composite
-                .open_read(field)
-                .ok_or_else(|| FastFieldNotAvailableError::new(field_entry))
-                .map(FastFieldReader::open)
-        } else {
-            Err(FastFieldNotAvailableError::new(field_entry))
-        }
-    }
-
-    pub(crate) fn fast_field_reader_with_idx<Item: FastValue>(
-        &self,
-        field: Field,
-        idx: usize,
-    ) -> fastfield::Result<FastFieldReader<Item>> {
-        if let Some(ff_source) = self.fast_fields_composite.open_read_with_idx(field, idx) {
-            Ok(FastFieldReader::open(ff_source))
-        } else {
-            let field_entry = self.schema.get_field_entry(field);
-            Err(FastFieldNotAvailableError::new(field_entry))
-        }
-    }
-
-    /// Accessor to the `MultiValueIntFastFieldReader` associated to a given `Field`.
-    /// May panick if the field is not a multivalued fastfield of the type `Item`.
-    pub fn multi_fast_field_reader<Item: FastValue>(
-        &self,
-        field: Field,
-    ) -> fastfield::Result<MultiValueIntFastFieldReader<Item>> {
-        let field_entry = self.schema.get_field_entry(field);
-        if Item::fast_field_cardinality(field_entry.field_type()) == Some(Cardinality::MultiValues)
-        {
-            let idx_reader = self.fast_field_reader_with_idx(field, 0)?;
-            let vals_reader = self.fast_field_reader_with_idx(field, 1)?;
-            Ok(MultiValueIntFastFieldReader::open(idx_reader, vals_reader))
-        } else {
-            Err(FastFieldNotAvailableError::new(field_entry))
-        }
-    }
-
-    /// Accessor to the `BytesFastFieldReader` associated to a given `Field`.
-    pub fn bytes_fast_field_reader(&self, field: Field) -> fastfield::Result<BytesFastFieldReader> {
-        let field_entry = self.schema.get_field_entry(field);
-        match *field_entry.field_type() {
-            FieldType::Bytes => {}
-            _ => return Err(FastFieldNotAvailableError::new(field_entry)),
-        }
-        let idx_reader = self
-            .fast_fields_composite
-            .open_read_with_idx(field, 0)
-            .ok_or_else(|| FastFieldNotAvailableError::new(field_entry))
-            .map(FastFieldReader::open)?;
-        let values = self
-            .fast_fields_composite
-            .open_read_with_idx(field, 1)
-            .ok_or_else(|| FastFieldNotAvailableError::new(field_entry))?;
-        Ok(BytesFastFieldReader::open(idx_reader, values))
+    pub fn fast_fields(&self) -> &FastFieldReaders {
+        &self.fast_fields_readers
     }
 
     /// Accessor to the `FacetReader` associated to a given `Field`.
-    pub fn facet_reader(&self, field: Field) -> Result<FacetReader> {
+    pub fn facet_reader(&self, field: Field) -> Option<FacetReader> {
         let field_entry = self.schema.get_field_entry(field);
         if field_entry.field_type() != &FieldType::HierarchicalFacet {
-            return Err(TantivyError::InvalidArgument(format!(
-                "The field {:?} is not a \
-                 hierarchical facet.",
-                field_entry
-            )));
+            return None;
         }
-        let term_ords_reader = self.multi_fast_field_reader(field)?;
-        let termdict_source = self.termdict_composite.open_read(field).ok_or_else(|| {
-            TantivyError::InvalidArgument(format!(
-                "The field \"{}\" is a hierarchical \
-                 but this segment does not seem to have the field term \
-                 dictionary.",
-                field_entry.name()
-            ))
-        })?;
+        let term_ords_reader = self.fast_fields().u64s(field)?;
+        let termdict_source = self.termdict_composite.open_read(field)?;
         let termdict = TermDictionary::from_source(&termdict_source);
         let facet_reader = FacetReader::new(term_ords_reader, termdict);
-        Ok(facet_reader)
+        Some(facet_reader)
     }
 
     /// Accessor to the segment's `Field norms`'s reader.
@@ -247,8 +171,12 @@ impl SegmentReader {
             }
         };
 
+        let schema = segment.schema();
+
         let fast_fields_data = segment.open_read(SegmentComponent::FASTFIELDS)?;
         let fast_fields_composite = CompositeFile::open(&fast_fields_data)?;
+        let fast_field_readers =
+            Arc::new(FastFieldReaders::load_all(&schema, &fast_fields_composite)?);
 
         let fieldnorms_data = segment.open_read(SegmentComponent::FIELDNORMS)?;
         let fieldnorms_composite = CompositeFile::open(&fieldnorms_data)?;
@@ -260,14 +188,13 @@ impl SegmentReader {
             None
         };
 
-        let schema = segment.schema();
         Ok(SegmentReader {
             inv_idx_reader_cache: Arc::new(RwLock::new(HashMap::new())),
             max_doc: segment.meta().max_doc(),
             num_docs: segment.meta().num_docs(),
             termdict_composite,
             postings_composite,
-            fast_fields_composite,
+            fast_fields_readers: fast_field_readers,
             fieldnorms_composite,
             segment_id: segment.id(),
             store_source,
@@ -381,12 +308,12 @@ impl SegmentReader {
             self.postings_composite.space_usage(),
             self.positions_composite.space_usage(),
             self.positions_idx_composite.space_usage(),
-            self.fast_fields_composite.space_usage(),
+            self.fast_fields_readers.space_usage(),
             self.fieldnorms_composite.space_usage(),
             self.get_store_reader().space_usage(),
             self.delete_bitset_opt
                 .as_ref()
-                .map(|x| x.space_usage())
+                .map(DeleteBitSet::space_usage)
                 .unwrap_or(0),
         )
     }

src/directory/directory.rs

@@ -48,14 +48,14 @@ impl RetryPolicy {
 ///
 /// It is transparently associated to a lock file, that gets deleted
 /// on `Drop.` The lock is released automatically on `Drop`.
-pub struct DirectoryLock(Box<Drop + Send + 'static>);
+pub struct DirectoryLock(Box<Drop + Send + Sync + 'static>);
 
 struct DirectoryLockGuard {
     directory: Box<Directory>,
     path: PathBuf,
 }
 
-impl<T: Drop + Send + 'static> From<Box<T>> for DirectoryLock {
+impl<T: Drop + Send + Sync + 'static> From<Box<T>> for DirectoryLock {
     fn from(underlying: Box<T>) -> Self {
         DirectoryLock(underlying)
     }

src/directory/ram_directory.rs

@@ -86,7 +86,7 @@ impl InnerDirectory {
         self.fs
             .get(path)
             .ok_or_else(|| OpenReadError::FileDoesNotExist(PathBuf::from(path)))
-            .map(|el| el.clone())
+            .map(Clone::clone)
     }
 
     fn delete(&mut self, path: &Path) -> result::Result<(), DeleteError> {

src/fastfield/bytes/mod.rs

@@ -23,14 +23,14 @@ mod tests {
         index_writer.add_document(doc!(field=>vec![0u8; 1000]));
         assert!(index_writer.commit().is_ok());
         let searcher = index.reader().unwrap().searcher();
-        let reader = searcher.segment_reader(0);
-        let bytes_reader = reader.bytes_fast_field_reader(field).unwrap();
+        let segment_reader = searcher.segment_reader(0);
+        let bytes_reader = segment_reader.fast_fields().bytes(field).unwrap();
 
-        assert_eq!(bytes_reader.get_val(0), &[0u8, 1, 2, 3]);
-        assert!(bytes_reader.get_val(1).is_empty());
-        assert_eq!(bytes_reader.get_val(2), &[255u8]);
-        assert_eq!(bytes_reader.get_val(3), &[1u8, 3, 5, 7, 9]);
+        assert_eq!(bytes_reader.get_bytes(0), &[0u8, 1, 2, 3]);
+        assert!(bytes_reader.get_bytes(1).is_empty());
+        assert_eq!(bytes_reader.get_bytes(2), &[255u8]);
+        assert_eq!(bytes_reader.get_bytes(3), &[1u8, 3, 5, 7, 9]);
         let long = vec![0u8; 1000];
-        assert_eq!(bytes_reader.get_val(4), long.as_slice());
+        assert_eq!(bytes_reader.get_bytes(4), long.as_slice());
     }
 }

src/fastfield/bytes/reader.rs

@@ -14,6 +14,7 @@ use DocId;
 ///
 /// Reading the value for a document is done by reading the start index for it,
 /// and the start index for the next document, and keeping the bytes in between.
+#[derive(Clone)]
 pub struct BytesFastFieldReader {
     idx_reader: FastFieldReader<u64>,
     values: OwningRef<ReadOnlySource, [u8]>,
@@ -28,10 +29,20 @@ impl BytesFastFieldReader {
         BytesFastFieldReader { idx_reader, values }
     }
 
-    /// Returns the bytes associated to the given `doc`
-    pub fn get_val(&self, doc: DocId) -> &[u8] {
+    fn range(&self, doc: DocId) -> (usize, usize) {
         let start = self.idx_reader.get(doc) as usize;
         let stop = self.idx_reader.get(doc + 1) as usize;
+        (start, stop)
+    }
+
+    /// Returns the bytes associated to the given `doc`
+    pub fn get_bytes(&self, doc: DocId) -> &[u8] {
+        let (start, stop) = self.range(doc);
         &self.values[start..stop]
     }
+
+    /// Returns the overall number of bytes in this bytes fast field.
+    pub fn total_num_bytes(&self) -> usize {
+        self.values.len()
+    }
 }

src/fastfield/delete.rs

@@ -53,16 +53,18 @@ impl DeleteBitSet {
         }
     }
 
-    /// Returns whether the document has been marked as deleted.
+    /// Returns true iff the document is still "alive". In other words, if it has not been deleted.
+    pub fn is_alive(&self, doc: DocId) -> bool {
+        !self.is_deleted(doc)
+    }
+
+    /// Returns true iff the document has been marked as deleted.
+    #[inline(always)]
     pub fn is_deleted(&self, doc: DocId) -> bool {
-        if self.len == 0 {
-            false
-        } else {
-            let byte_offset = doc / 8u32;
-            let b: u8 = (*self.data)[byte_offset as usize];
-            let shift = (doc & 7u32) as u8;
-            b & (1u8 << shift) != 0
-        }
+        let byte_offset = doc / 8u32;
+        let b: u8 = (*self.data)[byte_offset as usize];
+        let shift = (doc & 7u32) as u8;
+        b & (1u8 << shift) != 0
     }
 
     /// Summarize total space usage of this bitset.

src/fastfield/mod.rs

@@ -30,6 +30,7 @@ pub use self::error::{FastFieldNotAvailableError, Result};
 pub use self::facet_reader::FacetReader;
 pub use self::multivalued::{MultiValueIntFastFieldReader, MultiValueIntFastFieldWriter};
 pub use self::reader::FastFieldReader;
+pub use self::readers::FastFieldReaders;
 pub use self::serializer::FastFieldSerializer;
 pub use self::writer::{FastFieldsWriter, IntFastFieldWriter};
 use common;
@@ -43,6 +44,7 @@ mod error;
 mod facet_reader;
 mod multivalued;
 mod reader;
+mod readers;
 mod serializer;
 mod writer;
 
@@ -78,10 +80,6 @@ impl FastValue for u64 {
         *self
     }
 
-    fn as_u64(&self) -> u64 {
-        *self
-    }
-
     fn fast_field_cardinality(field_type: &FieldType) -> Option<Cardinality> {
         match *field_type {
             FieldType::U64(ref integer_options) => integer_options.get_fastfield_cardinality(),
@@ -89,6 +87,10 @@ impl FastValue for u64 {
             _ => None,
         }
     }
+
+    fn as_u64(&self) -> u64 {
+        *self
+    }
 }
 
 impl FastValue for i64 {

src/fastfield/multivalued/mod.rs

@@ -37,9 +37,7 @@ mod tests {
         let searcher = index.reader().unwrap().searcher();
         let segment_reader = searcher.segment_reader(0);
         let mut vals = Vec::new();
-        let multi_value_reader = segment_reader
-            .multi_fast_field_reader::<u64>(field)
-            .unwrap();
+        let multi_value_reader = segment_reader.fast_fields().u64s(field).unwrap();
         {
             multi_value_reader.get_vals(2, &mut vals);
             assert_eq!(&vals, &[4u64]);
@@ -198,9 +196,9 @@ mod tests {
         assert!(index_writer.commit().is_ok());
 
         let searcher = index.reader().unwrap().searcher();
-        let reader = searcher.segment_reader(0);
+        let segment_reader = searcher.segment_reader(0);
         let mut vals = Vec::new();
-        let multi_value_reader = reader.multi_fast_field_reader::<i64>(field).unwrap();
+        let multi_value_reader = segment_reader.fast_fields().i64s(field).unwrap();
         {
             multi_value_reader.get_vals(2, &mut vals);
             assert_eq!(&vals, &[-4i64]);

src/fastfield/multivalued/reader.rs

@@ -26,6 +26,13 @@ impl<Item: FastValue> MultiValueIntFastFieldReader<Item> {
         }
     }
 
+    pub(crate) fn into_u64s_reader(self) -> MultiValueIntFastFieldReader<u64> {
+        MultiValueIntFastFieldReader {
+            idx_reader: self.idx_reader,
+            vals_reader: self.vals_reader.into_u64_reader(),
+        }
+    }
+
     /// Returns `(start, stop)`, such that the values associated
     /// to the given document are `start..stop`.
     fn range(&self, doc: DocId) -> (u64, u64) {
@@ -41,13 +48,24 @@ impl<Item: FastValue> MultiValueIntFastFieldReader<Item> {
         vals.resize(len, Item::default());
         self.vals_reader.get_range_u64(start, &mut vals[..]);
     }
+
+    /// Returns the number of values associated with the document `DocId`.
+    pub fn num_vals(&self, doc: DocId) -> usize {
+        let (start, stop) = self.range(doc);
+        (stop - start) as usize
+    }
+
+    /// Returns the overall number of values in this field  .
+    pub fn total_num_vals(&self) -> u64 {
+        self.idx_reader.max_value()
+    }
 }
 
 #[cfg(test)]
 mod tests {
 
     use core::Index;
-    use schema::{Document, Facet, Schema};
+    use schema::{Facet, Schema};
 
     #[test]
     fn test_multifastfield_reader() {
@@ -58,22 +76,12 @@ mod tests {
         let mut index_writer = index
             .writer_with_num_threads(1, 30_000_000)
             .expect("Failed to create index writer.");
-        {
-            let mut doc = Document::new();
-            doc.add_facet(facet_field, "/category/cat2");
-            doc.add_facet(facet_field, "/category/cat1");
-            index_writer.add_document(doc);
-        }
-        {
-            let mut doc = Document::new();
-            doc.add_facet(facet_field, "/category/cat2");
-            index_writer.add_document(doc);
-        }
-        {
-            let mut doc = Document::new();
-            doc.add_facet(facet_field, "/category/cat3");
-            index_writer.add_document(doc);
-        }
+        index_writer.add_document(doc!(
+            facet_field => Facet::from("/category/cat2"),
+            facet_field => Facet::from("/category/cat1"),
+        ));
+        index_writer.add_document(doc!(facet_field => Facet::from("/category/cat2")));
+        index_writer.add_document(doc!(facet_field => Facet::from("/category/cat3")));
         index_writer.commit().expect("Commit failed");
         let searcher = index.reader().unwrap().searcher();
         let segment_reader = searcher.segment_reader(0);

src/fastfield/reader.rs

@@ -50,6 +50,15 @@ impl<Item: FastValue> FastFieldReader<Item> {
         }
     }
 
+    pub(crate) fn into_u64_reader(self) -> FastFieldReader<u64> {
+        FastFieldReader {
+            bit_unpacker: self.bit_unpacker,
+            min_value_u64: self.min_value_u64,
+            max_value_u64: self.max_value_u64,
+            _phantom: PhantomData,
+        }
+    }
+
     /// Return the value associated to the given document.
     ///
     /// This accessor should return as fast as possible.

src/fastfield/readers.rs

@@ -0,0 +1,191 @@
+use common::CompositeFile;
+use fastfield::BytesFastFieldReader;
+use fastfield::MultiValueIntFastFieldReader;
+use fastfield::{FastFieldNotAvailableError, FastFieldReader};
+use schema::{Cardinality, Field, FieldType, Schema};
+use space_usage::PerFieldSpaceUsage;
+use std::collections::HashMap;
+use Result;
+
+/// Provides access to all of the FastFieldReader.
+///
+/// Internally, `FastFieldReaders` have preloaded fast field readers,
+/// and just wraps several `HashMap`.
+pub struct FastFieldReaders {
+    fast_field_i64: HashMap<Field, FastFieldReader<i64>>,
+    fast_field_u64: HashMap<Field, FastFieldReader<u64>>,
+    fast_field_i64s: HashMap<Field, MultiValueIntFastFieldReader<i64>>,
+    fast_field_u64s: HashMap<Field, MultiValueIntFastFieldReader<u64>>,
+    fast_bytes: HashMap<Field, BytesFastFieldReader>,
+    fast_fields_composite: CompositeFile,
+}
+
+enum FastType {
+    I64,
+    U64,
+}
+
+fn type_and_cardinality(field_type: &FieldType) -> Option<(FastType, Cardinality)> {
+    match field_type {
+        FieldType::U64(options) => options
+            .get_fastfield_cardinality()
+            .map(|cardinality| (FastType::U64, cardinality)),
+        FieldType::I64(options) => options
+            .get_fastfield_cardinality()
+            .map(|cardinality| (FastType::I64, cardinality)),
+        FieldType::HierarchicalFacet => Some((FastType::U64, Cardinality::MultiValues)),
+        _ => None,
+    }
+}
+
+impl FastFieldReaders {
+    pub(crate) fn load_all(
+        schema: &Schema,
+        fast_fields_composite: &CompositeFile,
+    ) -> Result<FastFieldReaders> {
+        let mut fast_field_readers = FastFieldReaders {
+            fast_field_i64: Default::default(),
+            fast_field_u64: Default::default(),
+            fast_field_i64s: Default::default(),
+            fast_field_u64s: Default::default(),
+            fast_bytes: Default::default(),
+            fast_fields_composite: fast_fields_composite.clone(),
+        };
+        for (field_id, field_entry) in schema.fields().iter().enumerate() {
+            let field = Field(field_id as u32);
+            let field_type = field_entry.field_type();
+            if field_type == &FieldType::Bytes {
+                let idx_reader = fast_fields_composite
+                    .open_read_with_idx(field, 0)
+                    .ok_or_else(|| FastFieldNotAvailableError::new(field_entry))
+                    .map(FastFieldReader::open)?;
+                let data = fast_fields_composite
+                    .open_read_with_idx(field, 1)
+                    .ok_or_else(|| FastFieldNotAvailableError::new(field_entry))?;
+                fast_field_readers
+                    .fast_bytes
+                    .insert(field, BytesFastFieldReader::open(idx_reader, data));
+            } else if let Some((fast_type, cardinality)) = type_and_cardinality(field_type) {
+                match cardinality {
+                    Cardinality::SingleValue => {
+                        if let Some(fast_field_data) = fast_fields_composite.open_read(field) {
+                            match fast_type {
+                                FastType::U64 => {
+                                    let fast_field_reader = FastFieldReader::open(fast_field_data);
+                                    fast_field_readers
+                                        .fast_field_u64
+                                        .insert(field, fast_field_reader);
+                                }
+                                FastType::I64 => {
+                                    fast_field_readers.fast_field_i64.insert(
+                                        field,
+                                        FastFieldReader::open(fast_field_data.clone()),
+                                    );
+                                }
+                            }
+                        } else {
+                            return Err(From::from(FastFieldNotAvailableError::new(field_entry)));
+                        }
+                    }
+                    Cardinality::MultiValues => {
+                        let idx_opt = fast_fields_composite.open_read_with_idx(field, 0);
+                        let data_opt = fast_fields_composite.open_read_with_idx(field, 1);
+                        if let (Some(fast_field_idx), Some(fast_field_data)) = (idx_opt, data_opt) {
+                            let idx_reader = FastFieldReader::open(fast_field_idx);
+                            match fast_type {
+                                FastType::I64 => {
+                                    let vals_reader = FastFieldReader::open(fast_field_data);
+                                    let multivalued_int_fast_field =
+                                        MultiValueIntFastFieldReader::open(idx_reader, vals_reader);
+                                    fast_field_readers
+                                        .fast_field_i64s
+                                        .insert(field, multivalued_int_fast_field);
+                                }
+                                FastType::U64 => {
+                                    let vals_reader = FastFieldReader::open(fast_field_data);
+                                    let multivalued_int_fast_field =
+                                        MultiValueIntFastFieldReader::open(idx_reader, vals_reader);
+                                    fast_field_readers
+                                        .fast_field_u64s
+                                        .insert(field, multivalued_int_fast_field);
+                                }
+                            }
+                        } else {
+                            return Err(From::from(FastFieldNotAvailableError::new(field_entry)));
+                        }
+                    }
+                }
+            }
+        }
+        Ok(fast_field_readers)
+    }
+
+    pub(crate) fn space_usage(&self) -> PerFieldSpaceUsage {
+        self.fast_fields_composite.space_usage()
+    }
+
+    /// Returns the `u64` fast field reader reader associated to `field`.
+    ///
+    /// If `field` is not a u64 fast field, this method returns `None`.
+    pub fn u64(&self, field: Field) -> Option<FastFieldReader<u64>> {
+        self.fast_field_u64.get(&field).cloned()
+    }
+
+    /// If the field is a u64-fast field return the associated reader.
+    /// If the field is a i64-fast field, return the associated u64 reader. Values are
+    /// mapped from i64 to u64 using a (well the, it is unique) monotonic mapping.    ///
+    ///
+    /// This method is useful when merging segment reader.
+    pub(crate) fn u64_lenient(&self, field: Field) -> Option<FastFieldReader<u64>> {
+        if let Some(u64_ff_reader) = self.u64(field) {
+            return Some(u64_ff_reader);
+        }
+        if let Some(i64_ff_reader) = self.i64(field) {
+            return Some(i64_ff_reader.into_u64_reader());
+        }
+        None
+    }
+
+    /// Returns the `i64` fast field reader reader associated to `field`.
+    ///
+    /// If `field` is not a i64 fast field, this method returns `None`.
+    pub fn i64(&self, field: Field) -> Option<FastFieldReader<i64>> {
+        self.fast_field_i64.get(&field).cloned()
+    }
+
+    /// Returns a `u64s` multi-valued fast field reader reader associated to `field`.
+    ///
+    /// If `field` is not a u64 multi-valued fast field, this method returns `None`.
+    pub fn u64s(&self, field: Field) -> Option<MultiValueIntFastFieldReader<u64>> {
+        self.fast_field_u64s.get(&field).cloned()
+    }
+
+    /// If the field is a u64s-fast field return the associated reader.
+    /// If the field is a i64s-fast field, return the associated u64s reader. Values are
+    /// mapped from i64 to u64 using a (well the, it is unique) monotonic mapping.
+    ///
+    /// This method is useful when merging segment reader.
+    pub(crate) fn u64s_lenient(&self, field: Field) -> Option<MultiValueIntFastFieldReader<u64>> {
+        if let Some(u64s_ff_reader) = self.u64s(field) {
+            return Some(u64s_ff_reader);
+        }
+        if let Some(i64s_ff_reader) = self.i64s(field) {
+            return Some(i64s_ff_reader.into_u64s_reader());
+        }
+        None
+    }
+
+    /// Returns a `i64s` multi-valued fast field reader reader associated to `field`.
+    ///
+    /// If `field` is not a i64 multi-valued fast field, this method returns `None`.
+    pub fn i64s(&self, field: Field) -> Option<MultiValueIntFastFieldReader<i64>> {
+        self.fast_field_i64s.get(&field).cloned()
+    }
+
+    /// Returns the `bytes` fast field reader associated to `field`.
+    ///
+    /// If `field` is not a bytes fast field, returns `None`.
+    pub fn bytes(&self, field: Field) -> Option<BytesFastFieldReader> {
+        self.fast_bytes.get(&field).cloned()
+    }
+}

src/indexer/delete_queue.rs

@@ -2,6 +2,7 @@ use super::operation::DeleteOperation;
 use std::mem;
 use std::ops::DerefMut;
 use std::sync::{Arc, RwLock};
+use Opstamp;
 
 // The DeleteQueue is similar in conceptually to a multiple
 // consumer single producer broadcast channel.
@@ -184,7 +185,7 @@ impl DeleteCursor {
     ///   queue are consume and the next get will return None.
     /// - the next get will return the first operation with an
     /// `opstamp >= target_opstamp`.
-    pub fn skip_to(&mut self, target_opstamp: u64) {
+    pub fn skip_to(&mut self, target_opstamp: Opstamp) {
         // TODO Can be optimize as we work with block.
         while self.is_behind_opstamp(target_opstamp) {
             self.advance();
@@ -192,7 +193,7 @@ impl DeleteCursor {
     }
 
     #[cfg_attr(feature = "cargo-clippy", allow(clippy::wrong_self_convention))]
-    fn is_behind_opstamp(&mut self, target_opstamp: u64) -> bool {
+    fn is_behind_opstamp(&mut self, target_opstamp: Opstamp) -> bool {
         self.get()
             .map(|operation| operation.opstamp < target_opstamp)
             .unwrap_or(false)

src/indexer/doc_opstamp_mapping.rs

@@ -1,5 +1,6 @@
 use std::sync::Arc;
 use DocId;
+use Opstamp;
 
 // Doc to opstamp is used to identify which
 // document should be deleted.
@@ -23,7 +24,7 @@ pub enum DocToOpstampMapping {
 }
 
 impl From<Vec<u64>> for DocToOpstampMapping {
-    fn from(opstamps: Vec<u64>) -> DocToOpstampMapping {
+    fn from(opstamps: Vec<Opstamp>) -> DocToOpstampMapping {
         DocToOpstampMapping::WithMap(Arc::new(opstamps))
     }
 }
@@ -35,7 +36,7 @@ impl DocToOpstampMapping {
     //
     // The edge case opstamp = some doc opstamp is in practise
     // never called.
-    pub fn compute_doc_limit(&self, target_opstamp: u64) -> DocId {
+    pub fn compute_doc_limit(&self, target_opstamp: Opstamp) -> DocId {
         match *self {
             DocToOpstampMapping::WithMap(ref doc_opstamps) => {
                 match doc_opstamps.binary_search(&target_opstamp) {

src/indexer/index_writer.rs

@@ -30,6 +30,7 @@ use std::ops::Range;
 use std::sync::Arc;
 use std::thread;
 use std::thread::JoinHandle;
+use Opstamp;
 use Result;
 
 // Size of the margin for the heap. A segment is closed when the remaining memory
@@ -99,7 +100,7 @@ pub struct IndexWriter {
     delete_queue: DeleteQueue,
 
     stamper: Stamper,
-    committed_opstamp: u64,
+    committed_opstamp: Opstamp,
 }
 
 /// Open a new index writer. Attempts to acquire a lockfile.
@@ -177,7 +178,7 @@ pub fn compute_deleted_bitset(
     segment_reader: &SegmentReader,
     delete_cursor: &mut DeleteCursor,
     doc_opstamps: &DocToOpstampMapping,
-    target_opstamp: u64,
+    target_opstamp: Opstamp,
 ) -> Result<bool> {
     let mut might_have_changed = false;
 
@@ -219,7 +220,7 @@ pub fn compute_deleted_bitset(
 pub fn advance_deletes(
     mut segment: Segment,
     segment_entry: &mut SegmentEntry,
-    target_opstamp: u64,
+    target_opstamp: Opstamp,
 ) -> Result<()> {
     {
         if segment_entry.meta().delete_opstamp() == Some(target_opstamp) {
@@ -299,11 +300,11 @@ fn index_documents(
     // the worker thread.
     assert!(num_docs > 0);
 
-    let doc_opstamps: Vec<u64> = segment_writer.finalize()?;
+    let doc_opstamps: Vec<Opstamp> = segment_writer.finalize()?;
 
     let segment_meta = SegmentMeta::new(segment_id, num_docs);
 
-    let last_docstamp: u64 = *(doc_opstamps.last().unwrap());
+    let last_docstamp: Opstamp = *(doc_opstamps.last().unwrap());
 
     let delete_bitset_opt = if delete_cursor.get().is_some() {
         let doc_to_opstamps = DocToOpstampMapping::from(doc_opstamps);
@@ -494,7 +495,7 @@ impl IndexWriter {
     /// state as it was after the last commit.
     ///
     /// The opstamp at the last commit is returned.
-    pub fn rollback(&mut self) -> Result<()> {
+    pub fn rollback(&mut self) -> Result<Opstamp> {
         info!("Rolling back to opstamp {}", self.committed_opstamp);
 
         // marks the segment updater as killed. From now on, all
@@ -529,7 +530,7 @@ impl IndexWriter {
         // was dropped with the index_writer.
         for _ in document_receiver.clone() {}
 
-        Ok(())
+        Ok(self.committed_opstamp)
     }
 
     /// Prepares a commit.
@@ -567,7 +568,7 @@ impl IndexWriter {
         info!("Preparing commit");
 
         // this will drop the current document channel
-        // and recreate a new one channels.
+        // and recreate a new one.
         self.recreate_document_channel();
 
         let former_workers_join_handle = mem::replace(&mut self.workers_join_handle, Vec::new());
@@ -601,7 +602,7 @@ impl IndexWriter {
     /// Commit returns the `opstamp` of the last document
     /// that made it in the commit.
     ///
-    pub fn commit(&mut self) -> Result<u64> {
+    pub fn commit(&mut self) -> Result<Opstamp> {
         self.prepare_commit()?.commit()
     }
 
@@ -617,7 +618,7 @@ impl IndexWriter {
     ///
     /// Like adds, the deletion itself will be visible
     /// only after calling `commit()`.
-    pub fn delete_term(&mut self, term: Term) -> u64 {
+    pub fn delete_term(&self, term: Term) -> Opstamp {
         let opstamp = self.stamper.stamp();
         let delete_operation = DeleteOperation { opstamp, term };
         self.delete_queue.push(delete_operation);
@@ -631,7 +632,7 @@ impl IndexWriter {
     ///
     /// This is also the opstamp of the commit that is currently
     /// available for searchers.
-    pub fn commit_opstamp(&self) -> u64 {
+    pub fn commit_opstamp(&self) -> Opstamp {
         self.committed_opstamp
     }
 
@@ -645,7 +646,7 @@ impl IndexWriter {
     ///
     /// Currently it represents the number of documents that
     /// have been added since the creation of the index.
-    pub fn add_document(&mut self, document: Document) -> u64 {
+    pub fn add_document(&self, document: Document) -> Opstamp {
         let opstamp = self.stamper.stamp();
         let add_operation = AddOperation { opstamp, document };
         let send_result = self.operation_sender.send(vec![add_operation]);
@@ -662,7 +663,7 @@ impl IndexWriter {
     /// The total number of stamps generated by this method is `count + 1`;
     /// each operation gets a stamp from the `stamps` iterator and `last_opstamp`
     /// is for the batch itself.
-    fn get_batch_opstamps(&mut self, count: u64) -> (u64, Range<u64>) {
+    fn get_batch_opstamps(&self, count: Opstamp) -> (Opstamp, Range<Opstamp>) {
         let Range { start, end } = self.stamper.stamps(count + 1u64);
         let last_opstamp = end - 1;
         let stamps = Range {
@@ -688,7 +689,7 @@ impl IndexWriter {
     /// Like adds and deletes (see `IndexWriter.add_document` and
     /// `IndexWriter.delete_term`), the changes made by calling `run` will be
     /// visible to readers only after calling `commit()`.
-    pub fn run(&mut self, user_operations: Vec<UserOperation>) -> u64 {
+    pub fn run(&self, user_operations: Vec<UserOperation>) -> Opstamp {
         let count = user_operations.len() as u64;
         if count == 0 {
             return self.stamper.stamp();
@@ -739,7 +740,7 @@ mod tests {
         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();
+        let index_writer = index.writer_with_num_threads(1, 3_000_000).unwrap();
         let operations = vec![
             UserOperation::Add(doc!(text_field=>"a")),
             UserOperation::Add(doc!(text_field=>"b")),
@@ -801,7 +802,7 @@ mod tests {
     fn test_empty_operations_group() {
         let schema_builder = schema::Schema::builder();
         let index = Index::create_in_ram(schema_builder.build());
-        let mut index_writer = index.writer(3_000_000).unwrap();
+        let index_writer = index.writer(3_000_000).unwrap();
         let operations1 = vec![];
         let batch_opstamp1 = index_writer.run(operations1);
         assert_eq!(batch_opstamp1, 0u64);

src/indexer/log_merge_policy.rs

@@ -52,7 +52,7 @@ impl MergePolicy for LogMergePolicy {
 
         let mut size_sorted_tuples = segments
             .iter()
-            .map(|x| x.num_docs())
+            .map(SegmentMeta::num_docs)
             .enumerate()
             .collect::<Vec<(usize, u32)>>();
 

src/indexer/merge_operation.rs

@@ -1,5 +1,6 @@
 use census::{Inventory, TrackedObject};
 use std::collections::HashSet;
+use Opstamp;
 use SegmentId;
 
 #[derive(Default)]
@@ -17,8 +18,8 @@ impl MergeOperationInventory {
     }
 }
 
-/// A `MergeOperation` has two role.
-/// It carries all of the information required to describe a merge :
+/// A `MergeOperation` has two roles.
+/// It carries all of the information required to describe a merge:
 /// - `target_opstamp` is the opstamp up to which we want to consume the
 /// delete queue and reflect their deletes.
 /// - `segment_ids` is the list of segment to be merged.
@@ -35,14 +36,14 @@ pub struct MergeOperation {
 }
 
 struct InnerMergeOperation {
-    target_opstamp: u64,
+    target_opstamp: Opstamp,
     segment_ids: Vec<SegmentId>,
 }
 
 impl MergeOperation {
     pub fn new(
         inventory: &MergeOperationInventory,
-        target_opstamp: u64,
+        target_opstamp: Opstamp,
         segment_ids: Vec<SegmentId>,
     ) -> MergeOperation {
         let inner_merge_operation = InnerMergeOperation {
@@ -54,7 +55,7 @@ impl MergeOperation {
         }
     }
 
-    pub fn target_opstamp(&self) -> u64 {
+    pub fn target_opstamp(&self) -> Opstamp {
         self.inner.target_opstamp
     }
 

src/indexer/merger.rs

@@ -3,6 +3,7 @@ use core::Segment;
 use core::SegmentReader;
 use core::SerializableSegment;
 use docset::DocSet;
+use fastfield::BytesFastFieldReader;
 use fastfield::DeleteBitSet;
 use fastfield::FastFieldReader;
 use fastfield::FastFieldSerializer;
@@ -72,7 +73,7 @@ fn compute_min_max_val(
                 // some deleted documents,
                 // we need to recompute the max / min
                 (0..max_doc)
-                    .filter(|doc_id| !delete_bitset.is_deleted(*doc_id))
+                    .filter(|doc_id| delete_bitset.is_alive(*doc_id))
                     .map(|doc_id| u64_reader.get(doc_id))
                     .minmax()
                     .into_option()
@@ -239,7 +240,10 @@ impl IndexMerger {
         let mut max_value = u64::min_value();
 
         for reader in &self.readers {
-            let u64_reader: FastFieldReader<u64> = reader.fast_field_reader(field)?;
+            let u64_reader: FastFieldReader<u64> = reader
+                .fast_fields()
+                .u64_lenient(field)
+                .expect("Failed to find a reader for single fast field. This is a tantivy bug and it should never happen.");
             if let Some((seg_min_val, seg_max_val)) =
                 compute_min_max_val(&u64_reader, reader.max_doc(), reader.delete_bitset())
             {
@@ -282,24 +286,28 @@ impl IndexMerger {
         fast_field_serializer: &mut FastFieldSerializer,
     ) -> Result<()> {
         let mut total_num_vals = 0u64;
+        let mut u64s_readers: Vec<MultiValueIntFastFieldReader<u64>> = Vec::new();
 
         // In the first pass, we compute the total number of vals.
         //
         // This is required by the bitpacker, as it needs to know
         // what should be the bit length use for bitpacking.
         for reader in &self.readers {
-            let idx_reader = reader.fast_field_reader_with_idx::<u64>(field, 0)?;
+            let u64s_reader = reader.fast_fields()
+                .u64s_lenient(field)
+                .expect("Failed to find index for multivalued field. This is a bug in tantivy, please report.");
+
             if let Some(delete_bitset) = reader.delete_bitset() {
                 for doc in 0u32..reader.max_doc() {
-                    if !delete_bitset.is_deleted(doc) {
-                        let start = idx_reader.get(doc);
-                        let end = idx_reader.get(doc + 1);
-                        total_num_vals += end - start;
+                    if delete_bitset.is_alive(doc) {
+                        let num_vals = u64s_reader.num_vals(doc) as u64;
+                        total_num_vals += num_vals;
                     }
                 }
             } else {
-                total_num_vals += idx_reader.max_value();
+                total_num_vals += u64s_reader.total_num_vals();
             }
+            u64s_readers.push(u64s_reader);
         }
 
         // We can now create our `idx` serializer, and in a second pass,
@@ -307,13 +315,10 @@ impl IndexMerger {
         let mut serialize_idx =
             fast_field_serializer.new_u64_fast_field_with_idx(field, 0, total_num_vals, 0)?;
         let mut idx = 0;
-        for reader in &self.readers {
-            let idx_reader = reader.fast_field_reader_with_idx::<u64>(field, 0)?;
-            for doc in reader.doc_ids_alive() {
+        for (segment_reader, u64s_reader) in self.readers.iter().zip(&u64s_readers) {
+            for doc in segment_reader.doc_ids_alive() {
                 serialize_idx.add_val(idx)?;
-                let start = idx_reader.get(doc);
-                let end = idx_reader.get(doc + 1);
-                idx += end - start;
+                idx += u64s_reader.num_vals(doc) as u64;
             }
         }
         serialize_idx.add_val(idx)?;
@@ -344,8 +349,10 @@ impl IndexMerger {
             for (segment_ord, segment_reader) in self.readers.iter().enumerate() {
                 let term_ordinal_mapping: &[TermOrdinal] =
                     term_ordinal_mappings.get_segment(segment_ord);
-                let ff_reader: MultiValueIntFastFieldReader<u64> =
-                    segment_reader.multi_fast_field_reader(field)?;
+                let ff_reader: MultiValueIntFastFieldReader<u64> = segment_reader
+                    .fast_fields()
+                    .u64s(field)
+                    .expect("Could not find multivalued u64 fast value reader.");
                 // TODO optimize if no deletes
                 for doc in segment_reader.doc_ids_alive() {
                     ff_reader.get_vals(doc, &mut vals);
@@ -377,6 +384,8 @@ impl IndexMerger {
 
         let mut vals = Vec::with_capacity(100);
 
+        let mut ff_readers = Vec::new();
+
         // Our values are bitpacked and we need to know what should be
         // our bitwidth and our minimum value before serializing any values.
         //
@@ -385,7 +394,10 @@ impl IndexMerger {
         // maximum value and initialize our Serializer.
         for reader in &self.readers {
             let ff_reader: MultiValueIntFastFieldReader<u64> =
-                reader.multi_fast_field_reader(field)?;
+                reader.fast_fields().u64s_lenient(field).expect(
+                    "Failed to find multivalued fast field reader. This is a bug in \
+                     tantivy. Please report.",
+                );
             for doc in reader.doc_ids_alive() {
                 ff_reader.get_vals(doc, &mut vals);
                 for &val in &vals {
@@ -393,6 +405,7 @@ impl IndexMerger {
                     max_value = cmp::max(val, max_value);
                 }
             }
+            ff_readers.push(ff_reader);
             // TODO optimize when no deletes
         }
 
@@ -405,9 +418,7 @@ impl IndexMerger {
         {
             let mut serialize_vals = fast_field_serializer
                 .new_u64_fast_field_with_idx(field, min_value, max_value, 1)?;
-            for reader in &self.readers {
-                let ff_reader: MultiValueIntFastFieldReader<u64> =
-                    reader.multi_fast_field_reader(field)?;
+            for (reader, ff_reader) in self.readers.iter().zip(ff_readers) {
                 // TODO optimize if no deletes
                 for doc in reader.doc_ids_alive() {
                     ff_reader.get_vals(doc, &mut vals);
@@ -426,19 +437,53 @@ impl IndexMerger {
         field: Field,
         fast_field_serializer: &mut FastFieldSerializer,
     ) -> Result<()> {
-        self.write_fast_field_idx(field, fast_field_serializer)?;
+        let mut total_num_vals = 0u64;
+        let mut bytes_readers: Vec<BytesFastFieldReader> = Vec::new();
+
+        for reader in &self.readers {
+            let bytes_reader = reader.fast_fields().bytes(field).expect(
+                "Failed to find bytes fast field reader. This is a bug in tantivy, please report.",
+            );
+            if let Some(delete_bitset) = reader.delete_bitset() {
+                for doc in 0u32..reader.max_doc() {
+                    if delete_bitset.is_alive(doc) {
+                        let num_vals = bytes_reader.get_bytes(doc).len() as u64;
+                        total_num_vals += num_vals;
+                    }
+                }
+            } else {
+                total_num_vals += bytes_reader.total_num_bytes() as u64;
+            }
+            bytes_readers.push(bytes_reader);
+        }
+
+        {
+            // We can now create our `idx` serializer, and in a second pass,
+            // can effectively push the different indexes.
+            let mut serialize_idx =
+                fast_field_serializer.new_u64_fast_field_with_idx(field, 0, total_num_vals, 0)?;
+            let mut idx = 0;
+            for (segment_reader, bytes_reader) in self.readers.iter().zip(&bytes_readers) {
+                for doc in segment_reader.doc_ids_alive() {
+                    serialize_idx.add_val(idx)?;
+                    idx += bytes_reader.get_bytes(doc).len() as u64;
+                }
+            }
+            serialize_idx.add_val(idx)?;
+            serialize_idx.close_field()?;
+        }
 
         let mut serialize_vals = fast_field_serializer.new_bytes_fast_field_with_idx(field, 1)?;
-        for reader in &self.readers {
-            let bytes_reader = reader.bytes_fast_field_reader(field)?;
+        for segment_reader in &self.readers {
+            let bytes_reader = segment_reader.fast_fields().bytes(field)
+                .expect("Failed to find bytes field in fast field reader. This is a bug in tantivy. Please report.");
             // TODO: optimize if no deletes
-            for doc in reader.doc_ids_alive() {
-                let val = bytes_reader.get_val(doc);
+            for doc in segment_reader.doc_ids_alive() {
+                let val = bytes_reader.get_bytes(doc);
                 serialize_vals.write_all(val)?;
             }
         }
         serialize_vals.flush()?;
-
         Ok(())
     }
 
@@ -979,14 +1024,16 @@ mod tests {
 
             let score_field_reader = searcher
                 .segment_reader(0)
-                .fast_field_reader::<u64>(score_field)
+                .fast_fields()
+                .u64(score_field)
                 .unwrap();
             assert_eq!(score_field_reader.min_value(), 4000);
             assert_eq!(score_field_reader.max_value(), 7000);
 
             let score_field_reader = searcher
                 .segment_reader(1)
-                .fast_field_reader::<u64>(score_field)
+                .fast_fields()
+                .u64(score_field)
                 .unwrap();
             assert_eq!(score_field_reader.min_value(), 1);
             assert_eq!(score_field_reader.max_value(), 3);
@@ -1037,7 +1084,8 @@ mod tests {
             );
             let score_field_reader = searcher
                 .segment_reader(0)
-                .fast_field_reader::<u64>(score_field)
+                .fast_fields()
+                .u64(score_field)
                 .unwrap();
             assert_eq!(score_field_reader.min_value(), 3);
             assert_eq!(score_field_reader.max_value(), 7000);
@@ -1083,7 +1131,8 @@ mod tests {
             );
             let score_field_reader = searcher
                 .segment_reader(0)
-                .fast_field_reader::<u64>(score_field)
+                .fast_fields()
+                .u64(score_field)
                 .unwrap();
             assert_eq!(score_field_reader.min_value(), 3);
             assert_eq!(score_field_reader.max_value(), 7000);
@@ -1135,7 +1184,8 @@ mod tests {
             );
             let score_field_reader = searcher
                 .segment_reader(0)
-                .fast_field_reader::<u64>(score_field)
+                .fast_fields()
+                .u64(score_field)
                 .unwrap();
             assert_eq!(score_field_reader.min_value(), 6000);
             assert_eq!(score_field_reader.max_value(), 7000);
@@ -1381,7 +1431,7 @@ mod tests {
 
         {
             let segment = searcher.segment_reader(0u32);
-            let ff_reader = segment.multi_fast_field_reader(int_field).unwrap();
+            let ff_reader = segment.fast_fields().u64s(int_field).unwrap();
 
             ff_reader.get_vals(0, &mut vals);
             assert_eq!(&vals, &[1, 2]);
@@ -1416,7 +1466,7 @@ mod tests {
 
         {
             let segment = searcher.segment_reader(1u32);
-            let ff_reader = segment.multi_fast_field_reader(int_field).unwrap();
+            let ff_reader = segment.fast_fields().u64s(int_field).unwrap();
             ff_reader.get_vals(0, &mut vals);
             assert_eq!(&vals, &[28, 27]);
 
@@ -1426,7 +1476,7 @@ mod tests {
 
         {
             let segment = searcher.segment_reader(2u32);
-            let ff_reader = segment.multi_fast_field_reader(int_field).unwrap();
+            let ff_reader = segment.fast_fields().u64s(int_field).unwrap();
             ff_reader.get_vals(0, &mut vals);
             assert_eq!(&vals, &[20]);
         }
@@ -1459,7 +1509,7 @@ mod tests {
                     .collect::<Vec<_>>()
             );
             let segment = searcher.segment_reader(0u32);
-            let ff_reader = segment.multi_fast_field_reader(int_field).unwrap();
+            let ff_reader = segment.fast_fields().u64s(int_field).unwrap();
 
             ff_reader.get_vals(0, &mut vals);
             assert_eq!(&vals, &[1, 2]);

src/indexer/operation.rs

@@ -1,17 +1,18 @@
 use schema::Document;
 use schema::Term;
+use Opstamp;
 
 /// Timestamped Delete operation.
 #[derive(Clone, Eq, PartialEq, Debug)]
 pub struct DeleteOperation {
-    pub opstamp: u64,
+    pub opstamp: Opstamp,
     pub term: Term,
 }
 
 /// Timestamped Add operation.
 #[derive(Eq, PartialEq, Debug)]
 pub struct AddOperation {
-    pub opstamp: u64,
+    pub opstamp: Opstamp,
     pub document: Document,
 }
 

src/indexer/prepared_commit.rs

@@ -1,15 +1,16 @@
 use super::IndexWriter;
+use Opstamp;
 use Result;
 
 /// A prepared commit
 pub struct PreparedCommit<'a> {
     index_writer: &'a mut IndexWriter,
     payload: Option<String>,
-    opstamp: u64,
+    opstamp: Opstamp,
 }
 
 impl<'a> PreparedCommit<'a> {
-    pub(crate) fn new(index_writer: &'a mut IndexWriter, opstamp: u64) -> PreparedCommit {
+    pub(crate) fn new(index_writer: &'a mut IndexWriter, opstamp: Opstamp) -> PreparedCommit {
         PreparedCommit {
             index_writer,
             payload: None,
@@ -17,7 +18,7 @@ impl<'a> PreparedCommit<'a> {
         }
     }
 
-    pub fn opstamp(&self) -> u64 {
+    pub fn opstamp(&self) -> Opstamp {
         self.opstamp
     }
 
@@ -25,11 +26,11 @@ impl<'a> PreparedCommit<'a> {
         self.payload = Some(payload.to_string())
     }
 
-    pub fn abort(self) -> Result<()> {
+    pub fn abort(self) -> Result<Opstamp> {
         self.index_writer.rollback()
     }
 
-    pub fn commit(self) -> Result<u64> {
+    pub fn commit(self) -> Result<Opstamp> {
         info!("committing {}", self.opstamp);
         self.index_writer
             .segment_updater()

src/indexer/segment_register.rs

@@ -56,7 +56,7 @@ impl SegmentRegister {
             .values()
             .map(|segment_entry| segment_entry.meta().clone())
             .collect();
-        segment_ids.sort_by_key(|meta| meta.id());
+        segment_ids.sort_by_key(SegmentMeta::id);
         segment_ids
     }
 

src/indexer/segment_updater.rs

@@ -36,6 +36,7 @@ use std::sync::Arc;
 use std::sync::RwLock;
 use std::thread;
 use std::thread::JoinHandle;
+use Opstamp;
 use Result;
 
 /// Save the index meta file.
@@ -224,7 +225,7 @@ impl SegmentUpdater {
     ///
     /// Tne method returns copies of the segment entries,
     /// updated with the delete information.
-    fn purge_deletes(&self, target_opstamp: u64) -> Result<Vec<SegmentEntry>> {
+    fn purge_deletes(&self, target_opstamp: Opstamp) -> Result<Vec<SegmentEntry>> {
         let mut segment_entries = self.0.segment_manager.segment_entries();
         for segment_entry in &mut segment_entries {
             let segment = self.0.index.segment(segment_entry.meta().clone());
@@ -233,7 +234,7 @@ impl SegmentUpdater {
         Ok(segment_entries)
     }
 
-    pub fn save_metas(&self, opstamp: u64, commit_message: Option<String>) {
+    pub fn save_metas(&self, opstamp: Opstamp, commit_message: Option<String>) {
         if self.is_alive() {
             let index = &self.0.index;
             let directory = index.directory();
@@ -280,7 +281,7 @@ impl SegmentUpdater {
             .garbage_collect(|| self.0.segment_manager.list_files());
     }
 
-    pub fn commit(&self, opstamp: u64, payload: Option<String>) -> Result<()> {
+    pub fn commit(&self, opstamp: Opstamp, payload: Option<String>) -> Result<()> {
         self.run_async(move |segment_updater| {
             if segment_updater.is_alive() {
                 let segment_entries = segment_updater

src/indexer/segment_writer.rs

@@ -5,6 +5,7 @@ use fastfield::FastFieldsWriter;
 use fieldnorm::FieldNormsWriter;
 use indexer::segment_serializer::SegmentSerializer;
 use postings::MultiFieldPostingsWriter;
+use schema::FieldEntry;
 use schema::FieldType;
 use schema::Schema;
 use schema::Term;
@@ -15,6 +16,7 @@ use tokenizer::BoxedTokenizer;
 use tokenizer::FacetTokenizer;
 use tokenizer::{TokenStream, Tokenizer};
 use DocId;
+use Opstamp;
 use Result;
 
 /// A `SegmentWriter` is in charge of creating segment index from a
@@ -28,7 +30,7 @@ pub struct SegmentWriter {
     segment_serializer: SegmentSerializer,
     fast_field_writers: FastFieldsWriter,
     fieldnorms_writer: FieldNormsWriter,
-    doc_opstamps: Vec<u64>,
+    doc_opstamps: Vec<Opstamp>,
     tokenizers: Vec<Option<Box<BoxedTokenizer>>>,
 }
 
@@ -53,7 +55,7 @@ impl SegmentWriter {
             schema
                 .fields()
                 .iter()
-                .map(|field_entry| field_entry.field_type())
+                .map(FieldEntry::field_type)
                 .map(|field_type| match *field_type {
                     FieldType::Str(ref text_options) => text_options
                         .get_indexing_options()

src/indexer/stamper.rs

@@ -1,70 +1,27 @@
 use std::ops::Range;
-use std::sync::atomic::Ordering;
+use std::sync::atomic::{AtomicU64, Ordering};
 use std::sync::Arc;
+use Opstamp;
 
-// AtomicU64 have not landed in stable.
-// For the moment let's just use AtomicUsize on
-// x86/64 bit platform, and a mutex on other platform.
-#[cfg(target_arch = "x86_64")]
-mod archicture_impl {
-
-    use std::sync::atomic::{AtomicUsize, Ordering};
-
-    #[derive(Default)]
-    pub struct AtomicU64Ersatz(AtomicUsize);
-
-    impl AtomicU64Ersatz {
-        pub fn new(first_opstamp: u64) -> AtomicU64Ersatz {
-            AtomicU64Ersatz(AtomicUsize::new(first_opstamp as usize))
-        }
-
-        pub fn fetch_add(&self, val: u64, order: Ordering) -> u64 {
-            self.0.fetch_add(val as usize, order) as u64
-        }
-    }
-}
-
-#[cfg(not(target_arch = "x86_64"))]
-mod archicture_impl {
-
-    use std::sync::atomic::Ordering;
-    /// Under other architecture, we rely on a mutex.
-    use std::sync::RwLock;
-
-    #[derive(Default)]
-    pub struct AtomicU64Ersatz(RwLock<u64>);
-
-    impl AtomicU64Ersatz {
-        pub fn new(first_opstamp: u64) -> AtomicU64Ersatz {
-            AtomicU64Ersatz(RwLock::new(first_opstamp))
-        }
-
-        pub fn fetch_add(&self, incr: u64, _order: Ordering) -> u64 {
-            let mut lock = self.0.write().unwrap();
-            let previous_val = *lock;
-            *lock = previous_val + incr;
-            previous_val
-        }
-    }
-}
-
-use self::archicture_impl::AtomicU64Ersatz;
-
+/// Stamper provides Opstamps, which is just an auto-increment id to label
+/// an operation.
+///
+/// Cloning does not "fork" the stamp generation. The stamper actually wraps an `Arc`.
 #[derive(Clone, Default)]
-pub struct Stamper(Arc<AtomicU64Ersatz>);
+pub struct Stamper(Arc<AtomicU64>);
 
 impl Stamper {
-    pub fn new(first_opstamp: u64) -> Stamper {
-        Stamper(Arc::new(AtomicU64Ersatz::new(first_opstamp)))
+    pub fn new(first_opstamp: Opstamp) -> Stamper {
+        Stamper(Arc::new(AtomicU64::new(first_opstamp)))
     }
 
-    pub fn stamp(&self) -> u64 {
+    pub fn stamp(&self) -> Opstamp {
         self.0.fetch_add(1u64, Ordering::SeqCst) as u64
     }
 
     /// Given a desired count `n`, `stamps` returns an iterator that
     /// will supply `n` number of u64 stamps.
-    pub fn stamps(&self, n: u64) -> Range<u64> {
+    pub fn stamps(&self, n: u64) -> Range<Opstamp> {
         let start = self.0.fetch_add(n, Ordering::SeqCst);
         Range {
             start,
@@ -92,4 +49,5 @@ mod test {
         assert_eq!(stamper.stamps(3u64), (12..15));
         assert_eq!(stamper.stamp(), 15u64);
     }
+
 }

src/lib.rs

@@ -226,7 +226,7 @@ mod docset;
 pub use self::docset::{DocSet, SkipResult};
 
 pub use core::SegmentComponent;
-pub use core::{Index, Searcher, Segment, SegmentId, SegmentMeta};
+pub use core::{Index, Searcher, Segment, SegmentId, SegmentMeta, IndexMeta};
 pub use core::{InvertedIndexReader, SegmentReader};
 pub use directory::Directory;
 pub use indexer::IndexWriter;
@@ -254,6 +254,16 @@ pub mod merge_policy {
 /// as they are added in the segment.
 pub type DocId = u32;
 
+/// A u64 assigned to every operation incrementally
+///
+/// All operations modifying the index receives an monotonic Opstamp.
+/// The resulting state of the index is consistent with the opstamp ordering.
+///
+/// For instance, a commit with opstamp `32_423` will reflect all Add and Delete operations
+/// with an opstamp `<= 32_423`. A delete operation with opstamp n will no affect a document added
+/// with opstamp `n+1`.
+pub type Opstamp = u64;
+
 /// A f32 that represents the relevance of the document to the query
 ///
 /// This is modelled internally as a `f32`. The
@@ -876,28 +886,28 @@ mod tests {
         let searcher = reader.searcher();
         let segment_reader: &SegmentReader = searcher.segment_reader(0);
         {
-            let fast_field_reader_res = segment_reader.fast_field_reader::<u64>(text_field);
-            assert!(fast_field_reader_res.is_err());
+            let fast_field_reader_opt = segment_reader.fast_fields().u64(text_field);
+            assert!(fast_field_reader_opt.is_none());
         }
         {
-            let fast_field_reader_res = segment_reader.fast_field_reader::<u64>(stored_int_field);
-            assert!(fast_field_reader_res.is_err());
+            let fast_field_reader_opt = segment_reader.fast_fields().u64(stored_int_field);
+            assert!(fast_field_reader_opt.is_none());
         }
         {
-            let fast_field_reader_res = segment_reader.fast_field_reader::<u64>(fast_field_signed);
-            assert!(fast_field_reader_res.is_err());
+            let fast_field_reader_opt = segment_reader.fast_fields().u64(fast_field_signed);
+            assert!(fast_field_reader_opt.is_none());
         }
         {
-            let fast_field_reader_res = segment_reader.fast_field_reader::<i64>(fast_field_signed);
-            assert!(fast_field_reader_res.is_ok());
-            let fast_field_reader = fast_field_reader_res.unwrap();
+            let fast_field_reader_opt = segment_reader.fast_fields().i64(fast_field_signed);
+            assert!(fast_field_reader_opt.is_some());
+            let fast_field_reader = fast_field_reader_opt.unwrap();
             assert_eq!(fast_field_reader.get(0), 4i64)
         }
 
         {
-            let fast_field_reader_res = segment_reader.fast_field_reader::<i64>(fast_field_signed);
-            assert!(fast_field_reader_res.is_ok());
-            let fast_field_reader = fast_field_reader_res.unwrap();
+            let fast_field_reader_opt = segment_reader.fast_fields().i64(fast_field_signed);
+            assert!(fast_field_reader_opt.is_some());
+            let fast_field_reader = fast_field_reader_opt.unwrap();
             assert_eq!(fast_field_reader.get(0), 4i64)
         }
     }

src/postings/serializer.rs

@@ -175,7 +175,7 @@ impl<'a> FieldSerializer<'a> {
         let positions_idx = self
             .positions_serializer_opt
             .as_ref()
-            .map(|positions_serializer| positions_serializer.positions_idx())
+            .map(PositionSerializer::positions_idx)
             .unwrap_or(0u64);
         TermInfo {
             doc_freq: 0,

src/query/boolean_query/mod.rs

@@ -205,4 +205,332 @@ mod tests {
             assert_eq!(score_docs(&boolean_query), vec![0.977973, 0.84699446]);
         }
     }
+
+    /*
+            DoC 0
+            {
+            "_index": "test",
+            "_type": "_doc",
+            "_id": "0",
+            "matched": true,
+            "explanation": {
+                "value": 6.2610235,
+                "description": "max of:",
+                "details": [{
+                    "value": 6.1969156,
+                    "description": "sum of:",
+                    "details": [{
+                        "value": 6.1969156,
+                        "description": "weight(text:оксана in 561) [PerFieldSimilarity], result of:",
+                        "details": [{
+                            "value": 6.1969156,
+                            "description": "score(freq=1.0), product of:",
+                            "details": [{
+                                "value": 2.2,
+                                "description": "boost",
+                                "details": []
+                            }, {
+                                "value": 5.65998,
+                                "description": "idf, computed as log(1 + (N - n + 0.5) / (n + 0.5)) from:",
+                                "details": [{
+                                    "value": 3,
+                                    "description": "n, number of documents containing term",
+                                    "details": []
+                                }, {
+                                    "value": 1004,
+                                    "description": "N, total number of documents with field",
+                                    "details": []
+                                }]
+                            }, {
+                                "value": 0.49766606,
+                                "description": "tf, computed as freq / (freq + k1 * (1 - b + b * dl / avgdl)) from:",
+                                "details": [{
+                                    "value": 1.0,
+                                    "description": "freq, occurrences of term within document",
+                                    "details": []
+                                }, {
+                                    "value": 1.2,
+                                    "description": "k1, term saturation parameter",
+                                    "details": []
+                                }, {
+                                    "value": 0.75,
+                                    "description": "b, length normalization parameter",
+                                    "details": []
+                                }, {
+                                    "value": 19.0,
+                                    "description": "dl, length of field",
+                                    "details": []
+                                }, {
+                                    "value": 24.105577,
+                                    "description": "avgdl, average length of field",
+                                    "details": []
+                                }]
+                            }]
+                        }]
+                    }]
+                }, {
+                    "value": 6.2610235,
+                    "description": "sum of:",
+                    "details": [{
+                        "value": 6.2610235,
+                        "description": "weight(title:оксана in 561) [PerFieldSimilarity], result of:",
+                        "details": [{
+                            "value": 6.2610235,
+                            "description": "score(freq=1.0), product of:",
+                            "details": [{
+                                "value": 2.2,
+                                "description": "boost",
+                                "details": []
+                            }, {
+                                "value": 5.4086657,
+                                "description": "idf, computed as log(1 + (N - n + 0.5) / (n + 0.5)) from:",
+                                "details": [{
+                                    "value": 4,
+                                    "description": "n, number of documents containing term",
+                                    "details": []
+                                }, {
+                                    "value": 1004,
+                                    "description": "N, total number of documents with field",
+                                    "details": []
+                                }]
+                            }, {
+                                "value": 0.52617776,
+                                "description": "tf, computed as freq / (freq + k1 * (1 - b + b * dl / avgdl)) from:",
+                                "details": [{
+                                    "value": 1.0,
+                                    "description": "freq, occurrences of term within document",
+                                    "details": []
+                                }, {
+                                    "value": 1.2,
+                                    "description": "k1, term saturation parameter",
+                                    "details": []
+                                }, {
+                                    "value": 0.75,
+                                    "description": "b, length normalization parameter",
+                                    "details": []
+                                }, {
+                                    "value": 4.0,
+                                    "description": "dl, length of field",
+                                    "details": []
+                                }, {
+                                    "value": 5.99502,
+                                    "description": "avgdl, average length of field",
+                                    "details": []
+                                }]
+                            }]
+                        }]
+                    }]
+                }]
+            }
+        }
+
+
+
+
+        doc 2
+    {
+        "_index": "test",
+        "_type": "_doc",
+        "_id": "2",
+        "matched": true,
+        "explanation": {
+            "value": 11.911896,
+            "description": "max of:",
+            "details": [{
+                "value": 11.911896,
+                "description": "sum of:",
+                "details": [{
+                    "value": 5.4068284,
+                    "description": "weight(title:оксана in 0) [PerFieldSimilarity], result of:",
+                    "details": [{
+                        "value": 5.4068284,
+                        "description": "score(freq=1.0), product of:",
+                        "details": [{
+                            "value": 2.2,
+                            "description": "boost",
+                            "details": []
+                        }, {
+                            "value": 5.4086657,
+                            "description": "idf, computed as log(1 + (N - n + 0.5) / (n + 0.5)) from:",
+                            "details": [{
+                                "value": 4,
+                                "description": "n, number of documents containing term",
+                                "details": []
+                            }, {
+                                "value": 1004,
+                                "description": "N, total number of documents with field",
+                                "details": []
+                            }]
+                        }, {
+                            "value": 0.45439103,
+                            "description": "tf, computed as freq / (freq + k1 * (1 - b + b * dl / avgdl)) from:",
+                            "details": [{
+                                "value": 1.0,
+                                "description": "freq, occurrences of term within document",
+                                "details": []
+                            }, {
+                                "value": 1.2,
+                                "description": "k1, term saturation parameter",
+                                "details": []
+                            }, {
+                                "value": 0.75,
+                                "description": "b, length normalization parameter",
+                                "details": []
+                            }, {
+                                "value": 6.0,
+                                "description": "dl, length of field",
+                                "details": []
+                            }, {
+                                "value": 5.99502,
+                                "description": "avgdl, average length of field",
+                                "details": []
+                            }]
+                        }]
+                    }]
+                }, {
+                    "value": 6.505067,
+                    "description": "weight(title:лифенко in 0) [PerFieldSimilarity], result of:",
+                    "details": [{
+                        "value": 6.505067,
+                        "description": "score(freq=1.0), product of:",
+                        "details": [{
+                            "value": 2.2,
+                            "description": "boost",
+                            "details": []
+                        }, {
+                            "value": 6.5072775,
+                            "description": "idf, computed as log(1 + (N - n + 0.5) / (n + 0.5)) from:",
+                            "details": [{
+                                "value": 1,
+                                "description": "n, number of documents containing term",
+                                "details": []
+                            }, {
+                                "value": 1004,
+                                "description": "N, total number of documents with field",
+                                "details": []
+                            }]
+                        }, {
+                            "value": 0.45439103,
+                            "description": "tf, computed as freq / (freq + k1 * (1 - b + b * dl / avgdl)) from:",
+                            "details": [{
+                                "value": 1.0,
+                                "description": "freq, occurrences of term within document",
+                                "details": []
+                            }, {
+                                "value": 1.2,
+                                "description": "k1, term saturation parameter",
+                                "details": []
+                            }, {
+                                "value": 0.75,
+                                "description": "b, length normalization parameter",
+                                "details": []
+                            }, {
+                                "value": 6.0,
+                                "description": "dl, length of field",
+                                "details": []
+                            }, {
+                                "value": 5.99502,
+                                "description": "avgdl, average length of field",
+                                "details": []
+                            }]
+                        }]
+                    }]
+                }]
+            }]
+        }
+    }
+            */
+    // motivated by #554
+    #[test]
+    fn test_bm25_several_fields() {
+        let mut schema_builder = Schema::builder();
+        let title = schema_builder.add_text_field("title", TEXT);
+        let text = 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!(
+            // tf = 1 0
+            title =>  "Законы притяжения   Оксана Кулакова",
+            // tf = 1 0
+            text => "Законы притяжения Оксана Кулакова]  \n\nТема: Сексуальное искусство, Женственность\nТип товара: Запись вебинара (аудио)\nПродолжительность: 1,5 часа\n\nСсылка на вебинар:\n  ",
+        ));
+        index_writer.add_document(doc!(
+            // tf = 1 0
+            title => "Любимые русские пироги (Оксана Путан)",
+            // tf = 2 0
+            text => "http://i95.fastpic.ru/big/2017/0628/9a/615b9c8504d94a3893d7f496ac53539a.jpg \n\nОт издателя\nОксана Путан   профессиональный повар, автор кулинарных книг и известный кулинарный блогер. Ее рецепты отличаются практичностью, доступностью и пользуются огромной популярностью в русскоязычном интернете. Это третья книга автора о самом вкусном и ароматном   настоящих русских пирогах и выпечке!\nДаже новички на кухне легко готовят по ее рецептам. Оксана описывает процесс приготовления настолько подробно и понятно, что вам остается только наслаждаться готовкой и не тратить время на лишние усилия. Готовьте легко и просто!\n\nhttps://www.ozon.ru/context/detail/id/139872462/"
+        ));
+        index_writer.add_document(doc!(
+            // tf = 1 1
+            title =>  "PDF Мастер Класс \"Морячок\" (Оксана Лифенко)",
+            // tf = 0 0 
+            text => "https://i.ibb.co/pzvHrDN/I3d U T6 Gg TM.jpg\nhttps://i.ibb.co/NFrb6v6/N0ls Z9nwjb U.jpg\nВ описание входит штаны, кофта, берет, матросский воротник. Описание продается в формате PDF, состоит из 12 страниц формата А4 и может быть напечатано на любом принтере.\nОписание предназначено для кукол BJD RealPuki от FairyLand, но может подойти и другим подобным куклам. Также вы можете вязать этот наряд из обычной пряжи, и он подойдет для куколок побольше.\nhttps://vk.com/market 95724412?w=product 95724412_2212"
+        ));
+        for _ in 0..1_000 {
+            index_writer.add_document(doc!(
+                title =>  "a b d e f g",
+                text => "maitre corbeau sur un arbre perche tenait dans son bec un fromage Maitre rnard par lodeur alleche lui tint a peu pres ce langage."
+            ));
+        }
+        index_writer.commit().unwrap();
+        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 weight = query.weight(&searcher, true).unwrap();
+        let mut scorer = weight.scorer(searcher.segment_reader(0u32)).unwrap();
+        //        let mut scores = vec![];
+        // while
+        println!("=====|");
+        scorer.advance();
+        dbg!("scorer.score()");
+        assert!(false);
+
+        //        scores.push(scorer.score());
+        //        assert_eq!(scores, &[0.8017307, 0.72233325, 1.0300813]);
+    }
+
+    // motivated by #554
+    #[test]
+    fn test_bm25_several_fields_bbb() {
+        let mut schema_builder = Schema::builder();
+        let text = schema_builder.add_text_field("text", TEXT);
+        let schema = schema_builder.build();
+        let index = Index::create_in_ram(schema);
+
+        let mut index_writer = index.writer_with_num_threads(1, 3_000_000).unwrap();
+        index_writer.add_document(doc!(
+            text => "Законы притяжения Оксана Кулакова]  \n\nТема: Сексуальное искусство, Женственность\nТип товара: Запись вебинара (аудио)\nПродолжительность: 1,5 часа\n\nСсылка на вебинар:\n  ",
+        ));
+        index_writer.add_document(doc!(
+            text => "http://i95.fastpic.ru/big/2017/0628/9a/615b9c8504d94a3893d7f496ac53539a.jpg \n\nОт издателя\nОксана Путан   профессиональный повар, автор кулинарных книг и известный кулинарный блогер. Ее рецепты отличаются практичностью, доступностью и пользуются огромной популярностью в русскоязычном интернете. Это третья книга автора о самом вкусном и ароматном   настоящих русских пирогах и выпечке!\nДаже новички на кухне легко готовят по ее рецептам. Оксана описывает процесс приготовления настолько подробно и понятно, что вам остается только наслаждаться готовкой и не тратить время на лишние усилия. Готовьте легко и просто!\n\nhttps://www.ozon.ru/context/detail/id/139872462/"
+        ));
+        index_writer.add_document(doc!(
+            text => "https://i.ibb.co/pzvHrDN/I3d U T6 Gg TM.jpg\nhttps://i.ibb.co/NFrb6v6/N0ls Z9nwjb U.jpg\nВ описание входит штаны, кофта, берет, матросский воротник. Описание продается в формате PDF, состоит из 12 страниц формата А4 и может быть напечатано на любом принтере.\nОписание предназначено для кукол BJD RealPuki от FairyLand, но может подойти и другим подобным куклам. Также вы можете вязать этот наряд из обычной пряжи, и он подойдет для куколок побольше.\nhttps://vk.com/market 95724412?w=product 95724412_2212"
+        ));
+        for _ in 0..100 {
+            index_writer.add_document(doc!(
+                text => "maitre corbeau sur un arbre perche tenait dans son bec un fromage Maitre rnard par lodeur alleche lui tint a peu pres ce langage."
+            ));
+        }
+        index_writer.commit().unwrap();
+        let reader = index.reader().unwrap();
+        let searcher = reader.searcher();
+        let query_parser = QueryParser::for_index(&index, vec![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)).unwrap();
+        let mut scores = vec![];
+        while scorer.advance() {
+            scores.push(scorer.score());
+        }
+        assert_eq!(scores, &[0.8017307, 0.72233325, 1.0300813]);
+        index_writer.commit().unwrap();
+    }
 }

src/query/phrase_query/phrase_query.rs

@@ -4,6 +4,7 @@ use error::TantivyError;
 use query::bm25::BM25Weight;
 use query::Query;
 use query::Weight;
+use schema::IndexRecordOption;
 use schema::{Field, Term};
 use std::collections::BTreeSet;
 use Result;
@@ -83,7 +84,7 @@ impl Query for PhraseQuery {
         let has_positions = field_entry
             .field_type()
             .get_index_record_option()
-            .map(|index_record_option| index_record_option.has_positions())
+            .map(IndexRecordOption::has_positions)
             .unwrap_or(false);
         if !has_positions {
             let field_name = field_entry.name();

src/query/query_parser/query_grammar.rs

@@ -1,6 +1,7 @@
 #![cfg_attr(feature = "cargo-clippy", allow(clippy::unneeded_field_pattern))]
 #![cfg_attr(feature = "cargo-clippy", allow(clippy::toplevel_ref_arg))]
 
+use super::query_grammar;
 use super::user_input_ast::*;
 use combine::char::*;
 use combine::error::StreamError;
@@ -22,7 +23,7 @@ parser! {
 parser! {
     fn word[I]()(I) -> String
     where [I: Stream<Item = char>] {
-        many1(satisfy(|c: char| c.is_alphanumeric()))
+        many1(satisfy(char::is_alphanumeric))
                .and_then(|s: String| {
                    match s.as_str() {
                      "OR" => Err(StreamErrorFor::<I>::unexpected_static_message("OR")),
@@ -62,7 +63,7 @@ parser! {
     fn negative_number[I]()(I) -> String
     where [I: Stream<Item = char>]
     {
-            (char('-'), many1(satisfy(|c: char| c.is_numeric())))
+            (char('-'), many1(satisfy(char::is_numeric)))
                 .map(|(s1, s2): (char, String)| format!("{}{}", s1, s2))
     }
 }
@@ -184,7 +185,7 @@ parser! {
                         }
                     )
                 )
-                .map(|el| el.into_dnf())
+                .map(query_grammar::Element::into_dnf)
                 .map(|fnd| {
                     if fnd.len() == 1 {
                         UserInputAST::and(fnd.into_iter().next().unwrap()) //< safe

src/query/union.rs

@@ -96,7 +96,7 @@ fn refill<TScorer: Scorer, TScoreCombiner: ScoreCombiner>(
 
 impl<TScorer: Scorer, TScoreCombiner: ScoreCombiner> Union<TScorer, TScoreCombiner> {
     fn refill(&mut self) -> bool {
-        if let Some(min_doc) = self.docsets.iter_mut().map(|docset| docset.doc()).min() {
+        if let Some(min_doc) = self.docsets.iter().map(DocSet::doc).min() {
             self.offset = min_doc;
             self.cursor = 0;
             refill(

src/schema/document.rs

@@ -128,7 +128,7 @@ impl Document {
         self.field_values
             .iter()
             .filter(|field_value| field_value.field() == field)
-            .map(|field_value| field_value.value())
+            .map(FieldValue::value)
             .collect()
     }
 
@@ -137,7 +137,7 @@ impl Document {
         self.field_values
             .iter()
             .find(|field_value| field_value.field() == field)
-            .map(|field_value| field_value.value())
+            .map(FieldValue::value)
     }
 }
 

src/schema/field_type.rs

@@ -4,6 +4,7 @@ use schema::{IntOptions, TextOptions};
 
 use schema::Facet;
 use schema::IndexRecordOption;
+use schema::TextFieldIndexing;
 use schema::Value;
 use serde_json::Value as JsonValue;
 
@@ -94,7 +95,7 @@ impl FieldType {
         match *self {
             FieldType::Str(ref text_options) => text_options
                 .get_indexing_options()
-                .map(|indexing_options| indexing_options.index_option()),
+                .map(TextFieldIndexing::index_option),
             FieldType::U64(ref int_options)
             | FieldType::I64(ref int_options)
             | FieldType::Date(ref int_options) => {

src/schema/schema.rs

@@ -130,7 +130,16 @@ impl SchemaBuilder {
         self.add_field(field_entry)
     }
 
-    /// Adds a fast bytes field to the schema
+    /// Adds a fast bytes field to the schema.
+    ///
+    /// Bytes field are not searchable and are only used
+    /// as fast field, to associate any kind of payload
+    /// to a document.
+    ///
+    /// For instance, learning-to-rank often requires to access
+    /// some document features at scoring time.
+    /// These can be serializing and stored as a bytes field to
+    /// get access rapidly when scoring each document.
     pub fn add_bytes_field(&mut self, field_name: &str) -> Field {
         let field_entry = FieldEntry::new_bytes(field_name.to_string());
         self.add_field(field_entry)
@@ -224,7 +233,7 @@ impl Schema {
             let field_name = self.get_field_name(field);
             let values: Vec<Value> = field_values
                 .into_iter()
-                .map(|field_val| field_val.value())
+                .map(FieldValue::value)
                 .cloned()
                 .collect();
             field_map.insert(field_name.to_string(), values);

src/snippet/mod.rs

@@ -1,6 +1,7 @@
 use htmlescape::encode_minimal;
 use query::Query;
 use schema::Field;
+use schema::Value;
 use std::cmp::Ordering;
 use std::collections::BTreeMap;
 use std::collections::BTreeSet;
@@ -303,7 +304,7 @@ impl SnippetGenerator {
         let text: String = doc
             .get_all(self.field)
             .into_iter()
-            .flat_map(|val| val.text())
+            .flat_map(Value::text)
             .collect::<Vec<&str>>()
             .join(" ");
         self.snippet(&text)

src/space_usage/mod.rs

@@ -227,7 +227,7 @@ pub struct PerFieldSpaceUsage {
 
 impl PerFieldSpaceUsage {
     pub(crate) fn new(fields: HashMap<Field, FieldUsage>) -> PerFieldSpaceUsage {
-        let total = fields.values().map(|x| x.total()).sum();
+        let total = fields.values().map(FieldUsage::total).sum();
         PerFieldSpaceUsage { fields, total }
     }
 

src/tokenizer/lower_caser.rs

@@ -44,18 +44,17 @@ where
     }
 
     fn advance(&mut self) -> bool {
-        if self.tail.advance() {
-            if self.token_mut().text.is_ascii() {
-                // fast track for ascii.
-                self.token_mut().text.make_ascii_lowercase();
-            } else {
-                to_lowercase_unicode(&mut self.tail.token_mut().text, &mut self.buffer);
-                mem::swap(&mut self.tail.token_mut().text, &mut self.buffer);
-            }
-            true
-        } else {
-            false
+        if !self.tail.advance() {
+            return false;
         }
+        if self.token_mut().text.is_ascii() {
+            // fast track for ascii.
+            self.token_mut().text.make_ascii_lowercase();
+        } else {
+            to_lowercase_unicode(&mut self.tail.token_mut().text, &mut self.buffer);
+            mem::swap(&mut self.tail.token_mut().text, &mut self.buffer);
+        }
+        true
     }
 }
 

src/tokenizer/mod.rs

@@ -131,6 +131,7 @@
 //! ```
 //!
 mod alphanum_only;
+mod ascii_folding_filter;
 mod facet_tokenizer;
 mod lower_caser;
 mod ngram_tokenizer;
@@ -144,6 +145,7 @@ mod tokenizer;
 mod tokenizer_manager;
 
 pub use self::alphanum_only::AlphaNumOnlyFilter;
+pub use self::ascii_folding_filter::AsciiFoldingFilter;
 pub use self::facet_tokenizer::FacetTokenizer;
 pub use self::lower_caser::LowerCaser;
 pub use self::ngram_tokenizer::NgramTokenizer;

src/tokenizer/raw_tokenizer.rs

@@ -29,12 +29,9 @@ impl<'a> Tokenizer<'a> for RawTokenizer {
 
 impl TokenStream for RawTokenStream {
     fn advance(&mut self) -> bool {
-        if self.has_token {
-            self.has_token = false;
-            true
-        } else {
-            false
-        }
+        let result = self.has_token;
+        self.has_token = false;
+        result
     }
 
     fn token(&self) -> &Token {

src/tokenizer/remove_long.rs

@@ -91,7 +91,6 @@ where
                 return true;
             }
         }
-
         false
     }
 }

src/tokenizer/simple_tokenizer.rs

@@ -38,23 +38,16 @@ impl<'a> TokenStream for SimpleTokenStream<'a> {
     fn advance(&mut self) -> bool {
         self.token.text.clear();
         self.token.position = self.token.position.wrapping_add(1);
-
-        loop {
-            match self.chars.next() {
-                Some((offset_from, c)) => {
-                    if c.is_alphanumeric() {
-                        let offset_to = self.search_token_end();
-                        self.token.offset_from = offset_from;
-                        self.token.offset_to = offset_to;
-                        self.token.text.push_str(&self.text[offset_from..offset_to]);
-                        return true;
-                    }
-                }
-                None => {
-                    return false;
-                }
+        while let Some((offset_from, c)) = self.chars.next() {
+            if c.is_alphanumeric() {
+                let offset_to = self.search_token_end();
+                self.token.offset_from = offset_from;
+                self.token.offset_to = offset_to;
+                self.token.text.push_str(&self.text[offset_from..offset_to]);
+                return true;
             }
         }
+        false
     }
 
     fn token(&self) -> &Token {

src/tokenizer/stemmer.rs

@@ -108,15 +108,14 @@ where
     }
 
     fn advance(&mut self) -> bool {
-        if self.tail.advance() {
-            // TODO remove allocation
-            let stemmed_str: String = self.stemmer.stem(&self.token().text).into_owned();
-            self.token_mut().text.clear();
-            self.token_mut().text.push_str(&stemmed_str);
-            true
-        } else {
-            false
+        if !self.tail.advance() {
+            return false;
         }
+        // TODO remove allocation
+        let stemmed_str: String = self.stemmer.stem(&self.token().text).into_owned();
+        self.token_mut().text.clear();
+        self.token_mut().text.push_str(&stemmed_str);
+        true
     }
 }
 

src/tokenizer/stop_word_filter.rs

@@ -104,7 +104,6 @@ where
                 return true;
             }
         }
-
         false
     }
 }

src/tokenizer/tokenizer_manager.rs

@@ -1,4 +1,5 @@
 use std::collections::HashMap;
+use std::ops::Deref;
 use std::sync::{Arc, RwLock};
 use tokenizer::box_tokenizer;
 use tokenizer::stemmer::Language;
@@ -46,7 +47,8 @@ impl TokenizerManager {
             .read()
             .expect("Acquiring the lock should never fail")
             .get(tokenizer_name)
-            .map(|boxed_tokenizer| boxed_tokenizer.boxed_clone())
+            .map(Deref::deref)
+            .map(BoxedTokenizer::boxed_clone)
     }
 }