rust/tantivy
1
0
Fork 0
mirror of https://github.com/quickwit-oss/tantivy.git synced 2026-05-27 21:50:41 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge branch 'master' into issue/136

Browse Source 
Conflicts:
	src/datastruct/stacker/hashmap.rs
	src/datastruct/stacker/heap.rs
	src/datastruct/stacker/mod.rs
	src/indexer/index_writer.rs
	src/indexer/merger.rs
	src/indexer/segment_updater.rs
	src/indexer/segment_writer.rs
	src/postings/postings_writer.rs
	src/postings/recorder.rs
	src/schema/term.rs
This commit is contained in:
Paul Masurel
2017-05-17 18:40:09 +09:00
parent ecbdd70c37 b3f39f2343
commit 7b2b181652
118 changed files with 3234 additions and 2946 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
.travis.yml
View File

@@ -22,7 +22,9 @@ before_script:
- |
pip install 'travis-cargo<0.2' --user &&
export PATH=$HOME/.local/bin:$PATH
- (cargo install rustfmt || true)
script:
- cargo fmt -- --write-mode=diff
- |
travis-cargo build &&
travis-cargo test &&

1
CHANGELOG.md
View File

@@ -2,6 +2,7 @@ Tantivy 0.4.0
==========================
- Raise the limit of number of fields (previously 256 fields)
- Removed u32 fields. They are replaced by u64 and i64 fields (#65)
- Optimized skip in SegmentPostings (#130) (@lnicola)
- Replacing rustc_serialize by serde. Kudos to @KodrAus and @lnicola
- QueryParser:
- Explicit error returned when searched for a term that is not indexed

2
appveyor.yml
View File

@@ -22,4 +22,4 @@ build: false
test_script:
- REM SET RUST_LOG=tantivy,test & cargo test --verbose
- REM SET RUST_LOG=tantivy,test & cargo run --example simple_search
- REM SET RUST_BACKTRACE=1 & cargo run --example simple_search

2
build.rs
View File

@@ -46,6 +46,8 @@ mod build {
if !cfg!(debug_assertions) && cfg!(target_env = "msvc") {
println!("cargo:rustc-link-lib=dylib=simdcomp");
}
println!("cargo:rerun-if-changed=cpp");
}
}

58
examples/simple_search.rs
View File

@@ -1,6 +1,9 @@
extern crate tantivy;
extern crate tempdir;
#[macro_use]
extern crate serde_json;
use std::path::Path;
use tempdir::TempDir;
use tantivy::Index;
@@ -63,8 +66,7 @@ fn run_example(index_path: &Path) -> tantivy::Result<()> {
//
// This will actually just save a meta.json
// with our schema in the directory.
let index = try!(Index::create(index_path, schema.clone()));
let index = Index::create(index_path, schema.clone())?;
// To insert document we need an index writer.
@@ -74,7 +76,7 @@ fn run_example(index_path: &Path) -> tantivy::Result<()> {
//
// Here we use a buffer of 50MB per thread. Using a bigger
// heap for the indexer can increase its throughput.
let mut index_writer = try!(index.writer(50_000_000));
let mut index_writer = index.writer(50_000_000)?;
// Let's index our documents!
// We first need a handle on the title and the body field.
@@ -98,23 +100,37 @@ fn run_example(index_path: &Path) -> tantivy::Result<()> {
// ### Create a document directly from json.
//
// Alternatively, we can use our schema to parse
// a document object directly from json.
let mice_and_men_doc = try!(schema.parse_document(r#"{
// Alternatively, we can use our schema to parse a
// document object directly from json.
// The document is a string, but we use the `json` macro
// from `serde_json` for the convenience of multi-line support.
let json = json!({
"title": "Of Mice and Men",
"body": "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"
}"#));
"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"
});
let mice_and_men_doc = schema.parse_document(&json.to_string())?;
index_writer.add_document(mice_and_men_doc);
// Multi-valued field are allowed, they are
// expressed in JSON by an array.
// The following document has two titles.
let frankenstein_doc = try!(schema.parse_document(r#"{
"title": ["Frankenstein", "The Modern Promotheus"],
"body": "You will rejoice to hear that no disaster has accompanied the commencement of an enterprise which you have regarded with such evil forebodings. I arrived here yesterday, and my first task is to assure my dear sister of my welfare and increasing confidence in the success of my undertaking."
}"#));
let json = json!({
"title": ["Frankenstein", "The Modern Prometheus"],
"body": "You will rejoice to hear that no disaster has accompanied the commencement of an \
enterprise which you have regarded with such evil forebodings. I arrived here \
yesterday, and my first task is to assure my dear sister of my welfare and \
increasing confidence in the success of my undertaking."
});
let frankenstein_doc = schema.parse_document(&json.to_string())?;
index_writer.add_document(frankenstein_doc);
// This is an example, so we will only index 3 documents
@@ -135,7 +151,7 @@ fn run_example(index_path: &Path) -> tantivy::Result<()> {
// the existence of new documents.
//
// This call is blocking.
try!(index_writer.commit());
index_writer.commit()?;
// If `.commit()` returns correctly, then all of the
// documents that have been added are guaranteed to be
@@ -151,7 +167,7 @@ fn run_example(index_path: &Path) -> tantivy::Result<()> {
// Let's search our index. Start by reloading
// searchers in the index. This should be done
// after every commit().
try!(index.load_searchers());
index.load_searchers()?;
// Afterwards create one (or more) searchers.
//
@@ -168,7 +184,7 @@ fn run_example(index_path: &Path) -> tantivy::Result<()> {
// QueryParser may fail if the query is not in the right
// format. For user facing applications, this can be a problem.
// A ticket has been opened regarding this problem.
let query = try!(query_parser.parse_query("sea whale"));
let query = query_parser.parse_query("sea whale")?;
// A query defines a set of documents, as
@@ -186,7 +202,7 @@ fn run_example(index_path: &Path) -> tantivy::Result<()> {
let mut top_collector = TopCollector::with_limit(10);
// We can now perform our query.
try!(searcher.search(&*query, &mut top_collector));
searcher.search(&*query, &mut top_collector)?;
// Our top collector now contains the 10
// most relevant doc ids...
@@ -200,9 +216,15 @@ fn run_example(index_path: &Path) -> tantivy::Result<()> {
// a title.
for doc_address in doc_addresses {
let retrieved_doc = try!(searcher.doc(&doc_address));
let retrieved_doc = searcher.doc(&doc_address)?;
println!("{}", schema.to_json(&retrieved_doc));
}
// Wait for indexing and merging threads to shut down.
// Usually this isn't needed, but in `main` we try to
// delete the temporary directory and that fails on
// Windows if the files are still open.
index_writer.wait_merging_threads()?;
Ok(())
}

20
src/analyzer/mod.rs
View File

@@ -18,11 +18,10 @@ pub trait StreamingIterator<'a, T> {
impl<'a, 'b> TokenIter<'b> {
fn consume_token(&'a mut self) -> Option<&'a str> {
for c in &mut self.chars {
for c in &mut self.chars {
if c.is_alphanumeric() {
append_char_lowercase(c, &mut self.term_buffer);
}
else {
} else {
break;
}
}
@@ -32,9 +31,8 @@ impl<'a, 'b> TokenIter<'b> {
impl<'a, 'b> StreamingIterator<'a, &'a str> for TokenIter<'b> {
#[inline]
fn next(&'a mut self,) -> Option<&'a str> {
fn next(&'a mut self) -> Option<&'a str> {
self.term_buffer.clear();
// skipping non-letter characters.
loop {
@@ -45,24 +43,24 @@ impl<'a, 'b> StreamingIterator<'a, &'a str> for TokenIter<'b> {
return self.consume_token();
}
}
None => { return None; }
None => {
return None;
}
}
}
}
}
pub struct SimpleTokenizer;
impl SimpleTokenizer {
pub fn tokenize<'a>(&self, text: &'a str) -> TokenIter<'a> {
TokenIter {
term_buffer: String::new(),
chars: text.chars(),
term_buffer: String::new(),
chars: text.chars(),
}
}
}
}

19
src/collector/chained_collector.rs
View File

@@ -7,8 +7,8 @@ use Score;
/// Collector that does nothing.
/// This is used in the chain Collector and will hopefully
/// be optimized away by the compiler.
/// This is used in the chain Collector and will hopefully
/// be optimized away by the compiler.
pub struct DoNothingCollector;
impl Collector for DoNothingCollector {
#[inline]
@@ -24,10 +24,10 @@ impl Collector for DoNothingCollector {
/// are known at compile time.
pub struct ChainedCollector<Left: Collector, Right: Collector> {
left: Left,
right: Right
right: Right,
}
impl<Left: Collector, Right: Collector> ChainedCollector<Left, Right> {
impl<Left: Collector, Right: Collector> ChainedCollector<Left, Right> {
/// Adds a collector
pub fn push<C: Collector>(self, new_collector: &mut C) -> ChainedCollector<Self, &mut C> {
ChainedCollector {
@@ -38,7 +38,10 @@ impl<Left: Collector, Right: Collector> ChainedCollector<Left, Right> {
}
impl<Left: Collector, Right: Collector> Collector for ChainedCollector<Left, Right> {
fn set_segment(&mut self, segment_local_id: SegmentLocalId, segment: &SegmentReader) -> Result<()> {
fn set_segment(&mut self,
segment_local_id: SegmentLocalId,
segment: &SegmentReader)
-> Result<()> {
try!(self.left.set_segment(segment_local_id, segment));
try!(self.right.set_segment(segment_local_id, segment));
Ok(())
@@ -70,9 +73,7 @@ mod tests {
let mut top_collector = TopCollector::with_limit(2);
let mut count_collector = CountCollector::default();
{
let mut collectors = chain()
.push(&mut top_collector)
.push(&mut count_collector);
let mut collectors = chain().push(&mut top_collector).push(&mut count_collector);
collectors.collect(1, 0.2);
collectors.collect(2, 0.1);
collectors.collect(3, 0.5);
@@ -80,4 +81,4 @@ mod tests {
assert_eq!(count_collector.count(), 3);
assert!(top_collector.at_capacity());
}
}
}

20
src/collector/count_collector.rs
View File

@@ -6,7 +6,7 @@ use SegmentReader;
use SegmentLocalId;
/// `CountCollector` collector only counts how many
/// documents match the query.
/// documents match the query.
pub struct CountCollector {
count: usize,
}
@@ -14,20 +14,18 @@ pub struct CountCollector {
impl CountCollector {
/// Returns the count of documents that were
/// collected.
pub fn count(&self,) -> usize {
pub fn count(&self) -> usize {
self.count
}
}
impl Default for CountCollector {
fn default() -> CountCollector {
CountCollector {count: 0,
}
CountCollector { count: 0 }
}
}
impl Collector for CountCollector {
fn set_segment(&mut self, _: SegmentLocalId, _: &SegmentReader) -> Result<()> {
Ok(())
}
@@ -47,11 +45,11 @@ mod tests {
#[bench]
fn build_collector(b: &mut Bencher) {
b.iter(|| {
let mut count_collector = CountCollector::default();
for doc in 0..1_000_000 {
count_collector.collect(doc, 1f32);
}
count_collector.count()
});
let mut count_collector = CountCollector::default();
for doc in 0..1_000_000 {
count_collector.collect(doc, 1f32);
}
count_collector.count()
});
}
}

49
src/collector/mod.rs
View File

@@ -16,11 +16,11 @@ pub use self::top_collector::TopCollector;
mod chained_collector;
pub use self::chained_collector::chain;
/// Collectors are in charge of collecting and retaining relevant
/// Collectors are in charge of collecting and retaining relevant
/// information from the document found and scored by the query.
///
///
/// For instance,
/// For instance,
///
/// - keeping track of the top 10 best documents
/// - computing a breakdown over a fast field
@@ -29,7 +29,7 @@ pub use self::chained_collector::chain;
/// Queries are in charge of pushing the `DocSet` to the collector.
///
/// As they work on multiple segments, they first inform
/// the collector of a change in a segment and then
/// the collector of a change in a segment and then
/// call the `collect` method to push the document to the collector.
///
/// Temporally, our collector will receive calls
@@ -46,16 +46,22 @@ pub use self::chained_collector::chain;
///
/// Segments are not guaranteed to be visited in any specific order.
pub trait Collector {
/// `set_segment` is called before beginning to enumerate
/// `set_segment` is called before beginning to enumerate
/// on this segment.
fn set_segment(&mut self, segment_local_id: SegmentLocalId, segment: &SegmentReader) -> Result<()>;
fn set_segment(&mut self,
segment_local_id: SegmentLocalId,
segment: &SegmentReader)
-> Result<()>;
/// The query pushes the scored document to the collector via this method.
fn collect(&mut self, doc: DocId, score: Score);
}
impl<'a, C: Collector> Collector for &'a mut C {
fn set_segment(&mut self, segment_local_id: SegmentLocalId, segment: &SegmentReader) -> Result<()> {
fn set_segment(&mut self,
segment_local_id: SegmentLocalId,
segment: &SegmentReader)
-> Result<()> {
(*self).set_segment(segment_local_id, segment)
}
/// The query pushes the scored document to the collector via this method.
@@ -77,7 +83,7 @@ pub mod tests {
use fastfield::U64FastFieldReader;
use fastfield::FastFieldReader;
use schema::Field;
/// Stores all of the doc ids.
/// This collector is only used for tests.
/// It is unusable in practise, as it does not store
@@ -90,7 +96,7 @@ pub mod tests {
impl TestCollector {
/// Return the exhalist of documents.
pub fn docs(self,) -> Vec<DocId> {
pub fn docs(self) -> Vec<DocId> {
self.docs
}
}
@@ -106,7 +112,6 @@ pub mod tests {
}
impl Collector for TestCollector {
fn set_segment(&mut self, _: SegmentLocalId, reader: &SegmentReader) -> Result<()> {
self.offset += self.segment_max_doc;
self.segment_max_doc = reader.max_doc();
@@ -117,10 +122,10 @@ pub mod tests {
self.docs.push(doc + self.offset);
}
}
/// Collects in order all of the fast fields for all of the
/// doc in the `DocSet`
///
@@ -140,11 +145,11 @@ pub mod tests {
}
}
pub fn vals(self,) -> Vec<u64> {
pub fn vals(self) -> Vec<u64> {
self.vals
}
}
impl Collector for FastFieldTestCollector {
fn set_segment(&mut self, _: SegmentLocalId, reader: &SegmentReader) -> Result<()> {
self.ff_reader = Some(reader.get_fast_field_reader(self.field)?);
@@ -161,12 +166,12 @@ pub mod tests {
#[bench]
fn build_collector(b: &mut Bencher) {
b.iter(|| {
let mut count_collector = CountCollector::default();
let docs: Vec<u32> = (0..1_000_000).collect();
for doc in docs {
count_collector.collect(doc, 1f32);
}
count_collector.count()
});
let mut count_collector = CountCollector::default();
let docs: Vec<u32> = (0..1_000_000).collect();
for doc in docs {
count_collector.collect(doc, 1f32);
}
count_collector.count()
});
}
}

14
src/collector/multi_collector.rs
View File

@@ -7,7 +7,7 @@ use SegmentLocalId;
/// Multicollector makes it possible to collect on more than one collector.
/// It should only be used for use cases where the Collector types is unknown
/// It should only be used for use cases where the Collector types is unknown
/// at compile time.
/// If the type of the collectors is known, you should prefer to use `ChainedCollector`.
pub struct MultiCollector<'a> {
@@ -17,15 +17,16 @@ pub struct MultiCollector<'a> {
impl<'a> MultiCollector<'a> {
/// Constructor
pub fn from(collectors: Vec<&'a mut Collector>) -> MultiCollector {
MultiCollector {
collectors: collectors,
}
MultiCollector { collectors: collectors }
}
}
impl<'a> Collector for MultiCollector<'a> {
fn set_segment(&mut self, segment_local_id: SegmentLocalId, segment: &SegmentReader) -> Result<()> {
fn set_segment(&mut self,
segment_local_id: SegmentLocalId,
segment: &SegmentReader)
-> Result<()> {
for collector in &mut self.collectors {
try!(collector.set_segment(segment_local_id, segment));
}
@@ -52,7 +53,8 @@ mod tests {
let mut top_collector = TopCollector::with_limit(2);
let mut count_collector = CountCollector::default();
{
let mut collectors = MultiCollector::from(vec!(&mut top_collector, &mut count_collector));
let mut collectors = MultiCollector::from(vec![&mut top_collector,
&mut count_collector]);
collectors.collect(1, 0.2);
collectors.collect(2, 0.1);
collectors.collect(3, 0.5);

60
src/collector/top_collector.rs
View File

@@ -12,8 +12,7 @@ use Score;
#[derive(Clone, Copy)]
struct GlobalScoredDoc {
score: Score,
doc_address: DocAddress
doc_address: DocAddress,
}
impl PartialOrd for GlobalScoredDoc {
@@ -25,10 +24,10 @@ impl PartialOrd for GlobalScoredDoc {
impl Ord for GlobalScoredDoc {
#[inline]
fn cmp(&self, other: &GlobalScoredDoc) -> Ordering {
other.score.partial_cmp(&self.score)
.unwrap_or(
other.doc_address.cmp(&self.doc_address)
)
other
.score
.partial_cmp(&self.score)
.unwrap_or_else(|| other.doc_address.cmp(&self.doc_address))
}
}
@@ -53,7 +52,6 @@ pub struct TopCollector {
}
impl TopCollector {
/// Creates a top collector, with a number of documents equal to "limit".
///
/// # Panics
@@ -68,9 +66,9 @@ impl TopCollector {
segment_id: 0,
}
}
/// Returns K best documents sorted in decreasing order.
///
///
/// Calling this method triggers the sort.
/// The result of the sort is not cached.
pub fn docs(&self) -> Vec<DocAddress> {
@@ -81,30 +79,27 @@ impl TopCollector {
}
/// Returns K best ScoredDocument sorted in decreasing order.
///
///
/// Calling this method triggers the sort.
/// The result of the sort is not cached.
pub fn score_docs(&self) -> Vec<(Score, DocAddress)> {
let mut scored_docs: Vec<GlobalScoredDoc> = self.heap
.iter()
.cloned()
.collect();
let mut scored_docs: Vec<GlobalScoredDoc> = self.heap.iter().cloned().collect();
scored_docs.sort();
scored_docs.into_iter()
.map(|GlobalScoredDoc {score, doc_address}| (score, doc_address))
scored_docs
.into_iter()
.map(|GlobalScoredDoc { score, doc_address }| (score, doc_address))
.collect()
}
/// Return true iff at least K documents have gone through
/// the collector.
#[inline]
pub fn at_capacity(&self, ) -> bool {
pub fn at_capacity(&self) -> bool {
self.heap.len() >= self.limit
}
}
impl Collector for TopCollector {
fn set_segment(&mut self, segment_id: SegmentLocalId, _: &SegmentReader) -> Result<()> {
self.segment_id = segment_id;
Ok(())
@@ -113,17 +108,21 @@ impl Collector for TopCollector {
fn collect(&mut self, doc: DocId, score: Score) {
if self.at_capacity() {
// It's ok to unwrap as long as a limit of 0 is forbidden.
let limit_doc: GlobalScoredDoc = *self.heap.peek().expect("Top collector with size 0 is forbidden");
let limit_doc: GlobalScoredDoc =
*self.heap
.peek()
.expect("Top collector with size 0 is forbidden");
if limit_doc.score < score {
let mut mut_head = self.heap.peek_mut().expect("Top collector with size 0 is forbidden");
let mut mut_head = self.heap
.peek_mut()
.expect("Top collector with size 0 is forbidden");
mut_head.score = score;
mut_head.doc_address = DocAddress(self.segment_id, doc);
mut_head.doc_address = DocAddress(self.segment_id, doc);
}
}
else {
} else {
let wrapped_doc = GlobalScoredDoc {
score: score,
doc_address: DocAddress(self.segment_id, doc)
doc_address: DocAddress(self.segment_id, doc),
};
self.heap.push(wrapped_doc);
}
@@ -147,13 +146,12 @@ mod tests {
top_collector.collect(3, 0.2);
top_collector.collect(5, 0.3);
assert!(!top_collector.at_capacity());
let score_docs: Vec<(Score, DocId)> = top_collector.score_docs()
let score_docs: Vec<(Score, DocId)> = top_collector
.score_docs()
.into_iter()
.map(|(score, doc_address)| (score, doc_address.doc()))
.collect();
assert_eq!(score_docs, vec!(
(0.8, 1), (0.3, 5), (0.2, 3),
));
assert_eq!(score_docs, vec![(0.8, 1), (0.3, 5), (0.2, 3)]);
}
#[test]
@@ -171,9 +169,7 @@ mod tests {
.into_iter()
.map(|(score, doc_address)| (score, doc_address.doc()))
.collect();
assert_eq!(score_docs, vec!(
(0.9, 7), (0.8, 1), (0.3, 5), (0.2, 3)
));
assert_eq!(score_docs, vec![(0.9, 7), (0.8, 1), (0.3, 5), (0.2, 3)]);
}
{
let docs: Vec<DocId> = top_collector
@@ -181,7 +177,7 @@ mod tests {
.into_iter()
.map(|doc_address| doc_address.doc())
.collect();
assert_eq!(docs, vec!(7, 1, 5, 3));
assert_eq!(docs, vec![7, 1, 5, 3]);
}

91
src/common/bitpacker.rs
View File

@@ -6,19 +6,19 @@ use std::mem;
/// Computes the number of bits that will be used for bitpacking.
///
/// In general the target is the minimum number of bits
/// In general the target is the minimum number of bits
/// required to express the amplitude given in argument.
///
/// e.g. If the amplitude is 10, we can store all ints on simply 4bits.
///
///
/// The logic is slightly more convoluted here as for optimization
/// reasons, we want to ensure that a value spawns over at most 8 bytes
/// of aligns bytes.
///
/// Spawning over 9 bytes is possible for instance, if we do
///
/// Spawning over 9 bytes is possible for instance, if we do
/// bitpacking with an amplitude of 63 bits.
/// In this case, the second int will start on bit
/// 63 (which belongs to byte 7) and ends at byte 15;
/// 63 (which belongs to byte 7) and ends at byte 15;
/// Hence 9 bytes (from byte 7 to byte 15 included).
///
/// To avoid this, we force the number of bits to 64bits
@@ -30,12 +30,7 @@ use std::mem;
/// number of bits.
pub fn compute_num_bits(amplitude: u64) -> u8 {
let amplitude = (64u32 - amplitude.leading_zeros()) as u8;
if amplitude <= 64 - 8 {
amplitude
}
else {
64
}
if amplitude <= 64 - 8 { amplitude } else { 64 }
}
pub struct BitPacker {
@@ -46,7 +41,6 @@ pub struct BitPacker {
}
impl BitPacker {
pub fn new(num_bits: usize) -> BitPacker {
BitPacker {
mini_buffer: 0u64,
@@ -55,7 +49,7 @@ impl BitPacker {
written_size: 0,
}
}
pub fn write<TWrite: Write>(&mut self, val: u64, output: &mut TWrite) -> io::Result<()> {
let val_u64 = val as u64;
if self.mini_buffer_written + self.num_bits > 64 {
@@ -63,32 +57,34 @@ impl BitPacker {
self.written_size += self.mini_buffer.serialize(output)?;
self.mini_buffer = val_u64.wrapping_shr((64 - self.mini_buffer_written) as u32);
self.mini_buffer_written = self.mini_buffer_written + (self.num_bits as usize) - 64;
}
else {
} else {
self.mini_buffer |= val_u64 << self.mini_buffer_written;
self.mini_buffer_written += self.num_bits;
if self.mini_buffer_written == 64 {
self.written_size += self.mini_buffer.serialize(output)?;
self.mini_buffer_written = 0;
self.mini_buffer = 0u64;
}
}
}
Ok(())
}
fn flush<TWrite: Write>(&mut self, output: &mut TWrite) -> io::Result<()>{
fn flush<TWrite: Write>(&mut self, output: &mut TWrite) -> io::Result<()> {
if self.mini_buffer_written > 0 {
let num_bytes = (self.mini_buffer_written + 7) / 8;
let arr: [u8; 8] = unsafe { mem::transmute::<u64, [u8; 8]>(self.mini_buffer) };
let arr: [u8; 8] = unsafe { mem::transmute::<u64, [u8; 8]>(self.mini_buffer) };
output.write_all(&arr[..num_bytes])?;
self.written_size += num_bytes;
self.mini_buffer_written = 0;
}
Ok(())
}
pub fn close<TWrite: Write>(&mut self, output: &mut TWrite) -> io::Result<usize> {
self.flush(output)?;
// Padding the write file to simplify reads.
output.write_all(&[0u8; 7])?;
self.written_size += 7;
Ok(self.written_size)
}
}
@@ -99,26 +95,24 @@ pub struct BitUnpacker {
num_bits: usize,
mask: u64,
data_ptr: *const u8,
data_len: usize,
data_len: usize,
}
impl BitUnpacker {
pub fn new(data: &[u8], num_bits: usize) -> BitUnpacker {
let mask: u64 =
if num_bits == 64 {
!0u64
}
else {
(1u64 << num_bits) - 1u64
};
let mask: u64 = if num_bits == 64 {
!0u64
} else {
(1u64 << num_bits) - 1u64
};
BitUnpacker {
num_bits: num_bits,
mask: mask,
data_ptr: data.as_ptr(),
data_len: data.len()
data_len: data.len(),
}
}
pub fn get(&self, idx: usize) -> u64 {
if self.num_bits == 0 {
return 0;
@@ -126,22 +120,12 @@ impl BitUnpacker {
let addr = (idx * self.num_bits) / 8;
let bit_shift = idx * self.num_bits - addr * 8;
let val_unshifted_unmasked: u64;
if addr + 8 <= self.data_len {
val_unshifted_unmasked = unsafe { * (self.data_ptr.offset(addr as isize) as *const u64) };
}
else {
let mut arr = [0u8; 8];
if addr < self.data_len {
for i in 0..self.data_len - addr {
arr[i] = unsafe { *self.data_ptr.offset( (addr + i) as isize) };
}
}
val_unshifted_unmasked = unsafe { mem::transmute::<[u8; 8], u64>(arr) };
}
debug_assert!(addr + 8 <= self.data_len,
"The fast field field should have been padded with 7 bytes.");
val_unshifted_unmasked = unsafe { *(self.data_ptr.offset(addr as isize) as *const u64) };
let val_shifted = (val_unshifted_unmasked >> bit_shift) as u64;
(val_shifted & self.mask)
}
}
@@ -150,7 +134,7 @@ impl BitUnpacker {
#[cfg(test)]
mod test {
use super::{BitPacker, BitUnpacker, compute_num_bits};
#[test]
fn test_compute_num_bits() {
assert_eq!(compute_num_bits(1), 1u8);
@@ -162,31 +146,26 @@ mod test {
assert_eq!(compute_num_bits(256), 9u8);
assert_eq!(compute_num_bits(5_000_000_000), 33u8);
}
fn test_bitpacker_util(len: usize, num_bits: usize) {
let mut data = Vec::new();
let mut bitpacker = BitPacker::new(num_bits);
let max_val: u64 = (1 << num_bits) - 1;
let vals: Vec<u64> = (0u64..len as u64).map(|i| {
if max_val == 0 {
0
}
else {
i % max_val
}
}).collect();
let vals: Vec<u64> = (0u64..len as u64)
.map(|i| if max_val == 0 { 0 } else { i % max_val })
.collect();
for &val in &vals {
bitpacker.write(val, &mut data).unwrap();
}
let num_bytes = bitpacker.close(&mut data).unwrap();
assert_eq!(num_bytes, (num_bits * len + 7) / 8);
assert_eq!(num_bytes, (num_bits * len + 7) / 8 + 7);
assert_eq!(data.len(), num_bytes);
let bitunpacker = BitUnpacker::new(&data, num_bits);
for (i, val) in vals.iter().enumerate() {
assert_eq!(bitunpacker.get(i), *val);
}
}
#[test]
fn test_bitpacker() {
test_bitpacker_util(10, 3);
@@ -195,4 +174,4 @@ mod test {
test_bitpacker_util(6, 14);
test_bitpacker_util(1000, 14);
}
}
}

16
src/common/mod.rs
View File

@@ -19,10 +19,10 @@ pub fn make_io_err(msg: String) -> io::Error {
/// Has length trait
pub trait HasLen {
/// Return length
fn len(&self,) -> usize;
fn len(&self) -> usize;
/// Returns true iff empty.
fn is_empty(&self,) -> bool {
fn is_empty(&self) -> bool {
self.len() == 0
}
}
@@ -32,13 +32,13 @@ const HIGHEST_BIT: u64 = 1 << 63;
/// Maps `i64` to `u64` so that
/// `-2^63 .. 2^63-1` is mapped
/// to
/// `0 .. 2^64`
/// to
/// `0 .. 2^64-1`
/// in that order.
///
/// This is more suited than simply casting (`val as u64`)
/// because of bitpacking.
///
///
/// Imagine a list of `i64` ranging from -10 to 10.
/// When casting negative values, the negative values are projected
/// to values over 2^63, and all values end up requiring 64 bits.
@@ -47,7 +47,7 @@ pub fn i64_to_u64(val: i64) -> u64 {
(val as u64) ^ HIGHEST_BIT
}
/// Reverse the mapping given by
/// Reverse the mapping given by
/// `i64_to_u64`.
#[inline(always)]
pub fn u64_to_i64(val: u64) -> i64 {
@@ -76,4 +76,4 @@ mod test {
test_i64_converter_helper(i);
}
}
}
}

26
src/common/serialize.rs
View File

@@ -6,7 +6,7 @@ use std::io::Read;
use std::io;
use common::VInt;
pub trait BinarySerializable : fmt::Debug + Sized {
pub trait BinarySerializable: fmt::Debug + Sized {
fn serialize(&self, writer: &mut Write) -> io::Result<usize>;
fn deserialize(reader: &mut Read) -> io::Result<Self>;
}
@@ -45,14 +45,13 @@ impl<Left: BinarySerializable, Right: BinarySerializable> BinarySerializable for
Ok(try!(self.0.serialize(write)) + try!(self.1.serialize(write)))
}
fn deserialize(reader: &mut Read) -> io::Result<Self> {
Ok( (try!(Left::deserialize(reader)), try!(Right::deserialize(reader))) )
Ok((try!(Left::deserialize(reader)), try!(Right::deserialize(reader))))
}
}
impl BinarySerializable for u32 {
fn serialize(&self, writer: &mut Write) -> io::Result<usize> {
writer.write_u32::<Endianness>(*self)
.map(|_| 4)
writer.write_u32::<Endianness>(*self).map(|_| 4)
}
fn deserialize(reader: &mut Read) -> io::Result<u32> {
@@ -63,8 +62,7 @@ impl BinarySerializable for u32 {
impl BinarySerializable for u64 {
fn serialize(&self, writer: &mut Write) -> io::Result<usize> {
writer.write_u64::<Endianness>(*self)
.map(|_| 8)
writer.write_u64::<Endianness>(*self).map(|_| 8)
}
fn deserialize(reader: &mut Read) -> io::Result<u64> {
reader.read_u64::<Endianness>()
@@ -73,8 +71,7 @@ impl BinarySerializable for u64 {
impl BinarySerializable for i64 {
fn serialize(&self, writer: &mut Write) -> io::Result<usize> {
writer.write_i64::<Endianness>(*self)
.map(|_| 8)
writer.write_i64::<Endianness>(*self).map(|_| 8)
}
fn deserialize(reader: &mut Read) -> io::Result<i64> {
reader.read_i64::<Endianness>()
@@ -104,7 +101,9 @@ impl BinarySerializable for String {
fn deserialize(reader: &mut Read) -> io::Result<String> {
let string_length = try!(VInt::deserialize(reader)).val() as usize;
let mut result = String::with_capacity(string_length);
try!(reader.take(string_length as u64).read_to_string(&mut result));
try!(reader
.take(string_length as u64)
.read_to_string(&mut result));
Ok(result)
}
}
@@ -122,8 +121,7 @@ mod test {
if num_bytes != 0 {
assert_eq!(v.serialize(&mut buffer).unwrap(), num_bytes);
assert_eq!(buffer.len(), num_bytes);
}
else {
} else {
v.serialize(&mut buffer).unwrap();
}
let mut cursor = &buffer[..];
@@ -147,15 +145,15 @@ mod test {
#[test]
fn test_serialize_string() {
serialize_test(String::from(""), 1);
serialize_test(String::from("ぽよぽよ"), 1 + 3*4);
serialize_test(String::from("富士さん見える。"), 1 + 3*8);
serialize_test(String::from("ぽよぽよ"), 1 + 3 * 4);
serialize_test(String::from("富士さん見える。"), 1 + 3 * 8);
}
#[test]
fn test_serialize_vec() {
let v: Vec<u8> = Vec::new();
serialize_test(v, 1);
serialize_test(vec!(1u32, 3u32), 1 + 4*2);
serialize_test(vec![1u32, 3u32], 1 + 4 * 2);
}
#[test]

30
src/common/timer.rs
View File

@@ -10,7 +10,7 @@ pub struct OpenTimer<'a> {
impl<'a> OpenTimer<'a> {
/// Starts timing a new named subtask
///
/// The timer is stopped automatically
/// The timer is stopped automatically
/// when the `OpenTimer` is dropped.
pub fn open(&mut self, name: &'static str) -> OpenTimer {
OpenTimer {
@@ -23,12 +23,17 @@ impl<'a> OpenTimer<'a> {
}
impl<'a> Drop for OpenTimer<'a> {
fn drop(&mut self,) {
self.timer_tree.timings.push(Timing {
name: self.name,
duration: self.start.to(PreciseTime::now()).num_microseconds().unwrap(),
depth: self.depth,
});
fn drop(&mut self) {
self.timer_tree
.timings
.push(Timing {
name: self.name,
duration: self.start
.to(PreciseTime::now())
.num_microseconds()
.unwrap(),
depth: self.depth,
});
}
}
@@ -47,12 +52,11 @@ pub struct TimerTree {
}
impl TimerTree {
/// Returns the total time elapsed in microseconds
pub fn total_time(&self,) -> i64 {
/// Returns the total time elapsed in microseconds
pub fn total_time(&self) -> i64 {
self.timings.last().unwrap().duration
}
/// Open a new named subtask
pub fn open(&mut self, name: &'static str) -> OpenTimer {
OpenTimer {
@@ -66,9 +70,7 @@ impl TimerTree {
impl Default for TimerTree {
fn default() -> TimerTree {
TimerTree {
timings: Vec::new(),
}
TimerTree { timings: Vec::new() }
}
}

12
src/common/vint.rs
View File

@@ -5,12 +5,12 @@ use std::io::Read;
/// Wrapper over a `u64` that serializes as a variable int.
/// Wrapper over a `u64` that serializes as a variable int.
#[derive(Debug, Eq, PartialEq)]
pub struct VInt(pub u64);
impl VInt {
pub fn val(&self,) -> u64 {
pub fn val(&self) -> u64 {
self.0
}
}
@@ -27,8 +27,7 @@ impl BinarySerializable for VInt {
buffer[written] = next_byte | 128u8;
written += 1;
break;
}
else {
} else {
buffer[written] = next_byte;
written += 1;
}
@@ -50,12 +49,9 @@ impl BinarySerializable for VInt {
}
shift += 7;
}
_ => {
return Err(io::Error::new(io::ErrorKind::InvalidData, "Reach end of buffer"))
}
_ => return Err(io::Error::new(io::ErrorKind::InvalidData, "Reach end of buffer")),
}
}
Ok(VInt(result))
}
}

95
src/compression/composite.rs
View File

@@ -8,38 +8,40 @@ pub struct CompositeEncoder {
}
impl CompositeEncoder {
pub fn new() -> CompositeEncoder {
CompositeEncoder {
block_encoder: BlockEncoder::new(),
output: Vec::with_capacity(500_000),
}
}
pub fn compress_sorted(&mut self, vals: &[u32]) -> &[u8] {
self.output.clear();
let num_blocks = vals.len() / NUM_DOCS_PER_BLOCK;
let mut offset = 0u32;
for i in 0..num_blocks {
let vals_slice = &vals[i * NUM_DOCS_PER_BLOCK .. (i + 1) * NUM_DOCS_PER_BLOCK];
let vals_slice = &vals[i * NUM_DOCS_PER_BLOCK..(i + 1) * NUM_DOCS_PER_BLOCK];
let block_compressed = self.block_encoder.compress_block_sorted(vals_slice, offset);
offset = vals_slice[NUM_DOCS_PER_BLOCK - 1];
self.output.extend_from_slice(block_compressed);
}
let vint_compressed = self.block_encoder.compress_vint_sorted(&vals[num_blocks * NUM_DOCS_PER_BLOCK..], offset);
let vint_compressed =
self.block_encoder
.compress_vint_sorted(&vals[num_blocks * NUM_DOCS_PER_BLOCK..], offset);
self.output.extend_from_slice(vint_compressed);
&self.output
}
pub fn compress_unsorted(&mut self, vals: &[u32]) -> &[u8] {
self.output.clear();
let num_blocks = vals.len() / NUM_DOCS_PER_BLOCK;
for i in 0..num_blocks {
let vals_slice = &vals[i * NUM_DOCS_PER_BLOCK .. (i + 1) * NUM_DOCS_PER_BLOCK];
let vals_slice = &vals[i * NUM_DOCS_PER_BLOCK..(i + 1) * NUM_DOCS_PER_BLOCK];
let block_compressed = self.block_encoder.compress_block_unsorted(vals_slice);
self.output.extend_from_slice(block_compressed);
}
let vint_compressed = self.block_encoder.compress_vint_unsorted(&vals[num_blocks * NUM_DOCS_PER_BLOCK..]);
let vint_compressed = self.block_encoder
.compress_vint_unsorted(&vals[num_blocks * NUM_DOCS_PER_BLOCK..]);
self.output.extend_from_slice(vint_compressed);
&self.output
}
@@ -57,10 +59,13 @@ impl CompositeDecoder {
CompositeDecoder {
block_decoder: BlockDecoder::new(),
vals: Vec::with_capacity(500_000),
}
}
}
pub fn uncompress_sorted(&mut self, mut compressed_data: &[u8], uncompressed_len: usize) -> &[u32] {
pub fn uncompress_sorted(&mut self,
mut compressed_data: &[u8],
uncompressed_len: usize)
-> &[u32] {
if uncompressed_len > self.vals.capacity() {
let extra_capacity = uncompressed_len - self.vals.capacity();
self.vals.reserve(extra_capacity);
@@ -69,24 +74,37 @@ impl CompositeDecoder {
self.vals.clear();
let num_blocks = uncompressed_len / NUM_DOCS_PER_BLOCK;
for _ in 0..num_blocks {
compressed_data = self.block_decoder.uncompress_block_sorted(compressed_data, offset);
compressed_data = self.block_decoder
.uncompress_block_sorted(compressed_data, offset);
offset = self.block_decoder.output(NUM_DOCS_PER_BLOCK - 1);
self.vals.extend_from_slice(self.block_decoder.output_array());
self.vals
.extend_from_slice(self.block_decoder.output_array());
}
self.block_decoder.uncompress_vint_sorted(compressed_data, offset, uncompressed_len % NUM_DOCS_PER_BLOCK);
self.vals.extend_from_slice(self.block_decoder.output_array());
self.block_decoder
.uncompress_vint_sorted(compressed_data,
offset,
uncompressed_len % NUM_DOCS_PER_BLOCK);
self.vals
.extend_from_slice(self.block_decoder.output_array());
&self.vals
}
pub fn uncompress_unsorted(&mut self, mut compressed_data: &[u8], uncompressed_len: usize) -> &[u32] {
pub fn uncompress_unsorted(&mut self,
mut compressed_data: &[u8],
uncompressed_len: usize)
-> &[u32] {
self.vals.clear();
let num_blocks = uncompressed_len / NUM_DOCS_PER_BLOCK;
for _ in 0..num_blocks {
compressed_data = self.block_decoder.uncompress_block_unsorted(compressed_data);
self.vals.extend_from_slice(self.block_decoder.output_array());
compressed_data = self.block_decoder
.uncompress_block_unsorted(compressed_data);
self.vals
.extend_from_slice(self.block_decoder.output_array());
}
self.block_decoder.uncompress_vint_unsorted(compressed_data, uncompressed_len % NUM_DOCS_PER_BLOCK);
self.vals.extend_from_slice(self.block_decoder.output_array());
self.block_decoder
.uncompress_vint_unsorted(compressed_data, uncompressed_len % NUM_DOCS_PER_BLOCK);
self.vals
.extend_from_slice(self.block_decoder.output_array());
&self.vals
}
}
@@ -100,14 +118,14 @@ impl Into<Vec<u32>> for CompositeDecoder {
#[cfg(test)]
pub mod tests {
use test::Bencher;
use super::*;
use compression::tests::generate_array;
use tests;
#[test]
fn test_composite_unsorted() {
let data = generate_array(10_000, 0.1);
let data = tests::generate_array(10_000, 0.1);
let mut encoder = CompositeEncoder::new();
let compressed = encoder.compress_unsorted(&data);
assert!(compressed.len() <= 19_794);
@@ -115,12 +133,12 @@ pub mod tests {
let result = decoder.uncompress_unsorted(&compressed, data.len());
for i in 0..data.len() {
assert_eq!(data[i], result[i]);
}
}
}
#[test]
fn test_composite_sorted() {
let data = generate_array(10_000, 0.1);
let data = tests::generate_array(10_000, 0.1);
let mut encoder = CompositeEncoder::new();
let compressed = encoder.compress_sorted(&data);
assert!(compressed.len() <= 7_826);
@@ -128,32 +146,25 @@ pub mod tests {
let result = decoder.uncompress_sorted(&compressed, data.len());
for i in 0..data.len() {
assert_eq!(data[i], result[i]);
}
}
}
const BENCH_NUM_INTS: usize = 99_968;
#[bench]
fn bench_compress(b: &mut Bencher) {
let mut encoder = CompositeEncoder::new();
let data = generate_array(BENCH_NUM_INTS, 0.1);
b.iter(|| {
encoder.compress_sorted(&data);
});
let data = tests::generate_array(BENCH_NUM_INTS, 0.1);
b.iter(|| { encoder.compress_sorted(&data); });
}
#[bench]
fn bench_uncompress(b: &mut Bencher) {
let mut encoder = CompositeEncoder::new();
let data = generate_array(BENCH_NUM_INTS, 0.1);
let data = tests::generate_array(BENCH_NUM_INTS, 0.1);
let compressed = encoder.compress_sorted(&data);
let mut decoder = CompositeDecoder::new();
b.iter(|| {
decoder.uncompress_sorted(compressed, BENCH_NUM_INTS);
});
let mut decoder = CompositeDecoder::new();
b.iter(|| { decoder.uncompress_sorted(compressed, BENCH_NUM_INTS); });
}
}

116
src/compression/mod.rs
View File

@@ -38,39 +38,44 @@ pub trait VIntEncoder {
}
pub trait VIntDecoder {
fn uncompress_vint_sorted<'a>(&mut self, compressed_data: &'a [u8], offset: u32, num_els: usize) -> &'a [u8];
fn uncompress_vint_unsorted<'a>(&mut self, compressed_data: &'a [u8], num_els: usize) -> &'a [u8];
fn uncompress_vint_sorted<'a>(&mut self,
compressed_data: &'a [u8],
offset: u32,
num_els: usize)
-> &'a [u8];
fn uncompress_vint_unsorted<'a>(&mut self,
compressed_data: &'a [u8],
num_els: usize)
-> &'a [u8];
}
impl VIntEncoder for BlockEncoder {
fn compress_vint_sorted(&mut self, input: &[u32], offset: u32) -> &[u8] {
vint::compress_sorted(input, &mut self.output, offset)
}
fn compress_vint_unsorted(&mut self, input: &[u32]) -> &[u8] {
vint::compress_unsorted(input, &mut self.output)
}
}
impl VIntDecoder for BlockDecoder {
fn uncompress_vint_sorted<'a>(
&mut self,
compressed_data: &'a [u8],
offset: u32,
num_els: usize) -> &'a [u8] {
fn uncompress_vint_sorted<'a>(&mut self,
compressed_data: &'a [u8],
offset: u32,
num_els: usize)
-> &'a [u8] {
self.output_len = num_els;
vint::uncompress_sorted(compressed_data, &mut self.output[..num_els], offset)
}
fn uncompress_vint_unsorted<'a>(
&mut self,
compressed_data: &'a [u8],
num_els: usize) -> &'a [u8] {
fn uncompress_vint_unsorted<'a>(&mut self,
compressed_data: &'a [u8],
num_els: usize)
-> &'a [u8] {
self.output_len = num_els;
vint::uncompress_unsorted(compressed_data, &mut self.output[..num_els])
}
}
}
@@ -79,33 +84,18 @@ pub const NUM_DOCS_PER_BLOCK: usize = 128; //< should be a power of 2 to let the
#[cfg(test)]
pub mod tests {
use rand::Rng;
use rand::SeedableRng;
use rand::XorShiftRng;
use super::*;
use tests;
use test::Bencher;
fn generate_array_with_seed(n: usize, ratio: f32, seed_val: u32) -> Vec<u32> {
let seed: &[u32; 4] = &[1, 2, 3, seed_val];
let mut rng: XorShiftRng = XorShiftRng::from_seed(*seed);
(0..u32::max_value())
.filter(|_| rng.next_f32()< ratio)
.take(n)
.collect()
}
pub fn generate_array(n: usize, ratio: f32) -> Vec<u32> {
generate_array_with_seed(n, ratio, 4)
}
#[test]
fn test_encode_sorted_block() {
let vals: Vec<u32> = (0u32..128u32).map(|i| i*7).collect();
let vals: Vec<u32> = (0u32..128u32).map(|i| i * 7).collect();
let mut encoder = BlockEncoder::new();
let compressed_data = encoder.compress_block_sorted(&vals, 0);
let mut decoder = BlockDecoder::new();
{
let remaining_data = decoder.uncompress_block_sorted(compressed_data, 0);
let remaining_data = decoder.uncompress_block_sorted(compressed_data, 0);
assert_eq!(remaining_data.len(), 0);
}
for i in 0..128 {
@@ -115,31 +105,31 @@ pub mod tests {
#[test]
fn test_encode_sorted_block_with_offset() {
let vals: Vec<u32> = (0u32..128u32).map(|i| 11 + i*7).collect();
let vals: Vec<u32> = (0u32..128u32).map(|i| 11 + i * 7).collect();
let mut encoder = BlockEncoder::new();
let compressed_data = encoder.compress_block_sorted(&vals, 10);
let mut decoder = BlockDecoder::new();
{
let remaining_data = decoder.uncompress_block_sorted(compressed_data, 10);
let remaining_data = decoder.uncompress_block_sorted(compressed_data, 10);
assert_eq!(remaining_data.len(), 0);
}
for i in 0..128 {
assert_eq!(vals[i], decoder.output(i));
}
}
#[test]
fn test_encode_sorted_block_with_junk() {
let mut compressed: Vec<u8> = Vec::new();
let n = 128;
let vals: Vec<u32> = (0..n).map(|i| 11u32 + (i as u32)*7u32).collect();
let vals: Vec<u32> = (0..n).map(|i| 11u32 + (i as u32) * 7u32).collect();
let mut encoder = BlockEncoder::new();
let compressed_data = encoder.compress_block_sorted(&vals, 10);
compressed.extend_from_slice(compressed_data);
compressed.push(173u8);
let mut decoder = BlockDecoder::new();
{
let remaining_data = decoder.uncompress_block_sorted(&compressed, 10);
let remaining_data = decoder.uncompress_block_sorted(&compressed, 10);
assert_eq!(remaining_data.len(), 1);
assert_eq!(remaining_data[0], 173u8);
}
@@ -152,14 +142,14 @@ pub mod tests {
fn test_encode_unsorted_block_with_junk() {
let mut compressed: Vec<u8> = Vec::new();
let n = 128;
let vals: Vec<u32> = (0..n).map(|i| 11u32 + (i as u32)*7u32 % 12).collect();
let vals: Vec<u32> = (0..n).map(|i| 11u32 + (i as u32) * 7u32 % 12).collect();
let mut encoder = BlockEncoder::new();
let compressed_data = encoder.compress_block_unsorted(&vals);
compressed.extend_from_slice(compressed_data);
compressed.push(173u8);
let mut decoder = BlockDecoder::new();
{
let remaining_data = decoder.uncompress_block_unsorted(&compressed);
let remaining_data = decoder.uncompress_block_unsorted(&compressed);
assert_eq!(remaining_data.len(), 1);
assert_eq!(remaining_data[0], 173u8);
}
@@ -167,22 +157,20 @@ pub mod tests {
assert_eq!(vals[i], decoder.output(i));
}
}
#[test]
fn test_encode_vint() {
{
let expected_length = 154;
let mut encoder = BlockEncoder::new();
let input: Vec<u32> = (0u32..123u32)
.map(|i| 4 + i * 7 / 2)
.into_iter()
.collect();
let input: Vec<u32> = (0u32..123u32).map(|i| 4 + i * 7 / 2).into_iter().collect();
for offset in &[0u32, 1u32, 2u32] {
let encoded_data = encoder.compress_vint_sorted(&input, *offset);
assert!(encoded_data.len() <= expected_length);
let mut decoder = BlockDecoder::new();
let remaining_data = decoder.uncompress_vint_sorted(&encoded_data, *offset, input.len());
let remaining_data =
decoder.uncompress_vint_sorted(&encoded_data, *offset, input.len());
assert_eq!(0, remaining_data.len());
assert_eq!(input, decoder.output_array());
}
@@ -193,21 +181,17 @@ pub mod tests {
#[bench]
fn bench_compress(b: &mut Bencher) {
let mut encoder = BlockEncoder::new();
let data = generate_array(NUM_DOCS_PER_BLOCK, 0.1);
b.iter(|| {
encoder.compress_block_sorted(&data, 0u32);
});
let data = tests::generate_array(NUM_DOCS_PER_BLOCK, 0.1);
b.iter(|| { encoder.compress_block_sorted(&data, 0u32); });
}
#[bench]
fn bench_uncompress(b: &mut Bencher) {
let mut encoder = BlockEncoder::new();
let data = generate_array(NUM_DOCS_PER_BLOCK, 0.1);
let data = tests::generate_array(NUM_DOCS_PER_BLOCK, 0.1);
let compressed = encoder.compress_block_sorted(&data, 0u32);
let mut decoder = BlockDecoder::new();
b.iter(|| {
decoder.uncompress_block_sorted(compressed, 0u32);
});
let mut decoder = BlockDecoder::new();
b.iter(|| { decoder.uncompress_block_sorted(compressed, 0u32); });
}
@@ -216,21 +200,17 @@ pub mod tests {
#[bench]
fn bench_compress_vint(b: &mut Bencher) {
let mut encoder = BlockEncoder::new();
let data = generate_array(NUM_INTS_BENCH_VINT, 0.001);
b.iter(|| {
encoder.compress_vint_sorted(&data, 0u32);
});
let data = tests::generate_array(NUM_INTS_BENCH_VINT, 0.001);
b.iter(|| { encoder.compress_vint_sorted(&data, 0u32); });
}
#[bench]
fn bench_uncompress_vint(b: &mut Bencher) {
let mut encoder = BlockEncoder::new();
let data = generate_array(NUM_INTS_BENCH_VINT, 0.001);
let data = tests::generate_array(NUM_INTS_BENCH_VINT, 0.001);
let compressed = encoder.compress_vint_sorted(&data, 0u32);
let mut decoder = BlockDecoder::new();
b.iter(|| {
decoder.uncompress_vint_sorted(compressed, 0u32, NUM_INTS_BENCH_VINT);
});
let mut decoder = BlockDecoder::new();
b.iter(|| { decoder.uncompress_vint_sorted(compressed, 0u32, NUM_INTS_BENCH_VINT); });
}
}

54
src/compression/pack/compression_pack_nosimd.rs
View File

@@ -4,10 +4,10 @@ use std::cmp;
use std::io::Write;
use super::super::NUM_DOCS_PER_BLOCK;
const COMPRESSED_BLOCK_MAX_SIZE: usize = NUM_DOCS_PER_BLOCK * 4 + 1;
const COMPRESSED_BLOCK_MAX_SIZE: usize = NUM_DOCS_PER_BLOCK * 4 + 1;
pub fn compress_sorted(vals: &mut [u32], mut output: &mut [u8], offset: u32) -> usize {
let mut max_delta = 0;
let mut max_delta = 0;
{
let mut local_offset = offset;
for i in 0..NUM_DOCS_PER_BLOCK {
@@ -24,7 +24,10 @@ pub fn compress_sorted(vals: &mut [u32], mut output: &mut [u8], offset: u32) ->
for val in vals {
bit_packer.write(*val, &mut output).unwrap();
}
1 + bit_packer.close(&mut output).expect("packing in memory should never fail")
1 +
bit_packer
.close(&mut output)
.expect("packing in memory should never fail")
}
@@ -36,36 +39,40 @@ pub struct BlockEncoder {
}
impl BlockEncoder {
pub fn new() -> BlockEncoder {
BlockEncoder {
output: [0u8; COMPRESSED_BLOCK_MAX_SIZE],
output_len: 0,
input_buffer: [0u32; NUM_DOCS_PER_BLOCK],
}
}
}
pub fn compress_block_sorted(&mut self, vals: &[u32], offset: u32) -> &[u8] {
self.input_buffer.clone_from_slice(vals);
let compressed_size = compress_sorted(&mut self.input_buffer, &mut self.output, offset);
&self.output[..compressed_size]
}
pub fn compress_block_unsorted(&mut self, vals: &[u32]) -> &[u8] {
pub fn compress_block_unsorted(&mut self, vals: &[u32]) -> &[u8] {
let compressed_size: usize = {
let mut output: &mut [u8] = &mut self.output;
let max = vals.iter().cloned().max().expect("compress unsorted called with an empty array");
let mut output: &mut [u8] = &mut self.output;
let max = vals.iter()
.cloned()
.max()
.expect("compress unsorted called with an empty array");
let num_bits = compute_num_bits(max);
output.write_all(&[num_bits]).unwrap();
let mut bit_packer = BitPacker::new(num_bits as usize);
for val in vals {
bit_packer.write(*val, &mut output).unwrap();
}
1 + bit_packer.close(&mut output).expect("packing in memory should never fail")
1 +
bit_packer
.close(&mut output)
.expect("packing in memory should never fail")
};
&self.output[..compressed_size]
}
}
pub struct BlockDecoder {
@@ -78,15 +85,18 @@ impl BlockDecoder {
pub fn new() -> BlockDecoder {
BlockDecoder::with_val(0u32)
}
pub fn with_val(val: u32) -> BlockDecoder {
BlockDecoder {
output: [val; COMPRESSED_BLOCK_MAX_SIZE],
output_len: 0,
}
}
pub fn uncompress_block_sorted<'a>(&mut self, compressed_data: &'a [u8], mut offset: u32) -> &'a[u8] {
pub fn uncompress_block_sorted<'a>(&mut self,
compressed_data: &'a [u8],
mut offset: u32)
-> &'a [u8] {
let consumed_size = {
let num_bits = compressed_data[0];
let bit_unpacker = BitUnpacker::new(&compressed_data[1..], num_bits as usize);
@@ -96,13 +106,13 @@ impl BlockDecoder {
self.output[i] = val;
offset = val;
}
1 + (num_bits as usize * NUM_DOCS_PER_BLOCK + 7) / 8
1 + (num_bits as usize * NUM_DOCS_PER_BLOCK + 7) / 8
};
self.output_len = NUM_DOCS_PER_BLOCK;
&compressed_data[consumed_size..]
}
pub fn uncompress_block_unsorted<'a>(&mut self, compressed_data: &'a [u8]) -> &'a[u8] {
pub fn uncompress_block_unsorted<'a>(&mut self, compressed_data: &'a [u8]) -> &'a [u8] {
let num_bits = compressed_data[0];
let bit_unpacker = BitUnpacker::new(&compressed_data[1..], num_bits as usize);
for i in 0..NUM_DOCS_PER_BLOCK {
@@ -112,16 +122,14 @@ impl BlockDecoder {
self.output_len = NUM_DOCS_PER_BLOCK;
&compressed_data[consumed_size..]
}
#[inline]
pub fn output_array(&self,) -> &[u32] {
pub fn output_array(&self) -> &[u32] {
&self.output[..self.output_len]
}
#[inline]
pub fn output(&self, idx: usize) -> u32 {
self.output[idx]
}
}

67
src/compression/pack/compression_pack_simd.rs
View File

@@ -1,28 +1,21 @@
use super::super::NUM_DOCS_PER_BLOCK;
const COMPRESSED_BLOCK_MAX_SIZE: usize = NUM_DOCS_PER_BLOCK * 4 + 1;
const COMPRESSED_BLOCK_MAX_SIZE: usize = NUM_DOCS_PER_BLOCK * 4 + 1;
mod simdcomp {
use libc::size_t;
extern {
pub fn compress_sorted(
data: *const u32,
output: *mut u8,
offset: u32) -> size_t;
extern "C" {
pub fn compress_sorted(data: *const u32, output: *mut u8, offset: u32) -> size_t;
pub fn uncompress_sorted(
compressed_data: *const u8,
output: *mut u32,
offset: u32) -> size_t;
pub fn compress_unsorted(
data: *const u32,
output: *mut u8) -> size_t;
pub fn uncompress_sorted(compressed_data: *const u8,
output: *mut u32,
offset: u32)
-> size_t;
pub fn uncompress_unsorted(
compressed_data: *const u8,
output: *mut u32) -> size_t;
pub fn compress_unsorted(data: *const u32, output: *mut u8) -> size_t;
pub fn uncompress_unsorted(compressed_data: *const u8, output: *mut u32) -> size_t;
}
}
@@ -49,24 +42,22 @@ pub struct BlockEncoder {
}
impl BlockEncoder {
pub fn new() -> BlockEncoder {
BlockEncoder {
output: [0u8; COMPRESSED_BLOCK_MAX_SIZE],
output_len: 0,
}
}
}
pub fn compress_block_sorted(&mut self, vals: &[u32], offset: u32) -> &[u8] {
let compressed_size = compress_sorted(vals, &mut self.output, offset);
&self.output[..compressed_size]
}
pub fn compress_block_unsorted(&mut self, vals: &[u32]) -> &[u8] {
let compressed_size = compress_unsorted(vals, &mut self.output);
&self.output[..compressed_size]
}
}
pub struct BlockDecoder {
@@ -79,31 +70,34 @@ impl BlockDecoder {
pub fn new() -> BlockDecoder {
BlockDecoder::with_val(0u32)
}
pub fn with_val(val: u32) -> BlockDecoder {
BlockDecoder {
output: [val; COMPRESSED_BLOCK_MAX_SIZE],
output_len: 0,
}
}
pub fn uncompress_block_sorted<'a>(&mut self, compressed_data: &'a [u8], offset: u32) -> &'a[u8] {
pub fn uncompress_block_sorted<'a>(&mut self,
compressed_data: &'a [u8],
offset: u32)
-> &'a [u8] {
let consumed_size = uncompress_sorted(compressed_data, &mut self.output, offset);
self.output_len = NUM_DOCS_PER_BLOCK;
&compressed_data[consumed_size..]
}
pub fn uncompress_block_unsorted<'a>(&mut self, compressed_data: &'a [u8]) -> &'a[u8] {
pub fn uncompress_block_unsorted<'a>(&mut self, compressed_data: &'a [u8]) -> &'a [u8] {
let consumed_size = uncompress_unsorted(compressed_data, &mut self.output);
self.output_len = NUM_DOCS_PER_BLOCK;
&compressed_data[consumed_size..]
}
#[inline]
pub fn output_array(&self,) -> &[u32] {
pub fn output_array(&self) -> &[u32] {
&self.output[..self.output_len]
}
#[inline]
pub fn output(&self, idx: usize) -> u32 {
self.output[idx]
@@ -111,3 +105,16 @@ impl BlockDecoder {
}
#[cfg(test)]
mod tests {
use super::BlockEncoder;
#[test]
fn test_all_docs_compression_len() {
let data: Vec<u32> = (0u32..128u32).collect();
let mut encoder = BlockEncoder::new();
let compressed = encoder.compress_block_sorted(&data, 0u32);
assert_eq!(compressed.len(), 17);
}
}

18
src/compression/vint/compression_vint_nosimd.rs
View File

@@ -12,8 +12,7 @@ pub fn compress_sorted<'a>(input: &[u32], output: &'a mut [u8], mut offset: u32)
output[byte_written] = next_byte | 128u8;
byte_written += 1;
break;
}
else {
} else {
output[byte_written] = next_byte;
byte_written += 1;
}
@@ -34,8 +33,7 @@ pub fn compress_unsorted<'a>(input: &[u32], output: &'a mut [u8]) -> &'a [u8] {
output[byte_written] = next_byte | 128u8;
byte_written += 1;
break;
}
else {
} else {
output[byte_written] = next_byte;
byte_written += 1;
}
@@ -45,10 +43,10 @@ pub fn compress_unsorted<'a>(input: &[u32], output: &'a mut [u8]) -> &'a [u8] {
}
#[inline(always)]
pub fn uncompress_sorted<'a>(
compressed_data: &'a [u8],
output: &mut [u32],
offset: u32) -> &'a [u8] {
pub fn uncompress_sorted<'a>(compressed_data: &'a [u8],
output: &mut [u32],
offset: u32)
-> &'a [u8] {
let mut read_byte = 0;
let mut result = offset;
let num_els = output.len();
@@ -69,9 +67,7 @@ pub fn uncompress_sorted<'a>(
}
#[inline(always)]
pub fn uncompress_unsorted<'a>(
compressed_data: &'a [u8],
output: &mut [u32]) -> &'a [u8] {
pub fn uncompress_unsorted<'a>(compressed_data: &'a [u8], output: &mut [u32]) -> &'a [u8] {
let mut read_byte = 0;
let num_els = output.len();
for i in 0..num_els {

82
src/compression/vint/compression_vint_simd.rs
View File

@@ -3,28 +3,25 @@ mod streamvbyte {
use libc::size_t;
extern {
pub fn streamvbyte_delta_encode(
data: *const u32,
num_els: u32,
output: *mut u8,
offset: u32) -> size_t;
extern "C" {
pub fn streamvbyte_delta_encode(data: *const u32,
num_els: u32,
output: *mut u8,
offset: u32)
-> size_t;
pub fn streamvbyte_delta_decode(
compressed_data: *const u8,
output: *mut u32,
num_els: u32,
offset: u32) -> size_t;
pub fn streamvbyte_encode(
data: *const u32,
num_els: u32,
output: *mut u8) -> size_t;
pub fn streamvbyte_decode(
compressed_data: *const u8,
output: *mut u32,
num_els: usize) -> size_t;
pub fn streamvbyte_delta_decode(compressed_data: *const u8,
output: *mut u32,
num_els: u32,
offset: u32)
-> size_t;
pub fn streamvbyte_encode(data: *const u32, num_els: u32, output: *mut u8) -> size_t;
pub fn streamvbyte_decode(compressed_data: *const u8,
output: *mut u32,
num_els: usize)
-> size_t;
}
}
@@ -32,11 +29,10 @@ mod streamvbyte {
#[inline(always)]
pub fn compress_sorted<'a>(input: &[u32], output: &'a mut [u8], offset: u32) -> &'a [u8] {
let compress_length = unsafe {
streamvbyte::streamvbyte_delta_encode(
input.as_ptr(),
input.len() as u32,
output.as_mut_ptr(),
offset)
streamvbyte::streamvbyte_delta_encode(input.as_ptr(),
input.len() as u32,
output.as_mut_ptr(),
offset)
};
&output[..compress_length]
}
@@ -44,39 +40,29 @@ pub fn compress_sorted<'a>(input: &[u32], output: &'a mut [u8], offset: u32) ->
#[inline(always)]
pub fn compress_unsorted<'a>(input: &[u32], output: &'a mut [u8]) -> &'a [u8] {
let compress_length = unsafe {
streamvbyte::streamvbyte_encode(
input.as_ptr(),
input.len() as u32,
output.as_mut_ptr())
};
streamvbyte::streamvbyte_encode(input.as_ptr(), input.len() as u32, output.as_mut_ptr())
};
&output[..compress_length]
}
#[inline(always)]
pub fn uncompress_sorted<'a>(
compressed_data: &'a [u8],
output: &mut [u32],
offset: u32) -> &'a [u8] {
pub fn uncompress_sorted<'a>(compressed_data: &'a [u8],
output: &mut [u32],
offset: u32)
-> &'a [u8] {
let consumed_bytes = unsafe {
streamvbyte::streamvbyte_delta_decode(
compressed_data.as_ptr(),
output.as_mut_ptr(),
output.len() as u32,
offset)
streamvbyte::streamvbyte_delta_decode(compressed_data.as_ptr(),
output.as_mut_ptr(),
output.len() as u32,
offset)
};
&compressed_data[consumed_bytes..]
}
#[inline(always)]
pub fn uncompress_unsorted<'a>(
compressed_data: &'a [u8],
output: &mut [u32]) -> &'a [u8] {
pub fn uncompress_unsorted<'a>(compressed_data: &'a [u8], output: &mut [u32]) -> &'a [u8] {
let consumed_bytes = unsafe {
streamvbyte::streamvbyte_decode(
compressed_data.as_ptr(),
output.as_mut_ptr(),
output.len())
streamvbyte::streamvbyte_decode(compressed_data.as_ptr(), output.as_mut_ptr(), output.len())
};
&compressed_data[consumed_bytes..]
}

26
src/core/index.rs
View File

@@ -48,8 +48,11 @@ impl Index {
/// This should only be used for unit tests.
pub fn create_in_ram(schema: Schema) -> Index {
let ram_directory = RAMDirectory::create();
let directory = ManagedDirectory::new(ram_directory).expect("Creating a managed directory from a brand new RAM directory should never fail.");
Index::from_directory(directory, schema).expect("Creating a RAMDirectory should never fail") // unwrap is ok here
// unwrap is ok here
let directory = ManagedDirectory::new(ram_directory)
.expect("Creating a managed directory from a brand new RAM directory \
should never fail.");
Index::from_directory(directory, schema).expect("Creating a RAMDirectory should never fail")
}
/// Creates a new index in a given filepath.
@@ -153,11 +156,10 @@ impl Index {
/// Returns the list of segments that are searchable
pub fn searchable_segments(&self) -> Result<Vec<Segment>> {
Ok(self
.searchable_segment_metas()?
.into_iter()
.map(|segment_meta| self.segment(segment_meta))
.collect())
Ok(self.searchable_segment_metas()?
.into_iter()
.map(|segment_meta| self.segment(segment_meta))
.collect())
}
#[doc(hidden)]
@@ -186,13 +188,13 @@ impl Index {
pub fn searchable_segment_metas(&self) -> Result<Vec<SegmentMeta>> {
Ok(load_metas(self.directory())?.segments)
}
/// Returns the list of segment ids that are searchable.
pub fn searchable_segment_ids(&self) -> Result<Vec<SegmentId>> {
Ok(self.searchable_segment_metas()?
.iter()
.map(|segment_meta| segment_meta.id())
.collect())
.collect())
}
/// Creates a new generation of searchers after
@@ -203,9 +205,9 @@ impl Index {
pub fn load_searchers(&self) -> Result<()> {
let searchable_segments = self.searchable_segments()?;
let segment_readers: Vec<SegmentReader> = try!(searchable_segments
.into_iter()
.map(SegmentReader::open)
.collect());
.into_iter()
.map(SegmentReader::open)
.collect());
let searchers = (0..NUM_SEARCHERS)
.map(|_| Searcher::from(segment_readers.clone()))
.collect();

6
src/core/index_meta.rs
View File

@@ -2,9 +2,9 @@ use schema::Schema;
use core::SegmentMeta;
/// Meta information about the `Index`.
///
///
/// This object is serialized on disk in the `meta.json` file.
/// It keeps information about
/// It keeps information about
/// * the searchable segments,
/// * the index docstamp
/// * the schema
@@ -19,7 +19,7 @@ pub struct IndexMeta {
impl IndexMeta {
pub fn with_schema(schema: Schema) -> IndexMeta {
IndexMeta {
segments: vec!(),
segments: vec![],
schema: schema,
opstamp: 0u64,
}

4
src/core/mod.rs
View File

@@ -27,7 +27,7 @@ lazy_static! {
/// The meta file contains all the information about the list of segments and the schema
/// of the index.
pub static ref META_FILEPATH: PathBuf = PathBuf::from("meta.json");
/// The managed file contains a list of files that were created by the tantivy
/// and will therefore be garbage collected when they are deemed useless by tantivy.
///
@@ -40,4 +40,4 @@ lazy_static! {
///
/// If the process is killed and this file remains, it is safe to remove it manually.
pub static ref LOCKFILE_FILEPATH: PathBuf = PathBuf::from(".tantivy-indexer.lock");
}
}

83
src/core/pool.rs
View File

@@ -10,17 +10,44 @@ pub struct GenerationItem<T> {
item: T,
}
// See https://github.com/crossbeam-rs/crossbeam/issues/91
struct NonLeakingMsQueue<T> {
underlying_queue: MsQueue<T>,
}
impl<T> Default for NonLeakingMsQueue<T> {
fn default() -> NonLeakingMsQueue<T> {
NonLeakingMsQueue { underlying_queue: MsQueue::new() }
}
}
impl<T> NonLeakingMsQueue<T> {
fn pop(&self) -> T {
self.underlying_queue.pop()
}
fn push(&self, el: T) {
self.underlying_queue.push(el);
}
}
impl<T> Drop for NonLeakingMsQueue<T> {
fn drop(&mut self) {
while let Some(_popped_item_to_be_dropped) = self.underlying_queue.try_pop() {}
}
}
pub struct Pool<T> {
queue: Arc<MsQueue<GenerationItem<T>>>,
queue: Arc<NonLeakingMsQueue<GenerationItem<T>>>,
freshest_generation: AtomicUsize,
next_generation: AtomicUsize,
}
impl<T> Pool<T> {
pub fn new() -> Pool<T> {
Pool {
queue: Arc::new(MsQueue::new()),
queue: Arc::default(),
freshest_generation: AtomicUsize::default(),
next_generation: AtomicUsize::default(),
}
@@ -37,70 +64,74 @@ impl<T> Pool<T> {
}
self.advertise_generation(next_generation);
}
/// At the exit of this method,
/// At the exit of this method,
/// - freshest_generation has a value greater or equal than generation
/// - freshest_generation has a value that has been advertised
/// - freshest_generation has
/// - freshest_generation has
fn advertise_generation(&self, generation: usize) {
// not optimal at all but the easiest to read proof.
// not optimal at all but the easiest to read proof.
loop {
let former_generation = self.freshest_generation.load(Ordering::Acquire);
if former_generation >= generation {
break;
}
self.freshest_generation.compare_and_swap(former_generation, generation, Ordering::SeqCst);
}
self.freshest_generation
.compare_and_swap(former_generation, generation, Ordering::SeqCst);
}
}
fn generation(&self,) -> usize {
fn generation(&self) -> usize {
self.freshest_generation.load(Ordering::Acquire)
}
pub fn acquire(&self,) -> LeasedItem<T> {
pub fn acquire(&self) -> LeasedItem<T> {
let generation = self.generation();
loop {
let gen_item = self.queue.pop();
if gen_item.generation >= generation {
return LeasedItem {
gen_item: Some(gen_item),
recycle_queue: self.queue.clone(),
}
}
else {
gen_item: Some(gen_item),
recycle_queue: self.queue.clone(),
};
} else {
// this searcher is obsolete,
// removing it from the pool.
}
}
}
}
pub struct LeasedItem<T> {
gen_item: Option<GenerationItem<T>>,
recycle_queue: Arc<MsQueue<GenerationItem<T>>>,
recycle_queue: Arc<NonLeakingMsQueue<GenerationItem<T>>>,
}
impl<T> Deref for LeasedItem<T> {
type Target = T;
fn deref(&self) -> &T {
&self.gen_item.as_ref().expect("Unwrapping a leased item should never fail").item // unwrap is safe here
&self.gen_item
.as_ref()
.expect("Unwrapping a leased item should never fail")
.item // unwrap is safe here
}
}
impl<T> DerefMut for LeasedItem<T> {
fn deref_mut(&mut self) -> &mut T {
&mut self.gen_item.as_mut().expect("Unwrapping a mut leased item should never fail").item // unwrap is safe here
&mut self.gen_item
.as_mut()
.expect("Unwrapping a mut leased item should never fail")
.item // unwrap is safe here
}
}
impl<T> Drop for LeasedItem<T> {
fn drop(&mut self) {
let gen_item: GenerationItem<T> = mem::replace(&mut self.gen_item, None).expect("Unwrapping a leased item should never fail");
let gen_item: GenerationItem<T> = mem::replace(&mut self.gen_item, None)
.expect("Unwrapping a leased item should never fail");
self.recycle_queue.push(gen_item);
}
}
@@ -127,4 +158,4 @@ mod tests {
assert_eq!(*pool.acquire(), 11);
}
}
}
}

29
src/core/searcher.rs
View File

@@ -13,36 +13,35 @@ use std::fmt;
/// Holds a list of `SegmentReader`s ready for search.
///
/// It guarantees that the `Segment` will not be removed before
/// It guarantees that the `Segment` will not be removed before
/// the destruction of the `Searcher`.
///
///
pub struct Searcher {
segment_readers: Vec<SegmentReader>,
}
impl Searcher {
/// Fetches a document from tantivy's store given a `DocAddress`.
///
/// The searcher uses the segment ordinal to route the
/// the request to the right `Segment`.
/// the request to the right `Segment`.
pub fn doc(&self, doc_address: &DocAddress) -> Result<Document> {
let DocAddress(segment_local_id, doc_id) = *doc_address;
let segment_reader = &self.segment_readers[segment_local_id as usize];
segment_reader.doc(doc_id)
}
/// Returns the overall number of documents in the index.
pub fn num_docs(&self,) -> DocId {
pub fn num_docs(&self) -> DocId {
self.segment_readers
.iter()
.map(|segment_reader| segment_reader.num_docs())
.fold(0u32, |acc, val| acc + val)
}
/// Return the overall number of documents containing
/// the given term.
/// the given term.
pub fn doc_freq(&self, term: &Term) -> u32 {
self.segment_readers
.iter()
@@ -58,20 +57,20 @@ impl Searcher {
///
/// # Warning
/// This API is very likely to change in the future.
pub fn terms<'a>(&'a self) -> TermIterator<'a> {
pub fn terms(&self) -> TermIterator {
TermIterator::from(self.segment_readers())
}
/// Return the list of segment readers
pub fn segment_readers(&self,) -> &[SegmentReader] {
pub fn segment_readers(&self) -> &[SegmentReader] {
&self.segment_readers
}
/// Returns the segment_reader associated with the given segment_ordinal
pub fn segment_reader(&self, segment_ord: u32) -> &SegmentReader {
&self.segment_readers[segment_ord as usize]
}
/// Runs a query on the segment readers wrapped by the searcher
pub fn search<C: Collector>(&self, query: &Query, collector: &mut C) -> Result<TimerTree> {
query.search(self, collector)
@@ -81,9 +80,7 @@ impl Searcher {
impl From<Vec<SegmentReader>> for Searcher {
fn from(segment_readers: Vec<SegmentReader>) -> Searcher {
Searcher {
segment_readers: segment_readers,
}
Searcher { segment_readers: segment_readers }
}
}
@@ -95,4 +92,4 @@ impl fmt::Debug for Searcher {
.collect::<Vec<_>>();
write!(f, "Searcher({:?})", segment_ids)
}
}
}

29
src/core/segment.rs
View File

@@ -26,8 +26,8 @@ impl fmt::Debug for Segment {
}
/// Creates a new segment given an `Index` and a `SegmentId`
///
/// The function is here to make it private outside `tantivy`.
///
/// The function is here to make it private outside `tantivy`.
pub fn create_segment(index: Index, meta: SegmentMeta) -> Segment {
Segment {
index: index,
@@ -36,9 +36,8 @@ pub fn create_segment(index: Index, meta: SegmentMeta) -> Segment {
}
impl Segment {
/// Returns our index's schema.
pub fn schema(&self,) -> Schema {
pub fn schema(&self) -> Schema {
self.index.schema()
}
@@ -53,13 +52,13 @@ impl Segment {
}
/// Returns the segment's id.
pub fn id(&self,) -> SegmentId {
pub fn id(&self) -> SegmentId {
self.meta.id()
}
/// Returns the relative path of a component of our segment.
///
/// It just joins the segment id with the extension
///
/// It just joins the segment id with the extension
/// associated to a segment component.
pub fn relative_path(&self, component: SegmentComponent) -> PathBuf {
self.meta.relative_path(component)
@@ -77,14 +76,18 @@ impl Segment {
}
/// Open one of the component file for a *regular* read.
pub fn open_read(&self, component: SegmentComponent) -> result::Result<ReadOnlySource, OpenReadError> {
pub fn open_read(&self,
component: SegmentComponent)
-> result::Result<ReadOnlySource, OpenReadError> {
let path = self.relative_path(component);
let source = try!(self.index.directory().open_read(&path));
Ok(source)
}
/// Open one of the component file for *regular* write.
pub fn open_write(&mut self, component: SegmentComponent) -> result::Result<WritePtr, OpenWriteError> {
pub fn open_write(&mut self,
component: SegmentComponent)
-> result::Result<WritePtr, OpenWriteError> {
let path = self.relative_path(component);
let write = try!(self.index.directory_mut().open_write(&path));
Ok(write)
@@ -114,10 +117,10 @@ mod tests {
let mut index = Index::create_in_ram(SchemaBuilder::new().build());
let segment = index.new_segment();
let path = segment.relative_path(SegmentComponent::POSTINGS);
let directory = index.directory_mut();
directory.atomic_write(&*path, &vec!(0u8)).unwrap();
directory.atomic_write(&*path, &vec![0u8]).unwrap();
let living_files = HashSet::new();
{
let _file_protection = segment.protect_from_delete(SegmentComponent::POSTINGS);
@@ -130,4 +133,4 @@ mod tests {
assert!(!directory.exists(&*path));
}
}
}

24
src/core/segment_component.rs
View File

@@ -6,22 +6,18 @@ pub enum SegmentComponent {
FIELDNORMS,
TERMS,
STORE,
DELETE
DELETE,
}
impl SegmentComponent {
pub fn iterator() -> impl Iterator<Item=&'static SegmentComponent> {
static SEGMENT_COMPONENTS: [SegmentComponent; 7] = [
SegmentComponent::POSTINGS,
SegmentComponent::POSITIONS,
SegmentComponent::FASTFIELDS,
SegmentComponent::FIELDNORMS,
SegmentComponent::TERMS,
SegmentComponent::STORE,
SegmentComponent::DELETE
];
pub fn iterator() -> impl Iterator<Item = &'static SegmentComponent> {
static SEGMENT_COMPONENTS: [SegmentComponent; 7] = [SegmentComponent::POSTINGS,
SegmentComponent::POSITIONS,
SegmentComponent::FASTFIELDS,
SegmentComponent::FIELDNORMS,
SegmentComponent::TERMS,
SegmentComponent::STORE,
SegmentComponent::DELETE];
SEGMENT_COMPONENTS.into_iter()
}
}
}

12
src/core/segment_id.rs
View File

@@ -5,13 +5,13 @@ use std::cmp::{Ordering, Ord};
#[cfg(test)]
use std::sync::atomic;
/// Tantivy SegmentId.
/// Tantivy `SegmentId`.
///
/// Tantivy's segment are identified
/// Tantivy's segment are identified
/// by a UUID which is used to prefix the filenames
/// of all of the file associated with the segment.
///
/// In unit test, for reproducability, the SegmentId are
/// In unit test, for reproducability, the `SegmentId` are
/// simply generated in an autoincrement fashion.
#[derive(Clone, Copy, PartialEq, Eq, Hash, Serialize, Deserialize)]
pub struct SegmentId(Uuid);
@@ -52,14 +52,14 @@ impl SegmentId {
/// We are using UUID4, so only 6 bits are fixed,
/// and the rest is random.
///
/// Picking the first 8 chars is ok to identify
/// Picking the first 8 chars is ok to identify
/// segments in a display message.
pub fn short_uuid_string(&self,) -> String {
pub fn short_uuid_string(&self) -> String {
(&self.0.simple().to_string()[..8]).to_string()
}
/// Returns a segment uuid string.
pub fn uuid_string(&self,) -> String {
pub fn uuid_string(&self) -> String {
self.0.simple().to_string()
}
}

45
src/core/segment_meta.rs
View File

@@ -9,7 +9,7 @@ struct DeleteMeta {
opstamp: u64,
}
/// SegmentMeta contains simple meta information about a segment.
/// `SegmentMeta` contains simple meta information about a segment.
///
/// For instance the number of docs it contains,
/// how many are deleted, etc.
@@ -17,12 +17,11 @@ struct DeleteMeta {
pub struct SegmentMeta {
segment_id: SegmentId,
max_doc: u32,
deletes: Option<DeleteMeta>,
deletes: Option<DeleteMeta>,
}
impl SegmentMeta {
/// Creates a new segment meta for
/// Creates a new segment meta for
/// a segment with no deletes and no documents.
pub fn new(segment_id: SegmentId) -> SegmentMeta {
SegmentMeta {
@@ -53,28 +52,28 @@ impl SegmentMeta {
/// and are not used by any segment anymore.
pub fn list_files(&self) -> HashSet<PathBuf> {
SegmentComponent::iterator()
.map(|component| {
self.relative_path(*component)
})
.map(|component| self.relative_path(*component))
.collect::<HashSet<PathBuf>>()
}
/// Returns the relative path of a component of our segment.
///
/// It just joins the segment id with the extension
///
/// It just joins the segment id with the extension
/// associated to a segment component.
pub fn relative_path(&self, component: SegmentComponent) -> PathBuf {
let mut path = self.id().uuid_string();
path.push_str(&*match component {
SegmentComponent::POSITIONS => ".pos".to_string(),
SegmentComponent::POSTINGS => ".idx".to_string(),
SegmentComponent::TERMS => ".term".to_string(),
SegmentComponent::STORE => ".store".to_string(),
SegmentComponent::FASTFIELDS => ".fast".to_string(),
SegmentComponent::FIELDNORMS => ".fieldnorm".to_string(),
SegmentComponent::DELETE => {format!(".{}.del", self.delete_opstamp().unwrap_or(0))},
});
SegmentComponent::POSITIONS => ".pos".to_string(),
SegmentComponent::POSTINGS => ".idx".to_string(),
SegmentComponent::TERMS => ".term".to_string(),
SegmentComponent::STORE => ".store".to_string(),
SegmentComponent::FASTFIELDS => ".fast".to_string(),
SegmentComponent::FIELDNORMS => ".fieldnorm".to_string(),
SegmentComponent::DELETE => {
format!(".{}.del", self.delete_opstamp().unwrap_or(0))
}
});
PathBuf::from(path)
}
@@ -95,9 +94,7 @@ impl SegmentMeta {
/// Returns the opstamp of the last delete operation
/// taken in account in this segment.
pub fn delete_opstamp(&self) -> Option<u64> {
self.deletes
.as_ref()
.map(|delete_meta| delete_meta.opstamp)
self.deletes.as_ref().map(|delete_meta| delete_meta.opstamp)
}
/// Returns true iff the segment meta contains
@@ -114,8 +111,8 @@ impl SegmentMeta {
#[doc(hidden)]
pub fn set_delete_meta(&mut self, num_deleted_docs: u32, opstamp: u64) {
self.deletes = Some(DeleteMeta {
num_deleted_docs: num_deleted_docs,
opstamp: opstamp,
});
num_deleted_docs: num_deleted_docs,
opstamp: opstamp,
});
}
}

144
src/core/segment_reader.rs
View File

@@ -60,7 +60,7 @@ impl SegmentReader {
pub fn max_doc(&self) -> DocId {
self.segment_meta.max_doc()
}
/// Returns the number of documents.
/// Deleted documents are not counted.
///
@@ -69,7 +69,7 @@ impl SegmentReader {
pub fn num_docs(&self) -> DocId {
self.segment_meta.num_docs()
}
/// Return the number of documents that have been
/// deleted in the segment.
pub fn num_deleted_docs(&self) -> DocId {
@@ -91,39 +91,40 @@ impl SegmentReader {
///
/// # Panics
/// May panic if the index is corrupted.
pub fn get_fast_field_reader<TFastFieldReader: FastFieldReader>(&self, field: Field) -> fastfield::Result<TFastFieldReader> {
pub fn get_fast_field_reader<TFastFieldReader: FastFieldReader>
(&self,
field: Field)
-> fastfield::Result<TFastFieldReader> {
let field_entry = self.schema.get_field_entry(field);
if !TFastFieldReader::is_enabled(field_entry.field_type()) {
Err(FastFieldNotAvailableError::new(field_entry))
}
else {
Ok(
self.fast_fields_reader
.open_reader(field)
.expect("Fast field file corrupted.")
)
} else {
Ok(self.fast_fields_reader
.open_reader(field)
.expect("Fast field file corrupted."))
}
}
/// Accessor to the segment's `Field norms`'s reader.
///
/// Field norms are the length (in tokens) of the fields.
/// It is used in the computation of the [TfIdf](https://fulmicoton.gitbooks.io/tantivy-doc/content/tfidf.html).
/// It is used in the computation of the [TfIdf]
/// (https://fulmicoton.gitbooks.io/tantivy-doc/content/tfidf.html).
///
/// They are simply stored as a fast field, serialized in
/// the `.fieldnorm` file of the segment.
/// They are simply stored as a fast field, serialized in
/// the `.fieldnorm` file of the segment.
pub fn get_fieldnorms_reader(&self, field: Field) -> Option<U64FastFieldReader> {
self.fieldnorms_reader.open_reader(field)
self.fieldnorms_reader.open_reader(field)
}
/// Returns the number of documents containing the term.
pub fn doc_freq(&self, term: &Term) -> u32 {
match self.get_term_info(term) {
Some(term_info) => term_info.doc_freq,
None => 0,
}
}
}
/// Accessor to the segment's `StoreReader`.
pub fn get_store_reader(&self) -> &StoreReader {
&self.store_reader
@@ -136,46 +137,44 @@ impl SegmentReader {
let term_infos = try!(FstMap::from_source(source));
let store_reader = StoreReader::from(try!(segment.open_read(SegmentComponent::STORE)));
let postings_shared_mmap = try!(segment.open_read(SegmentComponent::POSTINGS));
let fast_field_data = try!(segment.open_read(SegmentComponent::FASTFIELDS));
let fast_fields_reader = try!(FastFieldsReader::open(fast_field_data));
let fieldnorms_data = try!(segment.open_read(SegmentComponent::FIELDNORMS));
let fieldnorms_reader = try!(FastFieldsReader::open(fieldnorms_data));
let positions_data = segment
.open_read(SegmentComponent::POSITIONS)
.unwrap_or_else(|_| ReadOnlySource::empty());
let delete_bitset =
if segment.meta().has_deletes() {
let delete_data = segment.open_read(SegmentComponent::DELETE)?;
DeleteBitSet::open(delete_data)
}
else {
DeleteBitSet::empty()
};
let delete_bitset = if segment.meta().has_deletes() {
let delete_data = segment.open_read(SegmentComponent::DELETE)?;
DeleteBitSet::open(delete_data)
} else {
DeleteBitSet::empty()
};
let schema = segment.schema();
Ok(SegmentReader {
segment_meta: segment.meta().clone(),
postings_data: postings_shared_mmap,
term_infos: Arc::new(term_infos),
segment_id: segment.id(),
store_reader: store_reader,
fast_fields_reader: Arc::new(fast_fields_reader),
fieldnorms_reader: Arc::new(fieldnorms_reader),
delete_bitset: delete_bitset,
positions_data: positions_data,
schema: schema,
})
segment_meta: segment.meta().clone(),
postings_data: postings_shared_mmap,
term_infos: Arc::new(term_infos),
segment_id: segment.id(),
store_reader: store_reader,
fast_fields_reader: Arc::new(fast_fields_reader),
fieldnorms_reader: Arc::new(fieldnorms_reader),
delete_bitset: delete_bitset,
positions_data: positions_data,
schema: schema,
})
}
/// Return the term dictionary datastructure.
pub fn term_infos(&self) -> &FstMap<TermInfo> {
&self.term_infos
}
/// Returns the document (or to be accurate, its stored field)
/// bearing the given doc id.
/// This method is slow and should seldom be called from
@@ -186,62 +185,62 @@ impl SegmentReader {
/// Returns the segment postings associated with the term, and with the given option,
/// or `None` if the term has never been encounterred and indexed.
///
/// If the field was not indexed with the indexing options that cover
/// or `None` if the term has never been encounterred and indexed.
///
/// If the field was not indexed with the indexing options that cover
/// the requested options, the returned `SegmentPostings` the method does not fail
/// and returns a `SegmentPostings` with as much information as possible.
///
/// For instance, requesting `SegmentPostingsOption::FreqAndPositions` for a `TextIndexingOptions`
/// that does not index position will return a `SegmentPostings` with `DocId`s and frequencies.
pub fn read_postings(&self, term: &Term, option: SegmentPostingsOption) -> Option<SegmentPostings> {
/// For instance, requesting `SegmentPostingsOption::FreqAndPositions` for a
/// `TextIndexingOptions` that does not index position will return a `SegmentPostings`
/// with `DocId`s and frequencies.
pub fn read_postings(&self,
term: &Term,
option: SegmentPostingsOption)
-> Option<SegmentPostings> {
let field = term.field();
let field_entry = self.schema.get_field_entry(field);
let term_info = get!(self.get_term_info(&term));
let term_info = get!(self.get_term_info(term));
let offset = term_info.postings_offset as usize;
let postings_data = &self.postings_data[offset..];
let freq_handler = match *field_entry.field_type() {
FieldType::Str(ref options) => {
let indexing_options = options.get_indexing_options();
match option {
SegmentPostingsOption::NoFreq => {
FreqHandler::new_without_freq()
}
SegmentPostingsOption::NoFreq => FreqHandler::new_without_freq(),
SegmentPostingsOption::Freq => {
if indexing_options.is_termfreq_enabled() {
FreqHandler::new_with_freq()
}
else {
} else {
FreqHandler::new_without_freq()
}
}
SegmentPostingsOption::FreqAndPositions => {
if indexing_options == TextIndexingOptions::TokenizedWithFreqAndPosition {
let offseted_position_data = &self.positions_data[term_info.positions_offset as usize ..];
let offset = term_info.positions_offset as usize;
let offseted_position_data = &self.positions_data[offset..];
FreqHandler::new_with_freq_and_position(offseted_position_data)
}
else if indexing_options.is_termfreq_enabled()
{
} else if indexing_options.is_termfreq_enabled() {
FreqHandler::new_with_freq()
}
else {
} else {
FreqHandler::new_without_freq()
}
}
}
}
_ => {
FreqHandler::new_without_freq()
}
_ => FreqHandler::new_without_freq(),
};
Some(SegmentPostings::from_data(term_info.doc_freq, postings_data, &self.delete_bitset, freq_handler))
Some(SegmentPostings::from_data(term_info.doc_freq,
postings_data,
&self.delete_bitset,
freq_handler))
}
/// Returns the posting list associated with a term.
///
/// If the term is not found, return None.
/// Even when non-null, because of deletes, the posting object
/// Even when non-null, because of deletes, the posting object
/// returned by this method may contain no documents.
pub fn read_postings_all_info(&self, term: &Term) -> Option<SegmentPostings> {
let field_entry = self.schema.get_field_entry(term.field());
@@ -249,15 +248,18 @@ impl SegmentReader {
FieldType::Str(ref text_options) => {
match text_options.get_indexing_options() {
TextIndexingOptions::TokenizedWithFreq => SegmentPostingsOption::Freq,
TextIndexingOptions::TokenizedWithFreqAndPosition => SegmentPostingsOption::FreqAndPositions,
TextIndexingOptions::TokenizedWithFreqAndPosition => {
SegmentPostingsOption::FreqAndPositions
}
_ => SegmentPostingsOption::NoFreq,
}
}
FieldType::U64(_) | FieldType::I64(_) => SegmentPostingsOption::NoFreq
FieldType::U64(_) |
FieldType::I64(_) => SegmentPostingsOption::NoFreq,
};
self.read_postings(term, segment_posting_option)
}
/// Returns the term info associated with the term.
pub fn get_term_info(&self, term: &Term) -> Option<TermInfo> {
self.term_infos.get(term.as_slice())

33
src/core/term_iterator.rs
View File

@@ -64,14 +64,15 @@ impl<'a> TermIterator<'a> {
loop {
match self.heap.peek() {
Some(&ref next_heap_it) if next_heap_it.term == self.current_term => {}
_ => { break; }
_ => {
break;
}
}
let next_heap_it = self.heap.pop().unwrap(); // safe : we peeked beforehand
self.current_segment_ords.push(next_heap_it.segment_ord);
}
true
}
else {
} else {
false
}
}
@@ -92,17 +93,18 @@ impl<'a> TermIterator<'a> {
/// This method may be called
/// iff advance() has been called before
/// and "true" was returned.
pub fn segment_ords(&self) -> &[usize]{
pub fn segment_ords(&self) -> &[usize] {
&self.current_segment_ords[..]
}
fn advance_segments(&mut self) {
for segment_ord in self.current_segment_ords.drain(..) {
if let Some(term) = self.key_streams[segment_ord].next() {
self.heap.push(HeapItem {
term: Term::from_bytes(term),
segment_ord: segment_ord,
});
self.heap
.push(HeapItem {
term: Term::from_bytes(term),
segment_ord: segment_ord,
});
}
}
}
@@ -114,8 +116,7 @@ impl<'a, 'f> Streamer<'a> for TermIterator<'f> {
fn next(&'a mut self) -> Option<Self::Item> {
if self.advance() {
Some(&self.current_term)
}
else {
} else {
None
}
}
@@ -123,12 +124,10 @@ impl<'a, 'f> Streamer<'a> for TermIterator<'f> {
impl<'a> From<&'a [SegmentReader]> for TermIterator<'a> {
fn from(segment_readers: &'a [SegmentReader]) -> TermIterator<'a> {
TermIterator::new(
segment_readers
.iter()
.map(|reader| reader.term_infos().keys())
.collect()
)
TermIterator::new(segment_readers
.iter()
.map(|reader| reader.term_infos().keys())
.collect())
}
}
@@ -180,4 +179,4 @@ mod tests {
assert_eq!(terms, "abcdef");
}
}
}

54
src/datastruct/fstmap.rs
View File

@@ -1,5 +1,3 @@
#![allow(should_implement_trait)]
use std::io;
use std::io::Write;
use fst;
@@ -20,18 +18,17 @@ pub struct FstMapBuilder<W: Write, V: BinarySerializable> {
}
impl<W: Write, V: BinarySerializable> FstMapBuilder<W, V> {
pub fn new(w: W) -> io::Result<FstMapBuilder<W, V>> {
let fst_builder = try!(fst::MapBuilder::new(w).map_err(convert_fst_error));
Ok(FstMapBuilder {
fst_builder: fst_builder,
data: Vec::new(),
_phantom_: PhantomData,
})
fst_builder: fst_builder,
data: Vec::new(),
_phantom_: PhantomData,
})
}
/// Horribly unsafe, nobody should ever do that... except me :)
///
///
/// If used, it must be used by systematically alternating calls
/// to insert_key and insert_value.
///
@@ -39,8 +36,8 @@ impl<W: Write, V: BinarySerializable> FstMapBuilder<W, V> {
/// in a nice way.
pub fn insert_key(&mut self, key: &[u8]) -> io::Result<()> {
try!(self.fst_builder
.insert(key, self.data.len() as u64)
.map_err(convert_fst_error));
.insert(key, self.data.len() as u64)
.map_err(convert_fst_error));
Ok(())
}
@@ -53,17 +50,14 @@ impl<W: Write, V: BinarySerializable> FstMapBuilder<W, V> {
#[cfg(test)]
pub fn insert(&mut self, key: &[u8], value: &V) -> io::Result<()> {
try!(self.fst_builder
.insert(key, self.data.len() as u64)
.map_err(convert_fst_error));
.insert(key, self.data.len() as u64)
.map_err(convert_fst_error));
try!(value.serialize(&mut self.data));
Ok(())
}
pub fn finish(self,) -> io::Result<W> {
let mut file = try!(
self.fst_builder
.into_inner()
.map_err(convert_fst_error));
pub fn finish(self) -> io::Result<W> {
let mut file = try!(self.fst_builder.into_inner().map_err(convert_fst_error));
let footer_size = self.data.len() as u32;
try!(file.write_all(&self.data));
try!((footer_size as u32).serialize(&mut file));
@@ -81,31 +75,35 @@ pub struct FstMap<V: BinarySerializable> {
fn open_fst_index(source: ReadOnlySource) -> io::Result<fst::Map> {
Ok(fst::Map::from(match source {
ReadOnlySource::Anonymous(data) => try!(Fst::from_shared_bytes(data.data, data.start, data.len).map_err(convert_fst_error)),
ReadOnlySource::Mmap(mmap_readonly) => try!(Fst::from_mmap(mmap_readonly).map_err(convert_fst_error)),
}))
ReadOnlySource::Anonymous(data) => {
try!(Fst::from_shared_bytes(data.data, data.start, data.len)
.map_err(convert_fst_error))
}
ReadOnlySource::Mmap(mmap_readonly) => {
try!(Fst::from_mmap(mmap_readonly).map_err(convert_fst_error))
}
}))
}
impl<V: BinarySerializable> FstMap<V> {
pub fn keys(&self,) -> fst::map::Keys {
pub fn keys(&self) -> fst::map::Keys {
self.fst_index.keys()
}
pub fn from_source(source: ReadOnlySource) -> io::Result<FstMap<V>> {
pub fn from_source(source: ReadOnlySource) -> io::Result<FstMap<V>> {
let total_len = source.len();
let length_offset = total_len - 4;
let mut split_len_buffer: &[u8] = &source.as_slice()[length_offset..];
let footer_size = try!(u32::deserialize(&mut split_len_buffer)) as usize;
let footer_size = try!(u32::deserialize(&mut split_len_buffer)) as usize;
let split_len = length_offset - footer_size;
let fst_source = source.slice(0, split_len);
let values_source = source.slice(split_len, length_offset);
let fst_index = try!(open_fst_index(fst_source));
Ok(FstMap {
fst_index: fst_index,
values_mmap: values_source,
_phantom_: PhantomData,
})
fst_index: fst_index,
values_mmap: values_source,
_phantom_: PhantomData,
})
}
fn read_value(&self, offset: u64) -> V {

4
src/datastruct/skip/mod.rs
View File

@@ -114,9 +114,9 @@ mod tests {
let mut skip_list: SkipList<()> = SkipList::from(output.as_slice());
assert_eq!(skip_list.next().unwrap(), (0, ()));
skip_list.seek(431);
assert_eq!(skip_list.next().unwrap(), (431,()) );
assert_eq!(skip_list.next().unwrap(), (431, ()));
skip_list.seek(1003);
assert_eq!(skip_list.next().unwrap(), (1004,()) );
assert_eq!(skip_list.next().unwrap(), (1004, ()));
assert_eq!(skip_list.next(), None);
}

55
src/datastruct/skip/skiplist.rs
View File

@@ -13,14 +13,12 @@ struct Layer<'a, T> {
}
impl<'a, T: BinarySerializable> Iterator for Layer<'a, T> {
type Item = (DocId, T);
fn next(&mut self,)-> Option<(DocId, T)> {
fn next(&mut self) -> Option<(DocId, T)> {
if self.next_id == u32::max_value() {
None
}
else {
} else {
let cur_val = T::deserialize(&mut self.cursor).unwrap();
let cur_id = self.next_id;
self.next_id = u32::deserialize(&mut self.cursor).unwrap_or(u32::max_value());
@@ -31,7 +29,7 @@ impl<'a, T: BinarySerializable> Iterator for Layer<'a, T> {
impl<'a, T: BinarySerializable> From<&'a [u8]> for Layer<'a, T> {
fn from(data: &'a [u8]) -> Layer<'a, T> {
let mut cursor = data;
let mut cursor = data;
let next_id = u32::deserialize(&mut cursor).unwrap_or(u32::max_value());
Layer {
data: data,
@@ -43,7 +41,6 @@ impl<'a, T: BinarySerializable> From<&'a [u8]> for Layer<'a, T> {
}
impl<'a, T: BinarySerializable> Layer<'a, T> {
fn empty() -> Layer<'a, T> {
Layer {
data: &EMPTY,
@@ -53,11 +50,11 @@ impl<'a, T: BinarySerializable> Layer<'a, T> {
}
}
fn seek_offset(&mut self, offset: usize) {
fn seek_offset(&mut self, offset: usize) {
self.cursor = &self.data[offset..];
self.next_id = u32::deserialize(&mut self.cursor).unwrap_or(u32::max_value());
}
// Returns the last element (key, val)
// such that (key < doc_id)
//
@@ -67,8 +64,12 @@ impl<'a, T: BinarySerializable> Layer<'a, T> {
let mut val = None;
while self.next_id < doc_id {
match self.next() {
None => { break; },
v => { val = v; }
None => {
break;
}
v => {
val = v;
}
}
}
val
@@ -82,16 +83,14 @@ pub struct SkipList<'a, T: BinarySerializable> {
}
impl<'a, T: BinarySerializable> Iterator for SkipList<'a, T> {
type Item = (DocId, T);
fn next(&mut self,)-> Option<(DocId, T)> {
fn next(&mut self) -> Option<(DocId, T)> {
self.data_layer.next()
}
}
impl<'a, T: BinarySerializable> SkipList<'a, T> {
pub fn seek(&mut self, doc_id: DocId) -> Option<(DocId, T)> {
let mut next_layer_skip: Option<(DocId, u32)> = None;
for skip_layer in &mut self.skip_layers {
@@ -99,39 +98,33 @@ impl<'a, T: BinarySerializable> SkipList<'a, T> {
skip_layer.seek_offset(offset as usize);
}
next_layer_skip = skip_layer.seek(doc_id);
}
if let Some((_, offset)) = next_layer_skip {
self.data_layer.seek_offset(offset as usize);
}
self.data_layer.seek(doc_id)
}
if let Some((_, offset)) = next_layer_skip {
self.data_layer.seek_offset(offset as usize);
}
self.data_layer.seek(doc_id)
}
}
impl<'a, T: BinarySerializable> From<&'a [u8]> for SkipList<'a, T> {
fn from(mut data: &'a [u8]) -> SkipList<'a, T> {
let offsets: Vec<u32> = Vec::deserialize(&mut data).unwrap();
let num_layers = offsets.len();
let layers_data: &[u8] = data;
let data_layer: Layer<'a, T> =
if num_layers == 0 { Layer::empty() }
else {
let first_layer_data: &[u8] = &layers_data[..offsets[0] as usize];
Layer::from(first_layer_data)
};
let data_layer: Layer<'a, T> = if num_layers == 0 {
Layer::empty()
} else {
let first_layer_data: &[u8] = &layers_data[..offsets[0] as usize];
Layer::from(first_layer_data)
};
let skip_layers = (0..max(1, num_layers) - 1)
.map(|i| (offsets[i] as usize, offsets[i + 1] as usize))
.map(|(start, stop)| {
Layer::from(&layers_data[start..stop])
})
.map(|(start, stop)| Layer::from(&layers_data[start..stop]))
.collect();
SkipList {
skip_layers: skip_layers,
data_layer: data_layer,
}
}
}

21
src/datastruct/skip/skiplist_builder.rs
View File

@@ -13,8 +13,7 @@ struct LayerBuilder<T: BinarySerializable> {
}
impl<T: BinarySerializable> LayerBuilder<T> {
fn written_size(&self,) -> usize {
fn written_size(&self) -> usize {
self.buffer.len()
}
@@ -42,8 +41,9 @@ impl<T: BinarySerializable> LayerBuilder<T> {
Ok(if self.remaining == 0 {
self.remaining = self.period;
Some((doc_id, offset))
}
else { None })
} else {
None
})
}
}
@@ -56,7 +56,6 @@ pub struct SkipListBuilder<T: BinarySerializable> {
impl<T: BinarySerializable> SkipListBuilder<T> {
pub fn new(period: usize) -> SkipListBuilder<T> {
SkipListBuilder {
period: period,
@@ -78,11 +77,13 @@ impl<T: BinarySerializable> SkipListBuilder<T> {
let mut skip_pointer = try!(self.data_layer.insert(doc_id, dest));
loop {
skip_pointer = match skip_pointer {
Some((skip_doc_id, skip_offset)) =>
try!(self
.get_skip_layer(layer_id)
.insert(skip_doc_id, &skip_offset)),
None => { return Ok(()); }
Some((skip_doc_id, skip_offset)) => {
try!(self.get_skip_layer(layer_id)
.insert(skip_doc_id, &skip_offset))
}
None => {
return Ok(());
}
};
layer_id += 1;
}

25
src/datastruct/stacker/expull.rs
View File

@@ -9,7 +9,7 @@ pub fn is_power_of_2(val: u32) -> bool {
#[inline]
pub fn jump_needed(val: u32) -> bool {
val > 3 && is_power_of_2(val)
val > 3 && is_power_of_2(val)
}
@@ -24,7 +24,6 @@ pub struct ExpUnrolledLinkedList {
}
impl ExpUnrolledLinkedList {
pub fn iter<'a>(&self, addr: u32, heap: &'a Heap) -> ExpUnrolledLinkedListIterator<'a> {
ExpUnrolledLinkedListIterator {
heap: heap,
@@ -42,10 +41,10 @@ impl ExpUnrolledLinkedList {
// the next block as a size of (length so far),
// and we need to add 1u32 to store the pointer
// to the next element.
let new_block_size: usize = (self.len as usize + 1) * mem::size_of::<u32>();
let new_block_size: usize = (self.len as usize + 1) * mem::size_of::<u32>();
let new_block_addr: u32 = heap.allocate_space(new_block_size);
heap.set(self.end, &new_block_addr);
self.end = new_block_addr;
self.end = new_block_addr;
}
heap.set(self.end, &val);
self.end += mem::size_of::<u32>() as u32;
@@ -77,23 +76,21 @@ pub struct ExpUnrolledLinkedListIterator<'a> {
impl<'a> Iterator for ExpUnrolledLinkedListIterator<'a> {
type Item = u32;
fn next(&mut self,) -> Option<u32> {
fn next(&mut self) -> Option<u32> {
if self.consumed == self.len {
None
}
else {
} else {
let addr: u32;
self.consumed += 1;
if jump_needed(self.consumed) {
addr = *self.heap.get_mut_ref(self.addr);
}
else {
} else {
addr = self.addr;
}
self.addr = addr + mem::size_of::<u32>() as u32;
Some(*self.heap.get_mut_ref(addr))
}
}
}
}
@@ -103,7 +100,7 @@ impl<'a> Iterator for ExpUnrolledLinkedListIterator<'a> {
#[cfg(test)]
mod tests {
use super::*;
use super::super::heap::Heap;
use test::Bencher;
@@ -147,7 +144,7 @@ mod tests {
#[bench]
fn bench_push_stack(bench: &mut Bencher) {
let heap = Heap::with_capacity(64_000_000);
let heap = Heap::with_capacity(64_000_000);
bench.iter(|| {
let mut stacks = Vec::with_capacity(100);
for _ in 0..NUM_STACK {
@@ -163,4 +160,4 @@ mod tests {
heap.clear();
});
}
}
}

38
src/datastruct/stacker/hashmap.rs
View File

@@ -3,7 +3,7 @@ use super::heap::{Heap, HeapAllocable, BytesRef};
/// dbj2 hash function
fn djb2(key: &[u8]) -> u64 {
let mut state: u64 = 5381;
let mut state: u64 = 5381;
for &b in key {
state = (state << 5).wrapping_add(state).wrapping_add(b as u64);
}
@@ -25,7 +25,7 @@ impl Default for BytesRef {
///
/// The key and the value are actually stored contiguously.
/// For this reason, the (start, stop) information is actually redundant
/// and can be simplified in the future
/// and can be simplified in the future
#[derive(Copy, Clone, Default)]
struct KeyValue {
key: BytesRef,
@@ -33,7 +33,7 @@ struct KeyValue {
}
impl KeyValue {
fn is_empty(&self,) -> bool {
fn is_empty(&self) -> bool {
self.key.stop == 0u32
}
}
@@ -45,7 +45,7 @@ pub enum Entry {
/// Customized `HashMap` with string keys
///
///
/// This `HashMap` takes String as keys. Keys are
/// stored in a user defined heap.
///
@@ -60,6 +60,7 @@ pub struct HashMap<'a> {
occupied: Vec<usize>,
}
struct QuadraticProbing {
hash: usize,
i: usize,
@@ -77,7 +78,7 @@ impl QuadraticProbing {
}
#[inline]
fn next(&mut self) -> usize {
fn next_probe(&mut self) -> usize {
self.i += 1;
(self.hash + self.i * self.i) & self.mask
}
@@ -88,9 +89,7 @@ impl<'a> HashMap<'a> {
pub fn new(num_bucket_power_of_2: usize, heap: &'a Heap) -> HashMap<'a> {
let table_size = 1 << num_bucket_power_of_2;
let table: Vec<KeyValue> = iter::repeat(KeyValue::default())
.take(table_size)
.collect();
let table: Vec<KeyValue> = iter::repeat(KeyValue::default()).take(table_size).collect();
HashMap {
table: table.into_boxed_slice(),
heap: heap,
@@ -136,9 +135,7 @@ impl<'a> HashMap<'a> {
pub fn get_or_create<S: AsRef<[u8]>, V: HeapAllocable>(&mut self, key: S) -> &mut V {
let entry = self.lookup(key.as_ref());
match entry {
Entry::Occupied(addr) => {
self.heap.get_mut_ref(addr)
}
Entry::Occupied(addr) => self.heap.get_mut_ref(addr),
Entry::Vacant(bucket) => {
let (addr, val): (u32, &mut V) = self.heap.allocate_object();
self.set_bucket(key.as_ref(), bucket, addr);
@@ -146,12 +143,12 @@ impl<'a> HashMap<'a> {
}
}
}
pub fn lookup<S: AsRef<[u8]>>(&self, key: S) -> Entry {
let key_bytes: &[u8] = key.as_ref();
let mut probe = self.probe(key_bytes);
loop {
let bucket = probe.next();
let bucket = probe.next_probe();
let kv: KeyValue = self.table[bucket];
if kv.is_empty() {
return Entry::Vacant(bucket);
@@ -166,7 +163,7 @@ impl<'a> HashMap<'a> {
#[cfg(test)]
mod tests {
use super::*;
use super::super::heap::{Heap, HeapAllocable};
use super::djb2;
@@ -227,20 +224,17 @@ mod tests {
#[bench]
fn bench_djb2(bench: &mut Bencher) {
let v = String::from("abwer");
bench.iter(|| {
djb2(v.as_bytes())
});
bench.iter(|| djb2(v.as_bytes()));
}
#[bench]
fn bench_siphasher(bench: &mut Bencher) {
let v = String::from("abwer");
bench.iter(|| {
let mut h = DefaultHasher::new();
h.write(v.as_bytes());
h.finish()
});
let mut h = DefaultHasher::new();
h.write(v.as_bytes());
h.finish()
});
}
}

55
src/datastruct/stacker/heap.rs
View File

@@ -19,20 +19,17 @@ pub struct Heap {
inner: UnsafeCell<InnerHeap>,
}
#[cfg_attr(feature = "cargo-clippy", allow(mut_from_ref))]
impl Heap {
/// Creates a new heap with a given capacity
pub fn with_capacity(num_bytes: usize) -> Heap {
Heap {
inner: UnsafeCell::new(
InnerHeap::with_capacity(num_bytes)
),
}
Heap { inner: UnsafeCell::new(InnerHeap::with_capacity(num_bytes)) }
}
fn inner(&self,) -> &mut InnerHeap {
unsafe { &mut *self.inner.get() }
fn inner(&self) -> &mut InnerHeap {
unsafe { &mut *self.inner.get() }
}
/// Clears the heap. All the underlying data is lost.
///
/// This heap does not support deallocation.
@@ -40,14 +37,14 @@ impl Heap {
pub fn clear(&self) {
self.inner().clear();
}
/// Return the heap capacity.
pub fn capacity(&self,) -> u32 {
pub fn capacity(&self) -> u32 {
self.inner().capacity()
}
/// Return amount of free space, in bytes.
pub fn num_free_bytes(&self,) -> u32 {
pub fn num_free_bytes(&self) -> u32 {
self.inner().num_free_bytes()
}
@@ -56,40 +53,40 @@ impl Heap {
pub fn allocate_space(&self, num_bytes: usize) -> u32 {
self.inner().allocate_space(num_bytes)
}
/// Allocate an object in the heap
pub fn allocate_object<V: HeapAllocable>(&self,) -> (u32, &mut V) {
pub fn allocate_object<V: HeapAllocable>(&self) -> (u32, &mut V) {
let addr = self.inner().allocate_space(mem::size_of::<V>());
let v: V = V::with_addr(addr);
self.inner().set(addr, &v);
(addr, self.inner().get_mut_ref(addr))
}
/// Stores a `&[u8]` in the heap and returns the destination BytesRef.
pub fn allocate_and_set(&self, data: &[u8]) -> BytesRef {
self.inner().allocate_and_set(data)
}
/// Fetches the `&[u8]` stored on the slice defined by the `BytesRef`
/// given as argumetn
pub fn get_slice(&self, bytes_ref: BytesRef) -> &[u8] {
self.inner().get_slice(bytes_ref.start, bytes_ref.stop)
}
/// Stores an item's data in the heap, at the given `address`.
pub fn set<Item>(&self, addr: u32, val: &Item) {
self.inner().set(addr, val);
}
/// Returns a reference to an `Item` at a given `addr`.
#[cfg(test)]
pub fn get_ref<Item>(&self, addr: u32) -> &Item {
/// Returns a mutable reference for an object at a given Item.
pub fn get_mut_ref<Item>(&self, addr: u32) -> &mut Item {
self.inner().get_mut_ref(addr)
}
/// Returns a mutable reference to an `Item` at a given `addr`.
pub fn get_mut_ref<Item>(&self, addr: u32) -> &mut Item {
self.inner().get_mut_ref(addr)
#[cfg(test)]
pub fn get_ref<Item>(&self, addr: u32) -> &mut Item {
self.get_mut_ref(addr)
}
}
@@ -102,7 +99,6 @@ struct InnerHeap {
impl InnerHeap {
pub fn with_capacity(num_bytes: usize) -> InnerHeap {
let buffer: Vec<u8> = vec![0u8; num_bytes];
InnerHeap {
@@ -116,7 +112,7 @@ impl InnerHeap {
self.used = 0u32;
}
pub fn capacity(&self,) -> u32 {
pub fn capacity(&self) -> u32 {
self.buffer.len() as u32
}
@@ -125,10 +121,9 @@ impl InnerHeap {
pub fn num_free_bytes(&self,) -> u32 {
if self.has_been_resized {
0u32
}
else {
} else {
(self.buffer.len() as u32) - self.used
}
}
}
pub fn allocate_space(&mut self, num_bytes: usize) -> u32 {
@@ -141,11 +136,11 @@ impl InnerHeap {
}
addr
}
fn get_slice(&self, start: u32, stop: u32) -> &[u8] {
&self.buffer[start as usize..stop as usize]
}
fn get_mut_slice(&mut self, start: u32, stop: u32) -> &mut [u8] {
&mut self.buffer[start as usize..stop as usize]
}
@@ -180,4 +175,4 @@ impl InnerHeap {
ptr::copy(v_ptr_u8, dest_ptr, mem::size_of::<Item>());
}
}
}
}

8
src/datastruct/stacker/mod.rs
View File

@@ -22,7 +22,7 @@ fn test_unrolled_linked_list() {
for j in 0..k {
for i in 0..500 {
let mut list: &mut ExpUnrolledLinkedList = hashmap.get_or_create(i.to_string());
list.push(i*j, &heap);
list.push(i * j, &heap);
}
}
for i in 0..500 {
@@ -31,7 +31,7 @@ fn test_unrolled_linked_list() {
let v: &mut ExpUnrolledLinkedList = heap.get_mut_ref(addr);
let mut it = v.iter(addr, &heap);
for j in 0..k {
assert_eq!(it.next().unwrap(), i*j);
assert_eq!(it.next().unwrap(), i * j);
}
assert!(!it.next().is_some());
}
@@ -41,6 +41,6 @@ fn test_unrolled_linked_list() {
}
}
}
}
}
}

41
src/directory/directory.rs
View File

@@ -8,31 +8,30 @@ use std::io;
use std::marker::Sync;
/// Write-once read many (WORM) abstraction for where
/// tantivy's data should be stored.
/// tantivy's data should be stored.
///
/// There are currently two implementations of `Directory`
///
///
/// - The [`MMapDirectory`](struct.MmapDirectory.html), this
/// should be your default choice.
/// - The [`RAMDirectory`](struct.RAMDirectory.html), which
/// should be your default choice.
/// - The [`RAMDirectory`](struct.RAMDirectory.html), which
/// should be used mostly for tests.
///
///
pub trait Directory: fmt::Debug + Send + Sync + 'static {
/// Opens a virtual file for read.
///
///
/// Once a virtual file is open, its data may not
/// change.
///
/// Specifically, subsequent writes or flushes should
/// have no effect on the returned `ReadOnlySource` object.
/// have no effect on the returned `ReadOnlySource` object.
fn open_read(&self, path: &Path) -> result::Result<ReadOnlySource, OpenReadError>;
/// Removes a file
///
/// Removing a file will not affect an eventual
/// existing ReadOnlySource pointing to it.
///
///
/// Removing a nonexistent file, yields a
/// `DeleteError::DoesNotExist`.
fn delete(&self, path: &Path) -> result::Result<(), DeleteError>;
@@ -40,18 +39,18 @@ pub trait Directory: fmt::Debug + Send + Sync + 'static {
/// Returns true iff the file exists
fn exists(&self, path: &Path) -> bool;
/// Opens a writer for the *virtual file* associated with
/// Opens a writer for the *virtual file* associated with
/// a Path.
///
/// Right after this call, the file should be created
/// and any subsequent call to `open_read` for the
/// and any subsequent call to `open_read` for the
/// same path should return a `ReadOnlySource`.
///
///
/// Write operations may be aggressively buffered.
/// The client of this trait is responsible for calling flush
/// to ensure that subsequent `read` operations
/// to ensure that subsequent `read` operations
/// will take into account preceding `write` operations.
///
///
/// Flush operation should also be persistent.
///
/// The user shall not rely on `Drop` triggering `flush`.
@@ -60,7 +59,7 @@ pub trait Directory: fmt::Debug + Send + Sync + 'static {
///
/// The file may not previously exist.
fn open_write(&mut self, path: &Path) -> Result<WritePtr, OpenWriteError>;
/// Reads the full content file that has been written using
/// atomic_write.
///
@@ -68,17 +67,13 @@ pub trait Directory: fmt::Debug + Send + Sync + 'static {
fn atomic_read(&self, path: &Path) -> Result<Vec<u8>, OpenReadError>;
/// Atomically replace the content of a file with data.
///
///
/// This calls ensure that reads can never *observe*
/// a partially written file.
///
///
/// The file may or may not previously exist.
fn atomic_write(&mut self, path: &Path, data: &[u8]) -> io::Result<()>;
/// Clones the directory and boxes the clone
/// Clones the directory and boxes the clone
fn box_clone(&self) -> Box<Directory>;
}

12
src/directory/error.rs
View File

@@ -4,7 +4,7 @@ use std::io;
/// Error that may occur when opening a directory
#[derive(Debug)]
pub enum OpenDirectoryError {
/// The underlying directory does not exists.
/// The underlying directory does not exists.
DoesNotExist(PathBuf),
/// The path exists but is not a directory.
NotADirectory(PathBuf),
@@ -14,9 +14,9 @@ pub enum OpenDirectoryError {
#[derive(Debug)]
pub enum OpenWriteError {
/// Our directory is WORM, writing an existing file is forbidden.
/// Checkout the `Directory` documentation.
/// Checkout the `Directory` documentation.
FileAlreadyExists(PathBuf),
/// Any kind of IO error that happens when
/// Any kind of IO error that happens when
/// writing in the underlying IO device.
IOError(io::Error),
}
@@ -32,7 +32,7 @@ impl From<io::Error> for OpenWriteError {
pub enum OpenReadError {
/// The file does not exists.
FileDoesNotExist(PathBuf),
/// Any kind of IO error that happens when
/// Any kind of IO error that happens when
/// interacting with the underlying IO device.
IOError(io::Error),
}
@@ -43,10 +43,10 @@ pub enum OpenReadError {
pub enum DeleteError {
/// The file does not exists.
FileDoesNotExist(PathBuf),
/// Any kind of IO error that happens when
/// Any kind of IO error that happens when
/// interacting with the underlying IO device.
IOError(io::Error),
/// The file may not be deleted because it is
/// The file may not be deleted because it is
/// protected.
FileProtected(PathBuf),
}

166
src/directory/managed_directory.rs
View File

@@ -7,6 +7,7 @@ use std::io;
use Directory;
use std::sync::{Arc, RwLock};
use std::collections::HashSet;
use std::sync::RwLockWriteGuard;
use std::io::Write;
use core::MANAGED_FILEPATH;
use std::collections::HashMap;
@@ -17,7 +18,7 @@ use Error;
/// Wrapper of directories that keeps track of files created by Tantivy.
///
/// A managed directory is just a wrapper of a directory
/// that keeps a (persisted) list of the files that
/// that keeps a (persisted) list of the files that
/// have been created (and not deleted) by tantivy so far.
///
/// Thanks to this list, it implements a `garbage_collect` method
@@ -45,19 +46,18 @@ pub struct FileProtection {
}
fn unprotect_file_from_delete(directory: &ManagedDirectory, path: &Path) {
let mut meta_informations_wlock = directory.meta_informations
let mut meta_informations_wlock = directory
.meta_informations
.write()
.expect("Managed file lock poisoned");
if let Some(counter_ref_mut) = meta_informations_wlock
.protected_files
.get_mut(path) {
if let Some(counter_ref_mut) = meta_informations_wlock.protected_files.get_mut(path) {
(*counter_ref_mut) -= 1;
}
}
impl fmt::Debug for FileProtection {
fn fmt(&self, formatter: &mut fmt::Formatter) -> result::Result<(), fmt::Error> {
write!(formatter, "FileProtectionFor({:?})", self.path)
write!(formatter, "FileProtectionFor({:?})", self.path)
}
}
@@ -67,33 +67,42 @@ impl Drop for FileProtection {
}
}
impl ManagedDirectory {
/// Saves the file containing the list of existing files
/// that were created by tantivy.
fn save_managed_paths(directory: &mut Directory,
wlock: &RwLockWriteGuard<MetaInformation>)
-> io::Result<()> {
let mut w = serde_json::to_vec(&wlock.managed_paths)?;
write!(&mut w, "\n")?;
directory.atomic_write(&MANAGED_FILEPATH, &w[..])?;
Ok(())
}
impl ManagedDirectory {
/// Wraps a directory as managed directory.
pub fn new<Dir: Directory>(directory: Dir) -> Result<ManagedDirectory> {
match directory.atomic_read(&MANAGED_FILEPATH) {
Ok(data) => {
let managed_files_json = String::from_utf8_lossy(&data);
let managed_files: HashSet<PathBuf> = serde_json::from_str(&managed_files_json)
.map_err(|e| Error::CorruptedFile(MANAGED_FILEPATH.clone(), Box::new(e)))?;
let managed_files: HashSet<PathBuf> =
serde_json::from_str(&managed_files_json)
.map_err(|e| Error::CorruptedFile(MANAGED_FILEPATH.clone(), Box::new(e)))?;
Ok(ManagedDirectory {
directory: box directory,
meta_informations: Arc::new(RwLock::new(
MetaInformation {
managed_paths: managed_files,
protected_files: HashMap::default()
})),
})
directory: box directory,
meta_informations: Arc::new(RwLock::new(MetaInformation {
managed_paths: managed_files,
protected_files:
HashMap::default(),
})),
})
}
Err(OpenReadError::FileDoesNotExist(_)) => {
Ok(ManagedDirectory {
directory: box directory,
meta_informations: Arc::default(),
})
}
Err(OpenReadError::IOError(e)) => {
Err(From::from(e))
directory: box directory,
meta_informations: Arc::default(),
})
}
Err(OpenReadError::IOError(e)) => Err(From::from(e)),
}
}
@@ -101,7 +110,7 @@ impl ManagedDirectory {
///
/// Removes the files that were created by `tantivy` and are not
/// used by any segment anymore.
///
///
/// * `living_files` - List of files that are still used by the index.
///
/// This method does not panick nor returns errors.
@@ -109,19 +118,21 @@ impl ManagedDirectory {
/// an error is simply logged, and the file remains in the list of managed
/// files.
pub fn garbage_collect(&mut self, living_files: HashSet<PathBuf>) {
let mut files_to_delete = vec!();
{ // releasing the lock as .delete() will use it too.
let meta_informations_rlock = self.meta_informations
.read()
.expect("Managed directory rlock poisoned in garbage collect.");
let mut files_to_delete = vec![];
{
// releasing the lock as .delete() will use it too.
let meta_informations_rlock =
self.meta_informations
.read()
.expect("Managed directory rlock poisoned in garbage collect.");
for managed_path in &meta_informations_rlock.managed_paths {
if !living_files.contains(managed_path) {
files_to_delete.push(managed_path.clone());
}
}
}
let mut deleted_files = vec!();
let mut deleted_files = vec![];
{
for file_to_delete in files_to_delete {
match self.delete(&file_to_delete) {
@@ -130,13 +141,14 @@ impl ManagedDirectory {
deleted_files.push(file_to_delete);
}
Err(file_error) => {
error!("Failed to delete {:?}", file_to_delete);
match file_error {
DeleteError::FileDoesNotExist(_) => {
deleted_files.push(file_to_delete);
}
DeleteError::IOError(_) => {
if !cfg!(target_os = "windows") {
// On windows, delete is expected to fail if the file
// is mmapped.
error!("Failed to delete {:?}", file_to_delete);
}
}
@@ -144,7 +156,7 @@ impl ManagedDirectory {
// this is expected.
}
}
}
}
}
@@ -152,18 +164,18 @@ impl ManagedDirectory {
if !deleted_files.is_empty() {
// update the list of managed files by removing
// update the list of managed files by removing
// the file that were removed.
let mut meta_informations_wlock = self.meta_informations
.write()
.expect("Managed directory wlock poisoned (2).");
{
let mut meta_informations_wlock = self.meta_informations
.write()
.expect("Managed directory wlock poisoned (2).");
let managed_paths_write = &mut meta_informations_wlock.managed_paths;
for delete_file in &deleted_files {
managed_paths_write.remove(delete_file);
}
}
if let Err(_) = self.save_managed_paths() {
if save_managed_paths(self.directory.as_mut(), &meta_informations_wlock).is_err() {
error!("Failed to save the list of managed files.");
}
}
@@ -175,7 +187,7 @@ impl ManagedDirectory {
///
/// The method returns a `FileProtection` object.
/// The file will not be garbage collected as long as the
/// `FileProtection` object is kept alive.
/// `FileProtection` object is kept alive.
pub fn protect_file_from_delete(&self, path: &Path) -> FileProtection {
let pathbuf = path.to_owned();
{
@@ -183,9 +195,9 @@ impl ManagedDirectory {
.write()
.expect("Managed file lock poisoned on protect");
*meta_informations_wlock
.protected_files
.entry(pathbuf.clone())
.or_insert(0) += 1;
.protected_files
.entry(pathbuf.clone())
.or_insert(0) += 1;
}
FileProtection {
directory: self.clone(),
@@ -193,46 +205,26 @@ impl ManagedDirectory {
}
}
/// Saves the file containing the list of existing files
/// that were created by tantivy.
fn save_managed_paths(&mut self,) -> io::Result<()> {
let managed_paths;
{
let meta_informations_rlock = self.meta_informations
.read()
.expect("Managed file lock poisoned");
managed_paths = meta_informations_rlock.managed_paths.clone();
}
let mut w = try!(serde_json::to_vec(&managed_paths));
try!(write!(&mut w, "\n"));
self.directory.atomic_write(&MANAGED_FILEPATH, &w[..])?;
Ok(())
}
/// Registers a file as managed
///
/// This method must be called before the file is
///
/// This method must be called before the file is
/// actually created to ensure that a failure between
/// registering the filepath and creating the file
/// will not lead to garbage files that will
/// will not lead to garbage files that will
/// never get removed.
fn register_file_as_managed(&mut self, filepath: &Path) -> io::Result<()> {
let has_changed = {
let mut meta_wlock = self.meta_informations
.write()
.expect("Managed file lock poisoned");
meta_wlock.managed_paths.insert(filepath.to_owned())
};
let mut meta_wlock = self.meta_informations
.write()
.expect("Managed file lock poisoned");
let has_changed = meta_wlock.managed_paths.insert(filepath.to_owned());
if has_changed {
self.save_managed_paths()?;
save_managed_paths(self.directory.as_mut(), &meta_wlock)?;
}
Ok(())
}
}
impl Directory for ManagedDirectory {
fn open_read(&self, path: &Path) -> result::Result<ReadOnlySource, OpenReadError> {
self.directory.open_read(path)
}
@@ -258,7 +250,7 @@ impl Directory for ManagedDirectory {
.expect("poisoned lock in managed directory meta");
if let Some(counter) = metas_rlock.protected_files.get(path) {
if *counter > 0 {
return Err(DeleteError::FileProtected(path.to_owned()))
return Err(DeleteError::FileProtected(path.to_owned()));
}
}
}
@@ -268,11 +260,10 @@ impl Directory for ManagedDirectory {
fn exists(&self, path: &Path) -> bool {
self.directory.exists(path)
}
fn box_clone(&self) -> Box<Directory> {
box self.clone()
}
}
impl Clone for ManagedDirectory {
@@ -292,10 +283,10 @@ mod tests {
use super::*;
use directory::MmapDirectory;
use std::path::Path;
use std::path::Path;
use std::io::Write;
use tempdir::TempDir;
lazy_static! {
static ref TEST_PATH1: &'static Path = Path::new("some_path_for_test");
static ref TEST_PATH2: &'static Path = Path::new("some_path_for_test2");
@@ -313,17 +304,17 @@ mod tests {
write_file.flush().unwrap();
}
{
managed_directory.atomic_write(*TEST_PATH2, &vec!(0u8,1u8)).unwrap();
managed_directory
.atomic_write(*TEST_PATH2, &vec![0u8, 1u8])
.unwrap();
}
{
assert!(managed_directory.exists(*TEST_PATH1));
assert!(managed_directory.exists(*TEST_PATH2));
}
{
let living_files: HashSet<PathBuf> = [TEST_PATH1.to_owned()]
.into_iter()
.cloned()
.collect();
let living_files: HashSet<PathBuf> =
[TEST_PATH1.to_owned()].into_iter().cloned().collect();
managed_directory.garbage_collect(living_files);
}
{
@@ -346,7 +337,7 @@ mod tests {
assert!(!managed_directory.exists(*TEST_PATH1));
assert!(!managed_directory.exists(*TEST_PATH2));
}
}
}
}
#[test]
@@ -357,10 +348,12 @@ mod tests {
let mmap_directory = MmapDirectory::open(&tempdir_path).unwrap();
let mut managed_directory = ManagedDirectory::new(mmap_directory).unwrap();
managed_directory.atomic_write(*TEST_PATH1, &vec!(0u8,1u8)).unwrap();
managed_directory
.atomic_write(*TEST_PATH1, &vec![0u8, 1u8])
.unwrap();
assert!(managed_directory.exists(*TEST_PATH1));
let _mmap_read = managed_directory.open_read(*TEST_PATH1).unwrap();
let _mmap_read = managed_directory.open_read(*TEST_PATH1).unwrap();
managed_directory.garbage_collect(living_files.clone());
if cfg!(target_os = "windows") {
// On Windows, gc should try and fail the file as it is mmapped.
@@ -371,8 +364,7 @@ mod tests {
// eventually be deleted once mmap is released.
managed_directory.garbage_collect(living_files);
assert!(!managed_directory.exists(*TEST_PATH1));
}
else {
} else {
assert!(!managed_directory.exists(*TEST_PATH1));
}
@@ -387,7 +379,9 @@ mod tests {
let mmap_directory = MmapDirectory::open(&tempdir_path).unwrap();
let mut managed_directory = ManagedDirectory::new(mmap_directory).unwrap();
managed_directory.atomic_write(*TEST_PATH1, &vec!(0u8,1u8)).unwrap();
managed_directory
.atomic_write(*TEST_PATH1, &vec![0u8, 1u8])
.unwrap();
assert!(managed_directory.exists(*TEST_PATH1));
{
@@ -398,7 +392,7 @@ mod tests {
managed_directory.garbage_collect(living_files.clone());
assert!(!managed_directory.exists(*TEST_PATH1));
}

225
src/directory/mmap_directory.rs
View File

@@ -24,33 +24,24 @@ use std::sync::Weak;
use tempdir::TempDir;
fn open_mmap(full_path: &PathBuf) -> result::Result<Option<Arc<Mmap>>, OpenReadError> {
let convert_file_error = |err: io::Error| {
if err.kind() == io::ErrorKind::NotFound {
OpenReadError::FileDoesNotExist(full_path.clone())
}
else {
OpenReadError::IOError(err)
}
let convert_file_error = |err: io::Error| if err.kind() == io::ErrorKind::NotFound {
OpenReadError::FileDoesNotExist(full_path.clone())
} else {
OpenReadError::IOError(err)
};
let file = File::open(&full_path).map_err(convert_file_error)?;
let meta_data = file
.metadata()
.map_err(|e| OpenReadError::IOError(e))?;
let meta_data = file.metadata().map_err(OpenReadError::IOError)?;
if meta_data.len() == 0 {
// if the file size is 0, it will not be possible
// if the file size is 0, it will not be possible
// to mmap the file, so we return an anonymous mmap_cache
// instead.
return Ok(None)
return Ok(None);
}
match Mmap::open(&file, Protection::Read) {
Ok(mmap) => {
Ok(Some(Arc::new(mmap)))
}
Err(e) => {
Err(OpenReadError::IOError(e))
}
Ok(mmap) => Ok(Some(Arc::new(mmap))),
Err(e) => Err(OpenReadError::IOError(e)),
}
}
#[derive(Default,Clone,Debug,Serialize,Deserialize)]
@@ -91,8 +82,7 @@ impl Default for MmapCache {
impl MmapCache {
fn cleanup(&mut self) {
fn cleanup(&mut self) {
let previous_cache_size = self.cache.len();
let mut new_cache = HashMap::new();
mem::swap(&mut new_cache, &mut self.cache);
@@ -107,9 +97,7 @@ impl MmapCache {
fn get_info(&mut self) -> CacheInfo {
self.cleanup();
let paths: Vec<PathBuf> = self.cache.keys()
.cloned()
.collect();
let paths: Vec<PathBuf> = self.cache.keys().cloned().collect();
CacheInfo {
counters: self.counters.clone(),
mmapped: paths,
@@ -123,68 +111,63 @@ impl MmapCache {
self.cleanup();
}
Ok(match self.cache.entry(full_path.clone()) {
HashMapEntry::Occupied(mut occupied_entry) => {
if let Some(mmap_arc) = occupied_entry.get().upgrade() {
self.counters.hit += 1;
Some(mmap_arc.clone())
}
else {
// The entry exists but the weak ref has been destroyed.
self.counters.miss_weak += 1;
if let Some(mmap_arc) = open_mmap(&full_path)? {
occupied_entry.insert(Arc::downgrade(&mmap_arc));
Some(mmap_arc)
}
else {
None
}
}
}
HashMapEntry::Vacant(vacant_entry) => {
self.counters.miss_empty += 1;
if let Some(mmap_arc) = open_mmap(&full_path)? {
vacant_entry.insert(Arc::downgrade(&mmap_arc));
Some(mmap_arc)
}
else {
None
}
}
})
HashMapEntry::Occupied(mut occupied_entry) => {
if let Some(mmap_arc) = occupied_entry.get().upgrade() {
self.counters.hit += 1;
Some(mmap_arc.clone())
} else {
// The entry exists but the weak ref has been destroyed.
self.counters.miss_weak += 1;
if let Some(mmap_arc) = open_mmap(&full_path)? {
occupied_entry.insert(Arc::downgrade(&mmap_arc));
Some(mmap_arc)
} else {
None
}
}
}
HashMapEntry::Vacant(vacant_entry) => {
self.counters.miss_empty += 1;
if let Some(mmap_arc) = open_mmap(&full_path)? {
vacant_entry.insert(Arc::downgrade(&mmap_arc));
Some(mmap_arc)
} else {
None
}
}
})
}
}
/// Directory storing data in files, read via mmap.
///
/// The Mmap object are cached to limit the
/// system calls.
/// The Mmap object are cached to limit the
/// system calls.
#[derive(Clone)]
pub struct MmapDirectory {
root_path: PathBuf,
mmap_cache: Arc<RwLock<MmapCache>>,
_temp_directory: Arc<Option<TempDir>>,
}
impl fmt::Debug for MmapDirectory {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
write!(f, "MmapDirectory({:?})", self.root_path)
}
write!(f, "MmapDirectory({:?})", self.root_path)
}
}
impl MmapDirectory {
/// Creates a new MmapDirectory in a temporary directory.
///
/// This is mostly useful to test the MmapDirectory itself.
/// For your unit tests, prefer the RAMDirectory.
/// For your unit tests, prefer the RAMDirectory.
pub fn create_from_tempdir() -> io::Result<MmapDirectory> {
let tempdir = try!(TempDir::new("index"));
let tempdir_path = PathBuf::from(tempdir.path());
let directory = MmapDirectory {
root_path: PathBuf::from(tempdir_path),
mmap_cache: Arc::new(RwLock::new(MmapCache::default())),
_temp_directory: Arc::new(Some(tempdir))
_temp_directory: Arc::new(Some(tempdir)),
};
Ok(directory)
}
@@ -196,16 +179,14 @@ impl MmapDirectory {
pub fn open(directory_path: &Path) -> Result<MmapDirectory, OpenDirectoryError> {
if !directory_path.exists() {
Err(OpenDirectoryError::DoesNotExist(PathBuf::from(directory_path)))
}
else if !directory_path.is_dir() {
} else if !directory_path.is_dir() {
Err(OpenDirectoryError::NotADirectory(PathBuf::from(directory_path)))
}
else {
} else {
Ok(MmapDirectory {
root_path: PathBuf::from(directory_path),
mmap_cache: Arc::new(RwLock::new(MmapCache::default())),
_temp_directory: Arc::new(None)
})
root_path: PathBuf::from(directory_path),
mmap_cache: Arc::new(RwLock::new(MmapCache::default())),
_temp_directory: Arc::new(None),
})
}
}
@@ -232,7 +213,8 @@ impl MmapDirectory {
use std::os::windows::fs::OpenOptionsExt;
use winapi::winbase;
open_opts.write(true)
open_opts
.write(true)
.custom_flags(winbase::FILE_FLAG_BACKUP_SEMANTICS);
}
@@ -242,8 +224,8 @@ impl MmapDirectory {
}
/// Returns some statistical information
/// about the Mmap cache.
///
/// The `MmapDirectory` embeds a `MmapDirectory`
///
/// The `MmapDirectory` embeds a `MmapDirectory`
/// to avoid multiplying the `mmap` system calls.
pub fn get_cache_info(&mut self) -> CacheInfo {
self.mmap_cache
@@ -251,12 +233,10 @@ impl MmapDirectory {
.expect("Mmap cache lock is poisoned.")
.get_info()
}
}
/// This Write wraps a File, but has the specificity of
/// call `sync_all` on flush.
/// This Write wraps a File, but has the specificity of
/// call `sync_all` on flush.
struct SafeFileWriter(File);
impl SafeFileWriter {
@@ -266,7 +246,6 @@ impl SafeFileWriter {
}
impl Write for SafeFileWriter {
fn write(&mut self, buf: &[u8]) -> io::Result<usize> {
self.0.write(buf)
}
@@ -285,51 +264,49 @@ impl Seek for SafeFileWriter {
impl Directory for MmapDirectory {
fn open_read(&self, path: &Path) -> result::Result<ReadOnlySource, OpenReadError> {
debug!("Open Read {:?}", path);
let full_path = self.resolve_path(path);
let mut mmap_cache = self.mmap_cache
.write()
.map_err(|_| OpenReadError::IOError(
make_io_err(format!("Failed to acquired write lock on mmap cache while reading {:?}", path))
))?;
Ok(mmap_cache.get_mmap(full_path)?
.map(MmapReadOnly::from)
.map(ReadOnlySource::Mmap)
.unwrap_or(ReadOnlySource::Anonymous(SharedVecSlice::empty()))
)
.map_err(|_| {
let msg = format!("Failed to acquired write lock \
on mmap cache while reading {:?}",
path);
OpenReadError::IOError(make_io_err(msg))
})?;
Ok(mmap_cache
.get_mmap(full_path)?
.map(MmapReadOnly::from)
.map(ReadOnlySource::Mmap)
.unwrap_or_else(|| ReadOnlySource::Anonymous(SharedVecSlice::empty())))
}
fn open_write(&mut self, path: &Path) -> Result<WritePtr, OpenWriteError> {
debug!("Open Write {:?}", path);
let full_path = self.resolve_path(path);
let open_res = OpenOptions::new()
.write(true)
.create_new(true)
.open(full_path);
let mut file = try!(
open_res.map_err(|err| {
if err.kind() == io::ErrorKind::AlreadyExists {
OpenWriteError::FileAlreadyExists(PathBuf::from(path))
}
else {
OpenWriteError::IOError(err)
}
})
);
let mut file = open_res
.map_err(|err| if err.kind() == io::ErrorKind::AlreadyExists {
OpenWriteError::FileAlreadyExists(PathBuf::from(path))
} else {
OpenWriteError::IOError(err)
})?;
// making sure the file is created.
try!(file.flush());
// Apparetntly, on some filesystem syncing the parent
// directory is required.
try!(self.sync_directory());
let writer = SafeFileWriter::new(file);
Ok(BufWriter::new(Box::new(writer)))
}
@@ -337,25 +314,24 @@ impl Directory for MmapDirectory {
fn delete(&self, path: &Path) -> result::Result<(), DeleteError> {
debug!("Deleting file {:?}", path);
let full_path = self.resolve_path(path);
let mut mmap_cache = try!(self.mmap_cache
let mut mmap_cache = self.mmap_cache
.write()
.map_err(|_|
DeleteError::IOError(make_io_err(format!("Failed to acquired write lock on mmap cache while deleting {:?}", path))))
);
.map_err(|_| {
let msg = format!("Failed to acquired write lock \
on mmap cache while deleting {:?}",
path);
DeleteError::IOError(make_io_err(msg))
})?;
// Removing the entry in the MMap cache.
// The munmap will appear on Drop,
// when the last reference is gone.
mmap_cache.cache.remove(&full_path);
match fs::remove_file(&full_path) {
Ok(_) => {
self.sync_directory()
.map_err(|e| DeleteError::IOError(e))
}
Ok(_) => self.sync_directory().map_err(DeleteError::IOError),
Err(e) => {
if e.kind() == io::ErrorKind::NotFound {
Err(DeleteError::FileDoesNotExist(path.to_owned()))
}
else {
} else {
Err(DeleteError::IOError(e))
}
}
@@ -373,32 +349,29 @@ impl Directory for MmapDirectory {
match File::open(&full_path) {
Ok(mut file) => {
file.read_to_end(&mut buffer)
.map_err(|e| OpenReadError::IOError(e))?;
.map_err(OpenReadError::IOError)?;
Ok(buffer)
}
Err(e) => {
if e.kind() == io::ErrorKind::NotFound {
Err(OpenReadError::FileDoesNotExist(path.to_owned()))
}
else {
} else {
Err(OpenReadError::IOError(e))
}
}
}
}
fn atomic_write(&mut self, path: &Path, data: &[u8]) -> io::Result<()> {
debug!("Atomic Write {:?}", path);
let full_path = self.resolve_path(path);
let meta_file = atomicwrites::AtomicFile::new(full_path, atomicwrites::AllowOverwrite);
try!(meta_file.write(|f| {
f.write_all(data)
}));
try!(meta_file.write(|f| f.write_all(data)));
Ok(())
}
fn box_clone(&self,) -> Box<Directory> {
fn box_clone(&self) -> Box<Directory> {
Box::new(self.clone())
}
}
@@ -457,9 +430,9 @@ mod tests {
}
}
assert_eq!(mmap_directory.get_cache_info().counters.miss_empty, 10);
{
{
// test weak miss
// the first pass create the weak refs.
for path in &paths {
@@ -475,7 +448,7 @@ mod tests {
}
{
let mut saved_readmmaps = vec!();
let mut saved_readmmaps = vec![];
// Keeps reference alive
for (i, path) in paths.iter().enumerate() {
let r = mmap_directory.open_read(path).unwrap();
@@ -494,7 +467,7 @@ mod tests {
}
assert_eq!(mmap_directory.get_cache_info().mmapped.len(), 0);
}
}

10
src/directory/mod.rs
View File

@@ -31,7 +31,7 @@ pub type WritePtr = BufWriter<Box<SeekableWrite>>;
mod tests {
use super::*;
use std::path::Path;
use std::path::Path;
use std::io::{Write, Seek, SeekFrom};
lazy_static! {
@@ -65,7 +65,7 @@ mod tests {
assert!(directory.exists(*TEST_PATH));
write_file.write_all(&[4]).unwrap();
write_file.write_all(&[3]).unwrap();
write_file.write_all(&[7,3,5]).unwrap();
write_file.write_all(&[7, 3, 5]).unwrap();
write_file.flush().unwrap();
}
let read_file = directory.open_read(*TEST_PATH).unwrap();
@@ -81,9 +81,9 @@ mod tests {
{
{
let mut write_file = directory.open_write(*TEST_PATH).unwrap();
write_file.write_all(&[4, 3, 7,3,5]).unwrap();
write_file.write_all(&[4, 3, 7, 3, 5]).unwrap();
write_file.seek(SeekFrom::Start(0)).unwrap();
write_file.write_all(&[3,1]).unwrap();
write_file.write_all(&[3, 1]).unwrap();
write_file.flush().unwrap();
}
let read_file = directory.open_read(*TEST_PATH).unwrap();
@@ -98,7 +98,7 @@ mod tests {
{
directory.open_write(*TEST_PATH).unwrap();
assert!(directory.exists(*TEST_PATH));
}
{
assert!(directory.open_write(*TEST_PATH).is_err());

89
src/directory/ram_directory.rs
View File

@@ -11,14 +11,14 @@ use directory::WritePtr;
use super::shared_vec_slice::SharedVecSlice;
/// Writer associated with the `RAMDirectory`
///
///
/// The Writer just writes a buffer.
///
/// # Panics
///
/// On drop, if the writer was left in a *dirty* state.
/// That is, if flush was not called after the last call
/// to write.
/// to write.
///
struct VecWriter {
path: PathBuf,
@@ -40,8 +40,9 @@ impl VecWriter {
impl Drop for VecWriter {
fn drop(&mut self) {
if !self.is_flushed {
panic!("You forgot to flush {:?} before its writter got Drop. Do not rely on drop.", self.path)
if !self.is_flushed {
panic!("You forgot to flush {:?} before its writter got Drop. Do not rely on drop.",
self.path)
}
}
}
@@ -61,7 +62,8 @@ impl Write for VecWriter {
fn flush(&mut self) -> io::Result<()> {
self.is_flushed = true;
try!(self.shared_directory.write(self.path.clone(), self.data.get_ref()));
try!(self.shared_directory
.write(self.path.clone(), self.data.get_ref()));
Ok(())
}
}
@@ -72,35 +74,36 @@ struct InnerDirectory(Arc<RwLock<HashMap<PathBuf, Arc<Vec<u8>>>>>);
impl InnerDirectory {
fn new() -> InnerDirectory {
InnerDirectory(Arc::new(RwLock::new(HashMap::new())))
}
fn write(&self, path: PathBuf, data: &[u8]) -> io::Result<bool> {
let mut map = try!(
self.0
.write()
.map_err(|_| make_io_err(format!("Failed to lock the directory, when trying to write {:?}", path)))
);
let mut map = try!(self.0
.write()
.map_err(|_| {
make_io_err(format!("Failed to lock the directory, when trying to write {:?}",
path))
}));
let prev_value = map.insert(path, Arc::new(Vec::from(data)));
Ok(prev_value.is_some())
}
fn open_read(&self, path: &Path) -> Result<ReadOnlySource, OpenReadError> {
fn open_read(&self, path: &Path) -> Result<ReadOnlySource, OpenReadError> {
self.0
.read()
.map_err(|_| {
let io_err = make_io_err(format!("Failed to acquire read lock for the directory, when trying to read {:?}", path));
OpenReadError::IOError(io_err)
})
let msg = format!("Failed to acquire read lock for the \
directory when trying to read {:?}",
path);
let io_err = make_io_err(msg);
OpenReadError::IOError(io_err)
})
.and_then(|readable_map| {
readable_map
.get(path)
.ok_or_else(|| OpenReadError::FileDoesNotExist(PathBuf::from(path)))
.map(|data| {
ReadOnlySource::Anonymous(SharedVecSlice::new(data.clone()))
})
.get(path)
.ok_or_else(|| OpenReadError::FileDoesNotExist(PathBuf::from(path)))
.map(|data| ReadOnlySource::Anonymous(SharedVecSlice::new(data.clone())))
})
}
@@ -108,19 +111,16 @@ impl InnerDirectory {
self.0
.write()
.map_err(|_| {
let io_err = make_io_err(format!("Failed to acquire write lock for the directory, when trying to delete {:?}", path));
DeleteError::IOError(io_err)
})
.and_then(|mut writable_map| {
match writable_map.remove(path) {
Some(_) => {
Ok(())
},
None => {
Err(DeleteError::FileDoesNotExist(PathBuf::from(path)))
}
}
})
let msg = format!("Failed to acquire write lock for the \
directory when trying to delete {:?}",
path);
let io_err = make_io_err(msg);
DeleteError::IOError(io_err)
})
.and_then(|mut writable_map| match writable_map.remove(path) {
Some(_) => Ok(()),
None => Err(DeleteError::FileDoesNotExist(PathBuf::from(path))),
})
}
fn exists(&self, path: &Path) -> bool {
@@ -129,13 +129,12 @@ impl InnerDirectory {
.expect("Failed to get read lock directory.")
.contains_key(path)
}
}
impl fmt::Debug for RAMDirectory {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
write!(f, "RAMDirectory")
}
write!(f, "RAMDirectory")
}
}
@@ -150,12 +149,9 @@ pub struct RAMDirectory {
}
impl RAMDirectory {
/// Constructor
pub fn create() -> RAMDirectory {
RAMDirectory {
fs: InnerDirectory::new()
}
RAMDirectory { fs: InnerDirectory::new() }
}
}
@@ -163,15 +159,14 @@ impl Directory for RAMDirectory {
fn open_read(&self, path: &Path) -> result::Result<ReadOnlySource, OpenReadError> {
self.fs.open_read(path)
}
fn open_write(&mut self, path: &Path) -> Result<WritePtr, OpenWriteError> {
let path_buf = PathBuf::from(path);
let vec_writer = VecWriter::new(path_buf.clone(), self.fs.clone());
// force the creation of the file to mimic the MMap directory.
if try!(self.fs.write(path_buf.clone(), &Vec::new())) {
Err(OpenWriteError::FileAlreadyExists(path_buf))
}
else {
} else {
Ok(BufWriter::new(Box::new(vec_writer)))
}
}
@@ -180,15 +175,14 @@ impl Directory for RAMDirectory {
self.fs.delete(path)
}
fn exists(&self, path: &Path) -> bool {
self.fs.exists(path)
}
fn atomic_read(&self, path: &Path) -> Result<Vec<u8>, OpenReadError> {
let read = self.open_read(path)?;
Ok(read.as_slice()
.to_owned())
Ok(read.as_slice().to_owned())
}
fn atomic_write(&mut self, path: &Path, data: &[u8]) -> io::Result<()> {
@@ -200,8 +194,7 @@ impl Directory for RAMDirectory {
Ok(())
}
fn box_clone(&self,) -> Box<Directory> {
fn box_clone(&self) -> Box<Directory> {
Box::new(self.clone())
}
}

30
src/directory/read_only_source.rs
View File

@@ -5,7 +5,7 @@ use common::HasLen;
/// Read object that represents files in tantivy.
///
///
/// These read objects are only in charge to deliver
/// the data in the form of a constant read-only `&[u8]`.
/// Whatever happens to the directory file, the data
@@ -13,12 +13,11 @@ use common::HasLen;
pub enum ReadOnlySource {
/// Mmap source of data
Mmap(MmapReadOnly),
/// Wrapping a `Vec<u8>`
/// Wrapping a `Vec<u8>`
Anonymous(SharedVecSlice),
}
impl Deref for ReadOnlySource {
type Target = [u8];
fn deref(&self) -> &[u8] {
@@ -27,35 +26,30 @@ impl Deref for ReadOnlySource {
}
impl ReadOnlySource {
/// Creates an empty ReadOnlySource
pub fn empty() -> ReadOnlySource {
ReadOnlySource::Anonymous(SharedVecSlice::empty())
}
/// Returns the data underlying the ReadOnlySource object.
pub fn as_slice(&self,) -> &[u8] {
pub fn as_slice(&self) -> &[u8] {
match *self {
ReadOnlySource::Mmap(ref mmap_read_only) => unsafe {
mmap_read_only.as_slice()
},
ReadOnlySource::Anonymous(ref shared_vec) => {
shared_vec.as_slice()
},
ReadOnlySource::Mmap(ref mmap_read_only) => unsafe { mmap_read_only.as_slice() },
ReadOnlySource::Anonymous(ref shared_vec) => shared_vec.as_slice(),
}
}
/// Creates a ReadOnlySource that is just a
/// Creates a ReadOnlySource that is just a
/// view over a slice of the data.
///
///
/// Keep in mind that any living slice extends
/// the lifetime of the original ReadOnlySource,
///
///
/// For instance, if `ReadOnlySource` wraps 500MB
/// worth of data in anonymous memory, and only a
/// 1KB slice is remaining, the whole `500MBs`
/// 1KB slice is remaining, the whole `500MBs`
/// are retained in memory.
pub fn slice(&self, from_offset:usize, to_offset:usize) -> ReadOnlySource {
pub fn slice(&self, from_offset: usize, to_offset: usize) -> ReadOnlySource {
match *self {
ReadOnlySource::Mmap(ref mmap_read_only) => {
let sliced_mmap = mmap_read_only.range(from_offset, to_offset - from_offset);
@@ -63,13 +57,13 @@ impl ReadOnlySource {
}
ReadOnlySource::Anonymous(ref shared_vec) => {
ReadOnlySource::Anonymous(shared_vec.slice(from_offset, to_offset))
},
}
}
}
}
impl HasLen for ReadOnlySource {
fn len(&self,) -> usize {
fn len(&self) -> usize {
self.as_slice().len()
}
}

9
src/directory/shared_vec_slice.rs
View File

@@ -4,12 +4,11 @@ use std::sync::Arc;
#[derive(Clone)]
pub struct SharedVecSlice {
pub data: Arc<Vec<u8>>,
pub start: usize,
pub len: usize
pub start: usize,
pub len: usize,
}
impl SharedVecSlice {
pub fn empty() -> SharedVecSlice {
SharedVecSlice::new(Arc::new(Vec::new()))
}
@@ -23,11 +22,11 @@ impl SharedVecSlice {
}
}
pub fn as_slice(&self,) -> &[u8] {
pub fn as_slice(&self) -> &[u8] {
&self.data[self.start..self.start + self.len]
}
pub fn slice(&self, from_offset: usize, to_offset:usize) -> SharedVecSlice {
pub fn slice(&self, from_offset: usize, to_offset: usize) -> SharedVecSlice {
SharedVecSlice {
data: self.data.clone(),
start: self.start + from_offset,

22
src/error.rs
View File

@@ -1,5 +1,3 @@
#![allow(enum_variant_names)]
/// Definition of Tantivy's error and result.
use std::io;
@@ -38,7 +36,7 @@ pub enum Error {
/// An Error appeared related to the lack of a field.
SchemaError(String),
/// Tried to access a fastfield reader for a field not configured accordingly.
FastFieldError(FastFieldNotAvailableError)
FastFieldError(FastFieldNotAvailableError),
}
impl From<FastFieldNotAvailableError> for Error {
@@ -83,10 +81,8 @@ impl From<schema::DocParsingError> for Error {
impl From<OpenWriteError> for Error {
fn from(error: OpenWriteError) -> Error {
match error {
OpenWriteError::FileAlreadyExists(filepath) =>
Error::FileAlreadyExists(filepath),
OpenWriteError::IOError(io_error) =>
Error::IOError(io_error),
OpenWriteError::FileAlreadyExists(filepath) => Error::FileAlreadyExists(filepath),
OpenWriteError::IOError(io_error) => Error::IOError(io_error),
}
}
}
@@ -94,10 +90,12 @@ impl From<OpenWriteError> for Error {
impl From<OpenDirectoryError> for Error {
fn from(error: OpenDirectoryError) -> Error {
match error {
OpenDirectoryError::DoesNotExist(directory_path) =>
Error::PathDoesNotExist(directory_path),
OpenDirectoryError::NotADirectory(directory_path) =>
Error::InvalidArgument(format!("{:?} is not a directory", directory_path)),
OpenDirectoryError::DoesNotExist(directory_path) => {
Error::PathDoesNotExist(directory_path)
}
OpenDirectoryError::NotADirectory(directory_path) => {
Error::InvalidArgument(format!("{:?} is not a directory", directory_path))
}
}
}
}
@@ -106,4 +104,4 @@ impl From<serde_json::Error> for Error {
fn from(error: serde_json::Error) -> Error {
Error::IOError(error.into())
}
}
}

18
src/fastfield/delete.rs
View File

@@ -6,7 +6,7 @@ use directory::ReadOnlySource;
use DocId;
use common::HasLen;
/// Write a delete BitSet
/// Write a delete `BitSet`
///
/// where `delete_bitset` is the set of deleted `DocId`.
pub fn write_delete_bitset(delete_bitset: &BitSet, writer: &mut WritePtr) -> io::Result<()> {
@@ -21,8 +21,7 @@ pub fn write_delete_bitset(delete_bitset: &BitSet, writer: &mut WritePtr) -> io:
writer.write_all(&[byte])?;
shift = 0;
byte = 0;
}
else {
} else {
shift += 1;
}
}
@@ -36,15 +35,14 @@ pub fn write_delete_bitset(delete_bitset: &BitSet, writer: &mut WritePtr) -> io:
#[derive(Clone)]
pub struct DeleteBitSet {
data: ReadOnlySource,
len: usize,
len: usize,
}
impl DeleteBitSet {
/// Opens a delete bitset given its data source.
pub fn open(data: ReadOnlySource) -> DeleteBitSet {
let num_deleted: usize = data
.as_slice()
let num_deleted: usize = data.as_slice()
.iter()
.map(|b| b.count_ones() as usize)
.sum();
@@ -71,15 +69,13 @@ impl DeleteBitSet {
pub fn is_deleted(&self, doc: DocId) -> bool {
if self.len == 0 {
false
}
else {
} 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
b & (1u8 << shift) != 0
}
}
}
@@ -132,4 +128,4 @@ mod tests {
test_delete_bitset_helper(&bitset);
}
}
}
}

11
src/fastfield/error.rs
View File

@@ -1,7 +1,7 @@
use std::result;
use schema::FieldEntry;
/// FastFieldNotAvailableError is returned when the
/// `FastFieldNotAvailableError` is returned when the
/// user requested for a fast field reader, and the field was not
/// defined in the schema as a fast field.
#[derive(Debug)]
@@ -10,17 +10,14 @@ pub struct FastFieldNotAvailableError {
}
impl FastFieldNotAvailableError {
/// Creates a `FastFieldNotAvailable` error.
/// `field_entry` is the configuration of the field
/// `field_entry` is the configuration of the field
/// for which fast fields are not available.
pub fn new(field_entry: &FieldEntry) -> FastFieldNotAvailableError {
FastFieldNotAvailableError {
field_name: field_entry.name().to_string(),
}
FastFieldNotAvailableError { field_name: field_entry.name().to_string() }
}
}
/// Result when trying to access a fast field reader.
pub type Result<R> = result::Result<R, FastFieldNotAvailableError>;
pub type Result<R> = result::Result<R, FastFieldNotAvailableError>;

135
src/fastfield/mod.rs
View File

@@ -3,20 +3,20 @@
//! Fast fields are the equivalent of `DocValues` in `Lucene`.
//! Fast fields is a non-compressed column-oriented fashion storage
//! of `tantivy`.
//!
//!
//! It is designed for the fast random access of some document
//! fields given a document id.
//!
//! `FastField` are useful when a field is required for all or most of
//! the `DocSet` : for instance for scoring, grouping, filtering, or facetting.
//!
//!
//!
//!
//! Fields have to be declared as `FAST` in the schema.
//! Currently only 64-bits integers (signed or unsigned) are
//! supported.
//!
//! They are stored in a bitpacked fashion so that their
//! memory usage is directly linear with the amplitude of the
//! memory usage is directly linear with the amplitude of the
//! values stored.
//!
//! Read access performance is comparable to that of an array lookup.
@@ -57,7 +57,7 @@ mod tests {
schema_builder.add_u64_field("field", FAST);
schema_builder.build()
};
static ref FIELD: Field = {
static ref FIELD: Field = {
SCHEMA.get_field("field").unwrap()
};
}
@@ -67,13 +67,13 @@ mod tests {
doc.add_u64(field, value);
fast_field_writers.add_document(&doc);
}
#[test]
pub fn test_fastfield() {
let test_fastfield = U64FastFieldReader::from(vec!(100,200,300));
let test_fastfield = U64FastFieldReader::from(vec![100, 200, 300]);
assert_eq!(test_fastfield.get(0), 100);
assert_eq!(test_fastfield.get(1), 200);
assert_eq!(test_fastfield.get(2), 300);
assert_eq!(test_fastfield.get(2), 300);
}
#[test]
@@ -92,11 +92,12 @@ mod tests {
}
let source = directory.open_read(&path).unwrap();
{
assert_eq!(source.len(), 31 as usize);
assert_eq!(source.len(), 38 as usize);
}
{
let fast_field_readers = FastFieldsReader::open(source).unwrap();
let fast_field_reader: U64FastFieldReader = fast_field_readers.open_reader(*FIELD).unwrap();
let fast_field_reader: U64FastFieldReader =
fast_field_readers.open_reader(*FIELD).unwrap();
assert_eq!(fast_field_reader.get(0), 13u64);
assert_eq!(fast_field_reader.get(1), 14u64);
assert_eq!(fast_field_reader.get(2), 2u64);
@@ -125,11 +126,12 @@ mod tests {
}
let source = directory.open_read(&path).unwrap();
{
assert_eq!(source.len(), 56 as usize);
assert_eq!(source.len(), 63 as usize);
}
{
let fast_field_readers = FastFieldsReader::open(source).unwrap();
let fast_field_reader: U64FastFieldReader = fast_field_readers.open_reader(*FIELD).unwrap();
let fast_field_reader: U64FastFieldReader =
fast_field_readers.open_reader(*FIELD).unwrap();
assert_eq!(fast_field_reader.get(0), 4u64);
assert_eq!(fast_field_reader.get(1), 14_082_001u64);
assert_eq!(fast_field_reader.get(2), 3_052u64);
@@ -141,9 +143,9 @@ mod tests {
assert_eq!(fast_field_reader.get(8), 215u64);
}
}
#[test]
fn test_intfastfield_null_amplitude() {
#[test]
fn test_intfastfield_null_amplitude() {
let path = Path::new("test");
let mut directory: RAMDirectory = RAMDirectory::create();
@@ -160,22 +162,23 @@ mod tests {
}
let source = directory.open_read(&path).unwrap();
{
assert_eq!(source.len(), 29 as usize);
assert_eq!(source.len(), 36 as usize);
}
{
let fast_field_readers = FastFieldsReader::open(source).unwrap();
let fast_field_reader: U64FastFieldReader = fast_field_readers.open_reader(*FIELD).unwrap();
let fast_field_reader: U64FastFieldReader =
fast_field_readers.open_reader(*FIELD).unwrap();
for doc in 0..10_000 {
assert_eq!(fast_field_reader.get(doc), 100_000u64);
}
}
}
#[test]
fn test_intfastfield_large_numbers() {
#[test]
fn test_intfastfield_large_numbers() {
let path = Path::new("test");
let mut directory: RAMDirectory = RAMDirectory::create();
{
let write: WritePtr = directory.open_write(Path::new("test")).unwrap();
let mut serializer = FastFieldSerializer::new(write).unwrap();
@@ -183,21 +186,25 @@ mod tests {
// forcing the amplitude to be high
add_single_field_doc(&mut fast_field_writers, *FIELD, 0u64);
for i in 0u64..10_000u64 {
add_single_field_doc(&mut fast_field_writers, *FIELD, 5_000_000_000_000_000_000u64 + i);
add_single_field_doc(&mut fast_field_writers,
*FIELD,
5_000_000_000_000_000_000u64 + i);
}
fast_field_writers.serialize(&mut serializer).unwrap();
serializer.close().unwrap();
}
let source = directory.open_read(&path).unwrap();
{
assert_eq!(source.len(), 80037 as usize);
assert_eq!(source.len(), 80044 as usize);
}
{
let fast_field_readers = FastFieldsReader::open(source).unwrap();
let fast_field_reader: U64FastFieldReader = fast_field_readers.open_reader(*FIELD).unwrap();
let fast_field_reader: U64FastFieldReader =
fast_field_readers.open_reader(*FIELD).unwrap();
assert_eq!(fast_field_reader.get(0), 0u64);
for doc in 1..10_001 {
assert_eq!(fast_field_reader.get(doc), 5_000_000_000_000_000_000u64 + doc as u64 - 1u64);
assert_eq!(fast_field_reader.get(doc),
5_000_000_000_000_000_000u64 + doc as u64 - 1u64);
}
}
}
@@ -225,11 +232,12 @@ mod tests {
}
let source = directory.open_read(&path).unwrap();
{
assert_eq!(source.len(), 17704 as usize);
assert_eq!(source.len(), 17711 as usize);
}
{
let fast_field_readers = FastFieldsReader::open(source).unwrap();
let fast_field_reader: I64FastFieldReader = fast_field_readers.open_reader(i64_field).unwrap();
let fast_field_reader: I64FastFieldReader =
fast_field_readers.open_reader(i64_field).unwrap();
assert_eq!(fast_field_reader.min_value(), -100i64);
assert_eq!(fast_field_reader.max_value(), 9_999i64);
for (doc, i) in (-100i64..10_000i64).enumerate() {
@@ -255,11 +263,12 @@ mod tests {
fast_field_writers.serialize(&mut serializer).unwrap();
serializer.close().unwrap();
}
let source = directory.open_read(&path).unwrap();
{
let fast_field_readers = FastFieldsReader::open(source).unwrap();
let fast_field_reader: I64FastFieldReader = fast_field_readers.open_reader(i64_field).unwrap();
let fast_field_reader: I64FastFieldReader =
fast_field_readers.open_reader(i64_field).unwrap();
assert_eq!(fast_field_reader.get(0u32), 0i64);
}
}
@@ -291,10 +300,14 @@ mod tests {
let source = directory.open_read(&path).unwrap();
{
let fast_field_readers = FastFieldsReader::open(source).unwrap();
let fast_field_reader: U64FastFieldReader = fast_field_readers.open_reader(*FIELD).unwrap();
let fast_field_reader: U64FastFieldReader =
fast_field_readers.open_reader(*FIELD).unwrap();
let mut a = 0u64;
for _ in 0..n {
println!("i {}=> {} {}", a, fast_field_reader.get(a as u32), permutation[a as usize]);
println!("i {}=> {} {}",
a,
fast_field_reader.get(a as u32),
permutation[a as usize]);
assert_eq!(fast_field_reader.get(a as u32), permutation[a as usize]);
a = fast_field_reader.get(a as u32);
}
@@ -305,26 +318,26 @@ mod tests {
fn bench_intfastfield_linear_veclookup(b: &mut Bencher) {
let permutation = generate_permutation();
b.iter(|| {
let n = test::black_box(7000u32);
let mut a = 0u64;
for i in (0u32..n).step_by(7) {
a ^= permutation[i as usize];
}
a
});
let n = test::black_box(7000u32);
let mut a = 0u64;
for i in (0u32..n).step_by(7) {
a ^= permutation[i as usize];
}
a
});
}
#[bench]
fn bench_intfastfield_veclookup(b: &mut Bencher) {
let permutation = generate_permutation();
b.iter(|| {
let n = test::black_box(1000u32);
let mut a = 0u64;
for _ in 0u32..n {
a = permutation[a as usize];
}
a
});
let n = test::black_box(1000u32);
let mut a = 0u64;
for _ in 0u32..n {
a = permutation[a as usize];
}
a
});
}
#[bench]
@@ -345,15 +358,16 @@ mod tests {
let source = directory.open_read(&path).unwrap();
{
let fast_field_readers = FastFieldsReader::open(source).unwrap();
let fast_field_reader: U64FastFieldReader = fast_field_readers.open_reader(*FIELD).unwrap();
let fast_field_reader: U64FastFieldReader =
fast_field_readers.open_reader(*FIELD).unwrap();
b.iter(|| {
let n = test::black_box(7000u32);
let mut a = 0u64;
for i in (0u32..n).step_by(7) {
a ^= fast_field_reader.get(i);
}
a
});
let n = test::black_box(7000u32);
let mut a = 0u64;
for i in (0u32..n).step_by(7) {
a ^= fast_field_reader.get(i);
}
a
});
}
}
@@ -375,15 +389,16 @@ mod tests {
let source = directory.open_read(&path).unwrap();
{
let fast_field_readers = FastFieldsReader::open(source).unwrap();
let fast_field_reader: U64FastFieldReader = fast_field_readers.open_reader(*FIELD).unwrap();
let fast_field_reader: U64FastFieldReader =
fast_field_readers.open_reader(*FIELD).unwrap();
b.iter(|| {
let n = test::black_box(1000u32);
let mut a = 0u32;
for _ in 0u32..n {
a = fast_field_reader.get(a) as u32;
}
a
});
let n = test::black_box(1000u32);
let mut a = 0u32;
for _ in 0u32..n {
a = fast_field_reader.get(a) as u32;
}
a
});
}
}
}

95
src/fastfield/reader.rs
View File

@@ -19,7 +19,6 @@ use common;
/// Depending on the field type, a different
/// fast field is required.
pub trait FastFieldReader: Sized {
/// Type of the value stored in the fastfield.
type ValueType;
@@ -33,12 +32,12 @@ pub trait FastFieldReader: Sized {
fn open(source: ReadOnlySource) -> Self;
/// Returns true iff the given field_type makes
/// it possible to access the field values via a
/// it possible to access the field values via a
/// fastfield.
fn is_enabled(field_type: &FieldType) -> bool;
}
/// FastFieldReader for unsigned 64-bits integers.
/// `FastFieldReader` for unsigned 64-bits integers.
pub struct U64FastFieldReader {
_data: ReadOnlySource,
bit_unpacker: BitUnpacker,
@@ -47,37 +46,35 @@ pub struct U64FastFieldReader {
}
impl U64FastFieldReader {
/// Returns the minimum value for this fast field.
///
/// The min value does not take in account of possible
/// deleted document, and should be considered as a lower bound
/// deleted document, and should be considered as a lower bound
/// of the actual minimum value.
pub fn min_value(&self,) -> u64 {
pub fn min_value(&self) -> u64 {
self.min_value
}
}
/// Returns the maximum value for this fast field.
///
/// The max value does not take in account of possible
/// deleted document, and should be considered as an upper bound
/// deleted document, and should be considered as an upper bound
/// of the actual maximum value.
pub fn max_value(&self,) -> u64 {
pub fn max_value(&self) -> u64 {
self.max_value
}
}
impl FastFieldReader for U64FastFieldReader {
type ValueType = u64;
fn get(&self, doc: DocId) -> u64 {
self.min_value + self.bit_unpacker.get(doc as usize)
}
fn is_enabled(field_type: &FieldType) -> bool {
match field_type {
&FieldType::U64(ref integer_options) =>
integer_options.is_fast(),
match *field_type {
FieldType::U64(ref integer_options) => integer_options.is_fast(),
_ => false,
}
}
@@ -90,11 +87,13 @@ impl FastFieldReader for U64FastFieldReader {
let min_value: u64;
let max_value: u64;
let bit_unpacker: BitUnpacker;
{
let mut cursor: &[u8] = data.as_slice();
min_value = u64::deserialize(&mut cursor).expect("Failed to read the min_value of fast field.");
let amplitude = u64::deserialize(&mut cursor).expect("Failed to read the amplitude of fast field.");
min_value = u64::deserialize(&mut cursor)
.expect("Failed to read the min_value of fast field.");
let amplitude = u64::deserialize(&mut cursor)
.expect("Failed to read the amplitude of fast field.");
max_value = min_value + amplitude;
let num_bits = compute_num_bits(amplitude);
bit_unpacker = BitUnpacker::new(cursor, num_bits as usize)
@@ -107,7 +106,6 @@ impl FastFieldReader for U64FastFieldReader {
max_value: max_value,
}
}
}
@@ -129,13 +127,13 @@ impl From<Vec<u64>> for U64FastFieldReader {
fast_field_writers.serialize(&mut serializer).unwrap();
serializer.close().unwrap();
}
let source = directory.open_read(&path).unwrap();
let source = directory.open_read(path).unwrap();
let fast_field_readers = FastFieldsReader::open(source).unwrap();
fast_field_readers.open_reader(field).unwrap()
}
}
}
/// FastFieldReader for signed 64-bits integers.
/// `FastFieldReader` for signed 64-bits integers.
pub struct I64FastFieldReader {
underlying: U64FastFieldReader,
}
@@ -144,61 +142,51 @@ impl I64FastFieldReader {
/// Returns the minimum value for this fast field.
///
/// The min value does not take in account of possible
/// deleted document, and should be considered as a lower bound
/// deleted document, and should be considered as a lower bound
/// of the actual minimum value.
pub fn min_value(&self,) -> i64 {
pub fn min_value(&self) -> i64 {
common::u64_to_i64(self.underlying.min_value())
}
/// Returns the maximum value for this fast field.
///
/// The max value does not take in account of possible
/// deleted document, and should be considered as an upper bound
/// deleted document, and should be considered as an upper bound
/// of the actual maximum value.
pub fn max_value(&self,) -> i64 {
pub fn max_value(&self) -> i64 {
common::u64_to_i64(self.underlying.max_value())
}
}
impl FastFieldReader for I64FastFieldReader {
type ValueType = i64;
fn get(&self, doc: DocId) -> i64 {
common::u64_to_i64(self.underlying.get(doc))
}
/// Opens a new fast field reader given a read only source.
///
/// # Panics
/// Panics if the data is corrupted.
fn open(data: ReadOnlySource) -> I64FastFieldReader {
I64FastFieldReader {
underlying: U64FastFieldReader::open(data)
}
I64FastFieldReader { underlying: U64FastFieldReader::open(data) }
}
fn is_enabled(field_type: &FieldType) -> bool {
match field_type {
&FieldType::I64(ref integer_options) => {
if integer_options.is_fast() {
true
}
else {
false
}
},
match *field_type {
FieldType::I64(ref integer_options) => integer_options.is_fast(),
_ => false,
}
}
}
/// The FastFieldsReader` is the datastructure containing
/// The `FastFieldsReader` is the datastructure containing
/// all of the fast fields' data.
///
/// It contains a mapping that associated these fields to
/// It contains a mapping that associated these fields to
/// the proper slice in the fastfield reader file.
pub struct FastFieldsReader {
source: ReadOnlySource,
@@ -206,11 +194,10 @@ pub struct FastFieldsReader {
}
impl FastFieldsReader {
/// Opens the `FastFieldsReader` file
///
/// When opening the fast field reader, the
/// the list of the offset is read (as a footer of the
/// the list of the offset is read (as a footer of the
/// data file).
pub fn open(source: ReadOnlySource) -> io::Result<FastFieldsReader> {
let header_offset;
@@ -223,23 +210,21 @@ impl FastFieldsReader {
}
{
let mut cursor = &buffer[header_offset as usize..];
field_offsets = Vec::deserialize(&mut cursor)?;
field_offsets = Vec::deserialize(&mut cursor)?;
}
}
let mut end_offsets: Vec<u32> = field_offsets
.iter()
.map(|&(_, offset)| offset)
.collect();
let mut end_offsets: Vec<u32> = field_offsets.iter().map(|&(_, offset)| offset).collect();
end_offsets.push(header_offset);
let mut field_offsets_map: HashMap<Field, (u32, u32)> = HashMap::new();
for (field_start_offsets, stop_offset) in field_offsets.iter().zip(end_offsets.iter().skip(1)) {
for (field_start_offsets, stop_offset) in
field_offsets.iter().zip(end_offsets.iter().skip(1)) {
let (field, start_offset) = *field_start_offsets;
field_offsets_map.insert(field, (start_offset, *stop_offset));
}
Ok(FastFieldsReader {
field_offsets: field_offsets_map,
source: source,
})
field_offsets: field_offsets_map,
source: source,
})
}
/// Returns the u64 fast value reader if the field
@@ -254,8 +239,8 @@ impl FastFieldsReader {
self.field_offsets
.get(&field)
.map(|&(start, stop)| {
let field_source = self.source.slice(start as usize, stop as usize);
FFReader::open(field_source)
})
let field_source = self.source.slice(start as usize, stop as usize);
FFReader::open(field_source)
})
}
}

42
src/fastfield/serializer.rs
View File

@@ -7,9 +7,9 @@ use std::io::{self, Write, Seek, SeekFrom};
/// `FastFieldSerializer` is in charge of serializing
/// fastfields on disk.
///
///
/// Fast fields are encoded using bit-packing.
///
///
/// `FastFieldWriter`s are in charge of pushing the data to
/// the serializer.
/// The serializer expects to receive the following calls.
@@ -41,17 +41,21 @@ impl FastFieldSerializer {
// just making room for the pointer to header.
let written_size: usize = try!(0u32.serialize(&mut write));
Ok(FastFieldSerializer {
write: write,
written_size: written_size,
fields: Vec::new(),
min_value: 0,
field_open: false,
bit_packer: BitPacker::new(0),
})
write: write,
written_size: written_size,
fields: Vec::new(),
min_value: 0,
field_open: false,
bit_packer: BitPacker::new(0),
})
}
/// Start serializing a new u64 fast field
pub fn new_u64_fast_field(&mut self, field: Field, min_value: u64, max_value: u64) -> io::Result<()> {
pub fn new_u64_fast_field(&mut self,
field: Field,
min_value: u64,
max_value: u64)
-> io::Result<()> {
if self.field_open {
return Err(io::Error::new(io::ErrorKind::Other, "Previous field not closed"));
}
@@ -68,15 +72,15 @@ impl FastFieldSerializer {
}
/// Pushes a new value to the currently open u64 fast field.
/// Pushes a new value to the currently open u64 fast field.
pub fn add_val(&mut self, val: u64) -> io::Result<()> {
let val_to_write: u64 = val - self.min_value;
self.bit_packer.write(val_to_write, &mut self.write)?;
Ok(())
}
/// Close the u64 fast field.
pub fn close_field(&mut self,) -> io::Result<()> {
/// Close the u64 fast field.
pub fn close_field(&mut self) -> io::Result<()> {
if !self.field_open {
return Err(io::Error::new(io::ErrorKind::Other, "Current field is already closed"));
}
@@ -87,12 +91,12 @@ impl FastFieldSerializer {
self.written_size += self.bit_packer.close(&mut self.write)?;
Ok(())
}
/// Closes the serializer
///
///
/// After this call the data must be persistently save on disk.
pub fn close(mut self,) -> io::Result<usize> {
pub fn close(mut self) -> io::Result<usize> {
if self.field_open {
return Err(io::Error::new(io::ErrorKind::Other, "Last field not closed"));
}

78
src/fastfield/writer.rs
View File

@@ -12,7 +12,6 @@ pub struct FastFieldsWriter {
}
impl FastFieldsWriter {
/// Create all `FastFieldWriter` required by the schema.
pub fn from_schema(schema: &Schema) -> FastFieldsWriter {
let field_writers: Vec<IntFastFieldWriter> = schema
@@ -21,46 +20,39 @@ impl FastFieldsWriter {
.enumerate()
.flat_map(|(field_id, field_entry)| {
let field = Field(field_id as u32);
match field_entry.field_type() {
&FieldType::I64(ref int_options) => {
match *field_entry.field_type() {
FieldType::I64(ref int_options) => {
if int_options.is_fast() {
let mut fast_field_writer = IntFastFieldWriter::new(field);
fast_field_writer.set_val_if_missing(common::i64_to_u64(0i64));
Some(fast_field_writer)
}
else {
} else {
None
}
}
&FieldType::U64(ref int_options) => {
FieldType::U64(ref int_options) => {
if int_options.is_fast() {
Some(IntFastFieldWriter::new(field))
}
else {
} else {
None
}
}
_ => None
_ => None,
}
})
})
.collect();
FastFieldsWriter {
field_writers: field_writers,
}
FastFieldsWriter { field_writers: field_writers }
}
/// Returns a `FastFieldsWriter`
/// with a `IntFastFieldWriter` for each
/// Returns a `FastFieldsWriter`
/// with a `IntFastFieldWriter` for each
/// of the field given in argument.
pub fn new(fields: Vec<Field>) -> FastFieldsWriter {
FastFieldsWriter {
field_writers: fields
.into_iter()
.map(IntFastFieldWriter::new)
.collect(),
field_writers: fields.into_iter().map(IntFastFieldWriter::new).collect(),
}
}
/// Get the `FastFieldWriter` associated to a field.
pub fn get_field_writer(&mut self, field: Field) -> Option<&mut IntFastFieldWriter> {
// TODO optimize
@@ -68,7 +60,7 @@ impl FastFieldsWriter {
.iter_mut()
.find(|field_writer| field_writer.field == field)
}
/// Indexes all of the fastfields of a new document.
pub fn add_document(&mut self, doc: &Document) {
@@ -77,7 +69,7 @@ impl FastFieldsWriter {
}
}
/// Serializes all of the `FastFieldWriter`s by pushing them in
/// Serializes all of the `FastFieldWriter`s by pushing them in
/// order to the fast field serializer.
pub fn serialize(&self, serializer: &mut FastFieldSerializer) -> io::Result<()> {
for field_writer in &self.field_writers {
@@ -85,10 +77,10 @@ impl FastFieldsWriter {
}
Ok(())
}
/// Ensures all of the fast field writers have
/// reached `doc`. (included)
///
///
/// The missing values will be filled with 0.
pub fn fill_val_up_to(&mut self, doc: DocId) {
for field_writer in &mut self.field_writers {
@@ -99,16 +91,16 @@ impl FastFieldsWriter {
/// Fast field writer for ints.
/// The fast field writer just keeps the values in memory.
///
///
/// Only when the segment writer can be closed and
/// persisted on disc, the fast field writer is
/// persisted on disc, the fast field writer is
/// sent to a `FastFieldSerializer` via the `.serialize(...)`
/// method.
///
/// We cannot serialize earlier as the values are
/// bitpacked and the number of bits required for bitpacking
/// We cannot serialize earlier as the values are
/// bitpacked and the number of bits required for bitpacking
/// can only been known once we have seen all of the values.
///
///
/// Both u64, and i64 use the same writer.
/// i64 are just remapped to the `0..2^64 - 1`
/// using `common::i64_to_u64`.
@@ -119,7 +111,6 @@ pub struct IntFastFieldWriter {
}
impl IntFastFieldWriter {
/// Creates a new `IntFastFieldWriter`
pub fn new(field: Field) -> IntFastFieldWriter {
IntFastFieldWriter {
@@ -128,10 +119,10 @@ impl IntFastFieldWriter {
val_if_missing: 0u64,
}
}
/// Sets the default value.
///
/// This default value is recorded for documents if
/// This default value is recorded for documents if
/// a document does not have any value.
fn set_val_if_missing(&mut self, val_if_missing: u64) {
self.val_if_missing = val_if_missing;
@@ -139,7 +130,7 @@ impl IntFastFieldWriter {
/// Ensures all of the fast field writer have
/// reached `doc`. (included)
///
///
/// The missing values will be filled with 0.
fn fill_val_up_to(&mut self, doc: DocId) {
let target = doc as usize + 1;
@@ -158,9 +149,9 @@ impl IntFastFieldWriter {
pub fn add_val(&mut self, val: u64) {
self.vals.push(val);
}
/// Extract the value associated to the fast field for
/// Extract the value associated to the fast field for
/// this document.
///
/// i64 are remapped to u64 using the logic
@@ -174,14 +165,12 @@ impl IntFastFieldWriter {
match doc.get_first(self.field) {
Some(v) => {
match *v {
Value::U64(ref val) => { *val },
Value::U64(ref val) => *val,
Value::I64(ref val) => common::i64_to_u64(*val),
_ => { panic!("Expected a u64field, got {:?} ", v) }
_ => panic!("Expected a u64field, got {:?} ", v),
}
},
None => {
self.val_if_missing
}
}
None => self.val_if_missing,
}
}
@@ -204,8 +193,3 @@ impl IntFastFieldWriter {
serializer.close_field()
}
}

11
src/functional_test.rs
View File

@@ -18,7 +18,7 @@ fn test_indexing() {
let mut schema_builder = SchemaBuilder::default();
let id_field = schema_builder.add_u64_field("id", INT_INDEXED);
let multiples_field = schema_builder.add_u64_field("multiples", INT_INDEXED);
let multiples_field = schema_builder.add_u64_field("multiples", INT_INDEXED);
let schema = schema_builder.build();
let index = Index::create_from_tempdir(schema).unwrap();
@@ -41,14 +41,11 @@ fn test_indexing() {
let searcher = index.searcher();
// check that everything is correct.
check_index_content(&searcher, &committed_docs);
}
else {
if committed_docs.remove(&random_val) ||
uncommitted_docs.remove(&random_val) {
} else {
if committed_docs.remove(&random_val) || uncommitted_docs.remove(&random_val) {
let doc_id_term = Term::from_field_u64(id_field, random_val);
index_writer.delete_term(doc_id_term);
}
else {
} else {
uncommitted_docs.insert(random_val);
let mut doc = Document::new();
doc.add_u64(id_field, random_val);

116
src/indexer/delete_queue.rs
View File

@@ -6,17 +6,17 @@ use std::ops::DerefMut;
// The DeleteQueue is similar in conceptually to a multiple
// consumer single producer broadcast channel.
//
//
// All consumer will receive all messages.
//
//
// Consumer of the delete queue are holding a `DeleteCursor`,
// which points to a specific place of the `DeleteQueue`.
//
//
// New consumer can be created in two ways
// - calling `delete_queue.cursor()` returns a cursor, that
// - calling `delete_queue.cursor()` returns a cursor, that
// will include all future delete operation (and no past operations).
// - cloning an existing cursor returns a new cursor, that
// is at the exact same position, and can now advance independantly
// is at the exact same position, and can now advance independantly
// from the original cursor.
#[derive(Default)]
struct InnerDeleteQueue {
@@ -31,32 +31,27 @@ pub struct DeleteQueue {
impl DeleteQueue {
// Creates a new delete queue.
pub fn new() -> DeleteQueue {
let delete_queue = DeleteQueue {
inner: Arc::default(),
};
let delete_queue = DeleteQueue { inner: Arc::default() };
let next_block = NextBlock::from(delete_queue.clone());
{
let mut delete_queue_wlock = delete_queue.inner.write().unwrap();
delete_queue_wlock.last_block = Some(
Arc::new(Block {
operations: Arc::default(),
next: next_block,
})
);
delete_queue_wlock.last_block = Some(Arc::new(Block {
operations: Arc::default(),
next: next_block,
}));
}
delete_queue
}
// Creates a new cursor that makes it possible to
// Creates a new cursor that makes it possible to
// consume future delete operations.
//
//
// Past delete operations are not accessible.
pub fn cursor(&self) -> DeleteCursor {
let last_block = self.inner
@@ -85,40 +80,37 @@ impl DeleteQueue {
// DeleteQueue is a linked list of blocks of
// delete operations.
//
//
// Writing happens by simply appending to a vec.
// `.flush()` takes this pending delete operations vec
// creates a new read-only block from it,
// creates a new read-only block from it,
// and appends it to the linked list.
//
// `.flush()` happens when, for instance,
//
// `.flush()` happens when, for instance,
// a consumer reaches the last read-only operations.
// It then ask the delete queue if there happen to
// It then ask the delete queue if there happen to
// be some unflushed operations.
//
fn flush(&self) -> Option<Arc<Block>> {
let mut self_wlock = self
.inner
let mut self_wlock = self.inner
.write()
.expect("Failed to acquire write lock on delete queue writer");
let delete_operations;
{
let writer: &mut Vec<DeleteOperation> = &mut self_wlock.writer;
if writer.is_empty() {
return None;
}
delete_operations = mem::replace(writer, vec!());
delete_operations = mem::replace(writer, vec![]);
}
let next_block = NextBlock::from(self.clone());
{
self_wlock.last_block = Some(
Arc::new(Block {
operations: Arc::new(delete_operations),
next: next_block,
})
);
self_wlock.last_block = Some(Arc::new(Block {
operations: Arc::new(delete_operations),
next: next_block,
}));
}
self_wlock.last_block.clone()
}
@@ -137,17 +129,14 @@ impl From<DeleteQueue> for NextBlock {
}
}
impl NextBlock {
impl NextBlock {
fn next_block(&self) -> Option<Arc<Block>> {
{
let next_read_lock = self.0
.read()
.expect("Failed to acquire write lock in delete queue");
match *next_read_lock {
InnerNextBlock::Closed(ref block) => {
return Some(block.clone());
}
_ => {}
if let InnerNextBlock::Closed(ref block) = *next_read_lock {
return Some(block.clone());
}
}
let next_block;
@@ -171,7 +160,7 @@ impl NextBlock {
}
}
*next_write_lock.deref_mut() = InnerNextBlock::Closed(next_block.clone());
return Some(next_block)
Some(next_block)
}
}
}
@@ -189,32 +178,31 @@ pub struct DeleteCursor {
}
impl DeleteCursor {
impl DeleteCursor {
/// Skips operations and position it so that
/// - either all of the delete operation currently in the
/// - either all of the delete operation currently in the
/// 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) {
// TODO Can be optimize as we work with block.
#[cfg_attr(feature = "cargo-clippy", allow(while_let_loop))]
loop {
if let Some(operation) = self.get() {
if operation.opstamp >= target_opstamp {
break;
}
}
else {
} else {
break;
}
self.advance();
}
}
/// If the current block has been entirely
/// If the current block has been entirely
/// consumed, try to load the next one.
///
/// Return `true`, if after this attempt,
///
/// Return `true`, if after this attempt,
/// the cursor is on a block that has not
/// been entirely consumed.
/// Return `false`, if we have reached the end of the queue.
@@ -229,24 +217,20 @@ impl DeleteCursor {
self.pos = 0;
true
}
None => {
false
}
None => false,
}
}
else {
} else {
true
}
}
/// Advance to the next delete operation.
/// Returns true iff there is such an operation.
pub fn advance(&mut self) -> bool {
if self.load_block_if_required() {
self.pos += 1;
true
}
else {
} else {
false
}
}
@@ -256,12 +240,10 @@ impl DeleteCursor {
pub fn get(&mut self) -> Option<&DeleteOperation> {
if self.load_block_if_required() {
Some(&self.block.operations[self.pos])
}
else {
} else {
None
}
}
}
@@ -278,12 +260,12 @@ mod tests {
#[test]
fn test_deletequeue() {
let delete_queue = DeleteQueue::new();
let make_op = |i: usize| {
let field = Field(1u32);
DeleteOperation {
opstamp: i as u64,
term: Term::from_field_u64(field, i as u64)
term: Term::from_field_u64(field, i as u64),
}
};
@@ -299,7 +281,7 @@ mod tests {
operations_it.advance();
assert!(operations_it.get().is_none());
operations_it.advance();
let mut snapshot2 = delete_queue.cursor();
assert!(snapshot2.get().is_none());
delete_queue.push(make_op(3));
@@ -310,7 +292,7 @@ mod tests {
assert!(operations_it.get().is_none());
operations_it.advance();
}
{
{
let mut operations_it = snapshot.clone();
assert_eq!(operations_it.get().unwrap().opstamp, 1);
operations_it.advance();
@@ -320,6 +302,6 @@ mod tests {
operations_it.advance();
assert!(operations_it.get().is_none());
}
}
}
}

2
src/indexer/directory_lock.rs
View File

@@ -26,4 +26,4 @@ impl Drop for DirectoryLock {
error!("Failed to remove the lock file. {:?}", e);
}
}
}
}

23
src/indexer/doc_opstamp_mapping.rs
View File

@@ -4,7 +4,7 @@ use DocId;
// Doc to opstamp is used to identify which
// document should be deleted.
//
//
// Since the docset matching the query of a delete operation
// is not computed right when the delete operation is received,
// we need to find a way to evaluate, for each document,
@@ -14,13 +14,13 @@ use DocId;
//
// The doc to opstamp mapping stores precisely an array
// indexed by doc id and storing the opstamp of the document.
//
//
// This mapping is (for the moment) stricly increasing
// because of the way document id are allocated.
#[derive(Clone)]
pub enum DocToOpstampMapping {
WithMap(Arc<Vec<u64>>),
None
None,
}
impl From<Vec<u64>> for DocToOpstampMapping {
@@ -30,9 +30,8 @@ impl From<Vec<u64>> for DocToOpstampMapping {
}
impl DocToOpstampMapping {
/// Given an opstamp return the limit doc id L
/// such that all doc id D such that
/// such that all doc id D such that
// D >= L iff opstamp(D) >= than `target_opstamp`.
//
// The edge case opstamp = some doc opstamp is in practise
@@ -41,8 +40,7 @@ impl DocToOpstampMapping {
match *self {
DocToOpstampMapping::WithMap(ref doc_opstamps) => {
match doc_opstamps.binary_search(&target_opstamp) {
Ok(doc_id) => doc_id as DocId,
Err(doc_id) => doc_id as DocId,
Ok(doc_id) | Err(doc_id) => doc_id as DocId,
}
}
DocToOpstampMapping::None => DocId::max_value(),
@@ -58,23 +56,24 @@ mod tests {
#[test]
fn test_doc_to_opstamp_mapping_none() {
let doc_to_opstamp_mapping = DocToOpstampMapping::None;
assert_eq!(doc_to_opstamp_mapping.compute_doc_limit(1), u32::max_value());
assert_eq!(doc_to_opstamp_mapping.compute_doc_limit(1),
u32::max_value());
}
#[test]
fn test_doc_to_opstamp_mapping_complex() {
{
let doc_to_opstamp_mapping = DocToOpstampMapping::from(vec!());
let doc_to_opstamp_mapping = DocToOpstampMapping::from(vec![]);
assert_eq!(doc_to_opstamp_mapping.compute_doc_limit(0u64), 0);
assert_eq!(doc_to_opstamp_mapping.compute_doc_limit(2u64), 0);
}
{
let doc_to_opstamp_mapping = DocToOpstampMapping::from(vec!(1u64));
let doc_to_opstamp_mapping = DocToOpstampMapping::from(vec![1u64]);
assert_eq!(doc_to_opstamp_mapping.compute_doc_limit(0u64), 0);
assert_eq!(doc_to_opstamp_mapping.compute_doc_limit(2u64), 1);
}
{
let doc_to_opstamp_mapping = DocToOpstampMapping::from(vec!(1u64, 12u64, 17u64, 23u64));
let doc_to_opstamp_mapping = DocToOpstampMapping::from(vec![1u64, 12u64, 17u64, 23u64]);
assert_eq!(doc_to_opstamp_mapping.compute_doc_limit(0u64), 0);
for i in 2u64..13u64 {
assert_eq!(doc_to_opstamp_mapping.compute_doc_limit(i), 1);
@@ -90,4 +89,4 @@ mod tests {
}
}
}
}
}

273
src/indexer/index_writer.rs
View File

@@ -27,7 +27,7 @@ use schema::Document;
use schema::Schema;
use schema::Term;
use std::mem;
use std::mem::swap;
use std::mem::swap;
use std::thread::JoinHandle;
use super::directory_lock::DirectoryLock;
use super::operation::AddOperation;
@@ -41,7 +41,8 @@ pub const MARGIN_IN_BYTES: u32 = 1_000_000u32;
// We impose the memory per thread to be at least 3 MB.
pub const HEAP_SIZE_LIMIT: u32 = MARGIN_IN_BYTES * 3u32;
// Add document will block if the number of docs waiting in the queue to be indexed reaches PIPELINE_MAX_SIZE_IN_DOCS
// Add document will block if the number of docs waiting in the queue to be indexed
// reaches `PIPELINE_MAX_SIZE_IN_DOCS`
const PIPELINE_MAX_SIZE_IN_DOCS: usize = 10_000;
type DocumentSender = chan::Sender<AddOperation>;
@@ -54,11 +55,10 @@ type DocumentReceiver = chan::Receiver<AddOperation>;
/// Each indexing thread builds its own independant `Segment`, via
/// a `SegmentWriter` object.
pub struct IndexWriter {
// the lock is just used to bind the
// the lock is just used to bind the
// lifetime of the lock with that of the IndexWriter.
_directory_lock: DirectoryLock,
_directory_lock: DirectoryLock,
index: Index,
heap_size_in_bytes_per_thread: usize,
@@ -96,42 +96,40 @@ impl !Sync for IndexWriter {}
/// `IndexWriter` on the system is accessing the index directory,
/// it is safe to manually delete the lockfile.
///
/// num_threads specifies the number of indexing workers that
/// `num_threads` specifies the number of indexing workers that
/// should work at the same time.
/// # Errors
/// If the lockfile already exists, returns `Error::FileAlreadyExists`.
/// # Panics
/// If the heap size per thread is too small, panics.
pub fn open_index_writer(
index: &Index,
num_threads: usize,
heap_size_in_bytes_per_thread: usize) -> Result<IndexWriter> {
pub fn open_index_writer(index: &Index,
num_threads: usize,
heap_size_in_bytes_per_thread: usize)
-> Result<IndexWriter> {
if heap_size_in_bytes_per_thread <= HEAP_SIZE_LIMIT as usize {
panic!(format!("The heap size per thread needs to be at least {}.",
HEAP_SIZE_LIMIT));
HEAP_SIZE_LIMIT));
}
let directory_lock = try!(DirectoryLock::lock(index.directory().box_clone()));
let (document_sender, document_receiver): (DocumentSender, DocumentReceiver) =
chan::sync(PIPELINE_MAX_SIZE_IN_DOCS);
let delete_queue = DeleteQueue::new();
let current_opstamp = index.opstamp();
let stamper = Stamper::new(current_opstamp);
let segment_updater = SegmentUpdater::new(index.clone(),
stamper.clone(),
delete_queue.cursor())?;
let segment_updater =
SegmentUpdater::new(index.clone(), stamper.clone(), delete_queue.cursor())?;
let mut index_writer = IndexWriter {
_directory_lock: directory_lock,
heap_size_in_bytes_per_thread: heap_size_in_bytes_per_thread,
index: index.clone(),
@@ -140,7 +138,7 @@ pub fn open_index_writer(
segment_updater: segment_updater,
workers_join_handle: vec!(),
workers_join_handle: vec![],
num_threads: num_threads,
delete_queue: delete_queue,
@@ -158,28 +156,29 @@ pub fn open_index_writer(
pub fn compute_deleted_bitset(
delete_bitset: &mut BitSet,
segment_reader: &SegmentReader,
delete_cursor: &mut DeleteCursor,
doc_opstamps: DocToOpstampMapping,
target_opstamp: u64) -> Result<bool> {
pub fn compute_deleted_bitset(delete_bitset: &mut BitSet,
segment_reader: &SegmentReader,
delete_cursor: &mut DeleteCursor,
doc_opstamps: &DocToOpstampMapping,
target_opstamp: u64)
-> Result<bool> {
let mut might_have_changed = false;
#[cfg_attr(feature = "cargo-clippy", allow(while_let_loop))]
loop {
if let Some(delete_op) = delete_cursor.get() {
if delete_op.opstamp > target_opstamp {
break;
}
else {
} else {
// A delete operation should only affect
// document that were inserted after it.
//
//
// Limit doc helps identify the first document
// that may be affected by the delete operation.
let limit_doc = doc_opstamps.compute_doc_limit(delete_op.opstamp);
if let Some(mut docset) = segment_reader.read_postings(&delete_op.term, SegmentPostingsOption::NoFreq) {
if let Some(mut docset) =
segment_reader.read_postings(&delete_op.term, SegmentPostingsOption::NoFreq) {
while docset.advance() {
let deleted_doc = docset.doc();
if deleted_doc < limit_doc {
@@ -189,8 +188,7 @@ pub fn compute_deleted_bitset(
}
}
}
}
else {
} else {
break;
}
delete_cursor.advance();
@@ -200,10 +198,10 @@ pub fn compute_deleted_bitset(
/// Advance delete for the given segment up
/// to the target opstamp.
pub fn advance_deletes(
mut segment: Segment,
segment_entry: &mut SegmentEntry,
target_opstamp: u64) -> Result<Option<FileProtection>> {
pub fn advance_deletes(mut segment: Segment,
segment_entry: &mut SegmentEntry,
target_opstamp: u64)
-> Result<Option<FileProtection>> {
let mut file_protect: Option<FileProtection> = None;
@@ -216,24 +214,20 @@ pub fn advance_deletes(
}
let segment_reader = SegmentReader::open(segment.clone())?;
let max_doc = segment_reader.max_doc();
let mut delete_bitset: BitSet =
match segment_entry.delete_bitset() {
Some(ref previous_delete_bitset) =>
(*previous_delete_bitset).clone(),
None =>
BitSet::with_capacity(max_doc as usize)
};
let mut delete_bitset: BitSet = match segment_entry.delete_bitset() {
Some(previous_delete_bitset) => (*previous_delete_bitset).clone(),
None => BitSet::with_capacity(max_doc as usize),
};
let delete_cursor = segment_entry.delete_cursor();
compute_deleted_bitset(
&mut delete_bitset,
&segment_reader,
delete_cursor,
DocToOpstampMapping::None,
target_opstamp)?;
compute_deleted_bitset(&mut delete_bitset,
&segment_reader,
delete_cursor,
&DocToOpstampMapping::None,
target_opstamp)?;
for doc in 0u32..max_doc {
if segment_reader.is_deleted(doc) {
delete_bitset.insert(doc as usize);
@@ -257,15 +251,15 @@ fn index_documents(heap: &mut Heap,
segment: Segment,
schema: &Schema,
generation: usize,
document_iterator: &mut Iterator<Item=AddOperation>,
document_iterator: &mut Iterator<Item = AddOperation>,
segment_updater: &mut SegmentUpdater,
mut delete_cursor: DeleteCursor)
-> Result<bool> {
heap.clear();
let segment_id = segment.id();
let mut segment_writer = SegmentWriter::for_segment(heap, segment.clone(), &schema)?;
let mut segment_writer = SegmentWriter::for_segment(heap, segment.clone(), schema)?;
for doc in document_iterator {
try!(segment_writer.add_document(&doc, &schema));
try!(segment_writer.add_document(&doc, schema));
// There is two possible conditions to close the segment.
// One is the memory arena dedicated to the segment is
// getting full.
@@ -286,10 +280,10 @@ fn index_documents(heap: &mut Heap,
}
}
let num_docs = segment_writer.max_doc();
// this is ensured by the call to peek before starting
// the worker thread.
assert!(num_docs > 0);
assert!(num_docs > 0);
let doc_opstamps: Vec<u64> = segment_writer.finalize()?;
@@ -297,59 +291,55 @@ fn index_documents(heap: &mut Heap,
segment_meta.set_max_doc(num_docs);
let last_docstamp: u64 = *(doc_opstamps.last().unwrap());
let doc_to_opstamps = DocToOpstampMapping::from(doc_opstamps);
let segment_reader = SegmentReader::open(segment)?;
let mut deleted_bitset = BitSet::with_capacity(num_docs as usize);
let may_have_deletes = compute_deleted_bitset(
&mut deleted_bitset,
&segment_reader,
&mut delete_cursor,
doc_to_opstamps,
last_docstamp,
)?;
let may_have_deletes = compute_deleted_bitset(&mut deleted_bitset,
&segment_reader,
&mut delete_cursor,
&doc_to_opstamps,
last_docstamp)?;
let segment_entry = SegmentEntry::new(
segment_meta,
delete_cursor,
{ if may_have_deletes { Some(deleted_bitset) }
else { None } }
);
Ok(
segment_updater
.add_segment(generation, segment_entry)
)
}
let segment_entry = SegmentEntry::new(segment_meta, delete_cursor, {
if may_have_deletes {
Some(deleted_bitset)
} else {
None
}
});
Ok(segment_updater.add_segment(generation, segment_entry))
}
impl IndexWriter {
/// The index writer
pub fn wait_merging_threads(mut self) -> Result<()> {
// this will stop the indexing thread,
// dropping the last reference to the segment_updater.
drop(self.document_sender);
let former_workers_handles = mem::replace(&mut self.workers_join_handle, vec!());
let former_workers_handles = mem::replace(&mut self.workers_join_handle, vec![]);
for join_handle in former_workers_handles {
try!(join_handle.join()
.expect("Indexing Worker thread panicked")
.map_err(|e| {
Error::ErrorInThread(format!("Error in indexing worker thread. {:?}", e))
}));
try!(join_handle
.join()
.expect("Indexing Worker thread panicked")
.map_err(|e| {
Error::ErrorInThread(format!("Error in indexing worker thread. {:?}", e))
}));
}
drop(self.workers_join_handle);
let result = self.segment_updater
.wait_merging_thread()
.map_err(|_|
Error::ErrorInThread("Failed to join merging thread.".to_string())
);
if let &Err(ref e) = &result {
let result =
self.segment_updater
.wait_merging_thread()
.map_err(|_| Error::ErrorInThread("Failed to join merging thread.".to_string()));
if let Err(ref e) = result {
error!("Some merging thread failed {:?}", e);
}
@@ -364,37 +354,34 @@ impl IndexWriter {
let document_receiver_clone = self.document_receiver.clone();
let mut segment_updater = self.segment_updater.clone();
let mut heap = Heap::with_capacity(self.heap_size_in_bytes_per_thread);
let generation = self.generation;
let mut delete_cursor = self.delete_queue.cursor();
let join_handle: JoinHandle<Result<()>> =
thread::Builder::new()
let join_handle: JoinHandle<Result<()>> = thread::Builder::new()
.name(format!("indexing thread {} for gen {}", self.worker_id, generation))
.spawn(move || {
loop {
let mut document_iterator = document_receiver_clone.clone()
.into_iter()
.peekable();
let mut document_iterator =
document_receiver_clone.clone().into_iter().peekable();
// the peeking here is to avoid
// creating a new segment's files
// if no document are available.
//
// this is a valid guarantee as the
// this is a valid guarantee as the
// peeked document now belongs to
// our local iterator.
if let Some(operation) = document_iterator.peek() {
delete_cursor.skip_to(operation.opstamp);
}
else {
} else {
// No more documents.
// Happens when there is a commit, or if the `IndexWriter`
// was dropped.
return Ok(())
return Ok(());
}
let segment = segment_updater.new_segment();
index_documents(&mut heap,
@@ -404,7 +391,7 @@ impl IndexWriter {
&mut document_iterator,
&mut segment_updater,
delete_cursor.clone())?;
}
})?;
self.worker_id += 1;
@@ -421,7 +408,7 @@ impl IndexWriter {
pub fn set_merge_policy(&self, merge_policy: Box<MergePolicy>) {
self.segment_updater.set_merge_policy(merge_policy);
}
fn start_workers(&mut self) -> Result<()> {
for _ in 0..self.num_threads {
try!(self.add_indexing_worker());
@@ -436,7 +423,9 @@ impl IndexWriter {
}
/// Merges a given list of segments
pub fn merge(&mut self, segment_ids: &[SegmentId]) -> impl Future<Item=SegmentMeta, Error=Canceled> {
pub fn merge(&mut self,
segment_ids: &[SegmentId])
-> impl Future<Item = SegmentMeta, Error = Canceled> {
self.segment_updater.start_merge(segment_ids)
}
@@ -449,7 +438,8 @@ impl IndexWriter {
///
/// Returns the former segment_ready channel.
fn recreate_document_channel(&mut self) -> DocumentReceiver {
let (mut document_sender, mut document_receiver): (DocumentSender, DocumentReceiver) =
let (mut document_sender, mut document_receiver): (DocumentSender,
DocumentReceiver) =
chan::sync(PIPELINE_MAX_SIZE_IN_DOCS);
swap(&mut self.document_sender, &mut document_sender);
swap(&mut self.document_receiver, &mut document_receiver);
@@ -477,12 +467,9 @@ impl IndexWriter {
let heap_size_in_bytes_per_thread = self.heap_size_in_bytes_per_thread;
drop(self);
for _ in receiver_clone {}
let index_writer = open_index_writer(
&index,
num_threads,
heap_size_in_bytes_per_thread)?;
let index_writer = open_index_writer(&index, num_threads, heap_size_in_bytes_per_thread)?;
Ok(index_writer)
}
@@ -524,23 +511,24 @@ impl IndexWriter {
let mut former_workers_join_handle = Vec::new();
swap(&mut former_workers_join_handle,
&mut self.workers_join_handle);
for worker_handle in former_workers_join_handle {
let indexing_worker_result = try!(worker_handle.join()
.map_err(|e| Error::ErrorInThread(format!("{:?}", e))));
let indexing_worker_result =
try!(worker_handle
.join()
.map_err(|e| Error::ErrorInThread(format!("{:?}", e))));
try!(indexing_worker_result);
// add a new worker for the next generation.
try!(self.add_indexing_worker());
}
// wait for the segment update thread to have processed the info
self.segment_updater
.commit(self.committed_opstamp)?;
self.segment_updater.commit(self.committed_opstamp)?;
Ok(self.committed_opstamp)
}
}
/// Delete all documents containing a given term.
///
@@ -548,7 +536,7 @@ impl IndexWriter {
/// were added in previous commits, and documents
/// that were added previously in the same commit.
///
/// Like adds, the deletion itself will be visible
/// Like adds, the deletion itself will be visible
/// only after calling `commit()`.
pub fn delete_term(&mut self, term: Term) -> u64 {
let opstamp = self.stamper.stamp();
@@ -561,7 +549,7 @@ impl IndexWriter {
}
/// Returns the opstamp of the last successful commit.
///
///
/// This is, for instance, the opstamp the index will
/// rollback to if there is a failure like a power surge.
///
@@ -615,16 +603,19 @@ mod tests {
_ => panic!("Expected FileAlreadyExists error"),
}
}
#[test]
fn test_set_merge_policy() {
let schema_builder = schema::SchemaBuilder::default();
let index = Index::create_in_ram(schema_builder.build());
let index_writer = index.writer(40_000_000).unwrap();
assert_eq!(format!("{:?}", index_writer.get_merge_policy()), "LogMergePolicy { min_merge_size: 8, min_layer_size: 10000, level_log_size: 0.75 }");
assert_eq!(format!("{:?}", index_writer.get_merge_policy()),
"LogMergePolicy { min_merge_size: 8, min_layer_size: 10000, \
level_log_size: 0.75 }");
let merge_policy = box NoMergePolicy::default();
index_writer.set_merge_policy(merge_policy);
assert_eq!(format!("{:?}", index_writer.get_merge_policy()), "NoMergePolicy");
assert_eq!(format!("{:?}", index_writer.get_merge_policy()),
"NoMergePolicy");
}
#[test]
@@ -633,12 +624,12 @@ mod tests {
let index = Index::create_in_ram(schema_builder.build());
{
let _index_writer = index.writer(40_000_000).unwrap();
// the lock should be released when the
// the lock should be released when the
// index_writer leaves the scope.
}
let _index_writer_two = index.writer(40_000_000).unwrap();
}
#[test]
fn test_commit_and_rollback() {
let mut schema_builder = schema::SchemaBuilder::default();
@@ -661,7 +652,7 @@ mod tests {
}
index_writer = index_writer.rollback().unwrap();
assert_eq!(index_writer.commit_opstamp(), 0u64);
assert_eq!(num_docs_containing("a"), 0);
@@ -714,12 +705,14 @@ mod tests {
}
// this should create 8 segments and trigger a merge.
index_writer.commit().expect("commit failed");
index_writer.wait_merging_threads().expect("waiting merging thread failed");
index_writer
.wait_merging_threads()
.expect("waiting merging thread failed");
index.load_searchers().unwrap();
assert_eq!(num_docs_containing("a"), 200);
assert!(index.searchable_segments().unwrap().len() < 8);
}
}

22
src/indexer/log_merge_policy.rs
View File

@@ -9,7 +9,7 @@ const DEFAULT_MIN_LAYER_SIZE: u32 = 10_000;
const DEFAULT_MIN_MERGE_SIZE: usize = 8;
/// LogMergePolicy tries tries to merge segments that have a similar number of
/// `LogMergePolicy` tries tries to merge segments that have a similar number of
/// documents.
#[derive(Debug, Clone)]
pub struct LogMergePolicy {
@@ -24,7 +24,7 @@ impl LogMergePolicy {
}
/// Set the minimum number of segment that may be merge together.
pub fn set_min_merge_size(&mut self, min_merge_size: usize) {
pub fn set_min_merge_size(&mut self, min_merge_size: usize) {
self.min_merge_size = min_merge_size;
}
@@ -52,14 +52,16 @@ impl MergePolicy for LogMergePolicy {
return Vec::new();
}
let mut size_sorted_tuples = segments.iter()
let mut size_sorted_tuples = segments
.iter()
.map(|x| x.num_docs())
.enumerate()
.collect::<Vec<(usize, u32)>>();
size_sorted_tuples.sort_by(|x, y| y.cmp(x));
let size_sorted_log_tuples: Vec<_> = size_sorted_tuples.into_iter()
let size_sorted_log_tuples: Vec<_> = size_sorted_tuples
.into_iter()
.map(|(ind, num_docs)| (ind, (self.clip_min_size(num_docs) as f64).log2()))
.collect();
@@ -77,14 +79,10 @@ impl MergePolicy for LogMergePolicy {
levels
.iter()
.filter(|level| level.len() >= self.min_merge_size)
.map(|ind_vec| {
MergeCandidate(ind_vec.iter()
.map(|&ind| segments[ind].id())
.collect())
})
.map(|ind_vec| MergeCandidate(ind_vec.iter().map(|&ind| segments[ind].id()).collect()))
.collect()
}
fn box_clone(&self) -> Box<MergePolicy> {
box self.clone()
}
@@ -128,9 +126,7 @@ mod tests {
#[test]
fn test_log_merge_policy_pair() {
let test_input = vec![seg_meta(10),
seg_meta(10),
seg_meta(10)];
let test_input = vec![seg_meta(10), seg_meta(10), seg_meta(10)];
let result_list = test_merge_policy().compute_merge_candidates(&test_input);
assert_eq!(result_list.len(), 1);
}

29
src/indexer/merge_policy.rs
View File

@@ -4,26 +4,26 @@ use std::marker;
use std::fmt::Debug;
/// Set of segment suggested for a merge.
/// Set of segment suggested for a merge.
#[derive(Debug, Clone)]
pub struct MergeCandidate(pub Vec<SegmentId>);
/// The Merge policy defines which segments should be merged.
///
/// The `MergePolicy` defines which segments should be merged.
///
/// Every time a the list of segments changes, the segment updater
/// asks the merge policy if some segments should be merged.
pub trait MergePolicy: marker::Send + marker::Sync + Debug {
/// Given the list of segment metas, returns the list of merge candidates.
/// Given the list of segment metas, returns the list of merge candidates.
///
/// This call happens on the segment updater thread, and will block
/// other segment updates, so all implementations should happen rapidly.
/// This call happens on the segment updater thread, and will block
/// other segment updates, so all implementations should happen rapidly.
fn compute_merge_candidates(&self, segments: &[SegmentMeta]) -> Vec<MergeCandidate>;
/// Returns a boxed clone of the MergePolicy.
fn box_clone(&self) -> Box<MergePolicy>;
}
/// Never merge segments.
/// Never merge segments.
#[derive(Debug)]
pub struct NoMergePolicy;
@@ -37,7 +37,7 @@ impl MergePolicy for NoMergePolicy {
fn compute_merge_candidates(&self, _segments: &[SegmentMeta]) -> Vec<MergeCandidate> {
Vec::new()
}
fn box_clone(&self) -> Box<MergePolicy> {
box NoMergePolicy
}
@@ -52,7 +52,7 @@ pub mod tests {
use core::SegmentMeta;
/// Merge policy useful for test purposes.
/// `MergePolicy` useful for test purposes.
///
/// Everytime there is more than one segment,
/// it will suggest to merge them.
@@ -66,15 +66,14 @@ pub mod tests {
.map(|segment_meta| segment_meta.id())
.collect::<Vec<SegmentId>>();
if segment_ids.len() > 1 {
vec!(MergeCandidate(segment_ids))
}
else {
vec!()
vec![MergeCandidate(segment_ids)]
} else {
vec![]
}
}
fn box_clone(&self) -> Box<MergePolicy> {
box MergeWheneverPossible
}
}
}
}

402
src/indexer/merger.rs
View File

@@ -31,9 +31,7 @@ struct DeltaPositionComputer {
impl DeltaPositionComputer {
fn new() -> DeltaPositionComputer {
DeltaPositionComputer {
buffer: vec![0u32; 512]
}
DeltaPositionComputer { buffer: vec![0u32; 512] }
}
fn compute_delta_positions(&mut self, positions: &[u32]) -> &[u32] {
@@ -50,16 +48,17 @@ impl DeltaPositionComputer {
}
fn compute_min_max_val(u64_reader: &U64FastFieldReader, max_doc: DocId, delete_bitset: &DeleteBitSet) -> Option<(u64, u64)> {
fn compute_min_max_val(u64_reader: &U64FastFieldReader,
max_doc: DocId,
delete_bitset: &DeleteBitSet)
-> Option<(u64, u64)> {
if max_doc == 0 {
None
}
else if !delete_bitset.has_deletes() {
// no deleted documents,
} else if !delete_bitset.has_deletes() {
// no deleted documents,
// we can use the previous min_val, max_val.
Some((u64_reader.min_value(), u64_reader.max_value()))
}
else {
} else {
// some deleted documents,
// we need to recompute the max / min
(0..max_doc)
@@ -70,19 +69,21 @@ fn compute_min_max_val(u64_reader: &U64FastFieldReader, max_doc: DocId, delete_b
}
}
fn extract_fieldnorm_reader(segment_reader: &SegmentReader, field: Field) -> Option<U64FastFieldReader> {
fn extract_fieldnorm_reader(segment_reader: &SegmentReader,
field: Field)
-> Option<U64FastFieldReader> {
segment_reader.get_fieldnorms_reader(field)
}
fn extract_fast_field_reader(segment_reader: &SegmentReader, field: Field) -> Option<U64FastFieldReader> {
segment_reader
.fast_fields_reader()
.open_reader(field)
fn extract_fast_field_reader(segment_reader: &SegmentReader,
field: Field)
-> Option<U64FastFieldReader> {
segment_reader.fast_fields_reader().open_reader(field)
}
impl IndexMerger {
pub fn open(schema: Schema, segments: &[Segment]) -> Result<IndexMerger> {
let mut readers = vec!();
let mut readers = vec![];
let mut max_doc: u32 = 0u32;
for segment in segments {
if segment.meta().num_docs() > 0 {
@@ -92,65 +93,75 @@ impl IndexMerger {
}
}
Ok(IndexMerger {
schema: schema,
readers: readers,
max_doc: max_doc,
})
schema: schema,
readers: readers,
max_doc: max_doc,
})
}
fn write_fieldnorms(&self,
fast_field_serializer: &mut FastFieldSerializer) -> Result<()> {
fn write_fieldnorms(&self, fast_field_serializer: &mut FastFieldSerializer) -> Result<()> {
let fieldnorm_fastfields: Vec<Field> = self.schema
.fields()
.iter()
.enumerate()
.filter(|&(_, field_entry)| field_entry.is_indexed())
.map(|(field_id, _)| Field(field_id as u32))
.collect();
self.generic_write_fast_field(fieldnorm_fastfields, &extract_fieldnorm_reader, fast_field_serializer)
.fields()
.iter()
.enumerate()
.filter(|&(_, field_entry)| field_entry.is_indexed())
.map(|(field_id, _)| Field(field_id as u32))
.collect();
self.generic_write_fast_field(fieldnorm_fastfields,
&extract_fieldnorm_reader,
fast_field_serializer)
}
fn write_fast_fields(&self, fast_field_serializer: &mut FastFieldSerializer) -> Result<()> {
let fast_fields: Vec<Field> = self.schema
.fields()
.iter()
.enumerate()
.filter(|&(_, field_entry)| field_entry.is_int_fast())
.map(|(field_id, _)| Field(field_id as u32))
.collect();
self.generic_write_fast_field(fast_fields, &extract_fast_field_reader, fast_field_serializer)
.fields()
.iter()
.enumerate()
.filter(|&(_, field_entry)| field_entry.is_int_fast())
.map(|(field_id, _)| Field(field_id as u32))
.collect();
self.generic_write_fast_field(fast_fields,
&extract_fast_field_reader,
fast_field_serializer)
}
// used both to merge field norms and regular u64 fast fields.
fn generic_write_fast_field(&self,
fields: Vec<Field>,
field_reader_extractor: &Fn(&SegmentReader, Field) -> Option<U64FastFieldReader>,
fast_field_serializer: &mut FastFieldSerializer) -> Result<()> {
fields: Vec<Field>,
field_reader_extractor: &Fn(&SegmentReader, Field)
-> Option<U64FastFieldReader>,
fast_field_serializer: &mut FastFieldSerializer)
-> Result<()> {
for field in fields {
let mut u64_readers = vec!();
let mut u64_readers = vec![];
let mut min_val = u64::max_value();
let mut max_val = u64::min_value();
for reader in &self.readers {
match field_reader_extractor(reader, field) {
Some(u64_reader) => {
if let Some((seg_min_val, seg_max_val)) = compute_min_max_val(&u64_reader, reader.max_doc(), reader.delete_bitset()) {
if let Some((seg_min_val, seg_max_val)) =
compute_min_max_val(&u64_reader,
reader.max_doc(),
reader.delete_bitset()) {
// the segment has some non-deleted documents
min_val = min(min_val, seg_min_val);
max_val = max(max_val, seg_max_val);
u64_readers.push((reader.max_doc(), u64_reader, reader.delete_bitset()));
}
u64_readers
.push((reader.max_doc(), u64_reader, reader.delete_bitset()));
}
}
None => {
let error_msg = format!("Failed to find a u64_reader for field {:?}", field);
let error_msg = format!("Failed to find a u64_reader for field {:?}",
field);
error!("{}", error_msg);
return Err(Error::SchemaError(error_msg))
return Err(Error::SchemaError(error_msg));
}
}
}
if u64_readers.is_empty() {
@@ -160,7 +171,7 @@ impl IndexMerger {
}
assert!(min_val <= max_val);
try!(fast_field_serializer.new_u64_fast_field(field, min_val, max_val));
for (max_doc, u64_reader, delete_bitset) in u64_readers {
for doc_id in 0..max_doc {
@@ -176,25 +187,22 @@ impl IndexMerger {
Ok(())
}
fn write_postings(
&self,
postings_serializer: &mut PostingsSerializer) -> Result<()> {
fn write_postings(&self, postings_serializer: &mut PostingsSerializer) -> Result<()> {
let mut merged_terms = TermIterator::from(&self.readers[..]);
let mut delta_position_computer = DeltaPositionComputer::new();
let mut max_doc = 0;
// map from segment doc ids to the resulting merged segment doc id.
let mut merged_doc_id_map: Vec<Vec<Option<DocId>>> = Vec::with_capacity(self.readers.len());
for reader in &self.readers {
let mut segment_local_map = Vec::with_capacity(reader.max_doc() as usize);
for doc_id in 0..reader.max_doc() {
if reader.is_deleted(doc_id) {
segment_local_map.push(None);
}
else {
} else {
segment_local_map.push(Some(max_doc));
max_doc += 1u32;
}
@@ -212,31 +220,34 @@ impl IndexMerger {
// segment are stacked so that :
// - Segment 0's doc ids become doc id [0, seg.max_doc]
// - Segment 1's doc ids become [seg0.max_doc, seg0.max_doc + seg.max_doc]
// - Segment 2's doc ids become [seg0.max_doc + seg1.max_doc, seg0.max_doc + seg1.max_doc + seg2.max_doc]
// - Segment 2's doc ids become [seg0.max_doc + seg1.max_doc,
// seg0.max_doc + seg1.max_doc + seg2.max_doc]
// ...
let term = merged_terms.term();
let mut term_written = false;
let segment_postings = merged_terms
.segment_ords()
.iter()
.cloned()
.flat_map(|segment_ord| {
self.readers[segment_ord]
.read_postings_all_info(&term)
.map(|segment_postings| (segment_ord, segment_postings))
})
.collect::<Vec<_>>();
.segment_ords()
.iter()
.cloned()
.flat_map(|segment_ord| {
self.readers[segment_ord]
.read_postings_all_info(term)
.map(|segment_postings| (segment_ord, segment_postings))
})
.collect::<Vec<_>>();
// We can remove the term if all documents which
// contained it have been deleted.
if segment_postings.len() > 0 {
if !segment_postings.is_empty() {
// We can now serialize this postings, by pushing each document to the
// postings serializer.
for (segment_ord, mut segment_postings) in segment_postings {
let old_to_new_doc_id = &merged_doc_id_map[segment_ord];
while segment_postings.advance() {
if let Some(remapped_doc_id) = old_to_new_doc_id[segment_postings.doc() as usize] {
if let Some(remapped_doc_id) =
old_to_new_doc_id[segment_postings.doc() as usize] {
if !term_written {
let current_field = term.field();
if last_field != Some(current_field) {
@@ -247,14 +258,15 @@ impl IndexMerger {
// we make sure to only write the term iff
// there is at least one document.
postings_serializer.new_term(term.as_slice())?;
term_written = true;
}
let delta_positions: &[u32] =
delta_position_computer.compute_delta_positions(segment_postings.positions());
try!(postings_serializer.write_doc(
remapped_doc_id,
segment_postings.term_freq(),
delta_positions));
delta_position_computer
.compute_delta_positions(segment_postings.positions());
try!(postings_serializer.write_doc(remapped_doc_id,
segment_postings.term_freq(),
delta_positions));
}
}
}
@@ -263,7 +275,7 @@ impl IndexMerger {
try!(postings_serializer.close_term());
}
}
}
Ok(())
}
@@ -274,12 +286,10 @@ impl IndexMerger {
for doc_id in 0..reader.max_doc() {
if !reader.is_deleted(doc_id) {
let doc = try!(store_reader.get(doc_id));
let field_values: Vec<&FieldValue> = doc.field_values()
.iter()
.collect();
let field_values: Vec<&FieldValue> = doc.field_values().iter().collect();
try!(store_writer.store(&field_values));
}
}
}
}
Ok(())
}
@@ -318,8 +328,8 @@ mod tests {
fn test_index_merger_no_deletes() {
let mut schema_builder = schema::SchemaBuilder::default();
let text_fieldtype = schema::TextOptions::default()
.set_indexing_options(TextIndexingOptions::TokenizedWithFreq)
.set_stored();
.set_indexing_options(TextIndexingOptions::TokenizedWithFreq)
.set_stored();
let text_field = schema_builder.add_text_field("text", text_fieldtype);
let score_fieldtype = schema::IntOptions::default().set_fast();
let score_field = schema_builder.add_u64_field("score", score_fieldtype);
@@ -368,11 +378,14 @@ mod tests {
}
}
{
let segment_ids = index.searchable_segment_ids().expect("Searchable segments failed.");
let segment_ids = index
.searchable_segment_ids()
.expect("Searchable segments failed.");
let mut index_writer = index.writer_with_num_threads(1, 40_000_000).unwrap();
index_writer.merge(&segment_ids)
.wait()
.expect("Merging failed");
index_writer
.merge(&segment_ids)
.wait()
.expect("Merging failed");
index_writer.wait_merging_threads().unwrap();
}
{
@@ -386,13 +399,13 @@ mod tests {
};
{
assert_eq!(get_doc_ids(vec![Term::from_field_text(text_field, "a")]),
vec!(1, 2, 4,));
vec![1, 2, 4]);
assert_eq!(get_doc_ids(vec![Term::from_field_text(text_field, "af")]),
vec!(0, 3,));
vec![0, 3]);
assert_eq!(get_doc_ids(vec![Term::from_field_text(text_field, "g")]),
vec!(4,));
vec![4]);
assert_eq!(get_doc_ids(vec![Term::from_field_text(text_field, "b")]),
vec!(0, 1, 2, 3, 4,));
vec![0, 1, 2, 3, 4]);
}
{
let doc = searcher.doc(&DocAddress(0, 0)).unwrap();
@@ -422,12 +435,12 @@ mod tests {
collector.vals()
};
assert_eq!(get_fast_vals(vec![Term::from_field_text(text_field, "a")]),
vec!(5, 7, 13,));
vec![5, 7, 13]);
}
}
}
fn search_term(searcher: &Searcher, term: Term) -> Vec<u64> {
fn search_term(searcher: &Searcher, term: Term) -> Vec<u64> {
let mut collector = FastFieldTestCollector::for_field(Field(1));
let term_query = TermQuery::new(term, SegmentPostingsOption::NoFreq);
searcher.search(&term_query, &mut collector).unwrap();
@@ -437,8 +450,7 @@ mod tests {
#[test]
fn test_index_merger_with_deletes() {
let mut schema_builder = schema::SchemaBuilder::default();
let text_fieldtype = schema::TextOptions
::default()
let text_fieldtype = schema::TextOptions::default()
.set_indexing_options(TextIndexingOptions::TokenizedWithFreq)
.set_stored();
let text_field = schema_builder.add_text_field("text", text_fieldtype);
@@ -448,21 +460,19 @@ mod tests {
let mut index_writer = index.writer_with_num_threads(1, 40_000_000).unwrap();
let empty_vec = Vec::<u64>::new();
{ // a first commit
index_writer.add_document(
doc!(
{
// a first commit
index_writer.add_document(doc!(
text_field => "a b d",
score_field => 1u64
));
index_writer.add_document(
doc!(
index_writer.add_document(doc!(
text_field => "b c",
score_field => 2u64
));
index_writer.delete_term(Term::from_field_text(text_field, "c"));
index_writer.add_document(
doc!(
index_writer.add_document(doc!(
text_field => "c d",
score_field => 3u64
));
@@ -472,31 +482,32 @@ mod tests {
assert_eq!(searcher.num_docs(), 2);
assert_eq!(searcher.segment_readers()[0].num_docs(), 2);
assert_eq!(searcher.segment_readers()[0].max_doc(), 3);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "a")), vec!(1));
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "b")), vec!(1));
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "c")), vec!(3));
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "d")), vec!(1, 3));
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "a")),
vec![1]);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "b")),
vec![1]);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "c")),
vec![3]);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "d")),
vec![1, 3]);
}
{ // a second commit
index_writer.add_document(
doc!(
{
// a second commit
index_writer.add_document(doc!(
text_field => "a d e",
score_field => 4_000u64
));
index_writer.add_document(
doc!(
index_writer.add_document(doc!(
text_field => "e f",
score_field => 5_000u64
));
index_writer.delete_term(Term::from_field_text(text_field, "a"));
index_writer.delete_term(Term::from_field_text(text_field, "f"));
index_writer.add_document(
doc!(
index_writer.add_document(doc!(
text_field => "f g",
score_field => 6_000u64
));
index_writer.add_document(
doc!(
index_writer.add_document(doc!(
text_field => "g h",
score_field => 7_000u64
));
@@ -510,71 +521,112 @@ mod tests {
assert_eq!(searcher.segment_readers()[0].max_doc(), 3);
assert_eq!(searcher.segment_readers()[1].num_docs(), 2);
assert_eq!(searcher.segment_readers()[1].max_doc(), 4);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "a")), empty_vec);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "b")), empty_vec);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "c")), vec!(3));
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "d")), vec!(3));
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "e")), empty_vec);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "f")), vec!(6_000));
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "g")), vec!(6_000, 7_000));
let score_field_reader: U64FastFieldReader = searcher.segment_reader(0).get_fast_field_reader(score_field).unwrap();
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "a")),
empty_vec);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "b")),
empty_vec);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "c")),
vec![3]);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "d")),
vec![3]);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "e")),
empty_vec);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "f")),
vec![6_000]);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "g")),
vec![6_000, 7_000]);
let score_field_reader: U64FastFieldReader = searcher
.segment_reader(0)
.get_fast_field_reader(score_field)
.unwrap();
assert_eq!(score_field_reader.min_value(), 1);
assert_eq!(score_field_reader.max_value(), 3);
let score_field_reader: U64FastFieldReader = searcher.segment_reader(1).get_fast_field_reader(score_field).unwrap();
let score_field_reader: U64FastFieldReader = searcher
.segment_reader(1)
.get_fast_field_reader(score_field)
.unwrap();
assert_eq!(score_field_reader.min_value(), 4000);
assert_eq!(score_field_reader.max_value(), 7000);
}
{ // merging the segments
let segment_ids = index.searchable_segment_ids().expect("Searchable segments failed.");
index_writer.merge(&segment_ids)
.wait()
.expect("Merging failed");
{
// merging the segments
let segment_ids = index
.searchable_segment_ids()
.expect("Searchable segments failed.");
index_writer
.merge(&segment_ids)
.wait()
.expect("Merging failed");
index.load_searchers().unwrap();
let ref searcher = *index.searcher();
assert_eq!(searcher.segment_readers().len(), 1);
assert_eq!(searcher.num_docs(), 3);
assert_eq!(searcher.segment_readers()[0].num_docs(), 3);
assert_eq!(searcher.segment_readers()[0].max_doc(), 3);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "a")), empty_vec);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "b")), empty_vec);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "c")), vec!(3));
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "d")), vec!(3));
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "e")), empty_vec);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "f")), vec!(6_000));
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "g")), vec!(6_000, 7_000));
let score_field_reader: U64FastFieldReader = searcher.segment_reader(0).get_fast_field_reader(score_field).unwrap();
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "a")),
empty_vec);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "b")),
empty_vec);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "c")),
vec![3]);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "d")),
vec![3]);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "e")),
empty_vec);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "f")),
vec![6_000]);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "g")),
vec![6_000, 7_000]);
let score_field_reader: U64FastFieldReader = searcher
.segment_reader(0)
.get_fast_field_reader(score_field)
.unwrap();
assert_eq!(score_field_reader.min_value(), 3);
assert_eq!(score_field_reader.max_value(), 7000);
}
{
{
// test a commit with only deletes
index_writer.delete_term(Term::from_field_text(text_field, "c"));
index_writer.commit().unwrap();
index.load_searchers().unwrap();
let ref searcher = *index.searcher();
assert_eq!(searcher.segment_readers().len(), 1);
assert_eq!(searcher.num_docs(), 2);
assert_eq!(searcher.segment_readers()[0].num_docs(), 2);
assert_eq!(searcher.segment_readers()[0].max_doc(), 3);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "a")), empty_vec);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "b")), empty_vec);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "c")), empty_vec);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "d")), empty_vec);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "e")), empty_vec);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "f")), vec!(6_000));
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "g")), vec!(6_000, 7_000));
let score_field_reader: U64FastFieldReader = searcher.segment_reader(0).get_fast_field_reader(score_field).unwrap();
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "a")),
empty_vec);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "b")),
empty_vec);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "c")),
empty_vec);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "d")),
empty_vec);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "e")),
empty_vec);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "f")),
vec![6_000]);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "g")),
vec![6_000, 7_000]);
let score_field_reader: U64FastFieldReader = searcher
.segment_reader(0)
.get_fast_field_reader(score_field)
.unwrap();
assert_eq!(score_field_reader.min_value(), 3);
assert_eq!(score_field_reader.max_value(), 7000);
}
{ // Test merging a single segment in order to remove deletes.
let segment_ids = index.searchable_segment_ids().expect("Searchable segments failed.");
index_writer.merge(&segment_ids)
.wait()
.expect("Merging failed");
{
// Test merging a single segment in order to remove deletes.
let segment_ids = index
.searchable_segment_ids()
.expect("Searchable segments failed.");
index_writer
.merge(&segment_ids)
.wait()
.expect("Merging failed");
index.load_searchers().unwrap();
let ref searcher = *index.searcher();
@@ -582,31 +634,45 @@ mod tests {
assert_eq!(searcher.num_docs(), 2);
assert_eq!(searcher.segment_readers()[0].num_docs(), 2);
assert_eq!(searcher.segment_readers()[0].max_doc(), 2);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "a")), empty_vec);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "b")), empty_vec);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "c")), empty_vec);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "d")), empty_vec);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "e")), empty_vec);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "f")), vec!(6_000));
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "g")), vec!(6_000, 7_000));
let score_field_reader: U64FastFieldReader = searcher.segment_reader(0).get_fast_field_reader(score_field).unwrap();
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "a")),
empty_vec);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "b")),
empty_vec);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "c")),
empty_vec);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "d")),
empty_vec);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "e")),
empty_vec);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "f")),
vec![6_000]);
assert_eq!(search_term(&searcher, Term::from_field_text(text_field, "g")),
vec![6_000, 7_000]);
let score_field_reader: U64FastFieldReader = searcher
.segment_reader(0)
.get_fast_field_reader(score_field)
.unwrap();
assert_eq!(score_field_reader.min_value(), 6000);
assert_eq!(score_field_reader.max_value(), 7000);
}
{ // Test removing all docs
{
// Test removing all docs
index_writer.delete_term(Term::from_field_text(text_field, "g"));
let segment_ids = index.searchable_segment_ids().expect("Searchable segments failed.");
index_writer.merge(&segment_ids)
.wait()
.expect("Merging failed");
let segment_ids = index
.searchable_segment_ids()
.expect("Searchable segments failed.");
index_writer
.merge(&segment_ids)
.wait()
.expect("Merging failed");
index.load_searchers().unwrap();
let ref searcher = *index.searcher();
assert_eq!(searcher.segment_readers().len(), 1);
assert_eq!(searcher.num_docs(), 0);
}
}
}

2
src/indexer/mod.rs
View File

@@ -22,5 +22,5 @@ pub use self::log_merge_policy::LogMergePolicy;
pub use self::merge_policy::{NoMergePolicy, MergeCandidate, MergePolicy};
pub use self::segment_manager::SegmentManager;
/// Alias for the default merge policy, which is the LogMergePolicy.
/// Alias for the default merge policy, which is the `LogMergePolicy`.
pub type DefaultMergePolicy = LogMergePolicy;

38
src/indexer/segment_entry.rs
View File

@@ -8,11 +8,11 @@ use std::fmt;
#[derive(Clone, Copy, PartialEq, Eq, Debug)]
pub enum SegmentState {
Ready,
InMerge,
InMerge,
}
impl SegmentState {
pub fn letter_code(&self,) -> char {
pub fn letter_code(&self) -> char {
match *self {
SegmentState::InMerge => 'M',
SegmentState::Ready => 'R',
@@ -21,35 +21,33 @@ impl SegmentState {
}
/// A segment entry describes the state of
/// A segment entry describes the state of
/// a given segment, at a given instant.
///
/// In addition to segment meta,
/// In addition to segment `meta`,
/// it contains a few transient states
/// - state expresses whether the segment is already in the
/// - `state` expresses whether the segment is already in the
/// middle of a merge
/// - delete_bitset is a bitset describing
/// - `delete_bitset` is a bitset describing
/// documents that were deleted during the commit
/// itself.
/// - Delete cursor, is the position in the delete queue.
/// - `delete_cursor` is the position in the delete queue.
/// Deletes happening before the cursor are reflected either
/// in the .del file or in the delete_bitset.
/// in the .del file or in the `delete_bitset`.
#[derive(Clone)]
pub struct SegmentEntry {
meta: SegmentMeta,
state: SegmentState,
delete_bitset: Option<BitSet>,
delete_cursor: DeleteCursor,
}
impl SegmentEntry {
/// Create a new `SegmentEntry`
pub fn new(segment_meta: SegmentMeta,
pub fn new(segment_meta: SegmentMeta,
delete_cursor: DeleteCursor,
delete_bitset: Option<BitSet>) -> SegmentEntry {
delete_bitset: Option<BitSet>)
-> SegmentEntry {
SegmentEntry {
meta: segment_meta,
state: SegmentState::Ready,
@@ -62,7 +60,7 @@ impl SegmentEntry {
/// Return a reference to the segment entry deleted bitset.
///
/// `DocId` in this bitset are flagged as deleted.
pub fn delete_bitset(&self,) -> Option<&BitSet> {
pub fn delete_bitset(&self) -> Option<&BitSet> {
self.delete_bitset.as_ref()
}
@@ -77,7 +75,7 @@ impl SegmentEntry {
&mut self.delete_cursor
}
/// Return the `SegmentEntry`.
/// Return the `SegmentEntry`.
///
/// The state describes whether the segment is available for
/// a merge or not.
@@ -89,7 +87,7 @@ impl SegmentEntry {
pub fn segment_id(&self) -> SegmentId {
self.meta.id()
}
/// Accessor to the `SegmentMeta`
pub fn meta(&self) -> &SegmentMeta {
@@ -99,9 +97,9 @@ impl SegmentEntry {
/// Mark the `SegmentEntry` as in merge.
///
/// Only segments that are not already
/// Only segments that are not already
/// in a merge are elligible for future merge.
pub fn start_merge(&mut self,) {
pub fn start_merge(&mut self) {
self.state = SegmentState::InMerge;
}
@@ -110,14 +108,14 @@ impl SegmentEntry {
/// If a merge fails, it is important to switch
/// the segment back to a idle state, so that it
/// may be elligible for future merges.
pub fn cancel_merge(&mut self,) {
pub fn cancel_merge(&mut self) {
self.state = SegmentState::Ready;
}
/// Returns true iff a segment should
/// be considered for a merge.
pub fn is_ready(&self,) -> bool {
pub fn is_ready(&self) -> bool {
self.state == SegmentState::Ready
}
}

134
src/indexer/segment_manager.rs
View File

@@ -14,7 +14,7 @@ use indexer::delete_queue::DeleteCursor;
struct SegmentRegisters {
uncommitted: SegmentRegister,
committed: SegmentRegister,
writing: HashSet<SegmentId>,
writing: HashSet<SegmentId>,
}
@@ -22,7 +22,7 @@ struct SegmentRegisters {
/// The segment manager stores the list of segments
/// as well as their state.
///
/// It guarantees the atomicity of the
/// It guarantees the atomicity of the
/// changes (merges especially)
#[derive(Default)]
pub struct SegmentManager {
@@ -32,43 +32,43 @@ pub struct SegmentManager {
impl Debug for SegmentManager {
fn fmt(&self, f: &mut Formatter) -> Result<(), fmt::Error> {
let lock = self.read();
write!(f, "{{ uncommitted: {:?}, committed: {:?} }}", lock.uncommitted, lock.committed)
write!(f,
"{{ uncommitted: {:?}, committed: {:?} }}",
lock.uncommitted,
lock.committed)
}
}
pub fn get_mergeable_segments(segment_manager: &SegmentManager,) -> (Vec<SegmentMeta>, Vec<SegmentMeta>) {
pub fn get_mergeable_segments(segment_manager: &SegmentManager)
-> (Vec<SegmentMeta>, Vec<SegmentMeta>) {
let registers_lock = segment_manager.read();
(registers_lock.committed.get_mergeable_segments(),
registers_lock.uncommitted.get_mergeable_segments())
}
impl SegmentManager {
pub fn from_segments(segment_metas: Vec<SegmentMeta>, delete_cursor: DeleteCursor) -> SegmentManager {
pub fn from_segments(segment_metas: Vec<SegmentMeta>,
delete_cursor: DeleteCursor)
-> SegmentManager {
SegmentManager {
registers: RwLock::new(SegmentRegisters {
uncommitted: SegmentRegister::default(),
committed: SegmentRegister::new(segment_metas, delete_cursor),
writing: HashSet::new(),
}),
uncommitted: SegmentRegister::default(),
committed: SegmentRegister::new(segment_metas,
delete_cursor),
writing: HashSet::new(),
}),
}
}
/// Returns all of the segment entries (committed or uncommitted)
pub fn segment_entries(&self,) -> Vec<SegmentEntry> {
let mut segment_entries = self.read()
.uncommitted
.segment_entries();
segment_entries.extend(
self.read()
.committed
.segment_entries()
);
pub fn segment_entries(&self) -> Vec<SegmentEntry> {
let mut segment_entries = self.read().uncommitted.segment_entries();
segment_entries.extend(self.read().committed.segment_entries());
segment_entries
}
/// Returns the overall number of segments in the `SegmentManager`
pub fn num_segments(&self,) -> usize {
pub fn num_segments(&self) -> usize {
let registers_lock = self.read();
registers_lock.committed.len() + registers_lock.uncommitted.len()
}
@@ -78,19 +78,14 @@ impl SegmentManager {
let mut files = HashSet::new();
files.insert(META_FILEPATH.clone());
files.insert(LOCKFILE_FILEPATH.clone());
let segment_metas: Vec<SegmentMeta> =
registers_lock.committed
.get_all_segments()
.into_iter()
.chain(registers_lock.uncommitted
.get_all_segments()
.into_iter())
.chain(registers_lock.writing
.iter()
.cloned()
.map(SegmentMeta::new))
.collect();
let segment_metas: Vec<SegmentMeta> = registers_lock
.committed
.get_all_segments()
.into_iter()
.chain(registers_lock.uncommitted.get_all_segments().into_iter())
.chain(registers_lock.writing.iter().cloned().map(SegmentMeta::new))
.collect();
for segment_meta in segment_metas {
files.extend(segment_meta.list_files());
}
@@ -102,18 +97,22 @@ impl SegmentManager {
registers
.committed
.segment_entry(segment_id)
.or_else(|| registers.uncommitted.segment_entry(segment_id))
.or_else(|| registers.uncommitted.segment_entry(segment_id))
}
// Lock poisoning should never happen :
// The lock is acquired and released within this class,
// and the operations cannot panic.
fn read(&self,) -> RwLockReadGuard<SegmentRegisters> {
self.registers.read().expect("Failed to acquire read lock on SegmentManager.")
// and the operations cannot panic.
fn read(&self) -> RwLockReadGuard<SegmentRegisters> {
self.registers
.read()
.expect("Failed to acquire read lock on SegmentManager.")
}
fn write(&self,) -> RwLockWriteGuard<SegmentRegisters> {
self.registers.write().expect("Failed to acquire write lock on SegmentManager.")
fn write(&self) -> RwLockWriteGuard<SegmentRegisters> {
self.registers
.write()
.expect("Failed to acquire write lock on SegmentManager.")
}
pub fn commit(&self, segment_entries: Vec<SegmentEntry>) {
@@ -124,42 +123,42 @@ impl SegmentManager {
registers_lock.committed.add_segment_entry(segment_entry);
}
}
pub fn start_merge(&self, segment_ids: &[SegmentId]) {
let mut registers_lock = self.write();
if registers_lock.uncommitted.contains_all(segment_ids) {
for segment_id in segment_ids {
registers_lock.uncommitted.start_merge(segment_id);
}
}
else if registers_lock.committed.contains_all(segment_ids) {
} else if registers_lock.committed.contains_all(segment_ids) {
for segment_id in segment_ids {
registers_lock.committed.start_merge(segment_id);
}
}
else {
} else {
error!("Merge operation sent for segments that are not all uncommited or commited.");
}
}
pub fn cancel_merge(&self,
before_merge_segment_ids: &[SegmentId],
after_merge_segment_id: SegmentId) {
before_merge_segment_ids: &[SegmentId],
after_merge_segment_id: SegmentId) {
let mut registers_lock = self.write();
// we mark all segments are ready for merge.
{
let target_segment_register: &mut SegmentRegister;
target_segment_register = {
if registers_lock.uncommitted.contains_all(&before_merge_segment_ids) {
if registers_lock
.uncommitted
.contains_all(before_merge_segment_ids) {
&mut registers_lock.uncommitted
}
else if registers_lock.committed.contains_all(&before_merge_segment_ids) {
} else if registers_lock
.committed
.contains_all(before_merge_segment_ids) {
&mut registers_lock.committed
}
else {
} else {
warn!("couldn't find segment in SegmentManager");
return;
}
@@ -185,19 +184,24 @@ impl SegmentManager {
registers_lock.writing.remove(&segment_entry.segment_id());
registers_lock.uncommitted.add_segment_entry(segment_entry);
}
pub fn end_merge(&self,
before_merge_segment_ids: &[SegmentId],
after_merge_segment_entry: SegmentEntry) {
before_merge_segment_ids: &[SegmentId],
after_merge_segment_entry: SegmentEntry) {
let mut registers_lock = self.write();
registers_lock.writing.remove(&after_merge_segment_entry.segment_id());
registers_lock
.writing
.remove(&after_merge_segment_entry.segment_id());
let mut target_register: &mut SegmentRegister = {
if registers_lock.uncommitted.contains_all(&before_merge_segment_ids) {
if registers_lock
.uncommitted
.contains_all(before_merge_segment_ids) {
&mut registers_lock.uncommitted
}
else if registers_lock.committed.contains_all(&before_merge_segment_ids) {
} else if registers_lock
.committed
.contains_all(before_merge_segment_ids) {
&mut registers_lock.committed
} else {
warn!("couldn't find segment in SegmentManager");
@@ -209,12 +213,12 @@ impl SegmentManager {
}
target_register.add_segment_entry(after_merge_segment_entry);
}
pub fn committed_segment_metas(&self,) -> Vec<SegmentMeta> {
pub fn committed_segment_metas(&self) -> Vec<SegmentMeta> {
let registers_lock = self.read();
registers_lock.committed.segment_metas()
}

103
src/indexer/segment_register.rs
View File

@@ -9,14 +9,14 @@ use indexer::delete_queue::DeleteCursor;
/// The segment register keeps track
/// of the list of segment, their size as well
/// as the state they are in.
///
/// It is consumed by indexes to get the list of
///
/// It is consumed by indexes to get the list of
/// segments that are currently searchable,
/// and by the index merger to identify
/// and by the index merger to identify
/// merge candidates.
#[derive(Default)]
pub struct SegmentRegister {
segment_states: HashMap<SegmentId, SegmentEntry>,
segment_states: HashMap<SegmentId, SegmentEntry>,
}
@@ -33,8 +33,7 @@ impl Debug for SegmentRegister {
impl SegmentRegister {
pub fn clear(&mut self,) {
pub fn clear(&mut self) {
self.segment_states.clear();
}
@@ -42,29 +41,26 @@ impl SegmentRegister {
self.segment_states.len()
}
pub fn get_all_segments(&self,) -> Vec<SegmentMeta> {
pub fn get_all_segments(&self) -> Vec<SegmentMeta> {
self.segment_states
.values()
.map(|segment_entry| segment_entry.meta().clone())
.collect()
}
pub fn get_mergeable_segments(&self,) -> Vec<SegmentMeta> {
pub fn get_mergeable_segments(&self) -> Vec<SegmentMeta> {
self.segment_states
.values()
.filter(|segment_entry| segment_entry.is_ready())
.map(|segment_entry| segment_entry.meta().clone())
.collect()
}
pub fn segment_entries(&self,) -> Vec<SegmentEntry> {
self.segment_states
.values()
.cloned()
.collect()
pub fn segment_entries(&self) -> Vec<SegmentEntry> {
self.segment_states.values().cloned().collect()
}
pub fn segment_metas(&self,) -> Vec<SegmentMeta> {
pub fn segment_metas(&self) -> Vec<SegmentMeta> {
let mut segment_ids: Vec<SegmentMeta> = self.segment_states
.values()
.map(|segment_entry| segment_entry.meta().clone())
@@ -72,28 +68,26 @@ impl SegmentRegister {
segment_ids.sort_by_key(|meta| meta.id());
segment_ids
}
pub fn segment_entry(&self, segment_id: &SegmentId) -> Option<SegmentEntry> {
self.segment_states
.get(&segment_id)
.map(|segment_entry| segment_entry.clone())
self.segment_states.get(segment_id).cloned()
}
pub fn contains_all(&mut self, segment_ids: &[SegmentId]) -> bool {
segment_ids
.iter()
.all(|segment_id| self.segment_states.contains_key(segment_id))
}
pub fn add_segment_entry(&mut self, segment_entry: SegmentEntry) {
let segment_id = segment_entry.segment_id();
self.segment_states.insert(segment_id, segment_entry);
}
pub fn remove_segment(&mut self, segment_id: &SegmentId) {
self.segment_states.remove(segment_id);
}
}
pub fn cancel_merge(&mut self, segment_id: &SegmentId) {
self.segment_states
.get_mut(segment_id)
@@ -106,21 +100,16 @@ impl SegmentRegister {
.get_mut(segment_id)
.expect("Received a merge notification for a segment that is not registered")
.start_merge();
}
}
pub fn new(segment_metas: Vec<SegmentMeta>, delete_cursor: DeleteCursor) -> SegmentRegister {
let mut segment_states = HashMap::new();
for segment_meta in segment_metas {
let segment_id = segment_meta.id();
let segment_entry = SegmentEntry::new(
segment_meta,
delete_cursor.clone(),
None);
let segment_entry = SegmentEntry::new(segment_meta, delete_cursor.clone(), None);
segment_states.insert(segment_id, segment_entry);
}
SegmentRegister {
segment_states: segment_states
}
SegmentRegister { segment_states: segment_states }
}
}
@@ -140,7 +129,7 @@ mod tests {
.map(|segment_meta| segment_meta.id())
.collect()
}
#[test]
fn test_segment_register() {
let delete_queue = DeleteQueue::new();
@@ -149,32 +138,48 @@ mod tests {
let segment_id_a = SegmentId::generate_random();
let segment_id_b = SegmentId::generate_random();
let segment_id_merged = SegmentId::generate_random();
{
let segment_meta = SegmentMeta::new(segment_id_a);
let segment_entry = SegmentEntry::new(segment_meta, delete_queue.cursor(), None);
let segment_entry = SegmentEntry::new(segment_meta, delete_queue.cursor(), None);
segment_register.add_segment_entry(segment_entry);
}
assert_eq!(segment_register.segment_entry(&segment_id_a).unwrap().state(), SegmentState::Ready);
assert_eq!(segment_ids(&segment_register), vec!(segment_id_a));
assert_eq!(segment_register
.segment_entry(&segment_id_a)
.unwrap()
.state(),
SegmentState::Ready);
assert_eq!(segment_ids(&segment_register), vec![segment_id_a]);
{
let segment_meta = SegmentMeta::new(segment_id_b);
let segment_entry = SegmentEntry::new(segment_meta, delete_queue.cursor(), None);
let segment_entry = SegmentEntry::new(segment_meta, delete_queue.cursor(), None);
segment_register.add_segment_entry(segment_entry);
}
assert_eq!(segment_register.segment_entry(&segment_id_b).unwrap().state(), SegmentState::Ready);
assert_eq!(segment_register
.segment_entry(&segment_id_b)
.unwrap()
.state(),
SegmentState::Ready);
segment_register.start_merge(&segment_id_a);
segment_register.start_merge(&segment_id_b);
assert_eq!(segment_register.segment_entry(&segment_id_a).unwrap().state(), SegmentState::InMerge);
assert_eq!(segment_register.segment_entry(&segment_id_b).unwrap().state(), SegmentState::InMerge);
assert_eq!(segment_register
.segment_entry(&segment_id_a)
.unwrap()
.state(),
SegmentState::InMerge);
assert_eq!(segment_register
.segment_entry(&segment_id_b)
.unwrap()
.state(),
SegmentState::InMerge);
segment_register.remove_segment(&segment_id_a);
segment_register.remove_segment(&segment_id_b);
{
let segment_meta_merged = SegmentMeta::new(segment_id_merged);
let segment_entry = SegmentEntry::new(segment_meta_merged, delete_queue.cursor(), None);
segment_register.add_segment_entry(segment_entry);
let segment_entry = SegmentEntry::new(segment_meta_merged, delete_queue.cursor(), None);
segment_register.add_segment_entry(segment_entry);
}
assert_eq!(segment_ids(&segment_register), vec!(segment_id_merged));
assert_eq!(segment_ids(&segment_register), vec![segment_id_merged]);
}
}
}

10
src/indexer/segment_serializer.rs
View File

@@ -29,11 +29,11 @@ impl SegmentSerializer {
let postings_serializer = try!(PostingsSerializer::open(segment));
Ok(SegmentSerializer {
postings_serializer: postings_serializer,
store_writer: StoreWriter::new(store_write),
fast_field_serializer: fast_field_serializer,
fieldnorms_serializer: fieldnorms_serializer,
})
postings_serializer: postings_serializer,
store_writer: StoreWriter::new(store_write),
fast_field_serializer: fast_field_serializer,
fieldnorms_serializer: fieldnorms_serializer,
})
}
/// Accessor to the `PostingsSerializer`.

278
src/indexer/segment_updater.rs
View File

@@ -1,5 +1,3 @@
#![allow(for_kv_map)]
use core::Index;
use core::IndexMeta;
use core::META_FILEPATH;
@@ -49,17 +47,14 @@ use super::segment_manager::{SegmentManager, get_mergeable_segments};
/// and flushed.
///
/// This method is not part of tantivy's public API
pub fn save_new_metas(schema: Schema,
opstamp: u64,
directory: &mut Directory)
-> Result<()> {
save_metas(vec!(), schema, opstamp, directory)
pub fn save_new_metas(schema: Schema, opstamp: u64, directory: &mut Directory) -> Result<()> {
save_metas(vec![], schema, opstamp, directory)
}
/// Save the index meta file.
/// This operation is atomic :
/// This operation is atomic:
/// Either
// - it fails, in which case an error is returned,
/// and the `meta.json` remains untouched,
@@ -79,10 +74,9 @@ pub fn save_metas(segment_metas: Vec<SegmentMeta>,
};
let mut w = try!(serde_json::to_vec_pretty(&metas));
try!(write!(&mut w, "\n"));
let res = directory.atomic_write(&META_FILEPATH, &w[..])?;
directory.atomic_write(&META_FILEPATH, &w[..])?;
debug!("Saved metas {:?}", serde_json::to_string_pretty(&metas));
Ok(res)
Ok(())
}
@@ -90,7 +84,7 @@ pub fn save_metas(segment_metas: Vec<SegmentMeta>,
// of the `SegmentUpdate`s.
//
// All this processing happens on a single thread
// consuming a common queue.
// consuming a common queue.
#[derive(Clone)]
pub struct SegmentUpdater(Arc<InnerSegmentUpdater>);
@@ -99,56 +93,57 @@ pub struct SegmentUpdater(Arc<InnerSegmentUpdater>);
fn perform_merge(segment_ids: &[SegmentId],
segment_updater: &SegmentUpdater,
mut merged_segment: Segment,
target_opstamp: u64) -> Result<SegmentEntry> {
target_opstamp: u64)
-> Result<SegmentEntry> {
// first we need to apply deletes to our segment.
info!("Start merge: {:?}", segment_ids);
let ref index = segment_updater.0.index;
let schema = index.schema();
let mut segment_entries = vec!();
let mut file_protections: Vec<FileProtection> = vec!();
let index = &segment_updater.0.index;
let schema = index.schema();
let mut segment_entries = vec![];
let mut file_protections: Vec<FileProtection> = vec![];
for segment_id in segment_ids {
if let Some(mut segment_entry) = segment_updater.0
.segment_manager
.segment_entry(segment_id) {
if let Some(mut segment_entry) =
segment_updater.0.segment_manager.segment_entry(segment_id) {
let segment = index.segment(segment_entry.meta().clone());
if let Some(file_protection) = advance_deletes(segment, &mut segment_entry, target_opstamp)? {
if let Some(file_protection) =
advance_deletes(segment, &mut segment_entry, target_opstamp)? {
file_protections.push(file_protection);
}
segment_entries.push(segment_entry);
}
else {
error!("Error, had to abort merge as some of the segment is not managed anymore.a");
return Err(Error::InvalidArgument(format!("Segment {:?} requested for merge is not managed.", segment_id)));
} else {
error!("Error, had to abort merge as some of the segment is not managed anymore.");
let msg = format!("Segment {:?} requested for merge is not managed.",
segment_id);
return Err(Error::InvalidArgument(msg));
}
}
let delete_cursor = segment_entries[0].delete_cursor().clone();
let segments: Vec<Segment> = segment_entries
.iter()
.map(|segment_entry| {
index.segment(segment_entry.meta().clone())
})
.map(|segment_entry| index.segment(segment_entry.meta().clone()))
.collect();
// An IndexMerger is like a "view" of our merged segments.
let merger: IndexMerger = IndexMerger::open(schema, &segments[..])?;
// ... we just serialize this index merger in our new segment
// to merge the two segments.
let segment_serializer =
SegmentSerializer::for_segment(&mut merged_segment)
let segment_serializer = SegmentSerializer::for_segment(&mut merged_segment)
.expect("Creating index serializer failed");
let num_docs = merger.write(segment_serializer).expect("Serializing merged index failed");
let num_docs = merger
.write(segment_serializer)
.expect("Serializing merged index failed");
let mut segment_meta = SegmentMeta::new(merged_segment.id());
segment_meta.set_max_doc(num_docs);
let after_merge_segment_entry = SegmentEntry::new(segment_meta.clone(), delete_cursor, None);
Ok(after_merge_segment_entry)
}
@@ -162,30 +157,28 @@ struct InnerSegmentUpdater {
merging_thread_id: AtomicUsize,
merging_threads: RwLock<HashMap<usize, JoinHandle<Result<()>>>>,
generation: AtomicUsize,
killed: AtomicBool,
killed: AtomicBool,
stamper: Stamper,
}
impl SegmentUpdater {
pub fn new(index: Index,
stamper: Stamper,
delete_cursor: DeleteCursor) -> Result<SegmentUpdater> {
delete_cursor: DeleteCursor)
-> Result<SegmentUpdater> {
let segments = index.searchable_segment_metas()?;
let segment_manager = SegmentManager::from_segments(segments, delete_cursor);
Ok(
SegmentUpdater(Arc::new(InnerSegmentUpdater {
pool: CpuPool::new(1),
index: index,
segment_manager: segment_manager,
merge_policy: RwLock::new(box DefaultMergePolicy::default()),
merging_thread_id: AtomicUsize::default(),
merging_threads: RwLock::new(HashMap::new()),
generation: AtomicUsize::default(),
killed: AtomicBool::new(false),
stamper: stamper,
}))
)
Ok(SegmentUpdater(Arc::new(InnerSegmentUpdater {
pool: CpuPool::new(1),
index: index,
segment_manager: segment_manager,
merge_policy: RwLock::new(box DefaultMergePolicy::default()),
merging_thread_id: AtomicUsize::default(),
merging_threads: RwLock::new(HashMap::new()),
generation: AtomicUsize::default(),
killed: AtomicBool::new(false),
stamper: stamper,
})))
}
pub fn new_segment(&self) -> Segment {
@@ -200,40 +193,41 @@ impl SegmentUpdater {
}
pub fn set_merge_policy(&self, merge_policy: Box<MergePolicy>) {
*self.0.merge_policy.write().unwrap()= merge_policy;
*self.0.merge_policy.write().unwrap() = merge_policy;
}
fn get_merging_thread_id(&self) -> usize {
self.0.merging_thread_id.fetch_add(1, Ordering::SeqCst)
}
fn run_async<T: 'static + Send, F: 'static + Send + FnOnce(SegmentUpdater) -> T>(&self, f: F) -> CpuFuture<T, Error> {
fn run_async<T: 'static + Send, F: 'static + Send + FnOnce(SegmentUpdater) -> T>
(&self,
f: F)
-> CpuFuture<T, Error> {
let me_clone = self.clone();
self.0.pool.spawn_fn(move || {
Ok(f(me_clone))
})
self.0.pool.spawn_fn(move || Ok(f(me_clone)))
}
pub fn add_segment(&self, generation: usize, segment_entry: SegmentEntry) -> bool {
if generation >= self.0.generation.load(Ordering::Acquire) {
self.run_async(|segment_updater| {
segment_updater.0.segment_manager.add_segment(segment_entry);
segment_updater.consider_merge_options();
true
}).forget();
segment_updater.0.segment_manager.add_segment(segment_entry);
segment_updater.consider_merge_options();
true
})
.forget();
true
}
else {
} else {
false
}
}
pub fn kill(&mut self,) {
pub fn kill(&mut self) {
self.0.killed.store(true, Ordering::Release);
}
fn is_alive(&self,) -> bool {
fn is_alive(&self) -> bool {
!self.0.killed.load(Ordering::Acquire)
}
@@ -243,75 +237,78 @@ 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>> {
let mut segment_entries = self.0.segment_manager.segment_entries();
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());
advance_deletes(segment, segment_entry, target_opstamp)?;
}
Ok(segment_entries)
}
pub fn save_metas(&self, opstamp: u64) {
if self.is_alive() {
let index = &self.0.index;
let directory = index.directory();
save_metas(
self.0.segment_manager.committed_segment_metas(),
index.schema(),
opstamp,
directory.box_clone().borrow_mut()).expect("Could not save metas.");
save_metas(self.0.segment_manager.committed_segment_metas(),
index.schema(),
opstamp,
directory.box_clone().borrow_mut())
.expect("Could not save metas.");
}
}
pub fn garbage_collect_files(&self) -> Result<()> {
self.run_async(move |segment_updater| {
segment_updater.garbage_collect_files_exec();
}).wait()
self.run_async(move |segment_updater| { segment_updater.garbage_collect_files_exec(); })
.wait()
}
fn garbage_collect_files_exec(&self) {
fn garbage_collect_files_exec(&self) {
let living_files = self.0.segment_manager.list_files();
let mut index = self.0.index.clone();
index.directory_mut().garbage_collect(living_files);
}
pub fn commit(&self, opstamp: u64) -> Result<()> {
self.run_async(move |segment_updater| {
if segment_updater.is_alive() {
let segment_entries = segment_updater
.purge_deletes(opstamp)
.expect("Failed purge deletes");
segment_updater.0.segment_manager.commit(segment_entries);
segment_updater.save_metas(opstamp);
segment_updater.garbage_collect_files_exec();
segment_updater.consider_merge_options();
}
}).wait()
self.run_async(move |segment_updater| if segment_updater.is_alive() {
let segment_entries = segment_updater
.purge_deletes(opstamp)
.expect("Failed purge deletes");
segment_updater.0.segment_manager.commit(segment_entries);
segment_updater.save_metas(opstamp);
segment_updater.garbage_collect_files_exec();
segment_updater.consider_merge_options();
})
.wait()
}
pub fn start_merge(&self, segment_ids: &[SegmentId]) -> impl Future<Item=SegmentMeta, Error=Canceled> {
pub fn start_merge(&self,
segment_ids: &[SegmentId])
-> impl Future<Item = SegmentMeta, Error = Canceled> {
self.0.segment_manager.start_merge(segment_ids);
let segment_updater_clone = self.clone();
let segment_ids_vec = segment_ids.to_vec();
let segment_ids_vec = segment_ids.to_vec();
let merging_thread_id = self.get_merging_thread_id();
let (merging_future_send, merging_future_recv) = oneshot();
if segment_ids.is_empty() {
return merging_future_recv;
}
let target_opstamp = self.0.stamper.stamp();
let merging_join_handle = thread::spawn(move || {
// first we need to apply deletes to our segment.
let merged_segment = segment_updater_clone.new_segment();
let merged_segment = segment_updater_clone.new_segment();
let merged_segment_id = merged_segment.id();
let merge_result = perform_merge(&segment_ids_vec, &segment_updater_clone, merged_segment, target_opstamp);
let merge_result = perform_merge(&segment_ids_vec,
&segment_updater_clone,
merged_segment,
target_opstamp);
match merge_result {
Ok(after_merge_segment_entry) => {
@@ -319,11 +316,11 @@ impl SegmentUpdater {
segment_updater_clone
.end_merge(segment_ids_vec, after_merge_segment_entry)
.expect("Segment updater thread is corrupted.");
// the future may fail if the listener of the oneshot future
// the future may fail if the listener of the oneshot future
// has been destroyed.
//
// This is not a problem here, so we just ignore any
// This is not a problem here, so we just ignore any
// possible error.
let _merging_future_res = merging_future_send.send(merged_segment_meta);
}
@@ -337,16 +334,26 @@ impl SegmentUpdater {
// merging_future_send will be dropped, sending an error to the future.
}
}
segment_updater_clone.0.merging_threads.write().unwrap().remove(&merging_thread_id);
segment_updater_clone
.0
.merging_threads
.write()
.unwrap()
.remove(&merging_thread_id);
Ok(())
});
self.0.merging_threads.write().unwrap().insert(merging_thread_id, merging_join_handle);
self.0
.merging_threads
.write()
.unwrap()
.insert(merging_thread_id, merging_join_handle);
merging_future_recv
}
fn consider_merge_options(&self) {
let (committed_segments, uncommitted_segments) = get_mergeable_segments(&self.0.segment_manager);
let (committed_segments, uncommitted_segments) =
get_mergeable_segments(&self.0.segment_manager);
// Committed segments cannot be merged with uncommitted_segments.
// We therefore consider merges using these two sets of segments independently.
let merge_policy = self.get_merge_policy();
@@ -358,17 +365,20 @@ impl SegmentUpdater {
}
}
fn cancel_merge(&self,
before_merge_segment_ids: &[SegmentId],
after_merge_segment_entry: SegmentId) {
self.0.segment_manager.cancel_merge(&before_merge_segment_ids, after_merge_segment_entry);
fn cancel_merge(&self,
before_merge_segment_ids: &[SegmentId],
after_merge_segment_entry: SegmentId) {
self.0
.segment_manager
.cancel_merge(before_merge_segment_ids, after_merge_segment_entry);
}
fn end_merge(&self,
before_merge_segment_ids: Vec<SegmentId>,
mut after_merge_segment_entry: SegmentEntry) -> Result<()> {
fn end_merge(&self,
before_merge_segment_ids: Vec<SegmentId>,
mut after_merge_segment_entry: SegmentEntry)
-> Result<()> {
self.run_async(move |segment_updater| {
debug!("End merge {:?}", after_merge_segment_entry.meta());
let mut delete_cursor = after_merge_segment_entry.delete_cursor().clone();
@@ -376,24 +386,30 @@ impl SegmentUpdater {
if let Some(delete_operation) = delete_cursor.get() {
let committed_opstamp = segment_updater.0.index.opstamp();
if delete_operation.opstamp < committed_opstamp {
let segment = segment_updater.0.index.segment(after_merge_segment_entry.meta().clone());
match advance_deletes(segment, &mut after_merge_segment_entry, committed_opstamp) {
let index = &segment_updater.0.index;
let segment = index.segment(after_merge_segment_entry.meta().clone());
match advance_deletes(segment,
&mut after_merge_segment_entry,
committed_opstamp) {
Ok(file_protection_opt_res) => {
_file_protection_opt = file_protection_opt_res;
}
Err(e) => {
error!("Merge of {:?} was cancelled (advancing deletes failed): {:?}", before_merge_segment_ids, e);
error!("Merge of {:?} was cancelled (advancing deletes failed): {:?}",
before_merge_segment_ids, e);
// ... cancel merge
if cfg!(test) {
panic!("Merge failed.");
}
segment_updater.cancel_merge(&before_merge_segment_ids, after_merge_segment_entry.segment_id());
segment_updater.cancel_merge(&before_merge_segment_ids,
after_merge_segment_entry.segment_id());
return;
}
}
}
}
segment_updater.0.segment_manager.end_merge(&before_merge_segment_ids, after_merge_segment_entry);
segment_updater.0.segment_manager.end_merge(&before_merge_segment_ids,
after_merge_segment_entry);
segment_updater.consider_merge_options();
segment_updater.save_metas(segment_updater.0.index.opstamp());
}).wait()
@@ -404,22 +420,22 @@ impl SegmentUpdater {
///
/// Upon termination of the current merging threads,
/// merge opportunity may appear.
//
//
/// We keep waiting until the merge policy judges that
/// no opportunity is available.
///
/// Note that it is not required to call this
/// Note that it is not required to call this
/// method in your application.
/// Terminating your application without letting
/// Terminating your application without letting
/// merge terminate is perfectly safe.
///
///
/// Obsolete files will eventually be cleaned up
/// by the directory garbage collector.
pub fn wait_merging_thread(&self) -> Result<()> {
let mut num_segments: usize;
loop {
num_segments = self.0.segment_manager.num_segments();
let mut new_merging_threads = HashMap::new();
@@ -432,9 +448,7 @@ impl SegmentUpdater {
merging_thread_handle
.join()
.map(|_| ())
.map_err(|_| {
Error::ErrorInThread("Merging thread failed.".to_string())
})?
.map_err(|_| Error::ErrorInThread("Merging thread failed.".to_string()))?
}
// Our merging thread may have queued their completed
self.run_async(move |_| {}).wait()?;
@@ -444,10 +458,9 @@ impl SegmentUpdater {
if new_num_segments >= num_segments {
break;
}
}
}
Ok(())
}
}
@@ -467,7 +480,7 @@ mod tests {
let schema = schema_builder.build();
let index = Index::create_in_ram(schema);
// writing the segment
let mut index_writer = index.writer_with_num_threads(1, 40_000_000).unwrap();
index_writer.set_merge_policy(box MergeWheneverPossible);
@@ -479,7 +492,7 @@ mod tests {
}
assert!(index_writer.commit().is_ok());
}
{
for _ in 0..100 {
index_writer.add_document(doc!(text_field=>"c"));
@@ -487,7 +500,7 @@ mod tests {
}
assert!(index_writer.commit().is_ok());
}
{
index_writer.add_document(doc!(text_field=>"e"));
index_writer.add_document(doc!(text_field=>"f"));
@@ -504,8 +517,9 @@ mod tests {
assert_eq!(index.searcher().num_docs(), 302);
{
index_writer.wait_merging_threads()
.expect( "waiting for merging threads");
index_writer
.wait_merging_threads()
.expect("waiting for merging threads");
}
index.load_searchers().unwrap();

73
src/indexer/segment_writer.rs
View File

@@ -1,7 +1,7 @@
use Result;
use DocId;
use std::io;
use schema::Schema;
use schema::Schema;
use schema::Term;
use core::Segment;
use core::SerializableSegment;
@@ -18,11 +18,11 @@ use postings::MultiFieldPostingsWriter;
/// A `SegmentWriter` is in charge of creating segment index from a
/// documents.
///
///
/// They creates the postings list in anonymous memory.
/// The segment is layed on disk when the segment gets `finalized`.
pub struct SegmentWriter<'a> {
heap: &'a Heap,
heap: &'a Heap,
max_doc: DocId,
multifield_postings: MultiFieldPostingsWriter<'a>,
segment_serializer: SegmentSerializer,
@@ -33,17 +33,17 @@ pub struct SegmentWriter<'a> {
fn create_fieldnorms_writer(schema: &Schema) -> FastFieldsWriter {
let u64_fields: Vec<Field> = schema.fields()
.iter()
.enumerate()
.filter(|&(_, field_entry)| field_entry.is_indexed())
.map(|(field_id, _)| Field(field_id as u32))
.collect();
FastFieldsWriter::new(u64_fields)
let u64_fields: Vec<Field> = schema
.fields()
.iter()
.enumerate()
.filter(|&(_, field_entry)| field_entry.is_indexed())
.map(|(field_id, _)| Field(field_id as u32))
.collect();
FastFieldsWriter::new(u64_fields)
}
impl<'a> SegmentWriter<'a> {
/// Creates a new `SegmentWriter`
@@ -108,8 +108,8 @@ impl<'a> SegmentWriter<'a> {
/// As a user, you should rather use `IndexWriter`'s add_document.
pub fn add_document(&mut self, add_operation: &AddOperation, schema: &Schema) -> io::Result<()> {
let doc_id = self.max_doc;
let doc = &add_operation.document;
self.doc_opstamps.push(add_operation.opstamp);
let doc = &add_operation.document;
self.doc_opstamps.push(add_operation.opstamp);
for (field, field_values) in doc.get_sorted_field_values() {
let field_options = schema.get_field_entry(field);
match *field_options.field_type() {
@@ -151,7 +151,7 @@ impl<'a> SegmentWriter<'a> {
}
}
self.fieldnorms_writer.fill_val_up_to(doc_id);
self.fast_field_writers.add_document(&doc);
self.fast_field_writers.add_document(doc);
let stored_fieldvalues: Vec<&FieldValue> = doc
.field_values()
.iter()
@@ -160,34 +160,33 @@ impl<'a> SegmentWriter<'a> {
let doc_writer = self.segment_serializer.get_store_writer();
try!(doc_writer.store(&stored_fieldvalues));
self.max_doc += 1;
Ok(())
Ok(())
}
/// Max doc is
/// - the number of documents in the segment assuming there is no deletes
/// - the maximum document id (including deleted documents) + 1
///
/// Currently, **tantivy** does not handle deletes anyway,
/// so `max_doc == num_docs`
pub fn max_doc(&self,) -> u32 {
self.max_doc
}
/// Number of documents in the index.
/// Deleted documents are not counted.
///
/// Currently, **tantivy** does not handle deletes anyway,
/// so `max_doc == num_docs`
#[allow(dead_code)]
pub fn num_docs(&self,) -> u32 {
self.max_doc
}
/// Max doc is
/// - the number of documents in the segment assuming there is no deletes
/// - the maximum document id (including deleted documents) + 1
///
/// Currently, **tantivy** does not handle deletes anyway,
/// so `max_doc == num_docs`
pub fn max_doc(&self) -> u32 {
self.max_doc
}
/// Number of documents in the index.
/// Deleted documents are not counted.
///
/// Currently, **tantivy** does not handle deletes anyway,
/// so `max_doc == num_docs`
#[allow(dead_code)]
pub fn num_docs(&self) -> u32 {
self.max_doc
}
}
// This method is used as a trick to workaround the borrow checker
fn write<'a>(
fn write(
multifield_postings: &MultiFieldPostingsWriter,
fast_field_writers: &FastFieldsWriter,
fieldnorms_writer: &FastFieldsWriter,

5
src/indexer/stamper.rs
View File

@@ -6,12 +6,11 @@ use std::sync::Arc;
pub struct Stamper(Arc<AtomicU64>);
impl Stamper {
pub fn new(first_opstamp: u64) -> Stamper {
Stamper(Arc::new(AtomicU64::new(first_opstamp)))
}
pub fn stamp(&self,) -> u64 {
pub fn stamp(&self) -> u64 {
self.0.fetch_add(1u64, Ordering::SeqCst)
}
}
}

228
src/lib.rs
View File

@@ -1,6 +1,6 @@
#![doc(html_logo_url = "http://fulmicoton.com/tantivy-logo/tantivy-logo.png")]
#![allow(unknown_lints)] // for the clippy lint options
#![allow(module_inception)]
#![cfg_attr(feature = "cargo-clippy", allow(module_inception))]
#![cfg_attr(feature = "cargo-clippy", allow(inline_always))]
#![feature(box_syntax)]
#![feature(optin_builtin_traits)]
@@ -16,11 +16,13 @@
//! # `tantivy`
//!
//! Tantivy is a search engine library.
//! Tantivy is a search engine library.
//! Think `Lucene`, but in Rust.
//!
//! A good place for you to get started is to check out
//! the example code ( [literate programming](http://fulmicoton.com/tantivy-examples/simple_search.html) / [source code](https://github.com/fulmicoton/tantivy/blob/master/examples/simple_search.rs))
//! the example code (
//! [literate programming](http://fulmicoton.com/tantivy-examples/simple_search.html) /
//! [source code](https://github.com/fulmicoton/tantivy/blob/master/examples/simple_search.rs))
#[macro_use]
extern crate lazy_static;
@@ -64,8 +66,10 @@ extern crate libc;
#[cfg(windows)]
extern crate winapi;
#[cfg(test)] extern crate test;
#[cfg(test)] extern crate rand;
#[cfg(test)]
extern crate test;
#[cfg(test)]
extern crate rand;
#[cfg(test)]
@@ -136,15 +140,14 @@ pub use core::TermIterator;
/// Expose the current version of tantivy, as well
/// whether it was compiled with the simd compression.
pub fn version() -> &'static str {
if cfg!(feature="simdcompression") {
concat!(version!(), "-simd")
}
else {
concat!(version!(), "-nosimd")
if cfg!(feature = "simdcompression") {
concat!(version!(), "-simd")
} else {
concat!(version!(), "-nosimd")
}
}
/// Tantivy's makes it possible to personalize when
/// Tantivy's makes it possible to personalize when
/// the indexer should merge its segments
pub mod merge_policy {
pub use indexer::MergePolicy;
@@ -166,28 +169,27 @@ pub type Score = f32;
pub type SegmentLocalId = u32;
impl DocAddress {
/// Return the segment ordinal.
/// The segment ordinal is an id identifying the segment
/// hosting the document. It is only meaningful, in the context
/// of a searcher.
pub fn segment_ord(&self,) -> SegmentLocalId {
pub fn segment_ord(&self) -> SegmentLocalId {
self.0
}
/// Return the segment local `DocId`
pub fn doc(&self,) -> DocId {
pub fn doc(&self) -> DocId {
self.1
}
}
/// `DocAddress` contains all the necessary information
/// `DocAddress` contains all the necessary information
/// to identify a document given a `Searcher` object.
///
/// It consists in an id identifying its segment, and
///
/// It consists in an id identifying its segment, and
/// its segment-local `DocId`.
///
///
/// The id used for the segment is actually an ordinal
/// in the list of segment hold by a `Searcher`.
#[derive(Debug, Clone, Copy, PartialEq, Eq, PartialOrd, Ord)]
@@ -207,7 +209,30 @@ mod tests {
use IndexWriter;
use fastfield::{FastFieldReader, U64FastFieldReader, I64FastFieldReader};
use Postings;
use rand::{XorShiftRng, Rng, SeedableRng};
fn generate_array_with_seed(n: usize, ratio: f32, seed_val: u32) -> Vec<u32> {
let seed: &[u32; 4] = &[1, 2, 3, seed_val];
let mut rng: XorShiftRng = XorShiftRng::from_seed(*seed);
(0..u32::max_value())
.filter(|_| rng.next_f32() < ratio)
.take(n)
.collect()
}
pub fn generate_array(n: usize, ratio: f32) -> Vec<u32> {
generate_array_with_seed(n, ratio, 4)
}
fn sample_with_seed(n: u32, ratio: f32, seed_val: u32) -> Vec<u32> {
let seed: &[u32; 4] = &[1, 2, 3, seed_val];
let mut rng: XorShiftRng = XorShiftRng::from_seed(*seed);
(0..n).filter(|_| rng.next_f32() < ratio).collect()
}
pub fn sample(n: u32, ratio: f32) -> Vec<u32> {
sample_with_seed(n, ratio, 4)
}
#[test]
fn test_indexing() {
@@ -274,8 +299,8 @@ mod tests {
assert_eq!(searcher.doc_freq(&term_d), 0);
}
}
#[test]
fn test_fieldnorm() {
let mut schema_builder = SchemaBuilder::default();
@@ -318,19 +343,23 @@ mod tests {
{
// writing the segment
let mut index_writer = index.writer_with_num_threads(1, 40_000_000).unwrap();
{ // 0
{
// 0
let doc = doc!(text_field=>"a b");
index_writer.add_document(doc);
}
{ // 1
{
// 1
let doc = doc!(text_field=>" a c");
index_writer.add_document(doc);
}
{ // 2
{
// 2
let doc = doc!(text_field=>" b c");
index_writer.add_document(doc);
}
{ // 3
{
// 3
let doc = doc!(text_field=>" b d");
index_writer.add_document(doc);
}
@@ -340,11 +369,13 @@ mod tests {
{
index_writer.delete_term(Term::from_field_text(text_field, "a"));
}
{ // 4
{
// 4
let doc = doc!(text_field=>" b c");
index_writer.add_document(doc);
}
{ // 5
{
// 5
let doc = doc!(text_field=>" a");
index_writer.add_document(doc);
}
@@ -354,15 +385,21 @@ mod tests {
index.load_searchers().unwrap();
let searcher = index.searcher();
let reader = searcher.segment_reader(0);
assert!(reader.read_postings_all_info(&Term::from_field_text(text_field, "abcd")).is_none());
assert!(reader
.read_postings_all_info(&Term::from_field_text(text_field, "abcd"))
.is_none());
{
let mut postings = reader.read_postings_all_info(&Term::from_field_text(text_field, "a")).unwrap();
let mut postings = reader
.read_postings_all_info(&Term::from_field_text(text_field, "a"))
.unwrap();
assert!(postings.advance());
assert_eq!(postings.doc(), 5);
assert!(!postings.advance());
}
{
let mut postings = reader.read_postings_all_info(&Term::from_field_text(text_field, "b")).unwrap();
let mut postings = reader
.read_postings_all_info(&Term::from_field_text(text_field, "b"))
.unwrap();
assert!(postings.advance());
assert_eq!(postings.doc(), 3);
assert!(postings.advance());
@@ -373,11 +410,13 @@ mod tests {
{
// writing the segment
let mut index_writer = index.writer_with_num_threads(1, 40_000_000).unwrap();
{ // 0
{
// 0
let doc = doc!(text_field=>"a b");
index_writer.add_document(doc);
}
{ // 1
{
// 1
index_writer.delete_term(Term::from_field_text(text_field, "c"));
}
index_writer.rollback().unwrap();
@@ -386,15 +425,21 @@ mod tests {
index.load_searchers().unwrap();
let searcher = index.searcher();
let reader = searcher.segment_reader(0);
assert!(reader.read_postings_all_info(&Term::from_field_text(text_field, "abcd")).is_none());
assert!(reader
.read_postings_all_info(&Term::from_field_text(text_field, "abcd"))
.is_none());
{
let mut postings = reader.read_postings_all_info(&Term::from_field_text(text_field, "a")).unwrap();
let mut postings = reader
.read_postings_all_info(&Term::from_field_text(text_field, "a"))
.unwrap();
assert!(postings.advance());
assert_eq!(postings.doc(), 5);
assert!(!postings.advance());
}
{
let mut postings = reader.read_postings_all_info(&Term::from_field_text(text_field, "b")).unwrap();
let mut postings = reader
.read_postings_all_info(&Term::from_field_text(text_field, "b"))
.unwrap();
assert!(postings.advance());
assert_eq!(postings.doc(), 3);
assert!(postings.advance());
@@ -405,14 +450,14 @@ mod tests {
{
// writing the segment
let mut index_writer = index.writer_with_num_threads(1, 40_000_000).unwrap();
{
{
let doc = doc!(text_field=>"a b");
index_writer.add_document(doc);
}
{
{
index_writer.delete_term(Term::from_field_text(text_field, "c"));
}
index_writer = index_writer.rollback().unwrap();
index_writer = index_writer.rollback().unwrap();
index_writer.delete_term(Term::from_field_text(text_field, "a"));
index_writer.commit().unwrap();
}
@@ -420,13 +465,19 @@ mod tests {
index.load_searchers().unwrap();
let searcher = index.searcher();
let reader = searcher.segment_reader(0);
assert!(reader.read_postings_all_info(&Term::from_field_text(text_field, "abcd")).is_none());
assert!(reader
.read_postings_all_info(&Term::from_field_text(text_field, "abcd"))
.is_none());
{
let mut postings = reader.read_postings_all_info(&Term::from_field_text(text_field, "a")).unwrap();
let mut postings = reader
.read_postings_all_info(&Term::from_field_text(text_field, "a"))
.unwrap();
assert!(!postings.advance());
}
{
let mut postings = reader.read_postings_all_info(&Term::from_field_text(text_field, "b")).unwrap();
let mut postings = reader
.read_postings_all_info(&Term::from_field_text(text_field, "b"))
.unwrap();
assert!(postings.advance());
assert_eq!(postings.doc(), 3);
assert!(postings.advance());
@@ -434,7 +485,9 @@ mod tests {
assert!(!postings.advance());
}
{
let mut postings = reader.read_postings_all_info(&Term::from_field_text(text_field, "c")).unwrap();
let mut postings = reader
.read_postings_all_info(&Term::from_field_text(text_field, "c"))
.unwrap();
assert!(postings.advance());
assert_eq!(postings.doc(), 4);
assert!(!postings.advance());
@@ -448,17 +501,18 @@ mod tests {
let mut schema_builder = SchemaBuilder::default();
let field = schema_builder.add_u64_field("value", INT_INDEXED);
let schema = schema_builder.build();
let index = Index::create_in_ram(schema);
let mut index_writer = index.writer_with_num_threads(1, 40_000_000).unwrap();
index_writer.add_document(
doc!(field=>1u64)
);
index_writer.add_document(doc!(field=>1u64));
index_writer.commit().unwrap();
index.load_searchers().unwrap();
let searcher = index.searcher();
let term = Term::from_field_u64(field, 1u64);
let mut postings = searcher.segment_reader(0).read_postings(&term, SegmentPostingsOption::NoFreq).unwrap();
let mut postings = searcher
.segment_reader(0)
.read_postings(&term, SegmentPostingsOption::NoFreq)
.unwrap();
assert!(postings.advance());
assert_eq!(postings.doc(), 0);
assert!(!postings.advance());
@@ -469,20 +523,19 @@ mod tests {
let mut schema_builder = SchemaBuilder::default();
let value_field = schema_builder.add_i64_field("value", INT_INDEXED);
let schema = schema_builder.build();
let index = Index::create_in_ram(schema);
let mut index_writer = index.writer_with_num_threads(1, 40_000_000).unwrap();
let negative_val = -1i64;
index_writer.add_document(
doc!(value_field => negative_val)
);
index_writer.add_document(doc!(value_field => negative_val));
index_writer.commit().unwrap();
index.load_searchers().unwrap();
let searcher = index.searcher();
let term = Term::from_field_i64(value_field, negative_val);
let mut postings = searcher
.segment_reader(0)
.read_postings(&term, SegmentPostingsOption::NoFreq).unwrap();
.read_postings(&term, SegmentPostingsOption::NoFreq)
.unwrap();
assert!(postings.advance());
assert_eq!(postings.doc(), 0);
assert!(!postings.advance());
@@ -494,15 +547,15 @@ mod tests {
let text_field = schema_builder.add_text_field("text", TEXT);
let schema = schema_builder.build();
let index = Index::create_in_ram(schema);
// writing the segment
let mut index_writer = index.writer_with_num_threads(2, 40_000_000).unwrap();
let add_document = |index_writer: &mut IndexWriter, val: &'static str| {
let doc = doc!(text_field=>val);
index_writer.add_document(doc);
};
let remove_document = |index_writer: &mut IndexWriter, val: &'static str| {
let delterm = Term::from_field_text(text_field, val);
index_writer.delete_term(delterm);
@@ -543,8 +596,12 @@ mod tests {
index.load_searchers().unwrap();
let searcher = index.searcher();
let reader = searcher.segment_reader(0);
assert!(reader.read_postings_all_info(&Term::from_field_text(text_field, "abcd")).is_none());
let mut postings = reader.read_postings_all_info(&Term::from_field_text(text_field, "af")).unwrap();
assert!(reader
.read_postings_all_info(&Term::from_field_text(text_field, "abcd"))
.is_none());
let mut postings = reader
.read_postings_all_info(&Term::from_field_text(text_field, "af"))
.unwrap();
assert!(postings.advance());
assert_eq!(postings.doc(), 0);
assert_eq!(postings.term_freq(), 3);
@@ -586,35 +643,29 @@ mod tests {
collector.docs()
};
{
assert_eq!(
get_doc_ids(vec!(Term::from_field_text(text_field, "a"))),
vec!(1, 2));
assert_eq!(get_doc_ids(vec![Term::from_field_text(text_field, "a")]),
vec![1, 2]);
}
{
assert_eq!(
get_doc_ids(vec!(Term::from_field_text(text_field, "af"))),
vec!(0));
assert_eq!(get_doc_ids(vec![Term::from_field_text(text_field, "af")]),
vec![0]);
}
{
assert_eq!(
get_doc_ids(vec!(Term::from_field_text(text_field, "b"))),
vec!(0, 1, 2));
assert_eq!(get_doc_ids(vec![Term::from_field_text(text_field, "b")]),
vec![0, 1, 2]);
}
{
assert_eq!(
get_doc_ids(vec!(Term::from_field_text(text_field, "c"))),
vec!(1, 2));
assert_eq!(get_doc_ids(vec![Term::from_field_text(text_field, "c")]),
vec![1, 2]);
}
{
assert_eq!(
get_doc_ids(vec!(Term::from_field_text(text_field, "d"))),
vec!(2));
assert_eq!(get_doc_ids(vec![Term::from_field_text(text_field, "d")]),
vec![2]);
}
{
assert_eq!(
get_doc_ids(vec!(Term::from_field_text(text_field, "b"),
Term::from_field_text(text_field, "a"), )),
vec!(0, 1, 2));
assert_eq!(get_doc_ids(vec![Term::from_field_text(text_field, "b"),
Term::from_field_text(text_field, "a")]),
vec![0, 1, 2]);
}
}
}
@@ -651,7 +702,9 @@ mod tests {
let mut schema_builder = SchemaBuilder::default();
let text_field = schema_builder.add_text_field("text", TEXT);
let other_text_field = schema_builder.add_text_field("text2", TEXT);
let document = doc!(text_field => "tantivy", text_field => "some other value", other_text_field => "short");
let document = doc!(text_field => "tantivy",
text_field => "some other value",
other_text_field => "short");
assert_eq!(document.len(), 3);
let values = document.get_all(text_field);
assert_eq!(values.len(), 2);
@@ -683,30 +736,35 @@ mod tests {
let searcher = index.searcher();
let segment_reader: &SegmentReader = searcher.segment_reader(0);
{
let fast_field_reader_res = segment_reader.get_fast_field_reader::<U64FastFieldReader>(text_field);
let fast_field_reader_res =
segment_reader.get_fast_field_reader::<U64FastFieldReader>(text_field);
assert!(fast_field_reader_res.is_err());
}
{
let fast_field_reader_res = segment_reader.get_fast_field_reader::<U64FastFieldReader>(stored_int_field);
let fast_field_reader_res =
segment_reader.get_fast_field_reader::<U64FastFieldReader>(stored_int_field);
assert!(fast_field_reader_res.is_err());
}
{
let fast_field_reader_res = segment_reader.get_fast_field_reader::<U64FastFieldReader>(fast_field_signed);
let fast_field_reader_res =
segment_reader.get_fast_field_reader::<U64FastFieldReader>(fast_field_signed);
assert!(fast_field_reader_res.is_err());
}
{
let fast_field_reader_res = segment_reader.get_fast_field_reader::<I64FastFieldReader>(fast_field_signed);
let fast_field_reader_res =
segment_reader.get_fast_field_reader::<I64FastFieldReader>(fast_field_signed);
assert!(fast_field_reader_res.is_ok());
let fast_field_reader = fast_field_reader_res.unwrap();
assert_eq!(fast_field_reader.get(0), 4i64)
}
{
let fast_field_reader_res = segment_reader.get_fast_field_reader::<I64FastFieldReader>(fast_field_signed);
let fast_field_reader_res =
segment_reader.get_fast_field_reader::<I64FastFieldReader>(fast_field_signed);
assert!(fast_field_reader_res.is_ok());
let fast_field_reader = fast_field_reader_res.unwrap();
assert_eq!(fast_field_reader.get(0), 4i64)
}
}
}

2
src/postings/freq_handler.rs
View File

@@ -123,4 +123,4 @@ impl FreqHandler {
}
}
}
}
}

304
src/postings/mod.rs
View File

@@ -35,9 +35,9 @@ pub use common::HasLen;
#[cfg(test)]
mod tests {
use super::*;
use schema::{Document, TEXT, STRING, SchemaBuilder, Term};
use schema::{Document, INT_INDEXED, TEXT, STRING, SchemaBuilder, Term};
use core::SegmentComponent;
use indexer::SegmentWriter;
use core::SegmentReader;
@@ -49,9 +49,10 @@ mod tests {
use schema::Field;
use test::Bencher;
use indexer::operation::AddOperation;
use tests;
use rand::{XorShiftRng, Rng, SeedableRng};
#[test]
pub fn test_position_write() {
let mut schema_builder = SchemaBuilder::default();
@@ -63,7 +64,7 @@ mod tests {
posting_serializer.new_field(text_field);
posting_serializer.new_term("abc".as_bytes()).unwrap();
for doc_id in 0u32..3u32 {
let positions = vec!(1,2,3,2);
let positions = vec![1, 2, 3, 2];
posting_serializer.write_doc(doc_id, 2, &positions).unwrap();
}
posting_serializer.close_term().unwrap();
@@ -71,7 +72,7 @@ mod tests {
let read = segment.open_read(SegmentComponent::POSITIONS).unwrap();
assert!(read.len() <= 16);
}
#[test]
pub fn test_position_and_fieldnorm() {
let mut schema_builder = SchemaBuilder::default();
@@ -81,14 +82,16 @@ mod tests {
let segment = index.new_segment();
let heap = Heap::with_capacity(10_000_000);
{
let mut segment_writer = SegmentWriter::for_segment(&heap, segment.clone(), &schema).unwrap();
let mut segment_writer = SegmentWriter::for_segment(&heap, segment.clone(), &schema)
.unwrap();
{
let mut doc = Document::default();
// checking that position works if the field has two values
doc.add_text(text_field, "a b a c a d a a.");
doc.add_text(text_field, "d d d d a"); // checking that position works if the field has two values.
doc.add_text(text_field, "d d d d a");
let op = AddOperation {
opstamp: 0u64,
document: doc,
document: doc,
};
segment_writer.add_document(&op, &schema).unwrap();
}
@@ -97,7 +100,7 @@ mod tests {
doc.add_text(text_field, "b a");
let op = AddOperation {
opstamp: 1u64,
document: doc,
document: doc,
};
segment_writer.add_document(&op, &schema).unwrap();
}
@@ -108,7 +111,7 @@ mod tests {
doc.add_text(text_field, &text);
let op = AddOperation {
opstamp: 2u64,
document: doc,
document: doc,
};
segment_writer.add_document(&op, &schema).unwrap();
}
@@ -120,7 +123,7 @@ mod tests {
let fieldnorm_reader = segment_reader.get_fieldnorms_reader(text_field).unwrap();
assert_eq!(fieldnorm_reader.get(0), 8 + 5);
assert_eq!(fieldnorm_reader.get(1), 2);
for i in 2 .. 1000 {
for i in 2..1000 {
assert_eq!(fieldnorm_reader.get(i), (i + 1) as u64);
}
}
@@ -139,7 +142,7 @@ mod tests {
assert!(postings_a.advance());
assert_eq!(postings_a.doc(), 1u32);
assert_eq!(postings_a.term_freq(), 1);
for i in 2u32 .. 1000u32 {
for i in 2u32..1000u32 {
assert!(postings_a.advance());
assert_eq!(postings_a.term_freq(), 1);
assert_eq!(postings_a.positions(), [i]);
@@ -151,7 +154,7 @@ mod tests {
let term_e = Term::from_field_text(text_field, "e");
let mut postings_e = segment_reader.read_postings_all_info(&term_e).unwrap();
assert_eq!(postings_e.len(), 1000 - 2);
for i in 2u32 .. 1000u32 {
for i in 2u32..1000u32 {
assert!(postings_e.advance());
assert_eq!(postings_e.term_freq(), i);
let positions = postings_e.positions();
@@ -165,7 +168,7 @@ mod tests {
}
}
}
#[test]
pub fn test_position_and_fieldnorm2() {
let mut schema_builder = SchemaBuilder::default();
@@ -187,7 +190,8 @@ mod tests {
assert!(index_writer.commit().is_ok());
}
index.load_searchers().unwrap();
let term_query = TermQuery::new(Term::from_field_text(text_field, "a"), SegmentPostingsOption::NoFreq);
let term_query = TermQuery::new(Term::from_field_text(text_field, "a"),
SegmentPostingsOption::NoFreq);
let searcher = index.searcher();
let mut term_weight = term_query.specialized_weight(&*searcher);
term_weight.segment_postings_options = SegmentPostingsOption::FreqAndPositions;
@@ -197,13 +201,178 @@ mod tests {
assert_eq!(term_scorer.doc(), 1u32);
assert_eq!(term_scorer.postings().positions(), &[1u32, 4]);
}
#[test]
fn test_skip_next() {
let term_0 = Term::from_field_u64(Field(0), 0);
let term_1 = Term::from_field_u64(Field(0), 1);
let term_2 = Term::from_field_u64(Field(0), 2);
let num_docs = 300u32;
let index = {
let mut schema_builder = SchemaBuilder::default();
let value_field = schema_builder.add_u64_field("value", INT_INDEXED);
let schema = schema_builder.build();
let index = Index::create_in_ram(schema);
{
let mut index_writer = index.writer_with_num_threads(1, 40_000_000).unwrap();
for i in 0..num_docs {
let mut doc = Document::default();
doc.add_u64(value_field, 2);
doc.add_u64(value_field, (i % 2) as u64);
index_writer.add_document(doc);
}
assert!(index_writer.commit().is_ok());
}
index.load_searchers().unwrap();
index
};
let searcher = index.searcher();
let segment_reader = searcher.segment_reader(0);
// check that the basic usage works
for i in 0..num_docs - 1 {
for j in i + 1..num_docs {
let mut segment_postings = segment_reader
.read_postings(&term_2, SegmentPostingsOption::NoFreq)
.unwrap();
assert_eq!(segment_postings.skip_next(i), SkipResult::Reached);
assert_eq!(segment_postings.doc(), i);
assert_eq!(segment_postings.skip_next(j), SkipResult::Reached);
assert_eq!(segment_postings.doc(), j);
}
}
{
let mut segment_postings = segment_reader
.read_postings(&term_2, SegmentPostingsOption::NoFreq)
.unwrap();
// check that `skip_next` advances the iterator
assert!(segment_postings.advance());
assert_eq!(segment_postings.doc(), 0);
assert_eq!(segment_postings.skip_next(1), SkipResult::Reached);
assert_eq!(segment_postings.doc(), 1);
assert_eq!(segment_postings.skip_next(1), SkipResult::OverStep);
assert_eq!(segment_postings.doc(), 2);
// check that going beyond the end is handled
assert_eq!(segment_postings.skip_next(num_docs), SkipResult::End);
}
// check that filtering works
{
let mut segment_postings = segment_reader
.read_postings(&term_0, SegmentPostingsOption::NoFreq)
.unwrap();
for i in 0..num_docs / 2 {
assert_eq!(segment_postings.skip_next(i * 2), SkipResult::Reached);
assert_eq!(segment_postings.doc(), i * 2);
}
let mut segment_postings = segment_reader
.read_postings(&term_0, SegmentPostingsOption::NoFreq)
.unwrap();
for i in 0..num_docs / 2 - 1 {
assert_eq!(segment_postings.skip_next(i * 2 + 1), SkipResult::OverStep);
assert_eq!(segment_postings.doc(), (i + 1) * 2);
}
}
// delete some of the documents
{
let mut index_writer = index.writer_with_num_threads(1, 40_000_000).unwrap();
index_writer.delete_term(term_0);
assert!(index_writer.commit().is_ok());
}
index.load_searchers().unwrap();
let searcher = index.searcher();
let segment_reader = searcher.segment_reader(0);
// make sure seeking still works
for i in 0..num_docs {
let mut segment_postings = segment_reader
.read_postings(&term_2, SegmentPostingsOption::NoFreq)
.unwrap();
if i % 2 == 0 {
assert_eq!(segment_postings.skip_next(i), SkipResult::OverStep);
assert_eq!(segment_postings.doc(), i + 1);
} else {
assert_eq!(segment_postings.skip_next(i), SkipResult::Reached);
assert_eq!(segment_postings.doc(), i);
}
}
// now try with a longer sequence
{
let mut segment_postings = segment_reader
.read_postings(&term_2, SegmentPostingsOption::NoFreq)
.unwrap();
let mut last = 2; // start from 5 to avoid seeking to 3 twice
let mut cur = 3;
loop {
match segment_postings.skip_next(cur) {
SkipResult::End => break,
SkipResult::Reached => assert_eq!(segment_postings.doc(), cur),
SkipResult::OverStep => assert_eq!(segment_postings.doc(), cur + 1),
}
let next = cur + last;
last = cur;
cur = next;
}
assert_eq!(cur, 377);
}
// delete everything else
{
let mut index_writer = index.writer_with_num_threads(1, 40_000_000).unwrap();
index_writer.delete_term(term_1);
assert!(index_writer.commit().is_ok());
}
index.load_searchers().unwrap();
let searcher = index.searcher();
let segment_reader = searcher.segment_reader(0);
// finally, check that it's empty
{
let mut segment_postings = segment_reader
.read_postings(&term_2, SegmentPostingsOption::NoFreq)
.unwrap();
assert_eq!(segment_postings.skip_next(0), SkipResult::End);
let mut segment_postings = segment_reader
.read_postings(&term_2, SegmentPostingsOption::NoFreq)
.unwrap();
assert_eq!(segment_postings.skip_next(num_docs), SkipResult::End);
}
}
#[test]
fn test_intersection() {
{
let left = VecPostings::from(vec!(1, 3, 9));
let right = VecPostings::from(vec!(3, 4, 9, 18));
let mut intersection = IntersectionDocSet::from(vec!(left, right));
let left = VecPostings::from(vec![1, 3, 9]);
let right = VecPostings::from(vec![3, 4, 9, 18]);
let mut intersection = IntersectionDocSet::from(vec![left, right]);
assert!(intersection.advance());
assert_eq!(intersection.doc(), 3);
assert!(intersection.advance());
@@ -211,17 +380,17 @@ mod tests {
assert!(!intersection.advance());
}
{
let a = VecPostings::from(vec!(1, 3, 9));
let b = VecPostings::from(vec!(3, 4, 9, 18));
let c = VecPostings::from(vec!(1, 5, 9, 111));
let mut intersection = IntersectionDocSet::from(vec!(a, b, c));
let a = VecPostings::from(vec![1, 3, 9]);
let b = VecPostings::from(vec![3, 4, 9, 18]);
let c = VecPostings::from(vec![1, 5, 9, 111]);
let mut intersection = IntersectionDocSet::from(vec![a, b, c]);
assert!(intersection.advance());
assert_eq!(intersection.doc(), 9);
assert!(!intersection.advance());
}
}
lazy_static! {
static ref TERM_A: Term = {
let field = Field(0);
@@ -235,10 +404,10 @@ mod tests {
let mut schema_builder = SchemaBuilder::default();
let text_field = schema_builder.add_text_field("text", STRING);
let schema = schema_builder.build();
let seed: &[u32; 4] = &[1, 2, 3, 4];
let mut rng: XorShiftRng = XorShiftRng::from_seed(*seed);
let index = Index::create_in_ram(schema);
let mut count_a = 0;
let mut count_b = 0;
@@ -267,27 +436,84 @@ mod tests {
index
};
}
#[bench]
fn bench_segment_postings(b: &mut Bencher) {
let searcher = INDEX.searcher();
let segment_reader = searcher.segment_reader(0);
b.iter(|| {
let mut segment_postings = segment_reader.read_postings(&*TERM_A, SegmentPostingsOption::NoFreq).unwrap();
while segment_postings.advance() {}
});
}
let mut segment_postings = segment_reader
.read_postings(&*TERM_A, SegmentPostingsOption::NoFreq)
.unwrap();
while segment_postings.advance() {}
});
}
#[bench]
fn bench_segment_intersection(b: &mut Bencher) {
let searcher = INDEX.searcher();
let segment_reader = searcher.segment_reader(0);
b.iter(|| {
let segment_postings_a = segment_reader.read_postings(&*TERM_A, SegmentPostingsOption::NoFreq).unwrap();
let segment_postings_b = segment_reader.read_postings(&*TERM_B, SegmentPostingsOption::NoFreq).unwrap();
let mut intersection = IntersectionDocSet::from(vec!(segment_postings_a, segment_postings_b));
let segment_postings_a = segment_reader
.read_postings(&*TERM_A, SegmentPostingsOption::NoFreq)
.unwrap();
let segment_postings_b = segment_reader
.read_postings(&*TERM_B, SegmentPostingsOption::NoFreq)
.unwrap();
let mut intersection = IntersectionDocSet::from(vec![segment_postings_a,
segment_postings_b]);
while intersection.advance() {}
});
}
}
fn bench_skip_next(p: f32, b: &mut Bencher) {
let searcher = INDEX.searcher();
let segment_reader = searcher.segment_reader(0);
let docs = tests::sample(segment_reader.num_docs(), p);
let mut segment_postings = segment_reader
.read_postings(&*TERM_A, SegmentPostingsOption::NoFreq)
.unwrap();
let mut existing_docs = Vec::new();
for doc in &docs {
if *doc >= segment_postings.doc() {
existing_docs.push(*doc);
if segment_postings.skip_next(*doc) == SkipResult::End {
break;
}
}
}
b.iter(|| {
let mut segment_postings = segment_reader
.read_postings(&*TERM_A, SegmentPostingsOption::NoFreq)
.unwrap();
for doc in &existing_docs {
if segment_postings.skip_next(*doc) == SkipResult::End {
break;
}
}
});
}
#[bench]
fn bench_skip_next_p01(b: &mut Bencher) {
bench_skip_next(0.001, b);
}
#[bench]
fn bench_skip_next_p1(b: &mut Bencher) {
bench_skip_next(0.01, b);
}
#[bench]
fn bench_skip_next_p10(b: &mut Bencher) {
bench_skip_next(0.1, b);
}
#[bench]
fn bench_skip_next_p90(b: &mut Bencher) {
bench_skip_next(0.9, b);
}
}

8
src/postings/postings_writer.rs
View File

@@ -142,7 +142,7 @@ pub trait PostingsWriter {
/// * heap - heap used to store the postings informations as well as the terms
/// in the hashmap.
fn suscribe(&mut self, term_index: &mut HashMap, doc: DocId, pos: u32, term: &Term, heap: &Heap);
/// Serializes the postings on disk.
/// The actual serialization format is handled by the `PostingsSerializer`.
fn serialize(&self, field: Field, term_addrs: &[(&[u8], u32)], serializer: &mut PostingsSerializer, heap: &Heap) -> io::Result<()>;
@@ -226,7 +226,7 @@ impl<'a, Rec: Recorder + 'static> PostingsWriter for SpecializedPostingsWriter<'
}
recorder.record_position(position, heap);
}
fn serialize(&self,
field: Field,
term_addrs: &[(&[u8], u32)],
@@ -235,7 +235,7 @@ impl<'a, Rec: Recorder + 'static> PostingsWriter for SpecializedPostingsWriter<'
serializer.new_field(field);
for &(term_bytes, addr) in term_addrs {
let recorder: &mut Rec = self.heap.get_mut_ref(addr);
try!(serializer.new_term(&term_bytes));
try!(serializer.new_term(term_bytes));
try!(recorder.serialize(addr, serializer, heap));
try!(serializer.close_term());
}
@@ -250,4 +250,4 @@ fn test_hashmap_size() {
assert_eq!(hashmap_size_in_bits(0), 10);
assert_eq!(hashmap_size_in_bits(100_000), 11);
assert_eq!(hashmap_size_in_bits(300_000_000), 23);
}
}

1
src/postings/recorder.rs
View File

@@ -122,7 +122,6 @@ impl Recorder for TermFrequencyRecorder {
serializer: &mut PostingsSerializer,
heap: &Heap)
-> io::Result<()> {
// the last document has not been closed...
// its term freq is self.current_tf.
let mut doc_iter = self.stack

114
src/postings/segment_postings.rs
View File

@@ -1,6 +1,7 @@
use compression::{NUM_DOCS_PER_BLOCK, BlockDecoder, VIntDecoder};
use DocId;
use postings::{Postings, FreqHandler, DocSet, HasLen};
use postings::{Postings, FreqHandler, DocSet, HasLen, SkipResult};
use std::cmp;
use std::num::Wrapping;
use fastfield::DeleteBitSet;
@@ -14,6 +15,9 @@ const EMPTY_DATA: [u8; 0] = [0u8; 0];
/// Positions on the other hand, are optionally entirely decoded upfront.
pub struct SegmentPostings<'a> {
len: usize,
// Removing this makes the code slower
// See https://github.com/tantivy-search/tantivy/issues/89
block_len: usize,
doc_offset: u32,
block_decoder: BlockDecoder,
freq_handler: FreqHandler,
@@ -26,14 +30,21 @@ impl<'a> SegmentPostings<'a> {
fn load_next_block(&mut self) {
let num_remaining_docs = self.len - self.cur.0;
if num_remaining_docs >= NUM_DOCS_PER_BLOCK {
self.remaining_data = self.block_decoder
.uncompress_block_sorted(self.remaining_data, self.doc_offset);
self.remaining_data =
self.block_decoder
.uncompress_block_sorted(self.remaining_data, self.doc_offset);
self.remaining_data = self.freq_handler.read_freq_block(self.remaining_data);
self.doc_offset = self.block_decoder.output(NUM_DOCS_PER_BLOCK - 1);
self.block_len = NUM_DOCS_PER_BLOCK;
} else {
self.remaining_data = self.block_decoder
.uncompress_vint_sorted(self.remaining_data, self.doc_offset, num_remaining_docs);
self.freq_handler.read_freq_vint(self.remaining_data, num_remaining_docs);
self.remaining_data =
self.block_decoder
.uncompress_vint_sorted(self.remaining_data,
self.doc_offset,
num_remaining_docs);
self.freq_handler
.read_freq_vint(self.remaining_data, num_remaining_docs);
self.block_len = num_remaining_docs;
}
}
@@ -46,9 +57,11 @@ impl<'a> SegmentPostings<'a> {
pub fn from_data(len: u32,
data: &'a [u8],
delete_bitset: &'a DeleteBitSet,
freq_handler: FreqHandler) -> SegmentPostings<'a> {
freq_handler: FreqHandler)
-> SegmentPostings<'a> {
SegmentPostings {
len: len as usize,
block_len: len as usize,
doc_offset: 0,
block_decoder: BlockDecoder::new(),
freq_handler: freq_handler,
@@ -62,6 +75,7 @@ impl<'a> SegmentPostings<'a> {
pub fn empty() -> SegmentPostings<'static> {
SegmentPostings {
len: 0,
block_len: 0,
doc_offset: 0,
block_decoder: BlockDecoder::new(),
freq_handler: FreqHandler::new_without_freq(),
@@ -76,6 +90,13 @@ impl<'a> SegmentPostings<'a> {
fn index_within_block(&self) -> usize {
self.cur.0 % NUM_DOCS_PER_BLOCK
}
/// Sets the current position to a location relative
/// to the current block
#[inline]
fn set_within_block(&mut self, inner_pos: usize) {
self.cur = Wrapping(self.cur.0 & !(NUM_DOCS_PER_BLOCK - 1)) + Wrapping(inner_pos)
}
}
@@ -98,6 +119,85 @@ impl<'a> DocSet for SegmentPostings<'a> {
}
}
fn skip_next(&mut self, target: DocId) -> SkipResult {
if !self.advance() {
return SkipResult::End;
}
let mut pos = self.index_within_block();
// skip blocks until one that might contain the target
loop {
// check if we need to go to the next block
if target > self.block_decoder.output(self.block_len - 1) {
self.cur += Wrapping(self.block_len - pos);
self.load_next_block();
pos = 0;
// there was no more data
if self.cur.0 == self.len {
return SkipResult::End;
}
} else if target < self.block_decoder.output(pos) {
// We've overpassed the target after the first `advance` call
// or we're at the beginning of a block.
// Either way, we're on the first `DocId` greater than `target`
return SkipResult::OverStep;
} else {
break;
}
}
debug_assert!(target >= self.block_decoder.output(pos));
debug_assert!(target <= self.block_decoder.output(self.block_len - 1));
// we're in the right block now, start with an exponential search
let mut start = pos;
let mut end = self.block_len;
let mut count = 1;
loop {
let new = start + count;
if new < end && self.block_decoder.output(new) < target {
start = new;
count *= 2;
} else {
break;
}
}
end = cmp::min(start + count, end);
// now do a binary search
let mut count = end - start;
while count > 0 {
let step = count / 2;
let mid = start + step;
let doc = self.block_decoder.output(mid);
if doc < target {
start = mid + 1;
count -= step + 1;
} else {
count = step;
}
}
// `doc` is now >= `target`
let doc = self.block_decoder.output(start);
self.set_within_block(start);
if !self.delete_bitset.is_deleted(doc) {
if doc == target {
return SkipResult::Reached;
} else {
return SkipResult::OverStep;
}
}
if self.advance() {
SkipResult::OverStep
} else {
SkipResult::End
}
}
#[inline]
fn doc(&self) -> DocId {
self.block_decoder.output(self.index_within_block())

57
src/postings/serializer.rs
View File

@@ -76,22 +76,22 @@ impl PostingsSerializer {
-> Result<PostingsSerializer> {
let terms_fst_builder = try!(FstMapBuilder::new(terms_write));
Ok(PostingsSerializer {
terms_fst_builder: terms_fst_builder,
postings_write: postings_write,
positions_write: positions_write,
written_bytes_postings: 0,
written_bytes_positions: 0,
last_doc_id_encoded: 0u32,
positions_encoder: CompositeEncoder::new(),
block_encoder: BlockEncoder::new(),
doc_ids: Vec::new(),
term_freqs: Vec::new(),
position_deltas: Vec::new(),
schema: schema,
text_indexing_options: TextIndexingOptions::Unindexed,
term_open: false,
current_term_info: TermInfo::default(),
})
terms_fst_builder: terms_fst_builder,
postings_write: postings_write,
positions_write: positions_write,
written_bytes_postings: 0,
written_bytes_positions: 0,
last_doc_id_encoded: 0u32,
positions_encoder: CompositeEncoder::new(),
block_encoder: BlockEncoder::new(),
doc_ids: Vec::new(),
term_freqs: Vec::new(),
position_deltas: Vec::new(),
schema: schema,
text_indexing_options: TextIndexingOptions::Unindexed,
term_open: false,
current_term_info: TermInfo::default(),
})
}
@@ -114,20 +114,14 @@ impl PostingsSerializer {
let field_entry: &FieldEntry = self.schema.get_field_entry(field);
self.text_indexing_options = match *field_entry.field_type() {
FieldType::Str(ref text_options) => text_options.get_indexing_options(),
FieldType::U64(ref int_options) => {
if int_options.is_indexed() {
TextIndexingOptions::Unindexed
} else {
TextIndexingOptions::Untokenized
}
}
FieldType::U64(ref int_options) |
FieldType::I64(ref int_options) => {
if int_options.is_indexed() {
TextIndexingOptions::Unindexed
} else {
TextIndexingOptions::Untokenized
}
}
}
};
}
@@ -159,7 +153,8 @@ impl PostingsSerializer {
pub fn close_term(&mut self) -> io::Result<()> {
if self.term_open {
self.terms_fst_builder.insert_value(&self.current_term_info)?;
self.terms_fst_builder
.insert_value(&self.current_term_info)?;
if !self.doc_ids.is_empty() {
// we have doc ids waiting to be written
@@ -169,8 +164,9 @@ impl PostingsSerializer {
// In that case, the remaining part is encoded
// using variable int encoding.
{
let block_encoded = self.block_encoder
.compress_vint_sorted(&self.doc_ids, self.last_doc_id_encoded);
let block_encoded =
self.block_encoder
.compress_vint_sorted(&self.doc_ids, self.last_doc_id_encoded);
self.written_bytes_postings += block_encoded.len();
try!(self.postings_write.write_all(block_encoded));
self.doc_ids.clear();
@@ -190,7 +186,7 @@ impl PostingsSerializer {
// end of the term, at which point they are compressed and written.
if self.text_indexing_options.is_position_enabled() {
self.written_bytes_positions += try!(VInt(self.position_deltas.len() as u64)
.serialize(&mut self.positions_write));
.serialize(&mut self.positions_write));
let positions_encoded: &[u8] = self.positions_encoder
.compress_unsorted(&self.position_deltas[..]);
try!(self.positions_write.write_all(positions_encoded));
@@ -228,8 +224,9 @@ impl PostingsSerializer {
if self.doc_ids.len() == NUM_DOCS_PER_BLOCK {
{
// encode the doc ids
let block_encoded: &[u8] = self.block_encoder
.compress_block_sorted(&self.doc_ids, self.last_doc_id_encoded);
let block_encoded: &[u8] =
self.block_encoder
.compress_block_sorted(&self.doc_ids, self.last_doc_id_encoded);
self.last_doc_id_encoded = self.doc_ids[self.doc_ids.len() - 1];
try!(self.postings_write.write_all(block_encoded));
self.written_bytes_postings += block_encoded.len();

23
src/postings/term_info.rs
View File

@@ -2,21 +2,21 @@ use common::BinarySerializable;
use std::io;
/// `TermInfo` contains all of the information
/// `TermInfo` contains all of the information
/// associated to terms in the `.term` file.
///
///
/// It consists of
/// * `doc_freq` : the number of document in the segment
/// containing this term. It is also the length of the
/// posting list associated to this term
/// * `postings_offset` : an offset in the `.idx` file
/// * `postings_offset` : an offset in the `.idx` file
/// addressing the start of the posting list associated
/// to this term.
#[derive(Debug,Default,Ord,PartialOrd,Eq,PartialEq,Clone)]
pub struct TermInfo {
/// Number of documents in the segment containing the term
pub doc_freq: u32,
/// Offset within the postings (`.idx`) file.
/// Offset within the postings (`.idx`) file.
pub postings_offset: u32,
/// Offset within the position (`.pos`) file.
pub positions_offset: u32,
@@ -25,20 +25,17 @@ pub struct TermInfo {
impl BinarySerializable for TermInfo {
fn serialize(&self, writer: &mut io::Write) -> io::Result<usize> {
Ok(
try!(self.doc_freq.serialize(writer)) +
try!(self.postings_offset.serialize(writer)) +
try!(self.positions_offset.serialize(writer))
)
Ok(try!(self.doc_freq.serialize(writer)) + try!(self.postings_offset.serialize(writer)) +
try!(self.positions_offset.serialize(writer)))
}
fn deserialize(reader: &mut io::Read) -> io::Result<Self> {
let doc_freq = try!(u32::deserialize(reader));
let postings_offset = try!(u32::deserialize(reader));
let positions_offset = try!(u32::deserialize(reader));
Ok(TermInfo {
doc_freq: doc_freq,
postings_offset: postings_offset,
positions_offset: positions_offset,
})
doc_freq: doc_freq,
postings_offset: postings_offset,
positions_offset: positions_offset,
})
}
}

21
src/query/boolean_query/boolean_query.rs
View File

@@ -37,10 +37,11 @@ impl Query for BooleanQuery {
}
fn weight(&self, searcher: &Searcher) -> Result<Box<Weight>> {
let sub_weights = try!(self.subqueries
.iter()
.map(|&(ref _occur, ref subquery)| subquery.weight(searcher))
.collect());
let sub_weights =
try!(self.subqueries
.iter()
.map(|&(ref _occur, ref subquery)| subquery.weight(searcher))
.collect());
let occurs: Vec<Occur> = self.subqueries
.iter()
.map(|&(ref occur, ref _subquery)| *occur)
@@ -54,12 +55,14 @@ impl BooleanQuery {
/// Helper method to create a boolean query matching a given list of terms.
/// The resulting query is a disjunction of the terms.
pub fn new_multiterms_query(terms: Vec<Term>) -> BooleanQuery {
let occur_term_queries: Vec<(Occur, Box<Query>)> = terms.into_iter()
let occur_term_queries: Vec<(Occur, Box<Query>)> = terms
.into_iter()
.map(|term| {
let term_query: Box<Query> = box TermQuery::new(term, SegmentPostingsOption::Freq);
(Occur::Should, term_query)
})
let term_query: Box<Query> = box TermQuery::new(term,
SegmentPostingsOption::Freq);
(Occur::Should, term_query)
})
.collect();
BooleanQuery::from(occur_term_queries)
}
}
}

50
src/query/boolean_query/boolean_scorer.rs
View File

@@ -10,7 +10,7 @@ use query::boolean_query::ScoreCombiner;
/// Each `HeapItem` represents the head of
/// one of scorer being merged.
///
/// * `doc` - is the current doc id for the given segment postings
/// * `doc` - is the current doc id for the given segment postings
/// * `ord` - is the ordinal used to identify to which segment postings
/// this heap item belong to.
#[derive(Eq, PartialEq)]
@@ -27,8 +27,8 @@ impl PartialOrd for HeapItem {
}
impl Ord for HeapItem {
fn cmp(&self, other:&Self) -> Ordering {
(other.doc).cmp(&self.doc)
fn cmp(&self, other: &Self) -> Ordering {
(other.doc).cmp(&self.doc)
}
}
@@ -41,9 +41,7 @@ pub struct BooleanScorer<TScorer: Scorer> {
}
impl<TScorer: Scorer> BooleanScorer<TScorer> {
pub fn new(scorers: Vec<TScorer>,
occur_filter: OccurFilter) -> BooleanScorer<TScorer> {
pub fn new(scorers: Vec<TScorer>, occur_filter: OccurFilter) -> BooleanScorer<TScorer> {
let score_combiner = ScoreCombiner::default_for_num_scorers(scorers.len());
let mut non_empty_scorers: Vec<TScorer> = Vec::new();
for mut posting in scorers {
@@ -57,11 +55,11 @@ impl<TScorer: Scorer> BooleanScorer<TScorer> {
.map(|posting| posting.doc())
.enumerate()
.map(|(ord, doc)| {
HeapItem {
doc: doc,
ord: ord as u32
}
})
HeapItem {
doc: doc,
ord: ord as u32,
}
})
.collect();
BooleanScorer {
scorers: non_empty_scorers,
@@ -69,20 +67,19 @@ impl<TScorer: Scorer> BooleanScorer<TScorer> {
doc: 0u32,
score_combiner: score_combiner,
occur_filter: occur_filter,
}
}
/// Advances the head of our heap (the segment posting with the lowest doc)
/// It will also update the new current `DocId` as well as the term frequency
/// associated with the segment postings.
///
///
/// After advancing the `SegmentPosting`, the postings is removed from the heap
/// if it has been entirely consumed, or pushed back into the heap.
///
///
/// # Panics
/// This method will panic if the head `SegmentPostings` is not empty.
fn advance_head(&mut self,) {
fn advance_head(&mut self) {
{
let mut mutable_head = self.queue.peek_mut().unwrap();
let cur_scorers = &mut self.scorers[mutable_head.ord as usize];
@@ -96,7 +93,7 @@ impl<TScorer: Scorer> BooleanScorer<TScorer> {
}
impl<TScorer: Scorer> DocSet for BooleanScorer<TScorer> {
fn advance(&mut self,) -> bool {
fn advance(&mut self) -> bool {
loop {
self.score_combiner.clear();
let mut ord_bitset = 0u64;
@@ -106,40 +103,37 @@ impl<TScorer: Scorer> DocSet for BooleanScorer<TScorer> {
self.doc = heap_item.doc;
let score = self.scorers[ord].score();
self.score_combiner.update(score);
ord_bitset |= 1 << ord;
ord_bitset |= 1 << ord;
}
None => {
return false;
}
}
self.advance_head();
while let Some(&HeapItem {doc, ord}) = self.queue.peek() {
while let Some(&HeapItem { doc, ord }) = self.queue.peek() {
if doc == self.doc {
let ord = ord as usize;
let score = self.scorers[ord].score();
self.score_combiner.update(score);
ord_bitset |= 1 << ord;
}
else {
} else {
break;
}
self.advance_head();
}
}
if self.occur_filter.accept(ord_bitset) {
return true;
}
}
}
fn doc(&self,) -> DocId {
}
fn doc(&self) -> DocId {
self.doc
}
}
impl<TScorer: Scorer> Scorer for BooleanScorer<TScorer> {
fn score(&self,) -> f32 {
fn score(&self) -> f32 {
self.score_combiner.score()
}
}

6
src/query/boolean_query/boolean_weight.rs
View File

@@ -23,9 +23,9 @@ impl BooleanWeight {
impl Weight for BooleanWeight {
fn scorer<'a>(&'a self, reader: &'a SegmentReader) -> Result<Box<Scorer + 'a>> {
let sub_scorers: Vec<Box<Scorer + 'a>> = try!(self.weights
.iter()
.map(|weight| weight.scorer(reader))
.collect());
.iter()
.map(|weight| weight.scorer(reader))
.collect());
let boolean_scorer = BooleanScorer::new(sub_scorers, self.occur_filter);
Ok(box boolean_scorer)
}

68
src/query/boolean_query/mod.rs
View File

@@ -11,7 +11,7 @@ pub use self::score_combiner::ScoreCombiner;
#[cfg(test)]
mod tests {
use super::*;
use postings::{DocSet, VecPostings};
use query::Scorer;
@@ -23,12 +23,12 @@ mod tests {
use collector::tests::TestCollector;
use Index;
use schema::*;
use fastfield::{U64FastFieldReader};
use fastfield::U64FastFieldReader;
use postings::SegmentPostingsOption;
fn abs_diff(left: f32, right: f32) -> f32 {
(right - left).abs()
}
}
#[test]
@@ -64,7 +64,8 @@ mod tests {
}
let make_term_query = |text: &str| {
let term_query = TermQuery::new(Term::from_field_text(text_field, text), SegmentPostingsOption::NoFreq);
let term_query = TermQuery::new(Term::from_field_text(text_field, text),
SegmentPostingsOption::NoFreq);
let query: Box<Query> = box term_query;
query
};
@@ -78,68 +79,71 @@ mod tests {
test_collector.docs()
};
{
let boolean_query = BooleanQuery::from(vec![(Occur::Must, make_term_query("a")) ]);
assert_eq!(matching_docs(&boolean_query), vec!(0, 1, 3));
let boolean_query = BooleanQuery::from(vec![(Occur::Must, make_term_query("a"))]);
assert_eq!(matching_docs(&boolean_query), vec![0, 1, 3]);
}
{
let boolean_query = BooleanQuery::from(vec![(Occur::Should, make_term_query("a")) ]);
assert_eq!(matching_docs(&boolean_query), vec!(0, 1, 3));
let boolean_query = BooleanQuery::from(vec![(Occur::Should, make_term_query("a"))]);
assert_eq!(matching_docs(&boolean_query), vec![0, 1, 3]);
}
{
let boolean_query = BooleanQuery::from(vec![(Occur::Should, make_term_query("a")), (Occur::Should, make_term_query("b"))]);
assert_eq!(matching_docs(&boolean_query), vec!(0, 1, 2, 3));
let boolean_query = BooleanQuery::from(vec![(Occur::Should, make_term_query("a")),
(Occur::Should, make_term_query("b"))]);
assert_eq!(matching_docs(&boolean_query), vec![0, 1, 2, 3]);
}
{
let boolean_query = BooleanQuery::from(vec![(Occur::Must, make_term_query("a")), (Occur::Should, make_term_query("b"))]);
assert_eq!(matching_docs(&boolean_query), vec!(0, 1, 3));
let boolean_query = BooleanQuery::from(vec![(Occur::Must, make_term_query("a")),
(Occur::Should, make_term_query("b"))]);
assert_eq!(matching_docs(&boolean_query), vec![0, 1, 3]);
}
{
let boolean_query = BooleanQuery::from(vec![(Occur::Must, make_term_query("a")),
let boolean_query = BooleanQuery::from(vec![(Occur::Must, make_term_query("a")),
(Occur::Should, make_term_query("b")),
(Occur::MustNot, make_term_query("d")),
]);
assert_eq!(matching_docs(&boolean_query), vec!(0, 1));
(Occur::MustNot, make_term_query("d"))]);
assert_eq!(matching_docs(&boolean_query), vec![0, 1]);
}
{
let boolean_query = BooleanQuery::from(vec![(Occur::MustNot, make_term_query("d")),]);
let boolean_query = BooleanQuery::from(vec![(Occur::MustNot, make_term_query("d"))]);
assert_eq!(matching_docs(&boolean_query), Vec::<u32>::new());
}
}
#[test]
pub fn test_boolean_scorer() {
let occurs = vec!(Occur::Should, Occur::Should);
let occurs = vec![Occur::Should, Occur::Should];
let occur_filter = OccurFilter::new(&occurs);
let left_fieldnorms = U64FastFieldReader::from(vec!(100,200,300));
let left = VecPostings::from(vec!(1, 2, 3));
let left_fieldnorms =
U64FastFieldReader::from((0u64..9u64).map(|doc| doc * 3).collect::<Vec<u64>>());
let left = VecPostings::from(vec![1, 2, 3]);
let left_scorer = TermScorer {
idf: 1f32,
fieldnorm_reader_opt: Some(left_fieldnorms),
postings: left,
};
let right_fieldnorms = U64FastFieldReader::from(vec!(15,25,35));
let right = VecPostings::from(vec!(1, 3, 8));
let right_fieldnorms =
U64FastFieldReader::from((0u64..9u64).map(|doc| doc * 5).collect::<Vec<u64>>());
let right = VecPostings::from(vec![1, 3, 8]);
let right_scorer = TermScorer {
idf: 4f32,
fieldnorm_reader_opt: Some(right_fieldnorms),
postings: right,
};
let mut boolean_scorer = BooleanScorer::new(vec!(left_scorer, right_scorer), occur_filter);
let mut boolean_scorer = BooleanScorer::new(vec![left_scorer, right_scorer], occur_filter);
assert_eq!(boolean_scorer.next(), Some(1u32));
assert!(abs_diff(boolean_scorer.score(), 0.8707107) < 0.001);
assert!(abs_diff(boolean_scorer.score(), 2.3662047) < 0.001);
assert_eq!(boolean_scorer.next(), Some(2u32));
assert!(abs_diff(boolean_scorer.score(), 0.028867513) < 0.001f32);
assert!(abs_diff(boolean_scorer.score(), 0.20412415) < 0.001f32);
assert_eq!(boolean_scorer.next(), Some(3u32));
assert_eq!(boolean_scorer.next(), Some(8u32));
assert!(abs_diff(boolean_scorer.score(), 0.5163978) < 0.001f32);
assert!(abs_diff(boolean_scorer.score(), 0.31622776) < 0.001f32);
assert!(!boolean_scorer.advance());
}
}

15
src/query/boolean_query/score_combiner.rs
View File

@@ -7,26 +7,25 @@ pub struct ScoreCombiner {
}
impl ScoreCombiner {
pub fn update(&mut self, score: Score) {
self.score += score;
self.num_fields += 1;
}
pub fn clear(&mut self,) {
pub fn clear(&mut self) {
self.score = 0f32;
self.num_fields = 0;
}
/// Compute the coord term
fn coord(&self,) -> f32 {
fn coord(&self) -> f32 {
self.coords[self.num_fields]
}
pub fn score(&self, ) -> Score {
pub fn score(&self) -> Score {
self.score * self.coord()
}
pub fn default_for_num_scorers(num_scorers: usize) -> ScoreCombiner {
let query_coords: Vec<Score> = (0..num_scorers + 1)
.map(|i| (i as Score) / (num_scorers as Score))
@@ -43,4 +42,4 @@ impl From<Vec<Score>> for ScoreCombiner {
score: 0f32,
}
}
}
}

2
src/query/mod.rs
View File

@@ -1,5 +1,5 @@
/// Query module
///
///
/// The query module regroups all of tantivy's query objects
///

19
src/query/occur_filter.rs
View File

@@ -1,25 +1,24 @@
use query::Occur;
/// An OccurFilter represents a filter over a bitset of
/// An `OccurFilter` represents a filter over a bitset of
// at most 64 elements.
///
/// It wraps some simple bitmask to compute the filter
/// rapidly.
/// rapidly.
#[derive(Clone, Copy)]
pub struct OccurFilter {
and_mask: u64,
result: u64,
result: u64,
}
impl OccurFilter {
/// Returns true if the bitset is matching the occur list.
pub fn accept(&self, ord_set: u64) -> bool {
(self.and_mask & ord_set) == self.result
}
/// Builds an `OccurFilter` from a list of `Occur`.
/// Builds an `OccurFilter` from a list of `Occur`.
pub fn new(occurs: &[Occur]) -> OccurFilter {
let mut and_mask = 0u64;
let mut result = 0u64;
@@ -29,16 +28,16 @@ impl OccurFilter {
Occur::Must => {
and_mask |= shift;
result |= shift;
},
}
Occur::MustNot => {
and_mask |= shift;
},
Occur::Should => {},
}
Occur::Should => {}
}
}
OccurFilter {
and_mask: and_mask,
result: result
result: result,
}
}
}

35
src/query/phrase_query/mod.rs
View File

@@ -9,7 +9,7 @@ pub use self::phrase_scorer::PhraseScorer;
#[cfg(test)]
mod tests {
use super::*;
use core::Index;
use schema::FieldValue;
@@ -18,30 +18,35 @@ mod tests {
#[test]
pub fn test_phrase_query() {
let mut schema_builder = SchemaBuilder::default();
let text_field = schema_builder.add_text_field("text", TEXT);
let schema = schema_builder.build();
let index = Index::create_in_ram(schema);
{
let mut index_writer = index.writer_with_num_threads(1, 40_000_000).unwrap();
{ // 0
{
// 0
let doc = doc!(text_field=>"b b b d c g c");
index_writer.add_document(doc);
}
{ // 1
{
// 1
let doc = doc!(text_field=>"a b b d c g c");
index_writer.add_document(doc);
}
{ // 2
{
// 2
let doc = doc!(text_field=>"a b a b c");
index_writer.add_document(doc);
}
{ // 3
{
// 3
let doc = doc!(text_field=>"c a b a d ga a");
index_writer.add_document(doc);
}
{ // 4
{
// 4
let doc = doc!(text_field=>"a b c");
index_writer.add_document(doc);
}
@@ -57,17 +62,19 @@ mod tests {
.map(|text| Term::from_field_text(text_field, text))
.collect();
let phrase_query = PhraseQuery::from(terms);
searcher.search(&phrase_query, &mut test_collector).expect("search should succeed");
searcher
.search(&phrase_query, &mut test_collector)
.expect("search should succeed");
test_collector.docs()
};
let empty_vec = Vec::<u32>::new();
assert_eq!(test_query(vec!("a", "b", "c")), vec!(2, 4));
assert_eq!(test_query(vec!("a", "b")), vec!(1, 2, 3, 4));
assert_eq!(test_query(vec!("b", "b")), vec!(0, 1));
assert_eq!(test_query(vec!("g", "ewrwer")), empty_vec);
assert_eq!(test_query(vec!("g", "a")), empty_vec);
assert_eq!(test_query(vec!["a", "b", "c"]), vec![2, 4]);
assert_eq!(test_query(vec!["a", "b"]), vec![1, 2, 3, 4]);
assert_eq!(test_query(vec!["b", "b"]), vec![0, 1]);
assert_eq!(test_query(vec!["g", "ewrwer"]), empty_vec);
assert_eq!(test_query(vec!["g", "a"]), empty_vec);
}
}

15
src/query/phrase_query/phrase_query.rs
View File

@@ -8,12 +8,12 @@ use Result;
/// `PhraseQuery` matches a specific sequence of word.
/// For instance the phrase query for `"part time"` will match
/// For instance the phrase query for `"part time"` will match
/// the sentence
///
///
/// **Alan just got a part time job.**
///
/// On the other hand it will not match the sentence.
/// On the other hand it will not match the sentence.
///
/// **This is my favorite part of the job.**
///
@@ -22,12 +22,10 @@ use Result;
///
#[derive(Debug)]
pub struct PhraseQuery {
phrase_terms: Vec<Term>,
phrase_terms: Vec<Term>,
}
impl Query for PhraseQuery {
/// Used to make it possible to cast Box<Query>
/// into a specific type. This is mostly useful for unit tests.
fn as_any(&self) -> &Any {
@@ -40,15 +38,12 @@ impl Query for PhraseQuery {
fn weight(&self, _searcher: &Searcher) -> Result<Box<Weight>> {
Ok(box PhraseWeight::from(self.phrase_terms.clone()))
}
}
impl From<Vec<Term>> for PhraseQuery {
fn from(phrase_terms: Vec<Term>) -> PhraseQuery {
assert!(phrase_terms.len() > 1);
PhraseQuery {
phrase_terms: phrase_terms,
}
PhraseQuery { phrase_terms: phrase_terms }
}
}

28
src/query/phrase_query/phrase_scorer.rs
View File

@@ -15,23 +15,21 @@ impl<'a> PhraseScorer<'a> {
let mut positions_arr: Vec<&[u32]> = self.intersection_docset
.docsets()
.iter()
.map(|posting| {
posting.positions()
})
.map(|posting| posting.positions())
.collect();
let num_postings = positions_arr.len() as u32;
let mut ord = 1u32;
let mut pos_candidate = positions_arr[0][0];
let mut pos_candidate = positions_arr[0][0];
positions_arr[0] = &(positions_arr[0])[1..];
let mut count_matching = 1;
#[cfg_attr(feature = "cargo-clippy", allow(never_loop))]
'outer: loop {
let target = pos_candidate + ord;
let positions = positions_arr[ord as usize];
for i in 0..positions.len() {
let pos_i = positions[i];
for (i, pos_i) in positions.iter().cloned().enumerate() {
if pos_i < target {
continue;
}
@@ -40,11 +38,10 @@ impl<'a> PhraseScorer<'a> {
if count_matching == num_postings {
return true;
}
}
else if pos_i > target {
} else if pos_i > target {
count_matching = 1;
pos_candidate = positions[i] - ord;
positions_arr[ord as usize] = &(positions_arr[ord as usize])[(i+1)..];
positions_arr[ord as usize] = &(positions_arr[ord as usize])[(i + 1)..];
}
ord += 1;
if ord == num_postings {
@@ -58,7 +55,7 @@ impl<'a> PhraseScorer<'a> {
}
impl<'a> DocSet for PhraseScorer<'a> {
fn advance(&mut self,) -> bool {
fn advance(&mut self) -> bool {
while self.intersection_docset.advance() {
if self.phrase_match() {
return true;
@@ -67,15 +64,14 @@ impl<'a> DocSet for PhraseScorer<'a> {
false
}
fn doc(&self,) -> DocId {
fn doc(&self) -> DocId {
self.intersection_docset.doc()
}
}
impl<'a> Scorer for PhraseScorer<'a> {
fn score(&self,) -> f32 {
fn score(&self) -> f32 {
1f32
}
}

16
src/query/phrase_query/phrase_weight.rs
View File

@@ -14,26 +14,22 @@ pub struct PhraseWeight {
impl From<Vec<Term>> for PhraseWeight {
fn from(phrase_terms: Vec<Term>) -> PhraseWeight {
PhraseWeight {
phrase_terms: phrase_terms
}
PhraseWeight { phrase_terms: phrase_terms }
}
}
impl Weight for PhraseWeight {
fn scorer<'a>(&'a self, reader: &'a SegmentReader) -> Result<Box<Scorer + 'a>> {
let mut term_postings_list = Vec::new();
for term in &self.phrase_terms {
let term_postings_option = reader.read_postings(term, SegmentPostingsOption::FreqAndPositions);
for term in &self.phrase_terms {
let term_postings_option =
reader.read_postings(term, SegmentPostingsOption::FreqAndPositions);
if let Some(term_postings) = term_postings_option {
term_postings_list.push(term_postings);
}
else {
} else {
return Ok(box EmptyScorer);
}
}
Ok(box PhraseScorer {
intersection_docset: IntersectionDocSet::from(term_postings_list),
})
Ok(box PhraseScorer { intersection_docset: IntersectionDocSet::from(term_postings_list) })
}
}

41
src/query/query.rs
View File

@@ -9,36 +9,36 @@ use std::any::Any;
/// Query trait are in charge of defining :
///
///
/// - a set of documents
/// - a way to score these documents
///
/// When performing a [search](#method.search), these documents will then
/// be pushed to a [Collector](../collector/trait.Collector.html),
/// which will in turn be in charge of deciding what to do with them.
///
/// which will in turn be in charge of deciding what to do with them.
///
/// Concretely, this scored docset is represented by the
/// [`Scorer`](./trait.Scorer.html) trait.
///
/// Because our index is actually split into segments, the
/// Because our index is actually split into segments, the
/// query does not actually directly creates `DocSet` object.
/// Instead, the query creates a [`Weight`](./trait.Weight.html)
/// object for a given searcher.
///
/// The weight object, in turn, makes it possible to create
/// object for a given searcher.
///
/// The weight object, in turn, makes it possible to create
/// a scorer for a specific [`SegmentReader`](../struct.SegmentReader.html).
///
///
/// So to sum it up :
/// - a `Query` is recipe to define a set of documents as well the way to score them.
/// - a `Weight` is this recipe tied to a specific `Searcher`. It may for instance
/// - a `Weight` is this recipe tied to a specific `Searcher`. It may for instance
/// hold statistics about the different term of the query. It is created by the query.
/// - a `Scorer` is a cursor over the set of matching documents, for a specific
/// [`SegmentReader`](../struct.SegmentReader.html). It is created by the [`Weight`](./trait.Weight.html).
/// - a `Scorer` is a cursor over the set of matching documents, for a specific
/// [`SegmentReader`](../struct.SegmentReader.html). It is created by the
/// [`Weight`](./trait.Weight.html).
///
/// When implementing a new type of `Query`, it is normal to implement a
/// dedicated `Query`, `Weight` and `Scorer`.
pub trait Query: fmt::Debug {
/// Used to make it possible to cast Box<Query>
/// into a specific type. This is mostly useful for unit tests.
fn as_any(&self) -> &Any;
@@ -47,31 +47,28 @@ pub trait Query: fmt::Debug {
///
/// See [Weight](./trait.Weight.html).
fn weight(&self, searcher: &Searcher) -> Result<Box<Weight>>;
/// Search works as follows :
///
/// First the weight object associated to the query is created.
///
///
/// Then, the query loops over the segments and for each segment :
/// - setup the collector and informs it that the segment being processed has changed.
/// - creates a `Scorer` object associated for this segment
/// - iterate throw the matched documents and push them to the collector.
///
fn search(
&self,
searcher: &Searcher,
collector: &mut Collector) -> Result<TimerTree> {
let mut timer_tree = TimerTree::default();
fn search(&self, searcher: &Searcher, collector: &mut Collector) -> Result<TimerTree> {
let mut timer_tree = TimerTree::default();
let weight = try!(self.weight(searcher));
{
let mut search_timer = timer_tree.open("search");
for (segment_ord, segment_reader) in searcher.segment_readers().iter().enumerate() {
let mut segment_search_timer = search_timer.open("segment_search");
{
let _ = segment_search_timer.open("set_segment");
try!(collector.set_segment(segment_ord as SegmentLocalId, &segment_reader));
try!(collector.set_segment(segment_ord as SegmentLocalId, segment_reader));
}
let mut scorer = try!(weight.scorer(segment_reader));
{

19
src/query/query_parser/logical_ast.rs
View File

@@ -9,9 +9,9 @@ pub enum LogicalLiteral {
}
#[derive(Clone)]
pub enum LogicalAST{
pub enum LogicalAST {
Clause(Vec<(Occur, LogicalAST)>),
Leaf(Box<LogicalLiteral>)
Leaf(Box<LogicalLiteral>),
}
fn occur_letter(occur: Occur) -> &'static str {
@@ -28,8 +28,7 @@ impl fmt::Debug for LogicalAST {
LogicalAST::Clause(ref clause) => {
if clause.is_empty() {
try!(write!(formatter, "<emptyclause>"));
}
else {
} else {
let (ref occur, ref subquery) = clause[0];
try!(write!(formatter, "({}{:?}", occur_letter(*occur), subquery));
for &(ref occur, ref subquery) in &clause[1..] {
@@ -39,9 +38,7 @@ impl fmt::Debug for LogicalAST {
}
Ok(())
}
LogicalAST::Leaf(ref literal) => {
write!(formatter, "{:?}", literal)
}
LogicalAST::Leaf(ref literal) => write!(formatter, "{:?}", literal),
}
}
}
@@ -55,12 +52,8 @@ impl From<LogicalLiteral> for LogicalAST {
impl fmt::Debug for LogicalLiteral {
fn fmt(&self, formatter: &mut fmt::Formatter) -> Result<(), fmt::Error> {
match *self {
LogicalLiteral::Term(ref term) => {
write!(formatter, "{:?}", term)
},
LogicalLiteral::Phrase(ref terms) => {
write!(formatter, "\"{:?}\"", terms)
}
LogicalLiteral::Term(ref term) => write!(formatter, "{:?}", term),
LogicalLiteral::Phrase(ref terms) => write!(formatter, "\"{:?}\"", terms),
}
}
}

2
src/query/query_parser/mod.rs
View File

@@ -4,4 +4,4 @@ mod user_input_ast;
pub mod logical_ast;
pub use self::query_parser::QueryParser;
pub use self::query_parser::QueryParserError;
pub use self::query_parser::QueryParserError;

46
src/query/query_parser/query_grammar.rs
View File

@@ -11,34 +11,30 @@ fn literal<I>(input: I) -> ParseResult<UserInputAST, I>
phrase.or(word)
};
let negative_numbers =
(char('-'), many1(satisfy(|c: char| c.is_numeric())))
let negative_numbers = (char('-'), many1(satisfy(|c: char| c.is_numeric())))
.map(|(s1, s2): (char, String)| format!("{}{}", s1, s2));
let field =
(
letter(),
many(satisfy(|c: char| c.is_alphanumeric() || c == '_'))
)
let field = (letter(), many(satisfy(|c: char| c.is_alphanumeric() || c == '_')))
.map(|(s1, s2): (char, String)| format!("{}{}", s1, s2));
let term_val_with_field = negative_numbers.or(term_val());
let term_query = (field, char(':'), term_val_with_field).map(|(field_name, _, phrase)| {
UserInputLiteral {
field_name: Some(field_name),
phrase: phrase,
}
});
UserInputLiteral {
field_name:
Some(field_name),
phrase: phrase,
}
});
let term_default_field = term_val().map(|phrase| {
UserInputLiteral {
field_name: None,
phrase: phrase,
}
});
UserInputLiteral {
field_name: None,
phrase: phrase,
}
});
try(term_query)
.or(term_default_field)
.map(|query_literal| UserInputAST::from(query_literal))
.map(UserInputAST::from)
.parse_stream(input)
}
@@ -58,13 +54,11 @@ pub fn parse_to_ast<I>(input: I) -> ParseResult<UserInputAST, I>
where I: Stream<Item = char>
{
sep_by(parser(leaf), spaces())
.map(|subqueries: Vec<UserInputAST>| {
if subqueries.len() == 1 {
subqueries.into_iter().next().unwrap()
} else {
UserInputAST::Clause(subqueries.into_iter().map(Box::new).collect())
}
})
.map(|subqueries: Vec<UserInputAST>| if subqueries.len() == 1 {
subqueries.into_iter().next().unwrap()
} else {
UserInputAST::Clause(subqueries.into_iter().map(Box::new).collect())
})
.parse_stream(input)
}

176
src/query/query_parser/query_parser.rs
View File

@@ -92,10 +92,10 @@ impl QueryParser {
analyzer: box SimpleTokenizer,
}
}
/// Set the default way to compose queries to a conjunction.
///
/// By default a ,
/// By default a ,
pub fn set_conjunction_by_default(&mut self) {
self.conjunction_by_default = true;
}
@@ -114,11 +114,11 @@ impl QueryParser {
let logical_ast = self.parse_query_to_logical_ast(query)?;
Ok(convert_to_query(logical_ast))
}
/// Parse the user query into an AST.
fn parse_query_to_logical_ast(&self, query: &str) -> Result<LogicalAST, QueryParserError> {
let (user_input_ast, _remaining) =
parse_to_ast(query).map_err(|_| QueryParserError::SyntaxError)?;
let (user_input_ast, _remaining) = parse_to_ast(query)
.map_err(|_| QueryParserError::SyntaxError)?;
self.compute_logical_ast(user_input_ast)
}
@@ -127,10 +127,10 @@ impl QueryParser {
.get_field(field_name)
.ok_or_else(|| QueryParserError::FieldDoesNotExist(String::from(field_name)))
}
fn compute_logical_ast(&self,
user_input_ast: UserInputAST)
-> Result<LogicalAST, QueryParserError> {
user_input_ast: UserInputAST)
-> Result<LogicalAST, QueryParserError> {
let (occur, ast) = self.compute_logical_ast_with_occur(user_input_ast)?;
if occur == Occur::MustNot {
return Err(QueryParserError::AllButQueryForbidden);
@@ -149,44 +149,38 @@ impl QueryParser {
let field_name = field_entry.name().to_string();
return Err(QueryParserError::FieldNotIndexed(field_name));
}
match field_type {
&FieldType::I64(_) => {
match *field_type {
FieldType::I64(_) => {
let val: i64 = i64::from_str(phrase)?;
let term = Term::from_field_i64(field, val);
return Ok(Some(LogicalLiteral::Term(term)));
Ok(Some(LogicalLiteral::Term(term)))
}
&FieldType::U64(_) => {
FieldType::U64(_) => {
let val: u64 = u64::from_str(phrase)?;
let term = Term::from_field_u64(field, val);
return Ok(Some(LogicalLiteral::Term(term)));
Ok(Some(LogicalLiteral::Term(term)))
}
&FieldType::Str(ref str_options) => {
FieldType::Str(ref str_options) => {
let mut terms: Vec<Term> = Vec::new();
if str_options.get_indexing_options().is_tokenized() {
let mut token_iter = self.analyzer.tokenize(phrase);
loop {
if let Some(token) = token_iter.next() {
let term = Term::from_field_text(field, token);
terms.push(term);
} else {
break;
}
while let Some(token) = token_iter.next() {
let term = Term::from_field_text(field, token);
terms.push(term);
}
}
else {
} else {
terms.push(Term::from_field_text(field, phrase));
}
if terms.is_empty() {
return Ok(None);
}
else if terms.len() == 1 {
return Ok(Some(LogicalLiteral::Term(terms.into_iter().next().unwrap())))
Ok(None)
} else if terms.len() == 1 {
Ok(Some(LogicalLiteral::Term(terms.into_iter().next().unwrap())))
} else {
return Ok(Some(LogicalLiteral::Phrase(terms)))
Ok(Some(LogicalLiteral::Phrase(terms)))
}
}
}
}
fn default_occur(&self) -> Occur {
@@ -198,8 +192,8 @@ impl QueryParser {
}
fn compute_logical_ast_with_occur(&self,
user_input_ast: UserInputAST)
-> Result<(Occur, LogicalAST), QueryParserError> {
user_input_ast: UserInputAST)
-> Result<(Occur, LogicalAST), QueryParserError> {
match user_input_ast {
UserInputAST::Clause(sub_queries) => {
let default_occur = self.default_occur();
@@ -224,11 +218,11 @@ impl QueryParser {
UserInputAST::Leaf(literal) => {
let term_phrases: Vec<(Field, String)> = match literal.field_name {
Some(ref field_name) => {
let field = try!(self.resolve_field_name(&field_name));
let field = try!(self.resolve_field_name(field_name));
vec![(field, literal.phrase.clone())]
}
None => {
if self.default_fields.len() == 0 {
if self.default_fields.is_empty() {
return Err(QueryParserError::NoDefaultFieldDeclared);
} else {
self.default_fields
@@ -244,17 +238,14 @@ impl QueryParser {
asts.push(LogicalAST::Leaf(box ast));
}
}
let result_ast =
if asts.len() == 0 {
// this should never happen
return Err(QueryParserError::SyntaxError);
} else if asts.len() == 1 {
asts[0].clone()
} else {
LogicalAST::Clause(asts.into_iter()
.map(|ast| (Occur::Should, ast))
.collect())
};
let result_ast = if asts.is_empty() {
// this should never happen
return Err(QueryParserError::SyntaxError);
} else if asts.len() == 1 {
asts[0].clone()
} else {
LogicalAST::Clause(asts.into_iter().map(|ast| (Occur::Should, ast)).collect())
};
Ok((Occur::Should, result_ast))
}
}
@@ -278,7 +269,7 @@ fn compose_occur(left: Occur, right: Occur) -> Occur {
} else {
Occur::MustNot
}
}
}
}
}
@@ -292,7 +283,8 @@ fn convert_literal_to_query(logical_literal: LogicalLiteral) -> Box<Query> {
fn convert_to_query(logical_ast: LogicalAST) -> Box<Query> {
match logical_ast {
LogicalAST::Clause(clause) => {
let occur_subqueries = clause.into_iter()
let occur_subqueries = clause
.into_iter()
.map(|(occur, subquery)| (occur, convert_to_query(subquery)))
.collect::<Vec<_>>();
box BooleanQuery::from(occur_subqueries)
@@ -311,7 +303,7 @@ mod test {
use super::QueryParser;
use super::QueryParserError;
use super::super::logical_ast::*;
fn make_query_parser() -> QueryParser {
let mut schema_builder = SchemaBuilder::default();
let title = schema_builder.add_text_field("title", TEXT);
@@ -331,7 +323,7 @@ mod test {
fn parse_query_to_logical_ast(query: &str,
default_conjunction: bool)
-> Result<LogicalAST, QueryParserError> {
let mut query_parser = make_query_parser();
let mut query_parser = make_query_parser();
if default_conjunction {
query_parser.set_conjunction_by_default();
}
@@ -345,72 +337,73 @@ mod test {
let query_str = format!("{:?}", query);
assert_eq!(query_str, expected);
}
#[test]
pub fn test_parse_query_simple() {
let query_parser = make_query_parser();
assert!(query_parser.parse_query("toto").is_ok());
assert!(query_parser.parse_query("toto").is_ok());
}
#[test]
pub fn test_parse_nonindexed_field_yields_error() {
let query_parser = make_query_parser();
let is_not_indexed_err = |query: &str| {
let result: Result<Box<Query>, QueryParserError> = query_parser.parse_query(query);
if let Err(QueryParserError::FieldNotIndexed(field_name)) = result {
Some(field_name.clone())
}
else {
} else {
None
}
};
assert_eq!(
is_not_indexed_err("notindexed_text:titi"),
Some(String::from("notindexed_text"))
);
assert_eq!(
is_not_indexed_err("notindexed_u64:23424"),
Some(String::from("notindexed_u64"))
);
assert_eq!(
is_not_indexed_err("notindexed_i64:-234324"),
Some(String::from("notindexed_i64"))
);
assert_eq!(is_not_indexed_err("notindexed_text:titi"),
Some(String::from("notindexed_text")));
assert_eq!(is_not_indexed_err("notindexed_u64:23424"),
Some(String::from("notindexed_u64")));
assert_eq!(is_not_indexed_err("notindexed_i64:-234324"),
Some(String::from("notindexed_i64")));
}
#[test]
pub fn test_parse_query_untokenized() {
test_parse_query_to_logical_ast_helper("nottokenized:\"wordone wordtwo\"",
"Term([0, 0, 0, 7, 119, 111, 114, 100, 111, 110, 101, 32, 119, 111, 114, 100, 116, 119, 111])",
"Term([0, 0, 0, 7, 119, 111, 114, 100, 111, 110, \
101, 32, 119, 111, 114, 100, 116, 119, 111])",
false);
}
#[test]
pub fn test_parse_query_ints() {
let query_parser = make_query_parser();
assert!(query_parser.parse_query("signed:2324").is_ok());
assert!(query_parser.parse_query("signed:\"22\"").is_ok());
assert!(query_parser.parse_query("signed:2324").is_ok());
assert!(query_parser.parse_query("signed:\"22\"").is_ok());
assert!(query_parser.parse_query("signed:\"-2234\"").is_ok());
assert!(query_parser.parse_query("signed:\"-9999999999999\"").is_ok());
assert!(query_parser
.parse_query("signed:\"-9999999999999\"")
.is_ok());
assert!(query_parser.parse_query("signed:\"a\"").is_err());
assert!(query_parser.parse_query("signed:\"2a\"").is_err());
assert!(query_parser.parse_query("signed:\"18446744073709551615\"").is_err());
assert!(query_parser
.parse_query("signed:\"18446744073709551615\"")
.is_err());
assert!(query_parser.parse_query("unsigned:\"2\"").is_ok());
assert!(query_parser.parse_query("unsigned:\"-2\"").is_err());
assert!(query_parser.parse_query("unsigned:\"18446744073709551615\"").is_ok());
assert!(query_parser
.parse_query("unsigned:\"18446744073709551615\"")
.is_ok());
test_parse_query_to_logical_ast_helper("unsigned:2324",
"Term([0, 0, 0, 3, 0, 0, 0, 0, 0, 0, 9, 20])",
false);
test_parse_query_to_logical_ast_helper("signed:-2324",
&format!("{:?}", Term::from_field_i64(Field(2u32), -2324)),
&format!("{:?}",
Term::from_field_i64(Field(2u32), -2324)),
false);
}
#[test]
pub fn test_parse_query_to_ast_disjunction() {
@@ -421,36 +414,49 @@ mod test {
"Term([0, 0, 0, 0, 116, 111, 116, 111])",
false);
test_parse_query_to_logical_ast_helper("+title:toto -titi",
"(+Term([0, 0, 0, 0, 116, 111, 116, 111]) -(Term([0, 0, 0, 0, 116, \
105, 116, 105]) Term([0, 0, 0, 1, 116, 105, 116, 105])))",
"(+Term([0, 0, 0, 0, 116, 111, 116, 111]) \
-(Term([0, 0, 0, 0, 116, 105, 116, 105]) \
Term([0, 0, 0, 1, 116, 105, 116, 105])))",
false);
assert_eq!(parse_query_to_logical_ast("-title:toto", false).err().unwrap(),
assert_eq!(parse_query_to_logical_ast("-title:toto", false)
.err()
.unwrap(),
QueryParserError::AllButQueryForbidden);
test_parse_query_to_logical_ast_helper("title:a b",
"(Term([0, 0, 0, 0, 97]) (Term([0, 0, 0, 0, 98]) Term([0, 0, 0, 1, 98])))",
"(Term([0, 0, 0, 0, 97]) (Term([0, 0, 0, 0, 98]) \
Term([0, 0, 0, 1, 98])))",
false);
test_parse_query_to_logical_ast_helper("title:\"a b\"",
"\"[Term([0, 0, 0, 0, 97]), Term([0, 0, 0, 0, 98])]\"",
"\"[Term([0, 0, 0, 0, 97]), \
Term([0, 0, 0, 0, 98])]\"",
false);
}
#[test]
pub fn test_parse_query_to_ast_conjunction() {
test_parse_query_to_logical_ast_helper("title:toto", "Term([0, 0, 0, 0, 116, 111, 116, 111])", true);
test_parse_query_to_logical_ast_helper("title:toto",
"Term([0, 0, 0, 0, 116, 111, 116, 111])",
true);
test_parse_query_to_logical_ast_helper("+title:toto",
"Term([0, 0, 0, 0, 116, 111, 116, 111])",
true);
test_parse_query_to_logical_ast_helper("+title:toto -titi",
"(+Term([0, 0, 0, 0, 116, 111, 116, 111]) -(Term([0, 0, 0, 0, 116, \
105, 116, 105]) Term([0, 0, 0, 1, 116, 105, 116, 105])))",
"(+Term([0, 0, 0, 0, 116, 111, 116, 111]) \
-(Term([0, 0, 0, 0, 116, 105, 116, 105]) \
Term([0, 0, 0, 1, 116, 105, 116, 105])))",
true);
assert_eq!(parse_query_to_logical_ast("-title:toto", true).err().unwrap(),
assert_eq!(parse_query_to_logical_ast("-title:toto", true)
.err()
.unwrap(),
QueryParserError::AllButQueryForbidden);
test_parse_query_to_logical_ast_helper("title:a b",
"(+Term([0, 0, 0, 0, 97]) +(Term([0, 0, 0, 0, 98]) Term([0, 0, 0, 1, 98])))",
"(+Term([0, 0, 0, 0, 97]) \
+(Term([0, 0, 0, 0, 98]) \
Term([0, 0, 0, 1, 98])))",
true);
test_parse_query_to_logical_ast_helper("title:\"a b\"",
"\"[Term([0, 0, 0, 0, 97]), Term([0, 0, 0, 0, 98])]\"",
"\"[Term([0, 0, 0, 0, 97]), \
Term([0, 0, 0, 0, 98])]\"",
true);
}
}

32
src/query/query_parser/user_input_ast.rs
View File

@@ -2,18 +2,14 @@ use std::fmt;
pub struct UserInputLiteral {
pub field_name: Option<String>,
pub phrase: String,
pub phrase: String,
}
impl fmt::Debug for UserInputLiteral {
fn fmt(&self, formatter: &mut fmt::Formatter) -> Result<(), fmt::Error> {
match self.field_name {
Some(ref field_name) => {
write!(formatter, "{}:\"{}\"", field_name, self.phrase)
}
None => {
write!(formatter, "\"{}\"", self.phrase)
}
Some(ref field_name) => write!(formatter, "{}:\"{}\"", field_name, self.phrase),
None => write!(formatter, "\"{}\"", self.phrase),
}
}
}
@@ -22,41 +18,33 @@ pub enum UserInputAST {
Clause(Vec<Box<UserInputAST>>),
Not(Box<UserInputAST>),
Must(Box<UserInputAST>),
Leaf(Box<UserInputLiteral>)
Leaf(Box<UserInputLiteral>),
}
impl From<UserInputLiteral> for UserInputAST {
fn from(literal: UserInputLiteral) -> UserInputAST {
UserInputAST::Leaf(box literal)
}
}
}
impl fmt::Debug for UserInputAST {
fn fmt(&self, formatter: &mut fmt::Formatter) -> Result<(), fmt::Error> {
match *self {
UserInputAST::Must(ref subquery) => {
write!(formatter, "+({:?})", subquery)
},
UserInputAST::Must(ref subquery) => write!(formatter, "+({:?})", subquery),
UserInputAST::Clause(ref subqueries) => {
if subqueries.is_empty() {
try!(write!(formatter, "<emptyclause>"));
}
else {
} else {
try!(write!(formatter, "{:?}", &subqueries[0]));
for subquery in &subqueries[1..] {
try!(write!(formatter, " {:?}", subquery));
}
}
Ok(())
},
UserInputAST::Not(ref subquery) => {
write!(formatter, "-({:?})", subquery)
}
UserInputAST::Leaf(ref subquery) => {
write!(formatter, "{:?}", subquery)
}
UserInputAST::Not(ref subquery) => write!(formatter, "-({:?})", subquery),
UserInputAST::Leaf(ref subquery) => write!(formatter, "{:?}", subquery),
}
}
}

27
src/query/scorer.rs
View File

@@ -5,30 +5,29 @@ use collector::Collector;
use std::ops::{Deref, DerefMut};
/// Scored set of documents matching a query within a specific segment.
///
///
/// See [Query](./trait.Query.html).
pub trait Scorer: DocSet {
/// Returns the score.
///
///
/// This method will perform a bit of computation and is not cached.
fn score(&self,) -> Score;
fn score(&self) -> Score;
/// Consumes the complete `DocSet` and
/// push the scored documents to the collector.
/// push the scored documents to the collector.
fn collect(&mut self, collector: &mut Collector) {
while self.advance() {
collector.collect(self.doc(), self.score());
}
}
}
}
impl<'a> Scorer for Box<Scorer + 'a> {
fn score(&self,) -> Score {
fn score(&self) -> Score {
self.deref().score()
}
fn collect(&mut self, collector: &mut Collector) {
let scorer = self.deref_mut();
while scorer.advance() {
@@ -37,23 +36,23 @@ impl<'a> Scorer for Box<Scorer + 'a> {
}
}
/// EmptyScorer is a dummy Scorer in which no document matches.
///
/// `EmptyScorer` is a dummy `Scorer` in which no document matches.
///
/// It is useful for tests and handling edge cases.
pub struct EmptyScorer;
impl DocSet for EmptyScorer {
fn advance(&mut self,) -> bool {
fn advance(&mut self) -> bool {
false
}
fn doc(&self,) -> DocId {
fn doc(&self) -> DocId {
DocId::max_value()
}
}
impl Scorer for EmptyScorer {
fn score(&self,) -> Score {
fn score(&self) -> Score {
0f32
}
}

15
src/query/term_query/mod.rs
View File

@@ -9,7 +9,7 @@ pub use self::term_scorer::TermScorer;
#[cfg(test)]
mod tests {
use postings::{DocSet, VecPostings};
use query::Scorer;
use query::term_query::TermScorer;
@@ -23,7 +23,7 @@ mod tests {
fn abs_diff(left: f32, right: f32) -> f32 {
(right - left).abs()
}
}
#[test]
@@ -44,7 +44,8 @@ mod tests {
index.load_searchers().unwrap();
let searcher = index.searcher();
let term_query = TermQuery::new(Term::from_field_text(text_field, "a"), SegmentPostingsOption::NoFreq);
let term_query = TermQuery::new(Term::from_field_text(text_field, "a"),
SegmentPostingsOption::NoFreq);
let term_weight = term_query.weight(&searcher).unwrap();
let segment_reader = searcher.segment_reader(0);
let mut term_scorer = term_weight.scorer(segment_reader).unwrap();
@@ -53,13 +54,13 @@ mod tests {
assert_eq!(term_scorer.score(), 0.30685282);
}
#[test]
pub fn test_term_scorer() {
let left_fieldnorms = U64FastFieldReader::from(vec!(10, 4));
let left_fieldnorms = U64FastFieldReader::from(vec![10, 4]);
assert_eq!(left_fieldnorms.get(0), 10);
assert_eq!(left_fieldnorms.get(1), 4);
let left = VecPostings::from(vec!(1));
let left = VecPostings::from(vec![1]);
let mut left_scorer = TermScorer {
idf: 0.30685282,
fieldnorm_reader_opt: Some(left_fieldnorms),
@@ -69,4 +70,4 @@ mod tests {
assert!(abs_diff(left_scorer.score(), 0.15342641) < 0.001f32);
}
}
}

11
src/query/term_query/term_query.rs
View File

@@ -13,9 +13,9 @@ use std::any::Any;
/// The score associated is defined as
/// `idf` * sqrt(`term_freq` / `field norm`)
/// in which :
/// * idf - inverse document frequency.
/// * term_freq - number of occurrences of the term in the field
/// * field norm - number of tokens in the field.
/// * `idf` - inverse document frequency.
/// * `term_freq` - number of occurrences of the term in the field
/// * `field norm` - number of tokens in the field.
#[derive(Debug)]
pub struct TermQuery {
term: Term,
@@ -31,9 +31,9 @@ impl TermQuery {
}
}
/// Returns a weight object.
///
///
/// While `.weight(...)` returns a boxed trait object,
/// this method return a specific implementation.
/// This is useful for optimization purpose.
@@ -55,5 +55,4 @@ impl Query for TermQuery {
fn weight(&self, searcher: &Searcher) -> Result<Box<Weight>> {
Ok(box self.specialized_weight(searcher))
}
}

Some files were not shown because too many files have changed in this diff Show More