lunr - npm Package Compare versions

Comparing version 2.1.0-alpha.1 to 2.1.0-alpha.2

lib/index.js

@@ -130,3 +130,3 @@ /*!
       matchingFields = Object.create(null),
-      queryVector = new lunr.Vector
+      queryVectors = Object.create(null)
 
@@ -174,9 +174,4 @@ fn.call(query, query)
         /*
-         * For each term calculate the score as the term relates to the
-         * query using the same calculation used to score documents during
-         * indexing. This score will be used to build a vector space
-         * representation  of the query.
-         *
-         * Also need to discover the terms index to insert into the query
-         * vector at the right position
+         * For each term get the posting and termIndex, this is required for
+         * building the query vector.
          */
@@ -187,15 +182,2 @@ var expandedTerm = expandedTerms[j],
 
-        /*
-         * Upserting the found query term, along with its term index
-         * into the vector representing the query. It is here that
-         * any boosts are applied to the score. They could have been
-         * applied when calculating the score above, but that expression
-         * is already quite busy.
-         *
-         * Using upsert because there could already be an entry in the vector
-         * for the term we are working with. In that case we just add the scores
-         * together.
-         */
-        queryVector.upsert(termIndex, 1 * clause.boost, function (a, b) { return a + b })
-
         for (var k = 0; k < clause.fields.length; k++) {
@@ -214,2 +196,21 @@ /*
 
+          /*
+           * To support field level boosts a query vector is created per
+           * field. This vector is populated using the termIndex found for
+           * the term and a unit value with the appropriate boost applied.
+           *
+           * If the query vector for this field does not exist yet it needs
+           * to be created.
+           */
+          if (!(field in queryVectors)) {
+            queryVectors[field] = new lunr.Vector
+          }
+
+          /*
+           * Using upsert because there could already be an entry in the vector
+           * for the term we are working with. In that case we just add the scores
+           * together.
+           */
+          queryVectors[field].upsert(termIndex, 1 * clause.boost, function (a, b) { return a + b })
+
           for (var l = 0; l < matchingDocumentRefs.length; l++) {
@@ -246,11 +247,8 @@ /*
     /*
-     * With all the matching documents found they now need
-     * to be sorted by their relevance to the query. This
-     * is done by retrieving the documents vector representation
-     * and then finding its similarity with the query vector
-     * that was constructed earlier.
+     * Currently we have document fields that match the query, but we
+     * need to return documents. The matchData and scores are combined
+     * from multiple fields belonging to the same document.
      *
-     * This score, along with the document ref and any metadata
-     * we collected into a lunr.MatchData instance are stored
-     * in the results array ready for returning to the caller
+     * Scores are calculated by field, using the query vectors created
+     * above, and combined into a final document score using addition.
      */
@@ -260,3 +258,3 @@ var fieldRef = lunr.FieldRef.fromString(matchingFieldRefs[i]),
         fieldVector = this.fieldVectors[fieldRef],
-        score = queryVector.similarity(fieldVector)
+        score = queryVectors[fieldRef.fieldName].similarity(fieldVector)
 
@@ -275,2 +273,6 @@ if (docRef in results) {
 
+  /*
+   * The results object needs to be converted into a list
+   * of results, sorted by score before being returned.
+   */
   return Object.keys(results)
@@ -277,0 +279,0 @@ .map(function (key) {

lib/query_lexer.js

@@ -7,2 +7,3 @@ lunr.QueryLexer = function (str) {
   this.start = 0
+  this.escapeCharPositions = []
 }
@@ -18,6 +19,23 @@
 
+lunr.QueryLexer.prototype.sliceString = function () {
+  var subSlices = [],
+      sliceStart = this.start,
+      sliceEnd = this.pos
+
+  for (var i = 0; i < this.escapeCharPositions.length; i++) {
+    sliceEnd = this.escapeCharPositions[i]
+    subSlices.push(this.str.slice(sliceStart, sliceEnd))
+    sliceStart = sliceEnd + 1
+  }
+
+  subSlices.push(this.str.slice(sliceStart, this.pos))
+  this.escapeCharPositions.length = 0
+
+  return subSlices.join('')
+}
+
 lunr.QueryLexer.prototype.emit = function (type) {
   this.lexemes.push({
     type: type,
-    str: this.str.slice(this.start, this.pos),
+    str: this.sliceString(),
     start: this.start,
@@ -30,4 +48,9 @@ end: this.pos
 
+lunr.QueryLexer.prototype.escapeCharacter = function () {
+  this.escapeCharPositions.push(this.pos - 1)
+  this.pos += 1
+}
+
 lunr.QueryLexer.prototype.next = function () {
-  if (this.pos == this.length) {
+  if (this.pos >= this.length) {
     return lunr.QueryLexer.EOS
@@ -141,2 +164,8 @@ }
 
+    // Escape character is '\'
+    if (char.charCodeAt(0) == 92) {
+      lexer.escapeCharacter()
+      continue
+    }
+
     if (char == ":") {
@@ -143,0 +172,0 @@ return lunr.QueryLexer.lexField

package.json

 {
   "name": "lunr",
   "description": "Simple full-text search in your browser.",
-  "version": "2.1.0-alpha.1",
+  "version": "2.1.0-alpha.2",
   "author": "Oliver Nightingale",
@@ -6,0 +6,0 @@ "keywords": ["search"],

test/query_lexer_test.js

@@ -42,2 +42,34 @@ suite('lunr.QueryLexer', function () {
 
+    suite('term escape char', function () {
+      setup(function () {
+        this.lexer = lex("foo\\:bar")
+      })
+
+      test('produces 1 lexeme', function () {
+        assert.lengthOf(this.lexer.lexemes, 1)
+      })
+
+      suite('lexeme', function () {
+        setup(function () {
+          this.lexeme = this.lexer.lexemes[0]
+        })
+
+        test('#type', function () {
+          assert.equal(lunr.QueryLexer.TERM, this.lexeme.type)
+        })
+
+        test('#str', function () {
+          assert.equal('foo:bar', this.lexeme.str)
+        })
+
+        test('#start', function () {
+          assert.equal(0, this.lexeme.start)
+        })
+
+        test('#end', function () {
+          assert.equal(8, this.lexeme.end)
+        })
+      })
+    })
+
     suite('multiple terms', function () {
@@ -164,2 +196,40 @@ setup(function () {
 
+    suite('term with field with escape char', function () {
+      setup(function () {
+        this.lexer = lex("ti\\:tle:foo")
+      })
+
+      test('produces 1 lexeme', function () {
+        assert.lengthOf(this.lexer.lexemes, 2)
+      })
+
+      suite('lexeme', function () {
+        setup(function () {
+          this.fieldLexeme = this.lexer.lexemes[0]
+          this.termLexeme = this.lexer.lexemes[1]
+        })
+
+        test('#type', function () {
+          assert.equal(lunr.QueryLexer.FIELD, this.fieldLexeme.type)
+          assert.equal(lunr.QueryLexer.TERM, this.termLexeme.type)
+        })
+
+        test('#str', function () {
+          assert.equal('ti:tle', this.fieldLexeme.str)
+          assert.equal('foo', this.termLexeme.str)
+        })
+
+        test('#start', function () {
+          assert.equal(0, this.fieldLexeme.start)
+          assert.equal(8, this.termLexeme.start)
+        })
+
+        test('#end', function () {
+          assert.equal(7, this.fieldLexeme.end)
+          assert.equal(11, this.termLexeme.end)
+        })
+      })
+    })
+
+
     suite('term with edit distance', function () {
@@ -166,0 +236,0 @@ setup(function () {

New alerts

License Policy Violation

License

This package is not allowed per your license policy. Review the package's license to ensure compliance.

Found 1 instance in 1 package

Fixed alerts

License Policy Violation

License

This package is not allowed per your license policy. Review the package's license to ensure compliance.

Found 1 instance in 1 package

Improved metrics

Total package byte prevSize

589360

increased by0.93%

Number of package files

78

increased by2.63%

Lines of code

18507

increased by0.63%

No dependency changes