From f17e46b97d692dcac5a90621a06beb2bf956338f Mon Sep 17 00:00:00 2001 From: Travis CI User Date: Sun, 25 Mar 2018 04:24:40 +0000 Subject: [PATCH] Automatic Travis documentation build --- .../tantivy/query/boolean_query/mod.rs.html | 54 +++++++++ .../query/phrase_query/phrase_query.rs.html | 8 +- .../query/phrase_query/phrase_scorer.rs.html | 104 ++++++++++++++++-- .../query/phrase_query/phrase_weight.rs.html | 26 ++--- master/tantivy/query/struct.PhraseQuery.html | 4 +- master/tantivy/query/trait.Query.html | 2 +- 6 files changed, 167 insertions(+), 31 deletions(-) diff --git a/master/src/tantivy/query/boolean_query/mod.rs.html b/master/src/tantivy/query/boolean_query/mod.rs.html index 25daeb759..8232dc84f 100644 --- a/master/src/tantivy/query/boolean_query/mod.rs.html +++ b/master/src/tantivy/query/boolean_query/mod.rs.html @@ -228,6 +228,33 @@ 171 172 173 +174 +175 +176 +177 +178 +179 +180 +181 +182 +183 +184 +185 +186 +187 +188 +189 +190 +191 +192 +193 +194 +195 +196 +197 +198 +199 +200 
 mod boolean_query;
 mod boolean_weight;
@@ -401,6 +428,33 @@
             assert_eq!(matching_docs(&boolean_query), Vec::<u32>::new());
         }
     }
+
+
+    #[test]
+    pub fn test_intersection_score() {
+        let (index, text_field) = aux_test_helper();
+
+        let make_term_query = |text: &str| {
+            let term_query = TermQuery::new(
+                Term::from_field_text(text_field, text),
+                IndexRecordOption::Basic,
+            );
+            let query: Box<Query> = box term_query;
+            query
+        };
+
+        let score_docs = |boolean_query: &Query| {
+            let searcher = index.searcher();
+            let mut test_collector = TestCollector::default();
+            searcher.search(boolean_query, &mut test_collector).unwrap();
+            test_collector.scores()
+        };
+
+        {
+            let boolean_query = BooleanQuery::from(vec![(Occur::Must, make_term_query("a")), (Occur::Must, make_term_query("b"))]);
+            assert_eq!(score_docs(&boolean_query), vec![0.977973, 0.84699446]);
+        }
+    }
 }
diff --git a/master/src/tantivy/query/phrase_query/phrase_query.rs.html b/master/src/tantivy/query/phrase_query/phrase_query.rs.html index 0c9ee6b1a..0b7b92a23 100644 --- a/master/src/tantivy/query/phrase_query/phrase_query.rs.html +++ b/master/src/tantivy/query/phrase_query/phrase_query.rs.html @@ -115,6 +115,7 @@ 58 59 60 +61 
 use schema::Term;
 use query::Query;
@@ -166,12 +167,13 @@
         let terms = self.phrase_terms.clone();
         if scoring_enabled {
             let bm25_weight = BM25Weight::for_terms(searcher, &terms);
-            Ok(box PhraseWeight::with_scoring(
+            Ok(box PhraseWeight::new(
                 terms,
-                bm25_weight
+                bm25_weight,
+                true
             ))
         } else {
-            Ok(box PhraseWeight::no_scoring(terms))
+            Ok(box PhraseWeight::new(terms, BM25Weight::null(), false))
         }
 
     }
diff --git a/master/src/tantivy/query/phrase_query/phrase_scorer.rs.html b/master/src/tantivy/query/phrase_query/phrase_scorer.rs.html
index a83535914..2bf966d9f 100644
--- a/master/src/tantivy/query/phrase_query/phrase_scorer.rs.html
+++ b/master/src/tantivy/query/phrase_query/phrase_scorer.rs.html
@@ -302,6 +302,51 @@
 245
 246
 247
+248
+249
+250
+251
+252
+253
+254
+255
+256
+257
+258
+259
+260
+261
+262
+263
+264
+265
+266
+267
+268
+269
+270
+271
+272
+273
+274
+275
+276
+277
+278
+279
+280
+281
+282
+283
+284
+285
+286
+287
+288
+289
+290
+291
+292
 
 use DocId;
 use docset::{DocSet, SkipResult};
@@ -354,10 +399,28 @@
     phrase_count: u32,
     fieldnorm_reader: FieldNormReader,
     similarity_weight: BM25Weight,
+    score_needed: bool
 }
 
 
-/// Computes the length of the intersection of two sorted arrays.
+/// Returns true iff the two sorted array contain a common element
+fn intersection_exists(left: &[u32], right: &[u32]) -> bool {
+    let mut left_i = 0;
+    let mut right_i = 0;
+    while left_i < left.len() && right_i < right.len() {
+        let left_val = left[left_i];
+        let right_val = right[right_i];
+        if left_val < right_val {
+            left_i += 1;
+        } else if right_val < left_val {
+            right_i += 1;
+        } else {
+            return true;
+        }
+    }
+    false
+}
+
 fn intersection_count(left: &[u32], right: &[u32]) -> usize {
     let mut left_i = 0;
     let mut right_i = 0;
@@ -410,7 +473,8 @@
 
     pub fn new(term_postings: Vec<TPostings>,
                similarity_weight: BM25Weight,
-               fieldnorm_reader: FieldNormReader) -> PhraseScorer<TPostings> {
+               fieldnorm_reader: FieldNormReader,
+               score_needed: bool) -> PhraseScorer<TPostings> {
         let num_docsets = term_postings.len();
         let postings_with_offsets = term_postings
             .into_iter()
@@ -425,13 +489,40 @@
             phrase_count: 0u32,
             similarity_weight,
             fieldnorm_reader,
+            score_needed,
         }
     }
 
     fn phrase_match(&mut self) -> bool {
-        let count = self.phrase_count();
-        self.phrase_count = count;
-        count > 0u32
+        if self.score_needed {
+            let count = self.phrase_count();
+            self.phrase_count = count;
+            count > 0u32
+        } else {
+            self.phrase_exists()
+        }
+    }
+
+
+    fn phrase_exists(&mut self) -> bool {
+        {
+            self.intersection_docset
+                .docset_mut_specialized(0)
+                .positions(&mut self.left);
+        }
+        let mut intersection_len = self.left.len();
+        for i in 1..self.num_docsets - 1 {
+            {
+                self.intersection_docset.docset_mut_specialized(i).positions(&mut self.right);
+            }
+            intersection_len = intersection(&mut self.left[..intersection_len], &self.right[..]);
+            if intersection_len == 0 {
+                return false;
+            }
+        }
+
+        self.intersection_docset.docset_mut_specialized(self.num_docsets - 1).positions(&mut self.right);
+        intersection_exists(&self.left[..intersection_len], &self.right[..])
     }
 
     fn phrase_count(&mut self) -> u32 {
@@ -452,8 +543,7 @@
         }
 
         self.intersection_docset.docset_mut_specialized(self.num_docsets - 1).positions(&mut self.right);
-        intersection_len = intersection_count(&mut self.left[..intersection_len], &self.right[..]);
-        intersection_len as u32
+        intersection_count(&self.left[..intersection_len], &self.right[..]) as u32
     }
 }
 
diff --git a/master/src/tantivy/query/phrase_query/phrase_weight.rs.html b/master/src/tantivy/query/phrase_query/phrase_weight.rs.html
index 0dcf18ebd..a8f97fe08 100644
--- a/master/src/tantivy/query/phrase_query/phrase_weight.rs.html
+++ b/master/src/tantivy/query/phrase_query/phrase_weight.rs.html
@@ -116,11 +116,6 @@
 59
 60
 61
-62
-63
-64
-65
-66
 
 use query::Weight;
 use query::Scorer;
@@ -135,23 +130,18 @@
 pub struct PhraseWeight {
     phrase_terms: Vec<Term>,
     similarity_weight: BM25Weight,
+    score_needed: bool,
 }
 
 impl PhraseWeight {
     /// Creates a new phrase weight.
-    pub fn with_scoring(phrase_terms: Vec<Term>,
-                        similarity_weight: BM25Weight) -> PhraseWeight {
+    pub fn new(phrase_terms: Vec<Term>,
+               similarity_weight: BM25Weight,
+               score_needed: bool) -> PhraseWeight {
         PhraseWeight {
             phrase_terms,
-            similarity_weight
-        }
-    }
-
-
-    pub fn no_scoring(phrase_terms: Vec<Term>) -> PhraseWeight {
-        PhraseWeight {
-            phrase_terms,
-            similarity_weight: BM25Weight::null()
+            similarity_weight,
+            score_needed
         }
     }
 }
@@ -172,7 +162,7 @@
                     return Ok(box EmptyScorer);
                 }
             }
-            Ok(box PhraseScorer::new(term_postings_list, similarity_weight, fieldnorm_reader))
+            Ok(box PhraseScorer::new(term_postings_list, similarity_weight, fieldnorm_reader, self.score_needed))
         } else {
             let mut term_postings_list = Vec::new();
             for term in &self.phrase_terms {
@@ -184,7 +174,7 @@
                     return Ok(box EmptyScorer);
                 }
             }
-            Ok(box PhraseScorer::new(term_postings_list, similarity_weight, fieldnorm_reader))
+            Ok(box PhraseScorer::new(term_postings_list, similarity_weight, fieldnorm_reader, self.score_needed))
         }
     }
 }
diff --git a/master/tantivy/query/struct.PhraseQuery.html b/master/tantivy/query/struct.PhraseQuery.html
index caf5b56f4..cf950eb36 100644
--- a/master/tantivy/query/struct.PhraseQuery.html
+++ b/master/tantivy/query/struct.PhraseQuery.html
@@ -86,8 +86,8 @@ must belong to the same field.

         
impl Debug for PhraseQuery
[src]

 
[src]

 
Formats the value using the given formatter. Read more

-
impl Query for PhraseQuery
[src]

-
[src]

+
impl Query for PhraseQuery
[src]

+
[src]

 
Create the weight associated to a query.

 
See Weight.

 
[src]

diff --git a/master/tantivy/query/trait.Query.html b/master/tantivy/query/trait.Query.html
index 20cb81bf3..02a1458e0 100644
--- a/master/tantivy/query/trait.Query.html
+++ b/master/tantivy/query/trait.Query.html
@@ -125,7 +125,7 @@ can increase performances.