| 1 |
efrain |
1 |
/**
|
|
|
2 |
* moodle readme
|
|
|
3 |
*
|
|
|
4 |
* Lunrjs can be downloaded from https://github.com/olivernn/lunr.js. To update this library get the lunr.js file
|
|
|
5 |
* from this project and replace the content below with the new content.
|
|
|
6 |
*/
|
|
|
7 |
|
|
|
8 |
/**
|
|
|
9 |
* lunr - http://lunrjs.com - A bit like Solr, but much smaller and not as bright - 2.3.9
|
|
|
10 |
* Copyright (C) 2020 Oliver Nightingale
|
|
|
11 |
* @license MIT
|
|
|
12 |
*/
|
|
|
13 |
|
|
|
14 |
;(function(){
|
|
|
15 |
|
|
|
16 |
/**
|
|
|
17 |
* A convenience function for configuring and constructing
|
|
|
18 |
* a new lunr Index.
|
|
|
19 |
*
|
|
|
20 |
* A lunr.Builder instance is created and the pipeline setup
|
|
|
21 |
* with a trimmer, stop word filter and stemmer.
|
|
|
22 |
*
|
|
|
23 |
* This builder object is yielded to the configuration function
|
|
|
24 |
* that is passed as a parameter, allowing the list of fields
|
|
|
25 |
* and other builder parameters to be customised.
|
|
|
26 |
*
|
|
|
27 |
* All documents _must_ be added within the passed config function.
|
|
|
28 |
*
|
|
|
29 |
* @example
|
|
|
30 |
* var idx = lunr(function () {
|
|
|
31 |
* this.field('title')
|
|
|
32 |
* this.field('body')
|
|
|
33 |
* this.ref('id')
|
|
|
34 |
*
|
|
|
35 |
* documents.forEach(function (doc) {
|
|
|
36 |
* this.add(doc)
|
|
|
37 |
* }, this)
|
|
|
38 |
* })
|
|
|
39 |
*
|
|
|
40 |
* @see {@link lunr.Builder}
|
|
|
41 |
* @see {@link lunr.Pipeline}
|
|
|
42 |
* @see {@link lunr.trimmer}
|
|
|
43 |
* @see {@link lunr.stopWordFilter}
|
|
|
44 |
* @see {@link lunr.stemmer}
|
|
|
45 |
* @namespace {function} lunr
|
|
|
46 |
*/
|
|
|
47 |
var lunr = function (config) {
|
|
|
48 |
var builder = new lunr.Builder
|
|
|
49 |
|
|
|
50 |
builder.pipeline.add(
|
|
|
51 |
lunr.trimmer,
|
|
|
52 |
lunr.stopWordFilter,
|
|
|
53 |
lunr.stemmer
|
|
|
54 |
)
|
|
|
55 |
|
|
|
56 |
builder.searchPipeline.add(
|
|
|
57 |
lunr.stemmer
|
|
|
58 |
)
|
|
|
59 |
|
|
|
60 |
config.call(builder, builder)
|
|
|
61 |
return builder.build()
|
|
|
62 |
}
|
|
|
63 |
|
|
|
64 |
lunr.version = "2.3.9"
|
|
|
65 |
/*!
|
|
|
66 |
* lunr.utils
|
|
|
67 |
* Copyright (C) 2020 Oliver Nightingale
|
|
|
68 |
*/
|
|
|
69 |
|
|
|
70 |
/**
|
|
|
71 |
* A namespace containing utils for the rest of the lunr library
|
|
|
72 |
* @namespace lunr.utils
|
|
|
73 |
*/
|
|
|
74 |
lunr.utils = {}
|
|
|
75 |
|
|
|
76 |
/**
|
|
|
77 |
* Print a warning message to the console.
|
|
|
78 |
*
|
|
|
79 |
* @param {String} message The message to be printed.
|
|
|
80 |
* @memberOf lunr.utils
|
|
|
81 |
* @function
|
|
|
82 |
*/
|
|
|
83 |
lunr.utils.warn = (function (global) {
|
|
|
84 |
/* eslint-disable no-console */
|
|
|
85 |
return function (message) {
|
|
|
86 |
if (global.console && console.warn) {
|
|
|
87 |
console.warn(message)
|
|
|
88 |
}
|
|
|
89 |
}
|
|
|
90 |
/* eslint-enable no-console */
|
|
|
91 |
})(this)
|
|
|
92 |
|
|
|
93 |
/**
|
|
|
94 |
* Convert an object to a string.
|
|
|
95 |
*
|
|
|
96 |
* In the case of `null` and `undefined` the function returns
|
|
|
97 |
* the empty string, in all other cases the result of calling
|
|
|
98 |
* `toString` on the passed object is returned.
|
|
|
99 |
*
|
|
|
100 |
* @param {Any} obj The object to convert to a string.
|
|
|
101 |
* @return {String} string representation of the passed object.
|
|
|
102 |
* @memberOf lunr.utils
|
|
|
103 |
*/
|
|
|
104 |
lunr.utils.asString = function (obj) {
|
|
|
105 |
if (obj === void 0 || obj === null) {
|
|
|
106 |
return ""
|
|
|
107 |
} else {
|
|
|
108 |
return obj.toString()
|
|
|
109 |
}
|
|
|
110 |
}
|
|
|
111 |
|
|
|
112 |
/**
|
|
|
113 |
* Clones an object.
|
|
|
114 |
*
|
|
|
115 |
* Will create a copy of an existing object such that any mutations
|
|
|
116 |
* on the copy cannot affect the original.
|
|
|
117 |
*
|
|
|
118 |
* Only shallow objects are supported, passing a nested object to this
|
|
|
119 |
* function will cause a TypeError.
|
|
|
120 |
*
|
|
|
121 |
* Objects with primitives, and arrays of primitives are supported.
|
|
|
122 |
*
|
|
|
123 |
* @param {Object} obj The object to clone.
|
|
|
124 |
* @return {Object} a clone of the passed object.
|
|
|
125 |
* @throws {TypeError} when a nested object is passed.
|
|
|
126 |
* @memberOf Utils
|
|
|
127 |
*/
|
|
|
128 |
lunr.utils.clone = function (obj) {
|
|
|
129 |
if (obj === null || obj === undefined) {
|
|
|
130 |
return obj
|
|
|
131 |
}
|
|
|
132 |
|
|
|
133 |
var clone = Object.create(null),
|
|
|
134 |
keys = Object.keys(obj)
|
|
|
135 |
|
|
|
136 |
for (var i = 0; i < keys.length; i++) {
|
|
|
137 |
var key = keys[i],
|
|
|
138 |
val = obj[key]
|
|
|
139 |
|
|
|
140 |
if (Array.isArray(val)) {
|
|
|
141 |
clone[key] = val.slice()
|
|
|
142 |
continue
|
|
|
143 |
}
|
|
|
144 |
|
|
|
145 |
if (typeof val === 'string' ||
|
|
|
146 |
typeof val === 'number' ||
|
|
|
147 |
typeof val === 'boolean') {
|
|
|
148 |
clone[key] = val
|
|
|
149 |
continue
|
|
|
150 |
}
|
|
|
151 |
|
|
|
152 |
throw new TypeError("clone is not deep and does not support nested objects")
|
|
|
153 |
}
|
|
|
154 |
|
|
|
155 |
return clone
|
|
|
156 |
}
|
|
|
157 |
lunr.FieldRef = function (docRef, fieldName, stringValue) {
|
|
|
158 |
this.docRef = docRef
|
|
|
159 |
this.fieldName = fieldName
|
|
|
160 |
this._stringValue = stringValue
|
|
|
161 |
}
|
|
|
162 |
|
|
|
163 |
lunr.FieldRef.joiner = "/"
|
|
|
164 |
|
|
|
165 |
lunr.FieldRef.fromString = function (s) {
|
|
|
166 |
var n = s.indexOf(lunr.FieldRef.joiner)
|
|
|
167 |
|
|
|
168 |
if (n === -1) {
|
|
|
169 |
throw "malformed field ref string"
|
|
|
170 |
}
|
|
|
171 |
|
|
|
172 |
var fieldRef = s.slice(0, n),
|
|
|
173 |
docRef = s.slice(n + 1)
|
|
|
174 |
|
|
|
175 |
return new lunr.FieldRef (docRef, fieldRef, s)
|
|
|
176 |
}
|
|
|
177 |
|
|
|
178 |
lunr.FieldRef.prototype.toString = function () {
|
|
|
179 |
if (this._stringValue == undefined) {
|
|
|
180 |
this._stringValue = this.fieldName + lunr.FieldRef.joiner + this.docRef
|
|
|
181 |
}
|
|
|
182 |
|
|
|
183 |
return this._stringValue
|
|
|
184 |
}
|
|
|
185 |
/*!
|
|
|
186 |
* lunr.Set
|
|
|
187 |
* Copyright (C) 2020 Oliver Nightingale
|
|
|
188 |
*/
|
|
|
189 |
|
|
|
190 |
/**
|
|
|
191 |
* A lunr set.
|
|
|
192 |
*
|
|
|
193 |
* @constructor
|
|
|
194 |
*/
|
|
|
195 |
lunr.Set = function (elements) {
|
|
|
196 |
this.elements = Object.create(null)
|
|
|
197 |
|
|
|
198 |
if (elements) {
|
|
|
199 |
this.length = elements.length
|
|
|
200 |
|
|
|
201 |
for (var i = 0; i < this.length; i++) {
|
|
|
202 |
this.elements[elements[i]] = true
|
|
|
203 |
}
|
|
|
204 |
} else {
|
|
|
205 |
this.length = 0
|
|
|
206 |
}
|
|
|
207 |
}
|
|
|
208 |
|
|
|
209 |
/**
|
|
|
210 |
* A complete set that contains all elements.
|
|
|
211 |
*
|
|
|
212 |
* @static
|
|
|
213 |
* @readonly
|
|
|
214 |
* @type {lunr.Set}
|
|
|
215 |
*/
|
|
|
216 |
lunr.Set.complete = {
|
|
|
217 |
intersect: function (other) {
|
|
|
218 |
return other
|
|
|
219 |
},
|
|
|
220 |
|
|
|
221 |
union: function () {
|
|
|
222 |
return this
|
|
|
223 |
},
|
|
|
224 |
|
|
|
225 |
contains: function () {
|
|
|
226 |
return true
|
|
|
227 |
}
|
|
|
228 |
}
|
|
|
229 |
|
|
|
230 |
/**
|
|
|
231 |
* An empty set that contains no elements.
|
|
|
232 |
*
|
|
|
233 |
* @static
|
|
|
234 |
* @readonly
|
|
|
235 |
* @type {lunr.Set}
|
|
|
236 |
*/
|
|
|
237 |
lunr.Set.empty = {
|
|
|
238 |
intersect: function () {
|
|
|
239 |
return this
|
|
|
240 |
},
|
|
|
241 |
|
|
|
242 |
union: function (other) {
|
|
|
243 |
return other
|
|
|
244 |
},
|
|
|
245 |
|
|
|
246 |
contains: function () {
|
|
|
247 |
return false
|
|
|
248 |
}
|
|
|
249 |
}
|
|
|
250 |
|
|
|
251 |
/**
|
|
|
252 |
* Returns true if this set contains the specified object.
|
|
|
253 |
*
|
|
|
254 |
* @param {object} object - Object whose presence in this set is to be tested.
|
|
|
255 |
* @returns {boolean} - True if this set contains the specified object.
|
|
|
256 |
*/
|
|
|
257 |
lunr.Set.prototype.contains = function (object) {
|
|
|
258 |
return !!this.elements[object]
|
|
|
259 |
}
|
|
|
260 |
|
|
|
261 |
/**
|
|
|
262 |
* Returns a new set containing only the elements that are present in both
|
|
|
263 |
* this set and the specified set.
|
|
|
264 |
*
|
|
|
265 |
* @param {lunr.Set} other - set to intersect with this set.
|
|
|
266 |
* @returns {lunr.Set} a new set that is the intersection of this and the specified set.
|
|
|
267 |
*/
|
|
|
268 |
|
|
|
269 |
lunr.Set.prototype.intersect = function (other) {
|
|
|
270 |
var a, b, elements, intersection = []
|
|
|
271 |
|
|
|
272 |
if (other === lunr.Set.complete) {
|
|
|
273 |
return this
|
|
|
274 |
}
|
|
|
275 |
|
|
|
276 |
if (other === lunr.Set.empty) {
|
|
|
277 |
return other
|
|
|
278 |
}
|
|
|
279 |
|
|
|
280 |
if (this.length < other.length) {
|
|
|
281 |
a = this
|
|
|
282 |
b = other
|
|
|
283 |
} else {
|
|
|
284 |
a = other
|
|
|
285 |
b = this
|
|
|
286 |
}
|
|
|
287 |
|
|
|
288 |
elements = Object.keys(a.elements)
|
|
|
289 |
|
|
|
290 |
for (var i = 0; i < elements.length; i++) {
|
|
|
291 |
var element = elements[i]
|
|
|
292 |
if (element in b.elements) {
|
|
|
293 |
intersection.push(element)
|
|
|
294 |
}
|
|
|
295 |
}
|
|
|
296 |
|
|
|
297 |
return new lunr.Set (intersection)
|
|
|
298 |
}
|
|
|
299 |
|
|
|
300 |
/**
|
|
|
301 |
* Returns a new set combining the elements of this and the specified set.
|
|
|
302 |
*
|
|
|
303 |
* @param {lunr.Set} other - set to union with this set.
|
|
|
304 |
* @return {lunr.Set} a new set that is the union of this and the specified set.
|
|
|
305 |
*/
|
|
|
306 |
|
|
|
307 |
lunr.Set.prototype.union = function (other) {
|
|
|
308 |
if (other === lunr.Set.complete) {
|
|
|
309 |
return lunr.Set.complete
|
|
|
310 |
}
|
|
|
311 |
|
|
|
312 |
if (other === lunr.Set.empty) {
|
|
|
313 |
return this
|
|
|
314 |
}
|
|
|
315 |
|
|
|
316 |
return new lunr.Set(Object.keys(this.elements).concat(Object.keys(other.elements)))
|
|
|
317 |
}
|
|
|
318 |
/**
|
|
|
319 |
* A function to calculate the inverse document frequency for
|
|
|
320 |
* a posting. This is shared between the builder and the index
|
|
|
321 |
*
|
|
|
322 |
* @private
|
|
|
323 |
* @param {object} posting - The posting for a given term
|
|
|
324 |
* @param {number} documentCount - The total number of documents.
|
|
|
325 |
*/
|
|
|
326 |
lunr.idf = function (posting, documentCount) {
|
|
|
327 |
var documentsWithTerm = 0
|
|
|
328 |
|
|
|
329 |
for (var fieldName in posting) {
|
|
|
330 |
if (fieldName == '_index') continue // Ignore the term index, its not a field
|
|
|
331 |
documentsWithTerm += Object.keys(posting[fieldName]).length
|
|
|
332 |
}
|
|
|
333 |
|
|
|
334 |
var x = (documentCount - documentsWithTerm + 0.5) / (documentsWithTerm + 0.5)
|
|
|
335 |
|
|
|
336 |
return Math.log(1 + Math.abs(x))
|
|
|
337 |
}
|
|
|
338 |
|
|
|
339 |
/**
|
|
|
340 |
* A token wraps a string representation of a token
|
|
|
341 |
* as it is passed through the text processing pipeline.
|
|
|
342 |
*
|
|
|
343 |
* @constructor
|
|
|
344 |
* @param {string} [str=''] - The string token being wrapped.
|
|
|
345 |
* @param {object} [metadata={}] - Metadata associated with this token.
|
|
|
346 |
*/
|
|
|
347 |
lunr.Token = function (str, metadata) {
|
|
|
348 |
this.str = str || ""
|
|
|
349 |
this.metadata = metadata || {}
|
|
|
350 |
}
|
|
|
351 |
|
|
|
352 |
/**
|
|
|
353 |
* Returns the token string that is being wrapped by this object.
|
|
|
354 |
*
|
|
|
355 |
* @returns {string}
|
|
|
356 |
*/
|
|
|
357 |
lunr.Token.prototype.toString = function () {
|
|
|
358 |
return this.str
|
|
|
359 |
}
|
|
|
360 |
|
|
|
361 |
/**
|
|
|
362 |
* A token update function is used when updating or optionally
|
|
|
363 |
* when cloning a token.
|
|
|
364 |
*
|
|
|
365 |
* @callback lunr.Token~updateFunction
|
|
|
366 |
* @param {string} str - The string representation of the token.
|
|
|
367 |
* @param {Object} metadata - All metadata associated with this token.
|
|
|
368 |
*/
|
|
|
369 |
|
|
|
370 |
/**
|
|
|
371 |
* Applies the given function to the wrapped string token.
|
|
|
372 |
*
|
|
|
373 |
* @example
|
|
|
374 |
* token.update(function (str, metadata) {
|
|
|
375 |
* return str.toUpperCase()
|
|
|
376 |
* })
|
|
|
377 |
*
|
|
|
378 |
* @param {lunr.Token~updateFunction} fn - A function to apply to the token string.
|
|
|
379 |
* @returns {lunr.Token}
|
|
|
380 |
*/
|
|
|
381 |
lunr.Token.prototype.update = function (fn) {
|
|
|
382 |
this.str = fn(this.str, this.metadata)
|
|
|
383 |
return this
|
|
|
384 |
}
|
|
|
385 |
|
|
|
386 |
/**
|
|
|
387 |
* Creates a clone of this token. Optionally a function can be
|
|
|
388 |
* applied to the cloned token.
|
|
|
389 |
*
|
|
|
390 |
* @param {lunr.Token~updateFunction} [fn] - An optional function to apply to the cloned token.
|
|
|
391 |
* @returns {lunr.Token}
|
|
|
392 |
*/
|
|
|
393 |
lunr.Token.prototype.clone = function (fn) {
|
|
|
394 |
fn = fn || function (s) { return s }
|
|
|
395 |
return new lunr.Token (fn(this.str, this.metadata), this.metadata)
|
|
|
396 |
}
|
|
|
397 |
/*!
|
|
|
398 |
* lunr.tokenizer
|
|
|
399 |
* Copyright (C) 2020 Oliver Nightingale
|
|
|
400 |
*/
|
|
|
401 |
|
|
|
402 |
/**
|
|
|
403 |
* A function for splitting a string into tokens ready to be inserted into
|
|
|
404 |
* the search index. Uses `lunr.tokenizer.separator` to split strings, change
|
|
|
405 |
* the value of this property to change how strings are split into tokens.
|
|
|
406 |
*
|
|
|
407 |
* This tokenizer will convert its parameter to a string by calling `toString` and
|
|
|
408 |
* then will split this string on the character in `lunr.tokenizer.separator`.
|
|
|
409 |
* Arrays will have their elements converted to strings and wrapped in a lunr.Token.
|
|
|
410 |
*
|
|
|
411 |
* Optional metadata can be passed to the tokenizer, this metadata will be cloned and
|
|
|
412 |
* added as metadata to every token that is created from the object to be tokenized.
|
|
|
413 |
*
|
|
|
414 |
* @static
|
|
|
415 |
* @param {?(string|object|object[])} obj - The object to convert into tokens
|
|
|
416 |
* @param {?object} metadata - Optional metadata to associate with every token
|
|
|
417 |
* @returns {lunr.Token[]}
|
|
|
418 |
* @see {@link lunr.Pipeline}
|
|
|
419 |
*/
|
|
|
420 |
lunr.tokenizer = function (obj, metadata) {
|
|
|
421 |
if (obj == null || obj == undefined) {
|
|
|
422 |
return []
|
|
|
423 |
}
|
|
|
424 |
|
|
|
425 |
if (Array.isArray(obj)) {
|
|
|
426 |
return obj.map(function (t) {
|
|
|
427 |
return new lunr.Token(
|
|
|
428 |
lunr.utils.asString(t).toLowerCase(),
|
|
|
429 |
lunr.utils.clone(metadata)
|
|
|
430 |
)
|
|
|
431 |
})
|
|
|
432 |
}
|
|
|
433 |
|
|
|
434 |
var str = obj.toString().toLowerCase(),
|
|
|
435 |
len = str.length,
|
|
|
436 |
tokens = []
|
|
|
437 |
|
|
|
438 |
for (var sliceEnd = 0, sliceStart = 0; sliceEnd <= len; sliceEnd++) {
|
|
|
439 |
var char = str.charAt(sliceEnd),
|
|
|
440 |
sliceLength = sliceEnd - sliceStart
|
|
|
441 |
|
|
|
442 |
if ((char.match(lunr.tokenizer.separator) || sliceEnd == len)) {
|
|
|
443 |
|
|
|
444 |
if (sliceLength > 0) {
|
|
|
445 |
var tokenMetadata = lunr.utils.clone(metadata) || {}
|
|
|
446 |
tokenMetadata["position"] = [sliceStart, sliceLength]
|
|
|
447 |
tokenMetadata["index"] = tokens.length
|
|
|
448 |
|
|
|
449 |
tokens.push(
|
|
|
450 |
new lunr.Token (
|
|
|
451 |
str.slice(sliceStart, sliceEnd),
|
|
|
452 |
tokenMetadata
|
|
|
453 |
)
|
|
|
454 |
)
|
|
|
455 |
}
|
|
|
456 |
|
|
|
457 |
sliceStart = sliceEnd + 1
|
|
|
458 |
}
|
|
|
459 |
|
|
|
460 |
}
|
|
|
461 |
|
|
|
462 |
return tokens
|
|
|
463 |
}
|
|
|
464 |
|
|
|
465 |
/**
|
|
|
466 |
* The separator used to split a string into tokens. Override this property to change the behaviour of
|
|
|
467 |
* `lunr.tokenizer` behaviour when tokenizing strings. By default this splits on whitespace and hyphens.
|
|
|
468 |
*
|
|
|
469 |
* @static
|
|
|
470 |
* @see lunr.tokenizer
|
|
|
471 |
*/
|
|
|
472 |
lunr.tokenizer.separator = /[\s\-]+/
|
|
|
473 |
/*!
|
|
|
474 |
* lunr.Pipeline
|
|
|
475 |
* Copyright (C) 2020 Oliver Nightingale
|
|
|
476 |
*/
|
|
|
477 |
|
|
|
478 |
/**
|
|
|
479 |
* lunr.Pipelines maintain an ordered list of functions to be applied to all
|
|
|
480 |
* tokens in documents entering the search index and queries being ran against
|
|
|
481 |
* the index.
|
|
|
482 |
*
|
|
|
483 |
* An instance of lunr.Index created with the lunr shortcut will contain a
|
|
|
484 |
* pipeline with a stop word filter and an English language stemmer. Extra
|
|
|
485 |
* functions can be added before or after either of these functions or these
|
|
|
486 |
* default functions can be removed.
|
|
|
487 |
*
|
|
|
488 |
* When run the pipeline will call each function in turn, passing a token, the
|
|
|
489 |
* index of that token in the original list of all tokens and finally a list of
|
|
|
490 |
* all the original tokens.
|
|
|
491 |
*
|
|
|
492 |
* The output of functions in the pipeline will be passed to the next function
|
|
|
493 |
* in the pipeline. To exclude a token from entering the index the function
|
|
|
494 |
* should return undefined, the rest of the pipeline will not be called with
|
|
|
495 |
* this token.
|
|
|
496 |
*
|
|
|
497 |
* For serialisation of pipelines to work, all functions used in an instance of
|
|
|
498 |
* a pipeline should be registered with lunr.Pipeline. Registered functions can
|
|
|
499 |
* then be loaded. If trying to load a serialised pipeline that uses functions
|
|
|
500 |
* that are not registered an error will be thrown.
|
|
|
501 |
*
|
|
|
502 |
* If not planning on serialising the pipeline then registering pipeline functions
|
|
|
503 |
* is not necessary.
|
|
|
504 |
*
|
|
|
505 |
* @constructor
|
|
|
506 |
*/
|
|
|
507 |
lunr.Pipeline = function () {
|
|
|
508 |
this._stack = []
|
|
|
509 |
}
|
|
|
510 |
|
|
|
511 |
lunr.Pipeline.registeredFunctions = Object.create(null)
|
|
|
512 |
|
|
|
513 |
/**
|
|
|
514 |
* A pipeline function maps lunr.Token to lunr.Token. A lunr.Token contains the token
|
|
|
515 |
* string as well as all known metadata. A pipeline function can mutate the token string
|
|
|
516 |
* or mutate (or add) metadata for a given token.
|
|
|
517 |
*
|
|
|
518 |
* A pipeline function can indicate that the passed token should be discarded by returning
|
|
|
519 |
* null, undefined or an empty string. This token will not be passed to any downstream pipeline
|
|
|
520 |
* functions and will not be added to the index.
|
|
|
521 |
*
|
|
|
522 |
* Multiple tokens can be returned by returning an array of tokens. Each token will be passed
|
|
|
523 |
* to any downstream pipeline functions and all will returned tokens will be added to the index.
|
|
|
524 |
*
|
|
|
525 |
* Any number of pipeline functions may be chained together using a lunr.Pipeline.
|
|
|
526 |
*
|
|
|
527 |
* @interface lunr.PipelineFunction
|
|
|
528 |
* @param {lunr.Token} token - A token from the document being processed.
|
|
|
529 |
* @param {number} i - The index of this token in the complete list of tokens for this document/field.
|
|
|
530 |
* @param {lunr.Token[]} tokens - All tokens for this document/field.
|
|
|
531 |
* @returns {(?lunr.Token|lunr.Token[])}
|
|
|
532 |
*/
|
|
|
533 |
|
|
|
534 |
/**
|
|
|
535 |
* Register a function with the pipeline.
|
|
|
536 |
*
|
|
|
537 |
* Functions that are used in the pipeline should be registered if the pipeline
|
|
|
538 |
* needs to be serialised, or a serialised pipeline needs to be loaded.
|
|
|
539 |
*
|
|
|
540 |
* Registering a function does not add it to a pipeline, functions must still be
|
|
|
541 |
* added to instances of the pipeline for them to be used when running a pipeline.
|
|
|
542 |
*
|
|
|
543 |
* @param {lunr.PipelineFunction} fn - The function to check for.
|
|
|
544 |
* @param {String} label - The label to register this function with
|
|
|
545 |
*/
|
|
|
546 |
lunr.Pipeline.registerFunction = function (fn, label) {
|
|
|
547 |
if (label in this.registeredFunctions) {
|
|
|
548 |
lunr.utils.warn('Overwriting existing registered function: ' + label)
|
|
|
549 |
}
|
|
|
550 |
|
|
|
551 |
fn.label = label
|
|
|
552 |
lunr.Pipeline.registeredFunctions[fn.label] = fn
|
|
|
553 |
}
|
|
|
554 |
|
|
|
555 |
/**
|
|
|
556 |
* Warns if the function is not registered as a Pipeline function.
|
|
|
557 |
*
|
|
|
558 |
* @param {lunr.PipelineFunction} fn - The function to check for.
|
|
|
559 |
* @private
|
|
|
560 |
*/
|
|
|
561 |
lunr.Pipeline.warnIfFunctionNotRegistered = function (fn) {
|
|
|
562 |
var isRegistered = fn.label && (fn.label in this.registeredFunctions)
|
|
|
563 |
|
|
|
564 |
if (!isRegistered) {
|
|
|
565 |
lunr.utils.warn('Function is not registered with pipeline. This may cause problems when serialising the index.\n', fn)
|
|
|
566 |
}
|
|
|
567 |
}
|
|
|
568 |
|
|
|
569 |
/**
|
|
|
570 |
* Loads a previously serialised pipeline.
|
|
|
571 |
*
|
|
|
572 |
* All functions to be loaded must already be registered with lunr.Pipeline.
|
|
|
573 |
* If any function from the serialised data has not been registered then an
|
|
|
574 |
* error will be thrown.
|
|
|
575 |
*
|
|
|
576 |
* @param {Object} serialised - The serialised pipeline to load.
|
|
|
577 |
* @returns {lunr.Pipeline}
|
|
|
578 |
*/
|
|
|
579 |
lunr.Pipeline.load = function (serialised) {
|
|
|
580 |
var pipeline = new lunr.Pipeline
|
|
|
581 |
|
|
|
582 |
serialised.forEach(function (fnName) {
|
|
|
583 |
var fn = lunr.Pipeline.registeredFunctions[fnName]
|
|
|
584 |
|
|
|
585 |
if (fn) {
|
|
|
586 |
pipeline.add(fn)
|
|
|
587 |
} else {
|
|
|
588 |
throw new Error('Cannot load unregistered function: ' + fnName)
|
|
|
589 |
}
|
|
|
590 |
})
|
|
|
591 |
|
|
|
592 |
return pipeline
|
|
|
593 |
}
|
|
|
594 |
|
|
|
595 |
/**
|
|
|
596 |
* Adds new functions to the end of the pipeline.
|
|
|
597 |
*
|
|
|
598 |
* Logs a warning if the function has not been registered.
|
|
|
599 |
*
|
|
|
600 |
* @param {lunr.PipelineFunction[]} functions - Any number of functions to add to the pipeline.
|
|
|
601 |
*/
|
|
|
602 |
lunr.Pipeline.prototype.add = function () {
|
|
|
603 |
var fns = Array.prototype.slice.call(arguments)
|
|
|
604 |
|
|
|
605 |
fns.forEach(function (fn) {
|
|
|
606 |
lunr.Pipeline.warnIfFunctionNotRegistered(fn)
|
|
|
607 |
this._stack.push(fn)
|
|
|
608 |
}, this)
|
|
|
609 |
}
|
|
|
610 |
|
|
|
611 |
/**
|
|
|
612 |
* Adds a single function after a function that already exists in the
|
|
|
613 |
* pipeline.
|
|
|
614 |
*
|
|
|
615 |
* Logs a warning if the function has not been registered.
|
|
|
616 |
*
|
|
|
617 |
* @param {lunr.PipelineFunction} existingFn - A function that already exists in the pipeline.
|
|
|
618 |
* @param {lunr.PipelineFunction} newFn - The new function to add to the pipeline.
|
|
|
619 |
*/
|
|
|
620 |
lunr.Pipeline.prototype.after = function (existingFn, newFn) {
|
|
|
621 |
lunr.Pipeline.warnIfFunctionNotRegistered(newFn)
|
|
|
622 |
|
|
|
623 |
var pos = this._stack.indexOf(existingFn)
|
|
|
624 |
if (pos == -1) {
|
|
|
625 |
throw new Error('Cannot find existingFn')
|
|
|
626 |
}
|
|
|
627 |
|
|
|
628 |
pos = pos + 1
|
|
|
629 |
this._stack.splice(pos, 0, newFn)
|
|
|
630 |
}
|
|
|
631 |
|
|
|
632 |
/**
|
|
|
633 |
* Adds a single function before a function that already exists in the
|
|
|
634 |
* pipeline.
|
|
|
635 |
*
|
|
|
636 |
* Logs a warning if the function has not been registered.
|
|
|
637 |
*
|
|
|
638 |
* @param {lunr.PipelineFunction} existingFn - A function that already exists in the pipeline.
|
|
|
639 |
* @param {lunr.PipelineFunction} newFn - The new function to add to the pipeline.
|
|
|
640 |
*/
|
|
|
641 |
lunr.Pipeline.prototype.before = function (existingFn, newFn) {
|
|
|
642 |
lunr.Pipeline.warnIfFunctionNotRegistered(newFn)
|
|
|
643 |
|
|
|
644 |
var pos = this._stack.indexOf(existingFn)
|
|
|
645 |
if (pos == -1) {
|
|
|
646 |
throw new Error('Cannot find existingFn')
|
|
|
647 |
}
|
|
|
648 |
|
|
|
649 |
this._stack.splice(pos, 0, newFn)
|
|
|
650 |
}
|
|
|
651 |
|
|
|
652 |
/**
|
|
|
653 |
* Removes a function from the pipeline.
|
|
|
654 |
*
|
|
|
655 |
* @param {lunr.PipelineFunction} fn The function to remove from the pipeline.
|
|
|
656 |
*/
|
|
|
657 |
lunr.Pipeline.prototype.remove = function (fn) {
|
|
|
658 |
var pos = this._stack.indexOf(fn)
|
|
|
659 |
if (pos == -1) {
|
|
|
660 |
return
|
|
|
661 |
}
|
|
|
662 |
|
|
|
663 |
this._stack.splice(pos, 1)
|
|
|
664 |
}
|
|
|
665 |
|
|
|
666 |
/**
|
|
|
667 |
* Runs the current list of functions that make up the pipeline against the
|
|
|
668 |
* passed tokens.
|
|
|
669 |
*
|
|
|
670 |
* @param {Array} tokens The tokens to run through the pipeline.
|
|
|
671 |
* @returns {Array}
|
|
|
672 |
*/
|
|
|
673 |
lunr.Pipeline.prototype.run = function (tokens) {
|
|
|
674 |
var stackLength = this._stack.length
|
|
|
675 |
|
|
|
676 |
for (var i = 0; i < stackLength; i++) {
|
|
|
677 |
var fn = this._stack[i]
|
|
|
678 |
var memo = []
|
|
|
679 |
|
|
|
680 |
for (var j = 0; j < tokens.length; j++) {
|
|
|
681 |
var result = fn(tokens[j], j, tokens)
|
|
|
682 |
|
|
|
683 |
if (result === null || result === void 0 || result === '') continue
|
|
|
684 |
|
|
|
685 |
if (Array.isArray(result)) {
|
|
|
686 |
for (var k = 0; k < result.length; k++) {
|
|
|
687 |
memo.push(result[k])
|
|
|
688 |
}
|
|
|
689 |
} else {
|
|
|
690 |
memo.push(result)
|
|
|
691 |
}
|
|
|
692 |
}
|
|
|
693 |
|
|
|
694 |
tokens = memo
|
|
|
695 |
}
|
|
|
696 |
|
|
|
697 |
return tokens
|
|
|
698 |
}
|
|
|
699 |
|
|
|
700 |
/**
|
|
|
701 |
* Convenience method for passing a string through a pipeline and getting
|
|
|
702 |
* strings out. This method takes care of wrapping the passed string in a
|
|
|
703 |
* token and mapping the resulting tokens back to strings.
|
|
|
704 |
*
|
|
|
705 |
* @param {string} str - The string to pass through the pipeline.
|
|
|
706 |
* @param {?object} metadata - Optional metadata to associate with the token
|
|
|
707 |
* passed to the pipeline.
|
|
|
708 |
* @returns {string[]}
|
|
|
709 |
*/
|
|
|
710 |
lunr.Pipeline.prototype.runString = function (str, metadata) {
|
|
|
711 |
var token = new lunr.Token (str, metadata)
|
|
|
712 |
|
|
|
713 |
return this.run([token]).map(function (t) {
|
|
|
714 |
return t.toString()
|
|
|
715 |
})
|
|
|
716 |
}
|
|
|
717 |
|
|
|
718 |
/**
|
|
|
719 |
* Resets the pipeline by removing any existing processors.
|
|
|
720 |
*
|
|
|
721 |
*/
|
|
|
722 |
lunr.Pipeline.prototype.reset = function () {
|
|
|
723 |
this._stack = []
|
|
|
724 |
}
|
|
|
725 |
|
|
|
726 |
/**
|
|
|
727 |
* Returns a representation of the pipeline ready for serialisation.
|
|
|
728 |
*
|
|
|
729 |
* Logs a warning if the function has not been registered.
|
|
|
730 |
*
|
|
|
731 |
* @returns {Array}
|
|
|
732 |
*/
|
|
|
733 |
lunr.Pipeline.prototype.toJSON = function () {
|
|
|
734 |
return this._stack.map(function (fn) {
|
|
|
735 |
lunr.Pipeline.warnIfFunctionNotRegistered(fn)
|
|
|
736 |
|
|
|
737 |
return fn.label
|
|
|
738 |
})
|
|
|
739 |
}
|
|
|
740 |
/*!
|
|
|
741 |
* lunr.Vector
|
|
|
742 |
* Copyright (C) 2020 Oliver Nightingale
|
|
|
743 |
*/
|
|
|
744 |
|
|
|
745 |
/**
|
|
|
746 |
* A vector is used to construct the vector space of documents and queries. These
|
|
|
747 |
* vectors support operations to determine the similarity between two documents or
|
|
|
748 |
* a document and a query.
|
|
|
749 |
*
|
|
|
750 |
* Normally no parameters are required for initializing a vector, but in the case of
|
|
|
751 |
* loading a previously dumped vector the raw elements can be provided to the constructor.
|
|
|
752 |
*
|
|
|
753 |
* For performance reasons vectors are implemented with a flat array, where an elements
|
|
|
754 |
* index is immediately followed by its value. E.g. [index, value, index, value]. This
|
|
|
755 |
* allows the underlying array to be as sparse as possible and still offer decent
|
|
|
756 |
* performance when being used for vector calculations.
|
|
|
757 |
*
|
|
|
758 |
* @constructor
|
|
|
759 |
* @param {Number[]} [elements] - The flat list of element index and element value pairs.
|
|
|
760 |
*/
|
|
|
761 |
lunr.Vector = function (elements) {
|
|
|
762 |
this._magnitude = 0
|
|
|
763 |
this.elements = elements || []
|
|
|
764 |
}
|
|
|
765 |
|
|
|
766 |
|
|
|
767 |
/**
|
|
|
768 |
* Calculates the position within the vector to insert a given index.
|
|
|
769 |
*
|
|
|
770 |
* This is used internally by insert and upsert. If there are duplicate indexes then
|
|
|
771 |
* the position is returned as if the value for that index were to be updated, but it
|
|
|
772 |
* is the callers responsibility to check whether there is a duplicate at that index
|
|
|
773 |
*
|
|
|
774 |
* @param {Number} insertIdx - The index at which the element should be inserted.
|
|
|
775 |
* @returns {Number}
|
|
|
776 |
*/
|
|
|
777 |
lunr.Vector.prototype.positionForIndex = function (index) {
|
|
|
778 |
// For an empty vector the tuple can be inserted at the beginning
|
|
|
779 |
if (this.elements.length == 0) {
|
|
|
780 |
return 0
|
|
|
781 |
}
|
|
|
782 |
|
|
|
783 |
var start = 0,
|
|
|
784 |
end = this.elements.length / 2,
|
|
|
785 |
sliceLength = end - start,
|
|
|
786 |
pivotPoint = Math.floor(sliceLength / 2),
|
|
|
787 |
pivotIndex = this.elements[pivotPoint * 2]
|
|
|
788 |
|
|
|
789 |
while (sliceLength > 1) {
|
|
|
790 |
if (pivotIndex < index) {
|
|
|
791 |
start = pivotPoint
|
|
|
792 |
}
|
|
|
793 |
|
|
|
794 |
if (pivotIndex > index) {
|
|
|
795 |
end = pivotPoint
|
|
|
796 |
}
|
|
|
797 |
|
|
|
798 |
if (pivotIndex == index) {
|
|
|
799 |
break
|
|
|
800 |
}
|
|
|
801 |
|
|
|
802 |
sliceLength = end - start
|
|
|
803 |
pivotPoint = start + Math.floor(sliceLength / 2)
|
|
|
804 |
pivotIndex = this.elements[pivotPoint * 2]
|
|
|
805 |
}
|
|
|
806 |
|
|
|
807 |
if (pivotIndex == index) {
|
|
|
808 |
return pivotPoint * 2
|
|
|
809 |
}
|
|
|
810 |
|
|
|
811 |
if (pivotIndex > index) {
|
|
|
812 |
return pivotPoint * 2
|
|
|
813 |
}
|
|
|
814 |
|
|
|
815 |
if (pivotIndex < index) {
|
|
|
816 |
return (pivotPoint + 1) * 2
|
|
|
817 |
}
|
|
|
818 |
}
|
|
|
819 |
|
|
|
820 |
/**
|
|
|
821 |
* Inserts an element at an index within the vector.
|
|
|
822 |
*
|
|
|
823 |
* Does not allow duplicates, will throw an error if there is already an entry
|
|
|
824 |
* for this index.
|
|
|
825 |
*
|
|
|
826 |
* @param {Number} insertIdx - The index at which the element should be inserted.
|
|
|
827 |
* @param {Number} val - The value to be inserted into the vector.
|
|
|
828 |
*/
|
|
|
829 |
lunr.Vector.prototype.insert = function (insertIdx, val) {
|
|
|
830 |
this.upsert(insertIdx, val, function () {
|
|
|
831 |
throw "duplicate index"
|
|
|
832 |
})
|
|
|
833 |
}
|
|
|
834 |
|
|
|
835 |
/**
|
|
|
836 |
* Inserts or updates an existing index within the vector.
|
|
|
837 |
*
|
|
|
838 |
* @param {Number} insertIdx - The index at which the element should be inserted.
|
|
|
839 |
* @param {Number} val - The value to be inserted into the vector.
|
|
|
840 |
* @param {function} fn - A function that is called for updates, the existing value and the
|
|
|
841 |
* requested value are passed as arguments
|
|
|
842 |
*/
|
|
|
843 |
lunr.Vector.prototype.upsert = function (insertIdx, val, fn) {
|
|
|
844 |
this._magnitude = 0
|
|
|
845 |
var position = this.positionForIndex(insertIdx)
|
|
|
846 |
|
|
|
847 |
if (this.elements[position] == insertIdx) {
|
|
|
848 |
this.elements[position + 1] = fn(this.elements[position + 1], val)
|
|
|
849 |
} else {
|
|
|
850 |
this.elements.splice(position, 0, insertIdx, val)
|
|
|
851 |
}
|
|
|
852 |
}
|
|
|
853 |
|
|
|
854 |
/**
|
|
|
855 |
* Calculates the magnitude of this vector.
|
|
|
856 |
*
|
|
|
857 |
* @returns {Number}
|
|
|
858 |
*/
|
|
|
859 |
lunr.Vector.prototype.magnitude = function () {
|
|
|
860 |
if (this._magnitude) return this._magnitude
|
|
|
861 |
|
|
|
862 |
var sumOfSquares = 0,
|
|
|
863 |
elementsLength = this.elements.length
|
|
|
864 |
|
|
|
865 |
for (var i = 1; i < elementsLength; i += 2) {
|
|
|
866 |
var val = this.elements[i]
|
|
|
867 |
sumOfSquares += val * val
|
|
|
868 |
}
|
|
|
869 |
|
|
|
870 |
return this._magnitude = Math.sqrt(sumOfSquares)
|
|
|
871 |
}
|
|
|
872 |
|
|
|
873 |
/**
|
|
|
874 |
* Calculates the dot product of this vector and another vector.
|
|
|
875 |
*
|
|
|
876 |
* @param {lunr.Vector} otherVector - The vector to compute the dot product with.
|
|
|
877 |
* @returns {Number}
|
|
|
878 |
*/
|
|
|
879 |
lunr.Vector.prototype.dot = function (otherVector) {
|
|
|
880 |
var dotProduct = 0,
|
|
|
881 |
a = this.elements, b = otherVector.elements,
|
|
|
882 |
aLen = a.length, bLen = b.length,
|
|
|
883 |
aVal = 0, bVal = 0,
|
|
|
884 |
i = 0, j = 0
|
|
|
885 |
|
|
|
886 |
while (i < aLen && j < bLen) {
|
|
|
887 |
aVal = a[i], bVal = b[j]
|
|
|
888 |
if (aVal < bVal) {
|
|
|
889 |
i += 2
|
|
|
890 |
} else if (aVal > bVal) {
|
|
|
891 |
j += 2
|
|
|
892 |
} else if (aVal == bVal) {
|
|
|
893 |
dotProduct += a[i + 1] * b[j + 1]
|
|
|
894 |
i += 2
|
|
|
895 |
j += 2
|
|
|
896 |
}
|
|
|
897 |
}
|
|
|
898 |
|
|
|
899 |
return dotProduct
|
|
|
900 |
}
|
|
|
901 |
|
|
|
902 |
/**
|
|
|
903 |
* Calculates the similarity between this vector and another vector.
|
|
|
904 |
*
|
|
|
905 |
* @param {lunr.Vector} otherVector - The other vector to calculate the
|
|
|
906 |
* similarity with.
|
|
|
907 |
* @returns {Number}
|
|
|
908 |
*/
|
|
|
909 |
lunr.Vector.prototype.similarity = function (otherVector) {
|
|
|
910 |
return this.dot(otherVector) / this.magnitude() || 0
|
|
|
911 |
}
|
|
|
912 |
|
|
|
913 |
/**
|
|
|
914 |
* Converts the vector to an array of the elements within the vector.
|
|
|
915 |
*
|
|
|
916 |
* @returns {Number[]}
|
|
|
917 |
*/
|
|
|
918 |
lunr.Vector.prototype.toArray = function () {
|
|
|
919 |
var output = new Array (this.elements.length / 2)
|
|
|
920 |
|
|
|
921 |
for (var i = 1, j = 0; i < this.elements.length; i += 2, j++) {
|
|
|
922 |
output[j] = this.elements[i]
|
|
|
923 |
}
|
|
|
924 |
|
|
|
925 |
return output
|
|
|
926 |
}
|
|
|
927 |
|
|
|
928 |
/**
|
|
|
929 |
* A JSON serializable representation of the vector.
|
|
|
930 |
*
|
|
|
931 |
* @returns {Number[]}
|
|
|
932 |
*/
|
|
|
933 |
lunr.Vector.prototype.toJSON = function () {
|
|
|
934 |
return this.elements
|
|
|
935 |
}
|
|
|
936 |
/* eslint-disable */
|
|
|
937 |
/*!
|
|
|
938 |
* lunr.stemmer
|
|
|
939 |
* Copyright (C) 2020 Oliver Nightingale
|
|
|
940 |
* Includes code from - http://tartarus.org/~martin/PorterStemmer/js.txt
|
|
|
941 |
*/
|
|
|
942 |
|
|
|
943 |
/**
|
|
|
944 |
* lunr.stemmer is an english language stemmer, this is a JavaScript
|
|
|
945 |
* implementation of the PorterStemmer taken from http://tartarus.org/~martin
|
|
|
946 |
*
|
|
|
947 |
* @static
|
|
|
948 |
* @implements {lunr.PipelineFunction}
|
|
|
949 |
* @param {lunr.Token} token - The string to stem
|
|
|
950 |
* @returns {lunr.Token}
|
|
|
951 |
* @see {@link lunr.Pipeline}
|
|
|
952 |
* @function
|
|
|
953 |
*/
|
|
|
954 |
lunr.stemmer = (function(){
|
|
|
955 |
var step2list = {
|
|
|
956 |
"ational" : "ate",
|
|
|
957 |
"tional" : "tion",
|
|
|
958 |
"enci" : "ence",
|
|
|
959 |
"anci" : "ance",
|
|
|
960 |
"izer" : "ize",
|
|
|
961 |
"bli" : "ble",
|
|
|
962 |
"alli" : "al",
|
|
|
963 |
"entli" : "ent",
|
|
|
964 |
"eli" : "e",
|
|
|
965 |
"ousli" : "ous",
|
|
|
966 |
"ization" : "ize",
|
|
|
967 |
"ation" : "ate",
|
|
|
968 |
"ator" : "ate",
|
|
|
969 |
"alism" : "al",
|
|
|
970 |
"iveness" : "ive",
|
|
|
971 |
"fulness" : "ful",
|
|
|
972 |
"ousness" : "ous",
|
|
|
973 |
"aliti" : "al",
|
|
|
974 |
"iviti" : "ive",
|
|
|
975 |
"biliti" : "ble",
|
|
|
976 |
"logi" : "log"
|
|
|
977 |
},
|
|
|
978 |
|
|
|
979 |
step3list = {
|
|
|
980 |
"icate" : "ic",
|
|
|
981 |
"ative" : "",
|
|
|
982 |
"alize" : "al",
|
|
|
983 |
"iciti" : "ic",
|
|
|
984 |
"ical" : "ic",
|
|
|
985 |
"ful" : "",
|
|
|
986 |
"ness" : ""
|
|
|
987 |
},
|
|
|
988 |
|
|
|
989 |
c = "[^aeiou]", // consonant
|
|
|
990 |
v = "[aeiouy]", // vowel
|
|
|
991 |
C = c + "[^aeiouy]*", // consonant sequence
|
|
|
992 |
V = v + "[aeiou]*", // vowel sequence
|
|
|
993 |
|
|
|
994 |
mgr0 = "^(" + C + ")?" + V + C, // [C]VC... is m>0
|
|
|
995 |
meq1 = "^(" + C + ")?" + V + C + "(" + V + ")?$", // [C]VC[V] is m=1
|
|
|
996 |
mgr1 = "^(" + C + ")?" + V + C + V + C, // [C]VCVC... is m>1
|
|
|
997 |
s_v = "^(" + C + ")?" + v; // vowel in stem
|
|
|
998 |
|
|
|
999 |
var re_mgr0 = new RegExp(mgr0);
|
|
|
1000 |
var re_mgr1 = new RegExp(mgr1);
|
|
|
1001 |
var re_meq1 = new RegExp(meq1);
|
|
|
1002 |
var re_s_v = new RegExp(s_v);
|
|
|
1003 |
|
|
|
1004 |
var re_1a = /^(.+?)(ss|i)es$/;
|
|
|
1005 |
var re2_1a = /^(.+?)([^s])s$/;
|
|
|
1006 |
var re_1b = /^(.+?)eed$/;
|
|
|
1007 |
var re2_1b = /^(.+?)(ed|ing)$/;
|
|
|
1008 |
var re_1b_2 = /.$/;
|
|
|
1009 |
var re2_1b_2 = /(at|bl|iz)$/;
|
|
|
1010 |
var re3_1b_2 = new RegExp("([^aeiouylsz])\\1$");
|
|
|
1011 |
var re4_1b_2 = new RegExp("^" + C + v + "[^aeiouwxy]$");
|
|
|
1012 |
|
|
|
1013 |
var re_1c = /^(.+?[^aeiou])y$/;
|
|
|
1014 |
var re_2 = /^(.+?)(ational|tional|enci|anci|izer|bli|alli|entli|eli|ousli|ization|ation|ator|alism|iveness|fulness|ousness|aliti|iviti|biliti|logi)$/;
|
|
|
1015 |
|
|
|
1016 |
var re_3 = /^(.+?)(icate|ative|alize|iciti|ical|ful|ness)$/;
|
|
|
1017 |
|
|
|
1018 |
var re_4 = /^(.+?)(al|ance|ence|er|ic|able|ible|ant|ement|ment|ent|ou|ism|ate|iti|ous|ive|ize)$/;
|
|
|
1019 |
var re2_4 = /^(.+?)(s|t)(ion)$/;
|
|
|
1020 |
|
|
|
1021 |
var re_5 = /^(.+?)e$/;
|
|
|
1022 |
var re_5_1 = /ll$/;
|
|
|
1023 |
var re3_5 = new RegExp("^" + C + v + "[^aeiouwxy]$");
|
|
|
1024 |
|
|
|
1025 |
var porterStemmer = function porterStemmer(w) {
|
|
|
1026 |
var stem,
|
|
|
1027 |
suffix,
|
|
|
1028 |
firstch,
|
|
|
1029 |
re,
|
|
|
1030 |
re2,
|
|
|
1031 |
re3,
|
|
|
1032 |
re4;
|
|
|
1033 |
|
|
|
1034 |
if (w.length < 3) { return w; }
|
|
|
1035 |
|
|
|
1036 |
firstch = w.substr(0,1);
|
|
|
1037 |
if (firstch == "y") {
|
|
|
1038 |
w = firstch.toUpperCase() + w.substr(1);
|
|
|
1039 |
}
|
|
|
1040 |
|
|
|
1041 |
// Step 1a
|
|
|
1042 |
re = re_1a
|
|
|
1043 |
re2 = re2_1a;
|
|
|
1044 |
|
|
|
1045 |
if (re.test(w)) { w = w.replace(re,"$1$2"); }
|
|
|
1046 |
else if (re2.test(w)) { w = w.replace(re2,"$1$2"); }
|
|
|
1047 |
|
|
|
1048 |
// Step 1b
|
|
|
1049 |
re = re_1b;
|
|
|
1050 |
re2 = re2_1b;
|
|
|
1051 |
if (re.test(w)) {
|
|
|
1052 |
var fp = re.exec(w);
|
|
|
1053 |
re = re_mgr0;
|
|
|
1054 |
if (re.test(fp[1])) {
|
|
|
1055 |
re = re_1b_2;
|
|
|
1056 |
w = w.replace(re,"");
|
|
|
1057 |
}
|
|
|
1058 |
} else if (re2.test(w)) {
|
|
|
1059 |
var fp = re2.exec(w);
|
|
|
1060 |
stem = fp[1];
|
|
|
1061 |
re2 = re_s_v;
|
|
|
1062 |
if (re2.test(stem)) {
|
|
|
1063 |
w = stem;
|
|
|
1064 |
re2 = re2_1b_2;
|
|
|
1065 |
re3 = re3_1b_2;
|
|
|
1066 |
re4 = re4_1b_2;
|
|
|
1067 |
if (re2.test(w)) { w = w + "e"; }
|
|
|
1068 |
else if (re3.test(w)) { re = re_1b_2; w = w.replace(re,""); }
|
|
|
1069 |
else if (re4.test(w)) { w = w + "e"; }
|
|
|
1070 |
}
|
|
|
1071 |
}
|
|
|
1072 |
|
|
|
1073 |
// Step 1c - replace suffix y or Y by i if preceded by a non-vowel which is not the first letter of the word (so cry -> cri, by -> by, say -> say)
|
|
|
1074 |
re = re_1c;
|
|
|
1075 |
if (re.test(w)) {
|
|
|
1076 |
var fp = re.exec(w);
|
|
|
1077 |
stem = fp[1];
|
|
|
1078 |
w = stem + "i";
|
|
|
1079 |
}
|
|
|
1080 |
|
|
|
1081 |
// Step 2
|
|
|
1082 |
re = re_2;
|
|
|
1083 |
if (re.test(w)) {
|
|
|
1084 |
var fp = re.exec(w);
|
|
|
1085 |
stem = fp[1];
|
|
|
1086 |
suffix = fp[2];
|
|
|
1087 |
re = re_mgr0;
|
|
|
1088 |
if (re.test(stem)) {
|
|
|
1089 |
w = stem + step2list[suffix];
|
|
|
1090 |
}
|
|
|
1091 |
}
|
|
|
1092 |
|
|
|
1093 |
// Step 3
|
|
|
1094 |
re = re_3;
|
|
|
1095 |
if (re.test(w)) {
|
|
|
1096 |
var fp = re.exec(w);
|
|
|
1097 |
stem = fp[1];
|
|
|
1098 |
suffix = fp[2];
|
|
|
1099 |
re = re_mgr0;
|
|
|
1100 |
if (re.test(stem)) {
|
|
|
1101 |
w = stem + step3list[suffix];
|
|
|
1102 |
}
|
|
|
1103 |
}
|
|
|
1104 |
|
|
|
1105 |
// Step 4
|
|
|
1106 |
re = re_4;
|
|
|
1107 |
re2 = re2_4;
|
|
|
1108 |
if (re.test(w)) {
|
|
|
1109 |
var fp = re.exec(w);
|
|
|
1110 |
stem = fp[1];
|
|
|
1111 |
re = re_mgr1;
|
|
|
1112 |
if (re.test(stem)) {
|
|
|
1113 |
w = stem;
|
|
|
1114 |
}
|
|
|
1115 |
} else if (re2.test(w)) {
|
|
|
1116 |
var fp = re2.exec(w);
|
|
|
1117 |
stem = fp[1] + fp[2];
|
|
|
1118 |
re2 = re_mgr1;
|
|
|
1119 |
if (re2.test(stem)) {
|
|
|
1120 |
w = stem;
|
|
|
1121 |
}
|
|
|
1122 |
}
|
|
|
1123 |
|
|
|
1124 |
// Step 5
|
|
|
1125 |
re = re_5;
|
|
|
1126 |
if (re.test(w)) {
|
|
|
1127 |
var fp = re.exec(w);
|
|
|
1128 |
stem = fp[1];
|
|
|
1129 |
re = re_mgr1;
|
|
|
1130 |
re2 = re_meq1;
|
|
|
1131 |
re3 = re3_5;
|
|
|
1132 |
if (re.test(stem) || (re2.test(stem) && !(re3.test(stem)))) {
|
|
|
1133 |
w = stem;
|
|
|
1134 |
}
|
|
|
1135 |
}
|
|
|
1136 |
|
|
|
1137 |
re = re_5_1;
|
|
|
1138 |
re2 = re_mgr1;
|
|
|
1139 |
if (re.test(w) && re2.test(w)) {
|
|
|
1140 |
re = re_1b_2;
|
|
|
1141 |
w = w.replace(re,"");
|
|
|
1142 |
}
|
|
|
1143 |
|
|
|
1144 |
// and turn initial Y back to y
|
|
|
1145 |
|
|
|
1146 |
if (firstch == "y") {
|
|
|
1147 |
w = firstch.toLowerCase() + w.substr(1);
|
|
|
1148 |
}
|
|
|
1149 |
|
|
|
1150 |
return w;
|
|
|
1151 |
};
|
|
|
1152 |
|
|
|
1153 |
return function (token) {
|
|
|
1154 |
return token.update(porterStemmer);
|
|
|
1155 |
}
|
|
|
1156 |
})();
|
|
|
1157 |
|
|
|
1158 |
lunr.Pipeline.registerFunction(lunr.stemmer, 'stemmer')
|
|
|
1159 |
/*!
|
|
|
1160 |
* lunr.stopWordFilter
|
|
|
1161 |
* Copyright (C) 2020 Oliver Nightingale
|
|
|
1162 |
*/
|
|
|
1163 |
|
|
|
1164 |
/**
|
|
|
1165 |
* lunr.generateStopWordFilter builds a stopWordFilter function from the provided
|
|
|
1166 |
* list of stop words.
|
|
|
1167 |
*
|
|
|
1168 |
* The built in lunr.stopWordFilter is built using this generator and can be used
|
|
|
1169 |
* to generate custom stopWordFilters for applications or non English languages.
|
|
|
1170 |
*
|
|
|
1171 |
* @function
|
|
|
1172 |
* @param {Array} token The token to pass through the filter
|
|
|
1173 |
* @returns {lunr.PipelineFunction}
|
|
|
1174 |
* @see lunr.Pipeline
|
|
|
1175 |
* @see lunr.stopWordFilter
|
|
|
1176 |
*/
|
|
|
1177 |
lunr.generateStopWordFilter = function (stopWords) {
|
|
|
1178 |
var words = stopWords.reduce(function (memo, stopWord) {
|
|
|
1179 |
memo[stopWord] = stopWord
|
|
|
1180 |
return memo
|
|
|
1181 |
}, {})
|
|
|
1182 |
|
|
|
1183 |
return function (token) {
|
|
|
1184 |
if (token && words[token.toString()] !== token.toString()) return token
|
|
|
1185 |
}
|
|
|
1186 |
}
|
|
|
1187 |
|
|
|
1188 |
/**
|
|
|
1189 |
* lunr.stopWordFilter is an English language stop word list filter, any words
|
|
|
1190 |
* contained in the list will not be passed through the filter.
|
|
|
1191 |
*
|
|
|
1192 |
* This is intended to be used in the Pipeline. If the token does not pass the
|
|
|
1193 |
* filter then undefined will be returned.
|
|
|
1194 |
*
|
|
|
1195 |
* @function
|
|
|
1196 |
* @implements {lunr.PipelineFunction}
|
|
|
1197 |
* @params {lunr.Token} token - A token to check for being a stop word.
|
|
|
1198 |
* @returns {lunr.Token}
|
|
|
1199 |
* @see {@link lunr.Pipeline}
|
|
|
1200 |
*/
|
|
|
1201 |
lunr.stopWordFilter = lunr.generateStopWordFilter([
|
|
|
1202 |
'a',
|
|
|
1203 |
'able',
|
|
|
1204 |
'about',
|
|
|
1205 |
'across',
|
|
|
1206 |
'after',
|
|
|
1207 |
'all',
|
|
|
1208 |
'almost',
|
|
|
1209 |
'also',
|
|
|
1210 |
'am',
|
|
|
1211 |
'among',
|
|
|
1212 |
'an',
|
|
|
1213 |
'and',
|
|
|
1214 |
'any',
|
|
|
1215 |
'are',
|
|
|
1216 |
'as',
|
|
|
1217 |
'at',
|
|
|
1218 |
'be',
|
|
|
1219 |
'because',
|
|
|
1220 |
'been',
|
|
|
1221 |
'but',
|
|
|
1222 |
'by',
|
|
|
1223 |
'can',
|
|
|
1224 |
'cannot',
|
|
|
1225 |
'could',
|
|
|
1226 |
'dear',
|
|
|
1227 |
'did',
|
|
|
1228 |
'do',
|
|
|
1229 |
'does',
|
|
|
1230 |
'either',
|
|
|
1231 |
'else',
|
|
|
1232 |
'ever',
|
|
|
1233 |
'every',
|
|
|
1234 |
'for',
|
|
|
1235 |
'from',
|
|
|
1236 |
'get',
|
|
|
1237 |
'got',
|
|
|
1238 |
'had',
|
|
|
1239 |
'has',
|
|
|
1240 |
'have',
|
|
|
1241 |
'he',
|
|
|
1242 |
'her',
|
|
|
1243 |
'hers',
|
|
|
1244 |
'him',
|
|
|
1245 |
'his',
|
|
|
1246 |
'how',
|
|
|
1247 |
'however',
|
|
|
1248 |
'i',
|
|
|
1249 |
'if',
|
|
|
1250 |
'in',
|
|
|
1251 |
'into',
|
|
|
1252 |
'is',
|
|
|
1253 |
'it',
|
|
|
1254 |
'its',
|
|
|
1255 |
'just',
|
|
|
1256 |
'least',
|
|
|
1257 |
'let',
|
|
|
1258 |
'like',
|
|
|
1259 |
'likely',
|
|
|
1260 |
'may',
|
|
|
1261 |
'me',
|
|
|
1262 |
'might',
|
|
|
1263 |
'most',
|
|
|
1264 |
'must',
|
|
|
1265 |
'my',
|
|
|
1266 |
'neither',
|
|
|
1267 |
'no',
|
|
|
1268 |
'nor',
|
|
|
1269 |
'not',
|
|
|
1270 |
'of',
|
|
|
1271 |
'off',
|
|
|
1272 |
'often',
|
|
|
1273 |
'on',
|
|
|
1274 |
'only',
|
|
|
1275 |
'or',
|
|
|
1276 |
'other',
|
|
|
1277 |
'our',
|
|
|
1278 |
'own',
|
|
|
1279 |
'rather',
|
|
|
1280 |
'said',
|
|
|
1281 |
'say',
|
|
|
1282 |
'says',
|
|
|
1283 |
'she',
|
|
|
1284 |
'should',
|
|
|
1285 |
'since',
|
|
|
1286 |
'so',
|
|
|
1287 |
'some',
|
|
|
1288 |
'than',
|
|
|
1289 |
'that',
|
|
|
1290 |
'the',
|
|
|
1291 |
'their',
|
|
|
1292 |
'them',
|
|
|
1293 |
'then',
|
|
|
1294 |
'there',
|
|
|
1295 |
'these',
|
|
|
1296 |
'they',
|
|
|
1297 |
'this',
|
|
|
1298 |
'tis',
|
|
|
1299 |
'to',
|
|
|
1300 |
'too',
|
|
|
1301 |
'twas',
|
|
|
1302 |
'us',
|
|
|
1303 |
'wants',
|
|
|
1304 |
'was',
|
|
|
1305 |
'we',
|
|
|
1306 |
'were',
|
|
|
1307 |
'what',
|
|
|
1308 |
'when',
|
|
|
1309 |
'where',
|
|
|
1310 |
'which',
|
|
|
1311 |
'while',
|
|
|
1312 |
'who',
|
|
|
1313 |
'whom',
|
|
|
1314 |
'why',
|
|
|
1315 |
'will',
|
|
|
1316 |
'with',
|
|
|
1317 |
'would',
|
|
|
1318 |
'yet',
|
|
|
1319 |
'you',
|
|
|
1320 |
'your'
|
|
|
1321 |
])
|
|
|
1322 |
|
|
|
1323 |
lunr.Pipeline.registerFunction(lunr.stopWordFilter, 'stopWordFilter')
|
|
|
1324 |
/*!
|
|
|
1325 |
* lunr.trimmer
|
|
|
1326 |
* Copyright (C) 2020 Oliver Nightingale
|
|
|
1327 |
*/
|
|
|
1328 |
|
|
|
1329 |
/**
|
|
|
1330 |
* lunr.trimmer is a pipeline function for trimming non word
|
|
|
1331 |
* characters from the beginning and end of tokens before they
|
|
|
1332 |
* enter the index.
|
|
|
1333 |
*
|
|
|
1334 |
* This implementation may not work correctly for non latin
|
|
|
1335 |
* characters and should either be removed or adapted for use
|
|
|
1336 |
* with languages with non-latin characters.
|
|
|
1337 |
*
|
|
|
1338 |
* @static
|
|
|
1339 |
* @implements {lunr.PipelineFunction}
|
|
|
1340 |
* @param {lunr.Token} token The token to pass through the filter
|
|
|
1341 |
* @returns {lunr.Token}
|
|
|
1342 |
* @see lunr.Pipeline
|
|
|
1343 |
*/
|
|
|
1344 |
lunr.trimmer = function (token) {
|
|
|
1345 |
return token.update(function (s) {
|
|
|
1346 |
return s.replace(/^\W+/, '').replace(/\W+$/, '')
|
|
|
1347 |
})
|
|
|
1348 |
}
|
|
|
1349 |
|
|
|
1350 |
lunr.Pipeline.registerFunction(lunr.trimmer, 'trimmer')
|
|
|
1351 |
/*!
|
|
|
1352 |
* lunr.TokenSet
|
|
|
1353 |
* Copyright (C) 2020 Oliver Nightingale
|
|
|
1354 |
*/
|
|
|
1355 |
|
|
|
1356 |
/**
|
|
|
1357 |
* A token set is used to store the unique list of all tokens
|
|
|
1358 |
* within an index. Token sets are also used to represent an
|
|
|
1359 |
* incoming query to the index, this query token set and index
|
|
|
1360 |
* token set are then intersected to find which tokens to look
|
|
|
1361 |
* up in the inverted index.
|
|
|
1362 |
*
|
|
|
1363 |
* A token set can hold multiple tokens, as in the case of the
|
|
|
1364 |
* index token set, or it can hold a single token as in the
|
|
|
1365 |
* case of a simple query token set.
|
|
|
1366 |
*
|
|
|
1367 |
* Additionally token sets are used to perform wildcard matching.
|
|
|
1368 |
* Leading, contained and trailing wildcards are supported, and
|
|
|
1369 |
* from this edit distance matching can also be provided.
|
|
|
1370 |
*
|
|
|
1371 |
* Token sets are implemented as a minimal finite state automata,
|
|
|
1372 |
* where both common prefixes and suffixes are shared between tokens.
|
|
|
1373 |
* This helps to reduce the space used for storing the token set.
|
|
|
1374 |
*
|
|
|
1375 |
* @constructor
|
|
|
1376 |
*/
|
|
|
1377 |
lunr.TokenSet = function () {
|
|
|
1378 |
this.final = false
|
|
|
1379 |
this.edges = {}
|
|
|
1380 |
this.id = lunr.TokenSet._nextId
|
|
|
1381 |
lunr.TokenSet._nextId += 1
|
|
|
1382 |
}
|
|
|
1383 |
|
|
|
1384 |
/**
|
|
|
1385 |
* Keeps track of the next, auto increment, identifier to assign
|
|
|
1386 |
* to a new tokenSet.
|
|
|
1387 |
*
|
|
|
1388 |
* TokenSets require a unique identifier to be correctly minimised.
|
|
|
1389 |
*
|
|
|
1390 |
* @private
|
|
|
1391 |
*/
|
|
|
1392 |
lunr.TokenSet._nextId = 1
|
|
|
1393 |
|
|
|
1394 |
/**
|
|
|
1395 |
* Creates a TokenSet instance from the given sorted array of words.
|
|
|
1396 |
*
|
|
|
1397 |
* @param {String[]} arr - A sorted array of strings to create the set from.
|
|
|
1398 |
* @returns {lunr.TokenSet}
|
|
|
1399 |
* @throws Will throw an error if the input array is not sorted.
|
|
|
1400 |
*/
|
|
|
1401 |
lunr.TokenSet.fromArray = function (arr) {
|
|
|
1402 |
var builder = new lunr.TokenSet.Builder
|
|
|
1403 |
|
|
|
1404 |
for (var i = 0, len = arr.length; i < len; i++) {
|
|
|
1405 |
builder.insert(arr[i])
|
|
|
1406 |
}
|
|
|
1407 |
|
|
|
1408 |
builder.finish()
|
|
|
1409 |
return builder.root
|
|
|
1410 |
}
|
|
|
1411 |
|
|
|
1412 |
/**
|
|
|
1413 |
* Creates a token set from a query clause.
|
|
|
1414 |
*
|
|
|
1415 |
* @private
|
|
|
1416 |
* @param {Object} clause - A single clause from lunr.Query.
|
|
|
1417 |
* @param {string} clause.term - The query clause term.
|
|
|
1418 |
* @param {number} [clause.editDistance] - The optional edit distance for the term.
|
|
|
1419 |
* @returns {lunr.TokenSet}
|
|
|
1420 |
*/
|
|
|
1421 |
lunr.TokenSet.fromClause = function (clause) {
|
|
|
1422 |
if ('editDistance' in clause) {
|
|
|
1423 |
return lunr.TokenSet.fromFuzzyString(clause.term, clause.editDistance)
|
|
|
1424 |
} else {
|
|
|
1425 |
return lunr.TokenSet.fromString(clause.term)
|
|
|
1426 |
}
|
|
|
1427 |
}
|
|
|
1428 |
|
|
|
1429 |
/**
|
|
|
1430 |
* Creates a token set representing a single string with a specified
|
|
|
1431 |
* edit distance.
|
|
|
1432 |
*
|
|
|
1433 |
* Insertions, deletions, substitutions and transpositions are each
|
|
|
1434 |
* treated as an edit distance of 1.
|
|
|
1435 |
*
|
|
|
1436 |
* Increasing the allowed edit distance will have a dramatic impact
|
|
|
1437 |
* on the performance of both creating and intersecting these TokenSets.
|
|
|
1438 |
* It is advised to keep the edit distance less than 3.
|
|
|
1439 |
*
|
|
|
1440 |
* @param {string} str - The string to create the token set from.
|
|
|
1441 |
* @param {number} editDistance - The allowed edit distance to match.
|
|
|
1442 |
* @returns {lunr.Vector}
|
|
|
1443 |
*/
|
|
|
1444 |
lunr.TokenSet.fromFuzzyString = function (str, editDistance) {
|
|
|
1445 |
var root = new lunr.TokenSet
|
|
|
1446 |
|
|
|
1447 |
var stack = [{
|
|
|
1448 |
node: root,
|
|
|
1449 |
editsRemaining: editDistance,
|
|
|
1450 |
str: str
|
|
|
1451 |
}]
|
|
|
1452 |
|
|
|
1453 |
while (stack.length) {
|
|
|
1454 |
var frame = stack.pop()
|
|
|
1455 |
|
|
|
1456 |
// no edit
|
|
|
1457 |
if (frame.str.length > 0) {
|
|
|
1458 |
var char = frame.str.charAt(0),
|
|
|
1459 |
noEditNode
|
|
|
1460 |
|
|
|
1461 |
if (char in frame.node.edges) {
|
|
|
1462 |
noEditNode = frame.node.edges[char]
|
|
|
1463 |
} else {
|
|
|
1464 |
noEditNode = new lunr.TokenSet
|
|
|
1465 |
frame.node.edges[char] = noEditNode
|
|
|
1466 |
}
|
|
|
1467 |
|
|
|
1468 |
if (frame.str.length == 1) {
|
|
|
1469 |
noEditNode.final = true
|
|
|
1470 |
}
|
|
|
1471 |
|
|
|
1472 |
stack.push({
|
|
|
1473 |
node: noEditNode,
|
|
|
1474 |
editsRemaining: frame.editsRemaining,
|
|
|
1475 |
str: frame.str.slice(1)
|
|
|
1476 |
})
|
|
|
1477 |
}
|
|
|
1478 |
|
|
|
1479 |
if (frame.editsRemaining == 0) {
|
|
|
1480 |
continue
|
|
|
1481 |
}
|
|
|
1482 |
|
|
|
1483 |
// insertion
|
|
|
1484 |
if ("*" in frame.node.edges) {
|
|
|
1485 |
var insertionNode = frame.node.edges["*"]
|
|
|
1486 |
} else {
|
|
|
1487 |
var insertionNode = new lunr.TokenSet
|
|
|
1488 |
frame.node.edges["*"] = insertionNode
|
|
|
1489 |
}
|
|
|
1490 |
|
|
|
1491 |
if (frame.str.length == 0) {
|
|
|
1492 |
insertionNode.final = true
|
|
|
1493 |
}
|
|
|
1494 |
|
|
|
1495 |
stack.push({
|
|
|
1496 |
node: insertionNode,
|
|
|
1497 |
editsRemaining: frame.editsRemaining - 1,
|
|
|
1498 |
str: frame.str
|
|
|
1499 |
})
|
|
|
1500 |
|
|
|
1501 |
// deletion
|
|
|
1502 |
// can only do a deletion if we have enough edits remaining
|
|
|
1503 |
// and if there are characters left to delete in the string
|
|
|
1504 |
if (frame.str.length > 1) {
|
|
|
1505 |
stack.push({
|
|
|
1506 |
node: frame.node,
|
|
|
1507 |
editsRemaining: frame.editsRemaining - 1,
|
|
|
1508 |
str: frame.str.slice(1)
|
|
|
1509 |
})
|
|
|
1510 |
}
|
|
|
1511 |
|
|
|
1512 |
// deletion
|
|
|
1513 |
// just removing the last character from the str
|
|
|
1514 |
if (frame.str.length == 1) {
|
|
|
1515 |
frame.node.final = true
|
|
|
1516 |
}
|
|
|
1517 |
|
|
|
1518 |
// substitution
|
|
|
1519 |
// can only do a substitution if we have enough edits remaining
|
|
|
1520 |
// and if there are characters left to substitute
|
|
|
1521 |
if (frame.str.length >= 1) {
|
|
|
1522 |
if ("*" in frame.node.edges) {
|
|
|
1523 |
var substitutionNode = frame.node.edges["*"]
|
|
|
1524 |
} else {
|
|
|
1525 |
var substitutionNode = new lunr.TokenSet
|
|
|
1526 |
frame.node.edges["*"] = substitutionNode
|
|
|
1527 |
}
|
|
|
1528 |
|
|
|
1529 |
if (frame.str.length == 1) {
|
|
|
1530 |
substitutionNode.final = true
|
|
|
1531 |
}
|
|
|
1532 |
|
|
|
1533 |
stack.push({
|
|
|
1534 |
node: substitutionNode,
|
|
|
1535 |
editsRemaining: frame.editsRemaining - 1,
|
|
|
1536 |
str: frame.str.slice(1)
|
|
|
1537 |
})
|
|
|
1538 |
}
|
|
|
1539 |
|
|
|
1540 |
// transposition
|
|
|
1541 |
// can only do a transposition if there are edits remaining
|
|
|
1542 |
// and there are enough characters to transpose
|
|
|
1543 |
if (frame.str.length > 1) {
|
|
|
1544 |
var charA = frame.str.charAt(0),
|
|
|
1545 |
charB = frame.str.charAt(1),
|
|
|
1546 |
transposeNode
|
|
|
1547 |
|
|
|
1548 |
if (charB in frame.node.edges) {
|
|
|
1549 |
transposeNode = frame.node.edges[charB]
|
|
|
1550 |
} else {
|
|
|
1551 |
transposeNode = new lunr.TokenSet
|
|
|
1552 |
frame.node.edges[charB] = transposeNode
|
|
|
1553 |
}
|
|
|
1554 |
|
|
|
1555 |
if (frame.str.length == 1) {
|
|
|
1556 |
transposeNode.final = true
|
|
|
1557 |
}
|
|
|
1558 |
|
|
|
1559 |
stack.push({
|
|
|
1560 |
node: transposeNode,
|
|
|
1561 |
editsRemaining: frame.editsRemaining - 1,
|
|
|
1562 |
str: charA + frame.str.slice(2)
|
|
|
1563 |
})
|
|
|
1564 |
}
|
|
|
1565 |
}
|
|
|
1566 |
|
|
|
1567 |
return root
|
|
|
1568 |
}
|
|
|
1569 |
|
|
|
1570 |
/**
|
|
|
1571 |
* Creates a TokenSet from a string.
|
|
|
1572 |
*
|
|
|
1573 |
* The string may contain one or more wildcard characters (*)
|
|
|
1574 |
* that will allow wildcard matching when intersecting with
|
|
|
1575 |
* another TokenSet.
|
|
|
1576 |
*
|
|
|
1577 |
* @param {string} str - The string to create a TokenSet from.
|
|
|
1578 |
* @returns {lunr.TokenSet}
|
|
|
1579 |
*/
|
|
|
1580 |
lunr.TokenSet.fromString = function (str) {
|
|
|
1581 |
var node = new lunr.TokenSet,
|
|
|
1582 |
root = node
|
|
|
1583 |
|
|
|
1584 |
/*
|
|
|
1585 |
* Iterates through all characters within the passed string
|
|
|
1586 |
* appending a node for each character.
|
|
|
1587 |
*
|
|
|
1588 |
* When a wildcard character is found then a self
|
|
|
1589 |
* referencing edge is introduced to continually match
|
|
|
1590 |
* any number of any characters.
|
|
|
1591 |
*/
|
|
|
1592 |
for (var i = 0, len = str.length; i < len; i++) {
|
|
|
1593 |
var char = str[i],
|
|
|
1594 |
final = (i == len - 1)
|
|
|
1595 |
|
|
|
1596 |
if (char == "*") {
|
|
|
1597 |
node.edges[char] = node
|
|
|
1598 |
node.final = final
|
|
|
1599 |
|
|
|
1600 |
} else {
|
|
|
1601 |
var next = new lunr.TokenSet
|
|
|
1602 |
next.final = final
|
|
|
1603 |
|
|
|
1604 |
node.edges[char] = next
|
|
|
1605 |
node = next
|
|
|
1606 |
}
|
|
|
1607 |
}
|
|
|
1608 |
|
|
|
1609 |
return root
|
|
|
1610 |
}
|
|
|
1611 |
|
|
|
1612 |
/**
|
|
|
1613 |
* Converts this TokenSet into an array of strings
|
|
|
1614 |
* contained within the TokenSet.
|
|
|
1615 |
*
|
|
|
1616 |
* This is not intended to be used on a TokenSet that
|
|
|
1617 |
* contains wildcards, in these cases the results are
|
|
|
1618 |
* undefined and are likely to cause an infinite loop.
|
|
|
1619 |
*
|
|
|
1620 |
* @returns {string[]}
|
|
|
1621 |
*/
|
|
|
1622 |
lunr.TokenSet.prototype.toArray = function () {
|
|
|
1623 |
var words = []
|
|
|
1624 |
|
|
|
1625 |
var stack = [{
|
|
|
1626 |
prefix: "",
|
|
|
1627 |
node: this
|
|
|
1628 |
}]
|
|
|
1629 |
|
|
|
1630 |
while (stack.length) {
|
|
|
1631 |
var frame = stack.pop(),
|
|
|
1632 |
edges = Object.keys(frame.node.edges),
|
|
|
1633 |
len = edges.length
|
|
|
1634 |
|
|
|
1635 |
if (frame.node.final) {
|
|
|
1636 |
/* In Safari, at this point the prefix is sometimes corrupted, see:
|
|
|
1637 |
* https://github.com/olivernn/lunr.js/issues/279 Calling any
|
|
|
1638 |
* String.prototype method forces Safari to "cast" this string to what
|
|
|
1639 |
* it's supposed to be, fixing the bug. */
|
|
|
1640 |
frame.prefix.charAt(0)
|
|
|
1641 |
words.push(frame.prefix)
|
|
|
1642 |
}
|
|
|
1643 |
|
|
|
1644 |
for (var i = 0; i < len; i++) {
|
|
|
1645 |
var edge = edges[i]
|
|
|
1646 |
|
|
|
1647 |
stack.push({
|
|
|
1648 |
prefix: frame.prefix.concat(edge),
|
|
|
1649 |
node: frame.node.edges[edge]
|
|
|
1650 |
})
|
|
|
1651 |
}
|
|
|
1652 |
}
|
|
|
1653 |
|
|
|
1654 |
return words
|
|
|
1655 |
}
|
|
|
1656 |
|
|
|
1657 |
/**
|
|
|
1658 |
* Generates a string representation of a TokenSet.
|
|
|
1659 |
*
|
|
|
1660 |
* This is intended to allow TokenSets to be used as keys
|
|
|
1661 |
* in objects, largely to aid the construction and minimisation
|
|
|
1662 |
* of a TokenSet. As such it is not designed to be a human
|
|
|
1663 |
* friendly representation of the TokenSet.
|
|
|
1664 |
*
|
|
|
1665 |
* @returns {string}
|
|
|
1666 |
*/
|
|
|
1667 |
lunr.TokenSet.prototype.toString = function () {
|
|
|
1668 |
// NOTE: Using Object.keys here as this.edges is very likely
|
|
|
1669 |
// to enter 'hash-mode' with many keys being added
|
|
|
1670 |
//
|
|
|
1671 |
// avoiding a for-in loop here as it leads to the function
|
|
|
1672 |
// being de-optimised (at least in V8). From some simple
|
|
|
1673 |
// benchmarks the performance is comparable, but allowing
|
|
|
1674 |
// V8 to optimize may mean easy performance wins in the future.
|
|
|
1675 |
|
|
|
1676 |
if (this._str) {
|
|
|
1677 |
return this._str
|
|
|
1678 |
}
|
|
|
1679 |
|
|
|
1680 |
var str = this.final ? '1' : '0',
|
|
|
1681 |
labels = Object.keys(this.edges).sort(),
|
|
|
1682 |
len = labels.length
|
|
|
1683 |
|
|
|
1684 |
for (var i = 0; i < len; i++) {
|
|
|
1685 |
var label = labels[i],
|
|
|
1686 |
node = this.edges[label]
|
|
|
1687 |
|
|
|
1688 |
str = str + label + node.id
|
|
|
1689 |
}
|
|
|
1690 |
|
|
|
1691 |
return str
|
|
|
1692 |
}
|
|
|
1693 |
|
|
|
1694 |
/**
|
|
|
1695 |
* Returns a new TokenSet that is the intersection of
|
|
|
1696 |
* this TokenSet and the passed TokenSet.
|
|
|
1697 |
*
|
|
|
1698 |
* This intersection will take into account any wildcards
|
|
|
1699 |
* contained within the TokenSet.
|
|
|
1700 |
*
|
|
|
1701 |
* @param {lunr.TokenSet} b - An other TokenSet to intersect with.
|
|
|
1702 |
* @returns {lunr.TokenSet}
|
|
|
1703 |
*/
|
|
|
1704 |
lunr.TokenSet.prototype.intersect = function (b) {
|
|
|
1705 |
var output = new lunr.TokenSet,
|
|
|
1706 |
frame = undefined
|
|
|
1707 |
|
|
|
1708 |
var stack = [{
|
|
|
1709 |
qNode: b,
|
|
|
1710 |
output: output,
|
|
|
1711 |
node: this
|
|
|
1712 |
}]
|
|
|
1713 |
|
|
|
1714 |
while (stack.length) {
|
|
|
1715 |
frame = stack.pop()
|
|
|
1716 |
|
|
|
1717 |
// NOTE: As with the #toString method, we are using
|
|
|
1718 |
// Object.keys and a for loop instead of a for-in loop
|
|
|
1719 |
// as both of these objects enter 'hash' mode, causing
|
|
|
1720 |
// the function to be de-optimised in V8
|
|
|
1721 |
var qEdges = Object.keys(frame.qNode.edges),
|
|
|
1722 |
qLen = qEdges.length,
|
|
|
1723 |
nEdges = Object.keys(frame.node.edges),
|
|
|
1724 |
nLen = nEdges.length
|
|
|
1725 |
|
|
|
1726 |
for (var q = 0; q < qLen; q++) {
|
|
|
1727 |
var qEdge = qEdges[q]
|
|
|
1728 |
|
|
|
1729 |
for (var n = 0; n < nLen; n++) {
|
|
|
1730 |
var nEdge = nEdges[n]
|
|
|
1731 |
|
|
|
1732 |
if (nEdge == qEdge || qEdge == '*') {
|
|
|
1733 |
var node = frame.node.edges[nEdge],
|
|
|
1734 |
qNode = frame.qNode.edges[qEdge],
|
|
|
1735 |
final = node.final && qNode.final,
|
|
|
1736 |
next = undefined
|
|
|
1737 |
|
|
|
1738 |
if (nEdge in frame.output.edges) {
|
|
|
1739 |
// an edge already exists for this character
|
|
|
1740 |
// no need to create a new node, just set the finality
|
|
|
1741 |
// bit unless this node is already final
|
|
|
1742 |
next = frame.output.edges[nEdge]
|
|
|
1743 |
next.final = next.final || final
|
|
|
1744 |
|
|
|
1745 |
} else {
|
|
|
1746 |
// no edge exists yet, must create one
|
|
|
1747 |
// set the finality bit and insert it
|
|
|
1748 |
// into the output
|
|
|
1749 |
next = new lunr.TokenSet
|
|
|
1750 |
next.final = final
|
|
|
1751 |
frame.output.edges[nEdge] = next
|
|
|
1752 |
}
|
|
|
1753 |
|
|
|
1754 |
stack.push({
|
|
|
1755 |
qNode: qNode,
|
|
|
1756 |
output: next,
|
|
|
1757 |
node: node
|
|
|
1758 |
})
|
|
|
1759 |
}
|
|
|
1760 |
}
|
|
|
1761 |
}
|
|
|
1762 |
}
|
|
|
1763 |
|
|
|
1764 |
return output
|
|
|
1765 |
}
|
|
|
1766 |
lunr.TokenSet.Builder = function () {
|
|
|
1767 |
this.previousWord = ""
|
|
|
1768 |
this.root = new lunr.TokenSet
|
|
|
1769 |
this.uncheckedNodes = []
|
|
|
1770 |
this.minimizedNodes = {}
|
|
|
1771 |
}
|
|
|
1772 |
|
|
|
1773 |
lunr.TokenSet.Builder.prototype.insert = function (word) {
|
|
|
1774 |
var node,
|
|
|
1775 |
commonPrefix = 0
|
|
|
1776 |
|
|
|
1777 |
if (word < this.previousWord) {
|
|
|
1778 |
throw new Error ("Out of order word insertion")
|
|
|
1779 |
}
|
|
|
1780 |
|
|
|
1781 |
for (var i = 0; i < word.length && i < this.previousWord.length; i++) {
|
|
|
1782 |
if (word[i] != this.previousWord[i]) break
|
|
|
1783 |
commonPrefix++
|
|
|
1784 |
}
|
|
|
1785 |
|
|
|
1786 |
this.minimize(commonPrefix)
|
|
|
1787 |
|
|
|
1788 |
if (this.uncheckedNodes.length == 0) {
|
|
|
1789 |
node = this.root
|
|
|
1790 |
} else {
|
|
|
1791 |
node = this.uncheckedNodes[this.uncheckedNodes.length - 1].child
|
|
|
1792 |
}
|
|
|
1793 |
|
|
|
1794 |
for (var i = commonPrefix; i < word.length; i++) {
|
|
|
1795 |
var nextNode = new lunr.TokenSet,
|
|
|
1796 |
char = word[i]
|
|
|
1797 |
|
|
|
1798 |
node.edges[char] = nextNode
|
|
|
1799 |
|
|
|
1800 |
this.uncheckedNodes.push({
|
|
|
1801 |
parent: node,
|
|
|
1802 |
char: char,
|
|
|
1803 |
child: nextNode
|
|
|
1804 |
})
|
|
|
1805 |
|
|
|
1806 |
node = nextNode
|
|
|
1807 |
}
|
|
|
1808 |
|
|
|
1809 |
node.final = true
|
|
|
1810 |
this.previousWord = word
|
|
|
1811 |
}
|
|
|
1812 |
|
|
|
1813 |
lunr.TokenSet.Builder.prototype.finish = function () {
|
|
|
1814 |
this.minimize(0)
|
|
|
1815 |
}
|
|
|
1816 |
|
|
|
1817 |
lunr.TokenSet.Builder.prototype.minimize = function (downTo) {
|
|
|
1818 |
for (var i = this.uncheckedNodes.length - 1; i >= downTo; i--) {
|
|
|
1819 |
var node = this.uncheckedNodes[i],
|
|
|
1820 |
childKey = node.child.toString()
|
|
|
1821 |
|
|
|
1822 |
if (childKey in this.minimizedNodes) {
|
|
|
1823 |
node.parent.edges[node.char] = this.minimizedNodes[childKey]
|
|
|
1824 |
} else {
|
|
|
1825 |
// Cache the key for this node since
|
|
|
1826 |
// we know it can't change anymore
|
|
|
1827 |
node.child._str = childKey
|
|
|
1828 |
|
|
|
1829 |
this.minimizedNodes[childKey] = node.child
|
|
|
1830 |
}
|
|
|
1831 |
|
|
|
1832 |
this.uncheckedNodes.pop()
|
|
|
1833 |
}
|
|
|
1834 |
}
|
|
|
1835 |
/*!
|
|
|
1836 |
* lunr.Index
|
|
|
1837 |
* Copyright (C) 2020 Oliver Nightingale
|
|
|
1838 |
*/
|
|
|
1839 |
|
|
|
1840 |
/**
|
|
|
1841 |
* An index contains the built index of all documents and provides a query interface
|
|
|
1842 |
* to the index.
|
|
|
1843 |
*
|
|
|
1844 |
* Usually instances of lunr.Index will not be created using this constructor, instead
|
|
|
1845 |
* lunr.Builder should be used to construct new indexes, or lunr.Index.load should be
|
|
|
1846 |
* used to load previously built and serialized indexes.
|
|
|
1847 |
*
|
|
|
1848 |
* @constructor
|
|
|
1849 |
* @param {Object} attrs - The attributes of the built search index.
|
|
|
1850 |
* @param {Object} attrs.invertedIndex - An index of term/field to document reference.
|
|
|
1851 |
* @param {Object<string, lunr.Vector>} attrs.fieldVectors - Field vectors
|
|
|
1852 |
* @param {lunr.TokenSet} attrs.tokenSet - An set of all corpus tokens.
|
|
|
1853 |
* @param {string[]} attrs.fields - The names of indexed document fields.
|
|
|
1854 |
* @param {lunr.Pipeline} attrs.pipeline - The pipeline to use for search terms.
|
|
|
1855 |
*/
|
|
|
1856 |
lunr.Index = function (attrs) {
|
|
|
1857 |
this.invertedIndex = attrs.invertedIndex
|
|
|
1858 |
this.fieldVectors = attrs.fieldVectors
|
|
|
1859 |
this.tokenSet = attrs.tokenSet
|
|
|
1860 |
this.fields = attrs.fields
|
|
|
1861 |
this.pipeline = attrs.pipeline
|
|
|
1862 |
}
|
|
|
1863 |
|
|
|
1864 |
/**
|
|
|
1865 |
* A result contains details of a document matching a search query.
|
|
|
1866 |
* @typedef {Object} lunr.Index~Result
|
|
|
1867 |
* @property {string} ref - The reference of the document this result represents.
|
|
|
1868 |
* @property {number} score - A number between 0 and 1 representing how similar this document is to the query.
|
|
|
1869 |
* @property {lunr.MatchData} matchData - Contains metadata about this match including which term(s) caused the match.
|
|
|
1870 |
*/
|
|
|
1871 |
|
|
|
1872 |
/**
|
|
|
1873 |
* Although lunr provides the ability to create queries using lunr.Query, it also provides a simple
|
|
|
1874 |
* query language which itself is parsed into an instance of lunr.Query.
|
|
|
1875 |
*
|
|
|
1876 |
* For programmatically building queries it is advised to directly use lunr.Query, the query language
|
|
|
1877 |
* is best used for human entered text rather than program generated text.
|
|
|
1878 |
*
|
|
|
1879 |
* At its simplest queries can just be a single term, e.g. `hello`, multiple terms are also supported
|
|
|
1880 |
* and will be combined with OR, e.g `hello world` will match documents that contain either 'hello'
|
|
|
1881 |
* or 'world', though those that contain both will rank higher in the results.
|
|
|
1882 |
*
|
|
|
1883 |
* Wildcards can be included in terms to match one or more unspecified characters, these wildcards can
|
|
|
1884 |
* be inserted anywhere within the term, and more than one wildcard can exist in a single term. Adding
|
|
|
1885 |
* wildcards will increase the number of documents that will be found but can also have a negative
|
|
|
1886 |
* impact on query performance, especially with wildcards at the beginning of a term.
|
|
|
1887 |
*
|
|
|
1888 |
* Terms can be restricted to specific fields, e.g. `title:hello`, only documents with the term
|
|
|
1889 |
* hello in the title field will match this query. Using a field not present in the index will lead
|
|
|
1890 |
* to an error being thrown.
|
|
|
1891 |
*
|
|
|
1892 |
* Modifiers can also be added to terms, lunr supports edit distance and boost modifiers on terms. A term
|
|
|
1893 |
* boost will make documents matching that term score higher, e.g. `foo^5`. Edit distance is also supported
|
|
|
1894 |
* to provide fuzzy matching, e.g. 'hello~2' will match documents with hello with an edit distance of 2.
|
|
|
1895 |
* Avoid large values for edit distance to improve query performance.
|
|
|
1896 |
*
|
|
|
1897 |
* Each term also supports a presence modifier. By default a term's presence in document is optional, however
|
|
|
1898 |
* this can be changed to either required or prohibited. For a term's presence to be required in a document the
|
|
|
1899 |
* term should be prefixed with a '+', e.g. `+foo bar` is a search for documents that must contain 'foo' and
|
|
|
1900 |
* optionally contain 'bar'. Conversely a leading '-' sets the terms presence to prohibited, i.e. it must not
|
|
|
1901 |
* appear in a document, e.g. `-foo bar` is a search for documents that do not contain 'foo' but may contain 'bar'.
|
|
|
1902 |
*
|
|
|
1903 |
* To escape special characters the backslash character '\' can be used, this allows searches to include
|
|
|
1904 |
* characters that would normally be considered modifiers, e.g. `foo\~2` will search for a term "foo~2" instead
|
|
|
1905 |
* of attempting to apply a boost of 2 to the search term "foo".
|
|
|
1906 |
*
|
|
|
1907 |
* @typedef {string} lunr.Index~QueryString
|
|
|
1908 |
* @example <caption>Simple single term query</caption>
|
|
|
1909 |
* hello
|
|
|
1910 |
* @example <caption>Multiple term query</caption>
|
|
|
1911 |
* hello world
|
|
|
1912 |
* @example <caption>term scoped to a field</caption>
|
|
|
1913 |
* title:hello
|
|
|
1914 |
* @example <caption>term with a boost of 10</caption>
|
|
|
1915 |
* hello^10
|
|
|
1916 |
* @example <caption>term with an edit distance of 2</caption>
|
|
|
1917 |
* hello~2
|
|
|
1918 |
* @example <caption>terms with presence modifiers</caption>
|
|
|
1919 |
* -foo +bar baz
|
|
|
1920 |
*/
|
|
|
1921 |
|
|
|
1922 |
/**
|
|
|
1923 |
* Performs a search against the index using lunr query syntax.
|
|
|
1924 |
*
|
|
|
1925 |
* Results will be returned sorted by their score, the most relevant results
|
|
|
1926 |
* will be returned first. For details on how the score is calculated, please see
|
|
|
1927 |
* the {@link https://lunrjs.com/guides/searching.html#scoring|guide}.
|
|
|
1928 |
*
|
|
|
1929 |
* For more programmatic querying use lunr.Index#query.
|
|
|
1930 |
*
|
|
|
1931 |
* @param {lunr.Index~QueryString} queryString - A string containing a lunr query.
|
|
|
1932 |
* @throws {lunr.QueryParseError} If the passed query string cannot be parsed.
|
|
|
1933 |
* @returns {lunr.Index~Result[]}
|
|
|
1934 |
*/
|
|
|
1935 |
lunr.Index.prototype.search = function (queryString) {
|
|
|
1936 |
return this.query(function (query) {
|
|
|
1937 |
var parser = new lunr.QueryParser(queryString, query)
|
|
|
1938 |
parser.parse()
|
|
|
1939 |
})
|
|
|
1940 |
}
|
|
|
1941 |
|
|
|
1942 |
/**
|
|
|
1943 |
* A query builder callback provides a query object to be used to express
|
|
|
1944 |
* the query to perform on the index.
|
|
|
1945 |
*
|
|
|
1946 |
* @callback lunr.Index~queryBuilder
|
|
|
1947 |
* @param {lunr.Query} query - The query object to build up.
|
|
|
1948 |
* @this lunr.Query
|
|
|
1949 |
*/
|
|
|
1950 |
|
|
|
1951 |
/**
|
|
|
1952 |
* Performs a query against the index using the yielded lunr.Query object.
|
|
|
1953 |
*
|
|
|
1954 |
* If performing programmatic queries against the index, this method is preferred
|
|
|
1955 |
* over lunr.Index#search so as to avoid the additional query parsing overhead.
|
|
|
1956 |
*
|
|
|
1957 |
* A query object is yielded to the supplied function which should be used to
|
|
|
1958 |
* express the query to be run against the index.
|
|
|
1959 |
*
|
|
|
1960 |
* Note that although this function takes a callback parameter it is _not_ an
|
|
|
1961 |
* asynchronous operation, the callback is just yielded a query object to be
|
|
|
1962 |
* customized.
|
|
|
1963 |
*
|
|
|
1964 |
* @param {lunr.Index~queryBuilder} fn - A function that is used to build the query.
|
|
|
1965 |
* @returns {lunr.Index~Result[]}
|
|
|
1966 |
*/
|
|
|
1967 |
lunr.Index.prototype.query = function (fn) {
|
|
|
1968 |
// for each query clause
|
|
|
1969 |
// * process terms
|
|
|
1970 |
// * expand terms from token set
|
|
|
1971 |
// * find matching documents and metadata
|
|
|
1972 |
// * get document vectors
|
|
|
1973 |
// * score documents
|
|
|
1974 |
|
|
|
1975 |
var query = new lunr.Query(this.fields),
|
|
|
1976 |
matchingFields = Object.create(null),
|
|
|
1977 |
queryVectors = Object.create(null),
|
|
|
1978 |
termFieldCache = Object.create(null),
|
|
|
1979 |
requiredMatches = Object.create(null),
|
|
|
1980 |
prohibitedMatches = Object.create(null)
|
|
|
1981 |
|
|
|
1982 |
/*
|
|
|
1983 |
* To support field level boosts a query vector is created per
|
|
|
1984 |
* field. An empty vector is eagerly created to support negated
|
|
|
1985 |
* queries.
|
|
|
1986 |
*/
|
|
|
1987 |
for (var i = 0; i < this.fields.length; i++) {
|
|
|
1988 |
queryVectors[this.fields[i]] = new lunr.Vector
|
|
|
1989 |
}
|
|
|
1990 |
|
|
|
1991 |
fn.call(query, query)
|
|
|
1992 |
|
|
|
1993 |
for (var i = 0; i < query.clauses.length; i++) {
|
|
|
1994 |
/*
|
|
|
1995 |
* Unless the pipeline has been disabled for this term, which is
|
|
|
1996 |
* the case for terms with wildcards, we need to pass the clause
|
|
|
1997 |
* term through the search pipeline. A pipeline returns an array
|
|
|
1998 |
* of processed terms. Pipeline functions may expand the passed
|
|
|
1999 |
* term, which means we may end up performing multiple index lookups
|
|
|
2000 |
* for a single query term.
|
|
|
2001 |
*/
|
|
|
2002 |
var clause = query.clauses[i],
|
|
|
2003 |
terms = null,
|
|
|
2004 |
clauseMatches = lunr.Set.empty
|
|
|
2005 |
|
|
|
2006 |
if (clause.usePipeline) {
|
|
|
2007 |
terms = this.pipeline.runString(clause.term, {
|
|
|
2008 |
fields: clause.fields
|
|
|
2009 |
})
|
|
|
2010 |
} else {
|
|
|
2011 |
terms = [clause.term]
|
|
|
2012 |
}
|
|
|
2013 |
|
|
|
2014 |
for (var m = 0; m < terms.length; m++) {
|
|
|
2015 |
var term = terms[m]
|
|
|
2016 |
|
|
|
2017 |
/*
|
|
|
2018 |
* Each term returned from the pipeline needs to use the same query
|
|
|
2019 |
* clause object, e.g. the same boost and or edit distance. The
|
|
|
2020 |
* simplest way to do this is to re-use the clause object but mutate
|
|
|
2021 |
* its term property.
|
|
|
2022 |
*/
|
|
|
2023 |
clause.term = term
|
|
|
2024 |
|
|
|
2025 |
/*
|
|
|
2026 |
* From the term in the clause we create a token set which will then
|
|
|
2027 |
* be used to intersect the indexes token set to get a list of terms
|
|
|
2028 |
* to lookup in the inverted index
|
|
|
2029 |
*/
|
|
|
2030 |
var termTokenSet = lunr.TokenSet.fromClause(clause),
|
|
|
2031 |
expandedTerms = this.tokenSet.intersect(termTokenSet).toArray()
|
|
|
2032 |
|
|
|
2033 |
/*
|
|
|
2034 |
* If a term marked as required does not exist in the tokenSet it is
|
|
|
2035 |
* impossible for the search to return any matches. We set all the field
|
|
|
2036 |
* scoped required matches set to empty and stop examining any further
|
|
|
2037 |
* clauses.
|
|
|
2038 |
*/
|
|
|
2039 |
if (expandedTerms.length === 0 && clause.presence === lunr.Query.presence.REQUIRED) {
|
|
|
2040 |
for (var k = 0; k < clause.fields.length; k++) {
|
|
|
2041 |
var field = clause.fields[k]
|
|
|
2042 |
requiredMatches[field] = lunr.Set.empty
|
|
|
2043 |
}
|
|
|
2044 |
|
|
|
2045 |
break
|
|
|
2046 |
}
|
|
|
2047 |
|
|
|
2048 |
for (var j = 0; j < expandedTerms.length; j++) {
|
|
|
2049 |
/*
|
|
|
2050 |
* For each term get the posting and termIndex, this is required for
|
|
|
2051 |
* building the query vector.
|
|
|
2052 |
*/
|
|
|
2053 |
var expandedTerm = expandedTerms[j],
|
|
|
2054 |
posting = this.invertedIndex[expandedTerm],
|
|
|
2055 |
termIndex = posting._index
|
|
|
2056 |
|
|
|
2057 |
for (var k = 0; k < clause.fields.length; k++) {
|
|
|
2058 |
/*
|
|
|
2059 |
* For each field that this query term is scoped by (by default
|
|
|
2060 |
* all fields are in scope) we need to get all the document refs
|
|
|
2061 |
* that have this term in that field.
|
|
|
2062 |
*
|
|
|
2063 |
* The posting is the entry in the invertedIndex for the matching
|
|
|
2064 |
* term from above.
|
|
|
2065 |
*/
|
|
|
2066 |
var field = clause.fields[k],
|
|
|
2067 |
fieldPosting = posting[field],
|
|
|
2068 |
matchingDocumentRefs = Object.keys(fieldPosting),
|
|
|
2069 |
termField = expandedTerm + "/" + field,
|
|
|
2070 |
matchingDocumentsSet = new lunr.Set(matchingDocumentRefs)
|
|
|
2071 |
|
|
|
2072 |
/*
|
|
|
2073 |
* if the presence of this term is required ensure that the matching
|
|
|
2074 |
* documents are added to the set of required matches for this clause.
|
|
|
2075 |
*
|
|
|
2076 |
*/
|
|
|
2077 |
if (clause.presence == lunr.Query.presence.REQUIRED) {
|
|
|
2078 |
clauseMatches = clauseMatches.union(matchingDocumentsSet)
|
|
|
2079 |
|
|
|
2080 |
if (requiredMatches[field] === undefined) {
|
|
|
2081 |
requiredMatches[field] = lunr.Set.complete
|
|
|
2082 |
}
|
|
|
2083 |
}
|
|
|
2084 |
|
|
|
2085 |
/*
|
|
|
2086 |
* if the presence of this term is prohibited ensure that the matching
|
|
|
2087 |
* documents are added to the set of prohibited matches for this field,
|
|
|
2088 |
* creating that set if it does not yet exist.
|
|
|
2089 |
*/
|
|
|
2090 |
if (clause.presence == lunr.Query.presence.PROHIBITED) {
|
|
|
2091 |
if (prohibitedMatches[field] === undefined) {
|
|
|
2092 |
prohibitedMatches[field] = lunr.Set.empty
|
|
|
2093 |
}
|
|
|
2094 |
|
|
|
2095 |
prohibitedMatches[field] = prohibitedMatches[field].union(matchingDocumentsSet)
|
|
|
2096 |
|
|
|
2097 |
/*
|
|
|
2098 |
* Prohibited matches should not be part of the query vector used for
|
|
|
2099 |
* similarity scoring and no metadata should be extracted so we continue
|
|
|
2100 |
* to the next field
|
|
|
2101 |
*/
|
|
|
2102 |
continue
|
|
|
2103 |
}
|
|
|
2104 |
|
|
|
2105 |
/*
|
|
|
2106 |
* The query field vector is populated using the termIndex found for
|
|
|
2107 |
* the term and a unit value with the appropriate boost applied.
|
|
|
2108 |
* Using upsert because there could already be an entry in the vector
|
|
|
2109 |
* for the term we are working with. In that case we just add the scores
|
|
|
2110 |
* together.
|
|
|
2111 |
*/
|
|
|
2112 |
queryVectors[field].upsert(termIndex, clause.boost, function (a, b) { return a + b })
|
|
|
2113 |
|
|
|
2114 |
/**
|
|
|
2115 |
* If we've already seen this term, field combo then we've already collected
|
|
|
2116 |
* the matching documents and metadata, no need to go through all that again
|
|
|
2117 |
*/
|
|
|
2118 |
if (termFieldCache[termField]) {
|
|
|
2119 |
continue
|
|
|
2120 |
}
|
|
|
2121 |
|
|
|
2122 |
for (var l = 0; l < matchingDocumentRefs.length; l++) {
|
|
|
2123 |
/*
|
|
|
2124 |
* All metadata for this term/field/document triple
|
|
|
2125 |
* are then extracted and collected into an instance
|
|
|
2126 |
* of lunr.MatchData ready to be returned in the query
|
|
|
2127 |
* results
|
|
|
2128 |
*/
|
|
|
2129 |
var matchingDocumentRef = matchingDocumentRefs[l],
|
|
|
2130 |
matchingFieldRef = new lunr.FieldRef (matchingDocumentRef, field),
|
|
|
2131 |
metadata = fieldPosting[matchingDocumentRef],
|
|
|
2132 |
fieldMatch
|
|
|
2133 |
|
|
|
2134 |
if ((fieldMatch = matchingFields[matchingFieldRef]) === undefined) {
|
|
|
2135 |
matchingFields[matchingFieldRef] = new lunr.MatchData (expandedTerm, field, metadata)
|
|
|
2136 |
} else {
|
|
|
2137 |
fieldMatch.add(expandedTerm, field, metadata)
|
|
|
2138 |
}
|
|
|
2139 |
|
|
|
2140 |
}
|
|
|
2141 |
|
|
|
2142 |
termFieldCache[termField] = true
|
|
|
2143 |
}
|
|
|
2144 |
}
|
|
|
2145 |
}
|
|
|
2146 |
|
|
|
2147 |
/**
|
|
|
2148 |
* If the presence was required we need to update the requiredMatches field sets.
|
|
|
2149 |
* We do this after all fields for the term have collected their matches because
|
|
|
2150 |
* the clause terms presence is required in _any_ of the fields not _all_ of the
|
|
|
2151 |
* fields.
|
|
|
2152 |
*/
|
|
|
2153 |
if (clause.presence === lunr.Query.presence.REQUIRED) {
|
|
|
2154 |
for (var k = 0; k < clause.fields.length; k++) {
|
|
|
2155 |
var field = clause.fields[k]
|
|
|
2156 |
requiredMatches[field] = requiredMatches[field].intersect(clauseMatches)
|
|
|
2157 |
}
|
|
|
2158 |
}
|
|
|
2159 |
}
|
|
|
2160 |
|
|
|
2161 |
/**
|
|
|
2162 |
* Need to combine the field scoped required and prohibited
|
|
|
2163 |
* matching documents into a global set of required and prohibited
|
|
|
2164 |
* matches
|
|
|
2165 |
*/
|
|
|
2166 |
var allRequiredMatches = lunr.Set.complete,
|
|
|
2167 |
allProhibitedMatches = lunr.Set.empty
|
|
|
2168 |
|
|
|
2169 |
for (var i = 0; i < this.fields.length; i++) {
|
|
|
2170 |
var field = this.fields[i]
|
|
|
2171 |
|
|
|
2172 |
if (requiredMatches[field]) {
|
|
|
2173 |
allRequiredMatches = allRequiredMatches.intersect(requiredMatches[field])
|
|
|
2174 |
}
|
|
|
2175 |
|
|
|
2176 |
if (prohibitedMatches[field]) {
|
|
|
2177 |
allProhibitedMatches = allProhibitedMatches.union(prohibitedMatches[field])
|
|
|
2178 |
}
|
|
|
2179 |
}
|
|
|
2180 |
|
|
|
2181 |
var matchingFieldRefs = Object.keys(matchingFields),
|
|
|
2182 |
results = [],
|
|
|
2183 |
matches = Object.create(null)
|
|
|
2184 |
|
|
|
2185 |
/*
|
|
|
2186 |
* If the query is negated (contains only prohibited terms)
|
|
|
2187 |
* we need to get _all_ fieldRefs currently existing in the
|
|
|
2188 |
* index. This is only done when we know that the query is
|
|
|
2189 |
* entirely prohibited terms to avoid any cost of getting all
|
|
|
2190 |
* fieldRefs unnecessarily.
|
|
|
2191 |
*
|
|
|
2192 |
* Additionally, blank MatchData must be created to correctly
|
|
|
2193 |
* populate the results.
|
|
|
2194 |
*/
|
|
|
2195 |
if (query.isNegated()) {
|
|
|
2196 |
matchingFieldRefs = Object.keys(this.fieldVectors)
|
|
|
2197 |
|
|
|
2198 |
for (var i = 0; i < matchingFieldRefs.length; i++) {
|
|
|
2199 |
var matchingFieldRef = matchingFieldRefs[i]
|
|
|
2200 |
var fieldRef = lunr.FieldRef.fromString(matchingFieldRef)
|
|
|
2201 |
matchingFields[matchingFieldRef] = new lunr.MatchData
|
|
|
2202 |
}
|
|
|
2203 |
}
|
|
|
2204 |
|
|
|
2205 |
for (var i = 0; i < matchingFieldRefs.length; i++) {
|
|
|
2206 |
/*
|
|
|
2207 |
* Currently we have document fields that match the query, but we
|
|
|
2208 |
* need to return documents. The matchData and scores are combined
|
|
|
2209 |
* from multiple fields belonging to the same document.
|
|
|
2210 |
*
|
|
|
2211 |
* Scores are calculated by field, using the query vectors created
|
|
|
2212 |
* above, and combined into a final document score using addition.
|
|
|
2213 |
*/
|
|
|
2214 |
var fieldRef = lunr.FieldRef.fromString(matchingFieldRefs[i]),
|
|
|
2215 |
docRef = fieldRef.docRef
|
|
|
2216 |
|
|
|
2217 |
if (!allRequiredMatches.contains(docRef)) {
|
|
|
2218 |
continue
|
|
|
2219 |
}
|
|
|
2220 |
|
|
|
2221 |
if (allProhibitedMatches.contains(docRef)) {
|
|
|
2222 |
continue
|
|
|
2223 |
}
|
|
|
2224 |
|
|
|
2225 |
var fieldVector = this.fieldVectors[fieldRef],
|
|
|
2226 |
score = queryVectors[fieldRef.fieldName].similarity(fieldVector),
|
|
|
2227 |
docMatch
|
|
|
2228 |
|
|
|
2229 |
if ((docMatch = matches[docRef]) !== undefined) {
|
|
|
2230 |
docMatch.score += score
|
|
|
2231 |
docMatch.matchData.combine(matchingFields[fieldRef])
|
|
|
2232 |
} else {
|
|
|
2233 |
var match = {
|
|
|
2234 |
ref: docRef,
|
|
|
2235 |
score: score,
|
|
|
2236 |
matchData: matchingFields[fieldRef]
|
|
|
2237 |
}
|
|
|
2238 |
matches[docRef] = match
|
|
|
2239 |
results.push(match)
|
|
|
2240 |
}
|
|
|
2241 |
}
|
|
|
2242 |
|
|
|
2243 |
/*
|
|
|
2244 |
* Sort the results objects by score, highest first.
|
|
|
2245 |
*/
|
|
|
2246 |
return results.sort(function (a, b) {
|
|
|
2247 |
return b.score - a.score
|
|
|
2248 |
})
|
|
|
2249 |
}
|
|
|
2250 |
|
|
|
2251 |
/**
|
|
|
2252 |
* Prepares the index for JSON serialization.
|
|
|
2253 |
*
|
|
|
2254 |
* The schema for this JSON blob will be described in a
|
|
|
2255 |
* separate JSON schema file.
|
|
|
2256 |
*
|
|
|
2257 |
* @returns {Object}
|
|
|
2258 |
*/
|
|
|
2259 |
lunr.Index.prototype.toJSON = function () {
|
|
|
2260 |
var invertedIndex = Object.keys(this.invertedIndex)
|
|
|
2261 |
.sort()
|
|
|
2262 |
.map(function (term) {
|
|
|
2263 |
return [term, this.invertedIndex[term]]
|
|
|
2264 |
}, this)
|
|
|
2265 |
|
|
|
2266 |
var fieldVectors = Object.keys(this.fieldVectors)
|
|
|
2267 |
.map(function (ref) {
|
|
|
2268 |
return [ref, this.fieldVectors[ref].toJSON()]
|
|
|
2269 |
}, this)
|
|
|
2270 |
|
|
|
2271 |
return {
|
|
|
2272 |
version: lunr.version,
|
|
|
2273 |
fields: this.fields,
|
|
|
2274 |
fieldVectors: fieldVectors,
|
|
|
2275 |
invertedIndex: invertedIndex,
|
|
|
2276 |
pipeline: this.pipeline.toJSON()
|
|
|
2277 |
}
|
|
|
2278 |
}
|
|
|
2279 |
|
|
|
2280 |
/**
|
|
|
2281 |
* Loads a previously serialized lunr.Index
|
|
|
2282 |
*
|
|
|
2283 |
* @param {Object} serializedIndex - A previously serialized lunr.Index
|
|
|
2284 |
* @returns {lunr.Index}
|
|
|
2285 |
*/
|
|
|
2286 |
lunr.Index.load = function (serializedIndex) {
|
|
|
2287 |
var attrs = {},
|
|
|
2288 |
fieldVectors = {},
|
|
|
2289 |
serializedVectors = serializedIndex.fieldVectors,
|
|
|
2290 |
invertedIndex = Object.create(null),
|
|
|
2291 |
serializedInvertedIndex = serializedIndex.invertedIndex,
|
|
|
2292 |
tokenSetBuilder = new lunr.TokenSet.Builder,
|
|
|
2293 |
pipeline = lunr.Pipeline.load(serializedIndex.pipeline)
|
|
|
2294 |
|
|
|
2295 |
if (serializedIndex.version != lunr.version) {
|
|
|
2296 |
lunr.utils.warn("Version mismatch when loading serialised index. Current version of lunr '" + lunr.version + "' does not match serialized index '" + serializedIndex.version + "'")
|
|
|
2297 |
}
|
|
|
2298 |
|
|
|
2299 |
for (var i = 0; i < serializedVectors.length; i++) {
|
|
|
2300 |
var tuple = serializedVectors[i],
|
|
|
2301 |
ref = tuple[0],
|
|
|
2302 |
elements = tuple[1]
|
|
|
2303 |
|
|
|
2304 |
fieldVectors[ref] = new lunr.Vector(elements)
|
|
|
2305 |
}
|
|
|
2306 |
|
|
|
2307 |
for (var i = 0; i < serializedInvertedIndex.length; i++) {
|
|
|
2308 |
var tuple = serializedInvertedIndex[i],
|
|
|
2309 |
term = tuple[0],
|
|
|
2310 |
posting = tuple[1]
|
|
|
2311 |
|
|
|
2312 |
tokenSetBuilder.insert(term)
|
|
|
2313 |
invertedIndex[term] = posting
|
|
|
2314 |
}
|
|
|
2315 |
|
|
|
2316 |
tokenSetBuilder.finish()
|
|
|
2317 |
|
|
|
2318 |
attrs.fields = serializedIndex.fields
|
|
|
2319 |
|
|
|
2320 |
attrs.fieldVectors = fieldVectors
|
|
|
2321 |
attrs.invertedIndex = invertedIndex
|
|
|
2322 |
attrs.tokenSet = tokenSetBuilder.root
|
|
|
2323 |
attrs.pipeline = pipeline
|
|
|
2324 |
|
|
|
2325 |
return new lunr.Index(attrs)
|
|
|
2326 |
}
|
|
|
2327 |
/*!
|
|
|
2328 |
* lunr.Builder
|
|
|
2329 |
* Copyright (C) 2020 Oliver Nightingale
|
|
|
2330 |
*/
|
|
|
2331 |
|
|
|
2332 |
/**
|
|
|
2333 |
* lunr.Builder performs indexing on a set of documents and
|
|
|
2334 |
* returns instances of lunr.Index ready for querying.
|
|
|
2335 |
*
|
|
|
2336 |
* All configuration of the index is done via the builder, the
|
|
|
2337 |
* fields to index, the document reference, the text processing
|
|
|
2338 |
* pipeline and document scoring parameters are all set on the
|
|
|
2339 |
* builder before indexing.
|
|
|
2340 |
*
|
|
|
2341 |
* @constructor
|
|
|
2342 |
* @property {string} _ref - Internal reference to the document reference field.
|
|
|
2343 |
* @property {string[]} _fields - Internal reference to the document fields to index.
|
|
|
2344 |
* @property {object} invertedIndex - The inverted index maps terms to document fields.
|
|
|
2345 |
* @property {object} documentTermFrequencies - Keeps track of document term frequencies.
|
|
|
2346 |
* @property {object} documentLengths - Keeps track of the length of documents added to the index.
|
|
|
2347 |
* @property {lunr.tokenizer} tokenizer - Function for splitting strings into tokens for indexing.
|
|
|
2348 |
* @property {lunr.Pipeline} pipeline - The pipeline performs text processing on tokens before indexing.
|
|
|
2349 |
* @property {lunr.Pipeline} searchPipeline - A pipeline for processing search terms before querying the index.
|
|
|
2350 |
* @property {number} documentCount - Keeps track of the total number of documents indexed.
|
|
|
2351 |
* @property {number} _b - A parameter to control field length normalization, setting this to 0 disabled normalization, 1 fully normalizes field lengths, the default value is 0.75.
|
|
|
2352 |
* @property {number} _k1 - A parameter to control how quickly an increase in term frequency results in term frequency saturation, the default value is 1.2.
|
|
|
2353 |
* @property {number} termIndex - A counter incremented for each unique term, used to identify a terms position in the vector space.
|
|
|
2354 |
* @property {array} metadataWhitelist - A list of metadata keys that have been whitelisted for entry in the index.
|
|
|
2355 |
*/
|
|
|
2356 |
lunr.Builder = function () {
|
|
|
2357 |
this._ref = "id"
|
|
|
2358 |
this._fields = Object.create(null)
|
|
|
2359 |
this._documents = Object.create(null)
|
|
|
2360 |
this.invertedIndex = Object.create(null)
|
|
|
2361 |
this.fieldTermFrequencies = {}
|
|
|
2362 |
this.fieldLengths = {}
|
|
|
2363 |
this.tokenizer = lunr.tokenizer
|
|
|
2364 |
this.pipeline = new lunr.Pipeline
|
|
|
2365 |
this.searchPipeline = new lunr.Pipeline
|
|
|
2366 |
this.documentCount = 0
|
|
|
2367 |
this._b = 0.75
|
|
|
2368 |
this._k1 = 1.2
|
|
|
2369 |
this.termIndex = 0
|
|
|
2370 |
this.metadataWhitelist = []
|
|
|
2371 |
}
|
|
|
2372 |
|
|
|
2373 |
/**
|
|
|
2374 |
* Sets the document field used as the document reference. Every document must have this field.
|
|
|
2375 |
* The type of this field in the document should be a string, if it is not a string it will be
|
|
|
2376 |
* coerced into a string by calling toString.
|
|
|
2377 |
*
|
|
|
2378 |
* The default ref is 'id'.
|
|
|
2379 |
*
|
|
|
2380 |
* The ref should _not_ be changed during indexing, it should be set before any documents are
|
|
|
2381 |
* added to the index. Changing it during indexing can lead to inconsistent results.
|
|
|
2382 |
*
|
|
|
2383 |
* @param {string} ref - The name of the reference field in the document.
|
|
|
2384 |
*/
|
|
|
2385 |
lunr.Builder.prototype.ref = function (ref) {
|
|
|
2386 |
this._ref = ref
|
|
|
2387 |
}
|
|
|
2388 |
|
|
|
2389 |
/**
|
|
|
2390 |
* A function that is used to extract a field from a document.
|
|
|
2391 |
*
|
|
|
2392 |
* Lunr expects a field to be at the top level of a document, if however the field
|
|
|
2393 |
* is deeply nested within a document an extractor function can be used to extract
|
|
|
2394 |
* the right field for indexing.
|
|
|
2395 |
*
|
|
|
2396 |
* @callback fieldExtractor
|
|
|
2397 |
* @param {object} doc - The document being added to the index.
|
|
|
2398 |
* @returns {?(string|object|object[])} obj - The object that will be indexed for this field.
|
|
|
2399 |
* @example <caption>Extracting a nested field</caption>
|
|
|
2400 |
* function (doc) { return doc.nested.field }
|
|
|
2401 |
*/
|
|
|
2402 |
|
|
|
2403 |
/**
|
|
|
2404 |
* Adds a field to the list of document fields that will be indexed. Every document being
|
|
|
2405 |
* indexed should have this field. Null values for this field in indexed documents will
|
|
|
2406 |
* not cause errors but will limit the chance of that document being retrieved by searches.
|
|
|
2407 |
*
|
|
|
2408 |
* All fields should be added before adding documents to the index. Adding fields after
|
|
|
2409 |
* a document has been indexed will have no effect on already indexed documents.
|
|
|
2410 |
*
|
|
|
2411 |
* Fields can be boosted at build time. This allows terms within that field to have more
|
|
|
2412 |
* importance when ranking search results. Use a field boost to specify that matches within
|
|
|
2413 |
* one field are more important than other fields.
|
|
|
2414 |
*
|
|
|
2415 |
* @param {string} fieldName - The name of a field to index in all documents.
|
|
|
2416 |
* @param {object} attributes - Optional attributes associated with this field.
|
|
|
2417 |
* @param {number} [attributes.boost=1] - Boost applied to all terms within this field.
|
|
|
2418 |
* @param {fieldExtractor} [attributes.extractor] - Function to extract a field from a document.
|
|
|
2419 |
* @throws {RangeError} fieldName cannot contain unsupported characters '/'
|
|
|
2420 |
*/
|
|
|
2421 |
lunr.Builder.prototype.field = function (fieldName, attributes) {
|
|
|
2422 |
if (/\//.test(fieldName)) {
|
|
|
2423 |
throw new RangeError ("Field '" + fieldName + "' contains illegal character '/'")
|
|
|
2424 |
}
|
|
|
2425 |
|
|
|
2426 |
this._fields[fieldName] = attributes || {}
|
|
|
2427 |
}
|
|
|
2428 |
|
|
|
2429 |
/**
|
|
|
2430 |
* A parameter to tune the amount of field length normalisation that is applied when
|
|
|
2431 |
* calculating relevance scores. A value of 0 will completely disable any normalisation
|
|
|
2432 |
* and a value of 1 will fully normalise field lengths. The default is 0.75. Values of b
|
|
|
2433 |
* will be clamped to the range 0 - 1.
|
|
|
2434 |
*
|
|
|
2435 |
* @param {number} number - The value to set for this tuning parameter.
|
|
|
2436 |
*/
|
|
|
2437 |
lunr.Builder.prototype.b = function (number) {
|
|
|
2438 |
if (number < 0) {
|
|
|
2439 |
this._b = 0
|
|
|
2440 |
} else if (number > 1) {
|
|
|
2441 |
this._b = 1
|
|
|
2442 |
} else {
|
|
|
2443 |
this._b = number
|
|
|
2444 |
}
|
|
|
2445 |
}
|
|
|
2446 |
|
|
|
2447 |
/**
|
|
|
2448 |
* A parameter that controls the speed at which a rise in term frequency results in term
|
|
|
2449 |
* frequency saturation. The default value is 1.2. Setting this to a higher value will give
|
|
|
2450 |
* slower saturation levels, a lower value will result in quicker saturation.
|
|
|
2451 |
*
|
|
|
2452 |
* @param {number} number - The value to set for this tuning parameter.
|
|
|
2453 |
*/
|
|
|
2454 |
lunr.Builder.prototype.k1 = function (number) {
|
|
|
2455 |
this._k1 = number
|
|
|
2456 |
}
|
|
|
2457 |
|
|
|
2458 |
/**
|
|
|
2459 |
* Adds a document to the index.
|
|
|
2460 |
*
|
|
|
2461 |
* Before adding fields to the index the index should have been fully setup, with the document
|
|
|
2462 |
* ref and all fields to index already having been specified.
|
|
|
2463 |
*
|
|
|
2464 |
* The document must have a field name as specified by the ref (by default this is 'id') and
|
|
|
2465 |
* it should have all fields defined for indexing, though null or undefined values will not
|
|
|
2466 |
* cause errors.
|
|
|
2467 |
*
|
|
|
2468 |
* Entire documents can be boosted at build time. Applying a boost to a document indicates that
|
|
|
2469 |
* this document should rank higher in search results than other documents.
|
|
|
2470 |
*
|
|
|
2471 |
* @param {object} doc - The document to add to the index.
|
|
|
2472 |
* @param {object} attributes - Optional attributes associated with this document.
|
|
|
2473 |
* @param {number} [attributes.boost=1] - Boost applied to all terms within this document.
|
|
|
2474 |
*/
|
|
|
2475 |
lunr.Builder.prototype.add = function (doc, attributes) {
|
|
|
2476 |
var docRef = doc[this._ref],
|
|
|
2477 |
fields = Object.keys(this._fields)
|
|
|
2478 |
|
|
|
2479 |
this._documents[docRef] = attributes || {}
|
|
|
2480 |
this.documentCount += 1
|
|
|
2481 |
|
|
|
2482 |
for (var i = 0; i < fields.length; i++) {
|
|
|
2483 |
var fieldName = fields[i],
|
|
|
2484 |
extractor = this._fields[fieldName].extractor,
|
|
|
2485 |
field = extractor ? extractor(doc) : doc[fieldName],
|
|
|
2486 |
tokens = this.tokenizer(field, {
|
|
|
2487 |
fields: [fieldName]
|
|
|
2488 |
}),
|
|
|
2489 |
terms = this.pipeline.run(tokens),
|
|
|
2490 |
fieldRef = new lunr.FieldRef (docRef, fieldName),
|
|
|
2491 |
fieldTerms = Object.create(null)
|
|
|
2492 |
|
|
|
2493 |
this.fieldTermFrequencies[fieldRef] = fieldTerms
|
|
|
2494 |
this.fieldLengths[fieldRef] = 0
|
|
|
2495 |
|
|
|
2496 |
// store the length of this field for this document
|
|
|
2497 |
this.fieldLengths[fieldRef] += terms.length
|
|
|
2498 |
|
|
|
2499 |
// calculate term frequencies for this field
|
|
|
2500 |
for (var j = 0; j < terms.length; j++) {
|
|
|
2501 |
var term = terms[j]
|
|
|
2502 |
|
|
|
2503 |
if (fieldTerms[term] == undefined) {
|
|
|
2504 |
fieldTerms[term] = 0
|
|
|
2505 |
}
|
|
|
2506 |
|
|
|
2507 |
fieldTerms[term] += 1
|
|
|
2508 |
|
|
|
2509 |
// add to inverted index
|
|
|
2510 |
// create an initial posting if one doesn't exist
|
|
|
2511 |
if (this.invertedIndex[term] == undefined) {
|
|
|
2512 |
var posting = Object.create(null)
|
|
|
2513 |
posting["_index"] = this.termIndex
|
|
|
2514 |
this.termIndex += 1
|
|
|
2515 |
|
|
|
2516 |
for (var k = 0; k < fields.length; k++) {
|
|
|
2517 |
posting[fields[k]] = Object.create(null)
|
|
|
2518 |
}
|
|
|
2519 |
|
|
|
2520 |
this.invertedIndex[term] = posting
|
|
|
2521 |
}
|
|
|
2522 |
|
|
|
2523 |
// add an entry for this term/fieldName/docRef to the invertedIndex
|
|
|
2524 |
if (this.invertedIndex[term][fieldName][docRef] == undefined) {
|
|
|
2525 |
this.invertedIndex[term][fieldName][docRef] = Object.create(null)
|
|
|
2526 |
}
|
|
|
2527 |
|
|
|
2528 |
// store all whitelisted metadata about this token in the
|
|
|
2529 |
// inverted index
|
|
|
2530 |
for (var l = 0; l < this.metadataWhitelist.length; l++) {
|
|
|
2531 |
var metadataKey = this.metadataWhitelist[l],
|
|
|
2532 |
metadata = term.metadata[metadataKey]
|
|
|
2533 |
|
|
|
2534 |
if (this.invertedIndex[term][fieldName][docRef][metadataKey] == undefined) {
|
|
|
2535 |
this.invertedIndex[term][fieldName][docRef][metadataKey] = []
|
|
|
2536 |
}
|
|
|
2537 |
|
|
|
2538 |
this.invertedIndex[term][fieldName][docRef][metadataKey].push(metadata)
|
|
|
2539 |
}
|
|
|
2540 |
}
|
|
|
2541 |
|
|
|
2542 |
}
|
|
|
2543 |
}
|
|
|
2544 |
|
|
|
2545 |
/**
|
|
|
2546 |
* Calculates the average document length for this index
|
|
|
2547 |
*
|
|
|
2548 |
* @private
|
|
|
2549 |
*/
|
|
|
2550 |
lunr.Builder.prototype.calculateAverageFieldLengths = function () {
|
|
|
2551 |
|
|
|
2552 |
var fieldRefs = Object.keys(this.fieldLengths),
|
|
|
2553 |
numberOfFields = fieldRefs.length,
|
|
|
2554 |
accumulator = {},
|
|
|
2555 |
documentsWithField = {}
|
|
|
2556 |
|
|
|
2557 |
for (var i = 0; i < numberOfFields; i++) {
|
|
|
2558 |
var fieldRef = lunr.FieldRef.fromString(fieldRefs[i]),
|
|
|
2559 |
field = fieldRef.fieldName
|
|
|
2560 |
|
|
|
2561 |
documentsWithField[field] || (documentsWithField[field] = 0)
|
|
|
2562 |
documentsWithField[field] += 1
|
|
|
2563 |
|
|
|
2564 |
accumulator[field] || (accumulator[field] = 0)
|
|
|
2565 |
accumulator[field] += this.fieldLengths[fieldRef]
|
|
|
2566 |
}
|
|
|
2567 |
|
|
|
2568 |
var fields = Object.keys(this._fields)
|
|
|
2569 |
|
|
|
2570 |
for (var i = 0; i < fields.length; i++) {
|
|
|
2571 |
var fieldName = fields[i]
|
|
|
2572 |
accumulator[fieldName] = accumulator[fieldName] / documentsWithField[fieldName]
|
|
|
2573 |
}
|
|
|
2574 |
|
|
|
2575 |
this.averageFieldLength = accumulator
|
|
|
2576 |
}
|
|
|
2577 |
|
|
|
2578 |
/**
|
|
|
2579 |
* Builds a vector space model of every document using lunr.Vector
|
|
|
2580 |
*
|
|
|
2581 |
* @private
|
|
|
2582 |
*/
|
|
|
2583 |
lunr.Builder.prototype.createFieldVectors = function () {
|
|
|
2584 |
var fieldVectors = {},
|
|
|
2585 |
fieldRefs = Object.keys(this.fieldTermFrequencies),
|
|
|
2586 |
fieldRefsLength = fieldRefs.length,
|
|
|
2587 |
termIdfCache = Object.create(null)
|
|
|
2588 |
|
|
|
2589 |
for (var i = 0; i < fieldRefsLength; i++) {
|
|
|
2590 |
var fieldRef = lunr.FieldRef.fromString(fieldRefs[i]),
|
|
|
2591 |
fieldName = fieldRef.fieldName,
|
|
|
2592 |
fieldLength = this.fieldLengths[fieldRef],
|
|
|
2593 |
fieldVector = new lunr.Vector,
|
|
|
2594 |
termFrequencies = this.fieldTermFrequencies[fieldRef],
|
|
|
2595 |
terms = Object.keys(termFrequencies),
|
|
|
2596 |
termsLength = terms.length
|
|
|
2597 |
|
|
|
2598 |
|
|
|
2599 |
var fieldBoost = this._fields[fieldName].boost || 1,
|
|
|
2600 |
docBoost = this._documents[fieldRef.docRef].boost || 1
|
|
|
2601 |
|
|
|
2602 |
for (var j = 0; j < termsLength; j++) {
|
|
|
2603 |
var term = terms[j],
|
|
|
2604 |
tf = termFrequencies[term],
|
|
|
2605 |
termIndex = this.invertedIndex[term]._index,
|
|
|
2606 |
idf, score, scoreWithPrecision
|
|
|
2607 |
|
|
|
2608 |
if (termIdfCache[term] === undefined) {
|
|
|
2609 |
idf = lunr.idf(this.invertedIndex[term], this.documentCount)
|
|
|
2610 |
termIdfCache[term] = idf
|
|
|
2611 |
} else {
|
|
|
2612 |
idf = termIdfCache[term]
|
|
|
2613 |
}
|
|
|
2614 |
|
|
|
2615 |
score = idf * ((this._k1 + 1) * tf) / (this._k1 * (1 - this._b + this._b * (fieldLength / this.averageFieldLength[fieldName])) + tf)
|
|
|
2616 |
score *= fieldBoost
|
|
|
2617 |
score *= docBoost
|
|
|
2618 |
scoreWithPrecision = Math.round(score * 1000) / 1000
|
|
|
2619 |
// Converts 1.23456789 to 1.234.
|
|
|
2620 |
// Reducing the precision so that the vectors take up less
|
|
|
2621 |
// space when serialised. Doing it now so that they behave
|
|
|
2622 |
// the same before and after serialisation. Also, this is
|
|
|
2623 |
// the fastest approach to reducing a number's precision in
|
|
|
2624 |
// JavaScript.
|
|
|
2625 |
|
|
|
2626 |
fieldVector.insert(termIndex, scoreWithPrecision)
|
|
|
2627 |
}
|
|
|
2628 |
|
|
|
2629 |
fieldVectors[fieldRef] = fieldVector
|
|
|
2630 |
}
|
|
|
2631 |
|
|
|
2632 |
this.fieldVectors = fieldVectors
|
|
|
2633 |
}
|
|
|
2634 |
|
|
|
2635 |
/**
|
|
|
2636 |
* Creates a token set of all tokens in the index using lunr.TokenSet
|
|
|
2637 |
*
|
|
|
2638 |
* @private
|
|
|
2639 |
*/
|
|
|
2640 |
lunr.Builder.prototype.createTokenSet = function () {
|
|
|
2641 |
this.tokenSet = lunr.TokenSet.fromArray(
|
|
|
2642 |
Object.keys(this.invertedIndex).sort()
|
|
|
2643 |
)
|
|
|
2644 |
}
|
|
|
2645 |
|
|
|
2646 |
/**
|
|
|
2647 |
* Builds the index, creating an instance of lunr.Index.
|
|
|
2648 |
*
|
|
|
2649 |
* This completes the indexing process and should only be called
|
|
|
2650 |
* once all documents have been added to the index.
|
|
|
2651 |
*
|
|
|
2652 |
* @returns {lunr.Index}
|
|
|
2653 |
*/
|
|
|
2654 |
lunr.Builder.prototype.build = function () {
|
|
|
2655 |
this.calculateAverageFieldLengths()
|
|
|
2656 |
this.createFieldVectors()
|
|
|
2657 |
this.createTokenSet()
|
|
|
2658 |
|
|
|
2659 |
return new lunr.Index({
|
|
|
2660 |
invertedIndex: this.invertedIndex,
|
|
|
2661 |
fieldVectors: this.fieldVectors,
|
|
|
2662 |
tokenSet: this.tokenSet,
|
|
|
2663 |
fields: Object.keys(this._fields),
|
|
|
2664 |
pipeline: this.searchPipeline
|
|
|
2665 |
})
|
|
|
2666 |
}
|
|
|
2667 |
|
|
|
2668 |
/**
|
|
|
2669 |
* Applies a plugin to the index builder.
|
|
|
2670 |
*
|
|
|
2671 |
* A plugin is a function that is called with the index builder as its context.
|
|
|
2672 |
* Plugins can be used to customise or extend the behaviour of the index
|
|
|
2673 |
* in some way. A plugin is just a function, that encapsulated the custom
|
|
|
2674 |
* behaviour that should be applied when building the index.
|
|
|
2675 |
*
|
|
|
2676 |
* The plugin function will be called with the index builder as its argument, additional
|
|
|
2677 |
* arguments can also be passed when calling use. The function will be called
|
|
|
2678 |
* with the index builder as its context.
|
|
|
2679 |
*
|
|
|
2680 |
* @param {Function} plugin The plugin to apply.
|
|
|
2681 |
*/
|
|
|
2682 |
lunr.Builder.prototype.use = function (fn) {
|
|
|
2683 |
var args = Array.prototype.slice.call(arguments, 1)
|
|
|
2684 |
args.unshift(this)
|
|
|
2685 |
fn.apply(this, args)
|
|
|
2686 |
}
|
|
|
2687 |
/**
|
|
|
2688 |
* Contains and collects metadata about a matching document.
|
|
|
2689 |
* A single instance of lunr.MatchData is returned as part of every
|
|
|
2690 |
* lunr.Index~Result.
|
|
|
2691 |
*
|
|
|
2692 |
* @constructor
|
|
|
2693 |
* @param {string} term - The term this match data is associated with
|
|
|
2694 |
* @param {string} field - The field in which the term was found
|
|
|
2695 |
* @param {object} metadata - The metadata recorded about this term in this field
|
|
|
2696 |
* @property {object} metadata - A cloned collection of metadata associated with this document.
|
|
|
2697 |
* @see {@link lunr.Index~Result}
|
|
|
2698 |
*/
|
|
|
2699 |
lunr.MatchData = function (term, field, metadata) {
|
|
|
2700 |
var clonedMetadata = Object.create(null),
|
|
|
2701 |
metadataKeys = Object.keys(metadata || {})
|
|
|
2702 |
|
|
|
2703 |
// Cloning the metadata to prevent the original
|
|
|
2704 |
// being mutated during match data combination.
|
|
|
2705 |
// Metadata is kept in an array within the inverted
|
|
|
2706 |
// index so cloning the data can be done with
|
|
|
2707 |
// Array#slice
|
|
|
2708 |
for (var i = 0; i < metadataKeys.length; i++) {
|
|
|
2709 |
var key = metadataKeys[i]
|
|
|
2710 |
clonedMetadata[key] = metadata[key].slice()
|
|
|
2711 |
}
|
|
|
2712 |
|
|
|
2713 |
this.metadata = Object.create(null)
|
|
|
2714 |
|
|
|
2715 |
if (term !== undefined) {
|
|
|
2716 |
this.metadata[term] = Object.create(null)
|
|
|
2717 |
this.metadata[term][field] = clonedMetadata
|
|
|
2718 |
}
|
|
|
2719 |
}
|
|
|
2720 |
|
|
|
2721 |
/**
|
|
|
2722 |
* An instance of lunr.MatchData will be created for every term that matches a
|
|
|
2723 |
* document. However only one instance is required in a lunr.Index~Result. This
|
|
|
2724 |
* method combines metadata from another instance of lunr.MatchData with this
|
|
|
2725 |
* objects metadata.
|
|
|
2726 |
*
|
|
|
2727 |
* @param {lunr.MatchData} otherMatchData - Another instance of match data to merge with this one.
|
|
|
2728 |
* @see {@link lunr.Index~Result}
|
|
|
2729 |
*/
|
|
|
2730 |
lunr.MatchData.prototype.combine = function (otherMatchData) {
|
|
|
2731 |
var terms = Object.keys(otherMatchData.metadata)
|
|
|
2732 |
|
|
|
2733 |
for (var i = 0; i < terms.length; i++) {
|
|
|
2734 |
var term = terms[i],
|
|
|
2735 |
fields = Object.keys(otherMatchData.metadata[term])
|
|
|
2736 |
|
|
|
2737 |
if (this.metadata[term] == undefined) {
|
|
|
2738 |
this.metadata[term] = Object.create(null)
|
|
|
2739 |
}
|
|
|
2740 |
|
|
|
2741 |
for (var j = 0; j < fields.length; j++) {
|
|
|
2742 |
var field = fields[j],
|
|
|
2743 |
keys = Object.keys(otherMatchData.metadata[term][field])
|
|
|
2744 |
|
|
|
2745 |
if (this.metadata[term][field] == undefined) {
|
|
|
2746 |
this.metadata[term][field] = Object.create(null)
|
|
|
2747 |
}
|
|
|
2748 |
|
|
|
2749 |
for (var k = 0; k < keys.length; k++) {
|
|
|
2750 |
var key = keys[k]
|
|
|
2751 |
|
|
|
2752 |
if (this.metadata[term][field][key] == undefined) {
|
|
|
2753 |
this.metadata[term][field][key] = otherMatchData.metadata[term][field][key]
|
|
|
2754 |
} else {
|
|
|
2755 |
this.metadata[term][field][key] = this.metadata[term][field][key].concat(otherMatchData.metadata[term][field][key])
|
|
|
2756 |
}
|
|
|
2757 |
|
|
|
2758 |
}
|
|
|
2759 |
}
|
|
|
2760 |
}
|
|
|
2761 |
}
|
|
|
2762 |
|
|
|
2763 |
/**
|
|
|
2764 |
* Add metadata for a term/field pair to this instance of match data.
|
|
|
2765 |
*
|
|
|
2766 |
* @param {string} term - The term this match data is associated with
|
|
|
2767 |
* @param {string} field - The field in which the term was found
|
|
|
2768 |
* @param {object} metadata - The metadata recorded about this term in this field
|
|
|
2769 |
*/
|
|
|
2770 |
lunr.MatchData.prototype.add = function (term, field, metadata) {
|
|
|
2771 |
if (!(term in this.metadata)) {
|
|
|
2772 |
this.metadata[term] = Object.create(null)
|
|
|
2773 |
this.metadata[term][field] = metadata
|
|
|
2774 |
return
|
|
|
2775 |
}
|
|
|
2776 |
|
|
|
2777 |
if (!(field in this.metadata[term])) {
|
|
|
2778 |
this.metadata[term][field] = metadata
|
|
|
2779 |
return
|
|
|
2780 |
}
|
|
|
2781 |
|
|
|
2782 |
var metadataKeys = Object.keys(metadata)
|
|
|
2783 |
|
|
|
2784 |
for (var i = 0; i < metadataKeys.length; i++) {
|
|
|
2785 |
var key = metadataKeys[i]
|
|
|
2786 |
|
|
|
2787 |
if (key in this.metadata[term][field]) {
|
|
|
2788 |
this.metadata[term][field][key] = this.metadata[term][field][key].concat(metadata[key])
|
|
|
2789 |
} else {
|
|
|
2790 |
this.metadata[term][field][key] = metadata[key]
|
|
|
2791 |
}
|
|
|
2792 |
}
|
|
|
2793 |
}
|
|
|
2794 |
/**
|
|
|
2795 |
* A lunr.Query provides a programmatic way of defining queries to be performed
|
|
|
2796 |
* against a {@link lunr.Index}.
|
|
|
2797 |
*
|
|
|
2798 |
* Prefer constructing a lunr.Query using the {@link lunr.Index#query} method
|
|
|
2799 |
* so the query object is pre-initialized with the right index fields.
|
|
|
2800 |
*
|
|
|
2801 |
* @constructor
|
|
|
2802 |
* @property {lunr.Query~Clause[]} clauses - An array of query clauses.
|
|
|
2803 |
* @property {string[]} allFields - An array of all available fields in a lunr.Index.
|
|
|
2804 |
*/
|
|
|
2805 |
lunr.Query = function (allFields) {
|
|
|
2806 |
this.clauses = []
|
|
|
2807 |
this.allFields = allFields
|
|
|
2808 |
}
|
|
|
2809 |
|
|
|
2810 |
/**
|
|
|
2811 |
* Constants for indicating what kind of automatic wildcard insertion will be used when constructing a query clause.
|
|
|
2812 |
*
|
|
|
2813 |
* This allows wildcards to be added to the beginning and end of a term without having to manually do any string
|
|
|
2814 |
* concatenation.
|
|
|
2815 |
*
|
|
|
2816 |
* The wildcard constants can be bitwise combined to select both leading and trailing wildcards.
|
|
|
2817 |
*
|
|
|
2818 |
* @constant
|
|
|
2819 |
* @default
|
|
|
2820 |
* @property {number} wildcard.NONE - The term will have no wildcards inserted, this is the default behaviour
|
|
|
2821 |
* @property {number} wildcard.LEADING - Prepend the term with a wildcard, unless a leading wildcard already exists
|
|
|
2822 |
* @property {number} wildcard.TRAILING - Append a wildcard to the term, unless a trailing wildcard already exists
|
|
|
2823 |
* @see lunr.Query~Clause
|
|
|
2824 |
* @see lunr.Query#clause
|
|
|
2825 |
* @see lunr.Query#term
|
|
|
2826 |
* @example <caption>query term with trailing wildcard</caption>
|
|
|
2827 |
* query.term('foo', { wildcard: lunr.Query.wildcard.TRAILING })
|
|
|
2828 |
* @example <caption>query term with leading and trailing wildcard</caption>
|
|
|
2829 |
* query.term('foo', {
|
|
|
2830 |
* wildcard: lunr.Query.wildcard.LEADING | lunr.Query.wildcard.TRAILING
|
|
|
2831 |
* })
|
|
|
2832 |
*/
|
|
|
2833 |
|
|
|
2834 |
lunr.Query.wildcard = new String ("*")
|
|
|
2835 |
lunr.Query.wildcard.NONE = 0
|
|
|
2836 |
lunr.Query.wildcard.LEADING = 1
|
|
|
2837 |
lunr.Query.wildcard.TRAILING = 2
|
|
|
2838 |
|
|
|
2839 |
/**
|
|
|
2840 |
* Constants for indicating what kind of presence a term must have in matching documents.
|
|
|
2841 |
*
|
|
|
2842 |
* @constant
|
|
|
2843 |
* @enum {number}
|
|
|
2844 |
* @see lunr.Query~Clause
|
|
|
2845 |
* @see lunr.Query#clause
|
|
|
2846 |
* @see lunr.Query#term
|
|
|
2847 |
* @example <caption>query term with required presence</caption>
|
|
|
2848 |
* query.term('foo', { presence: lunr.Query.presence.REQUIRED })
|
|
|
2849 |
*/
|
|
|
2850 |
lunr.Query.presence = {
|
|
|
2851 |
/**
|
|
|
2852 |
* Term's presence in a document is optional, this is the default value.
|
|
|
2853 |
*/
|
|
|
2854 |
OPTIONAL: 1,
|
|
|
2855 |
|
|
|
2856 |
/**
|
|
|
2857 |
* Term's presence in a document is required, documents that do not contain
|
|
|
2858 |
* this term will not be returned.
|
|
|
2859 |
*/
|
|
|
2860 |
REQUIRED: 2,
|
|
|
2861 |
|
|
|
2862 |
/**
|
|
|
2863 |
* Term's presence in a document is prohibited, documents that do contain
|
|
|
2864 |
* this term will not be returned.
|
|
|
2865 |
*/
|
|
|
2866 |
PROHIBITED: 3
|
|
|
2867 |
}
|
|
|
2868 |
|
|
|
2869 |
/**
|
|
|
2870 |
* A single clause in a {@link lunr.Query} contains a term and details on how to
|
|
|
2871 |
* match that term against a {@link lunr.Index}.
|
|
|
2872 |
*
|
|
|
2873 |
* @typedef {Object} lunr.Query~Clause
|
|
|
2874 |
* @property {string[]} fields - The fields in an index this clause should be matched against.
|
|
|
2875 |
* @property {number} [boost=1] - Any boost that should be applied when matching this clause.
|
|
|
2876 |
* @property {number} [editDistance] - Whether the term should have fuzzy matching applied, and how fuzzy the match should be.
|
|
|
2877 |
* @property {boolean} [usePipeline] - Whether the term should be passed through the search pipeline.
|
|
|
2878 |
* @property {number} [wildcard=lunr.Query.wildcard.NONE] - Whether the term should have wildcards appended or prepended.
|
|
|
2879 |
* @property {number} [presence=lunr.Query.presence.OPTIONAL] - The terms presence in any matching documents.
|
|
|
2880 |
*/
|
|
|
2881 |
|
|
|
2882 |
/**
|
|
|
2883 |
* Adds a {@link lunr.Query~Clause} to this query.
|
|
|
2884 |
*
|
|
|
2885 |
* Unless the clause contains the fields to be matched all fields will be matched. In addition
|
|
|
2886 |
* a default boost of 1 is applied to the clause.
|
|
|
2887 |
*
|
|
|
2888 |
* @param {lunr.Query~Clause} clause - The clause to add to this query.
|
|
|
2889 |
* @see lunr.Query~Clause
|
|
|
2890 |
* @returns {lunr.Query}
|
|
|
2891 |
*/
|
|
|
2892 |
lunr.Query.prototype.clause = function (clause) {
|
|
|
2893 |
if (!('fields' in clause)) {
|
|
|
2894 |
clause.fields = this.allFields
|
|
|
2895 |
}
|
|
|
2896 |
|
|
|
2897 |
if (!('boost' in clause)) {
|
|
|
2898 |
clause.boost = 1
|
|
|
2899 |
}
|
|
|
2900 |
|
|
|
2901 |
if (!('usePipeline' in clause)) {
|
|
|
2902 |
clause.usePipeline = true
|
|
|
2903 |
}
|
|
|
2904 |
|
|
|
2905 |
if (!('wildcard' in clause)) {
|
|
|
2906 |
clause.wildcard = lunr.Query.wildcard.NONE
|
|
|
2907 |
}
|
|
|
2908 |
|
|
|
2909 |
if ((clause.wildcard & lunr.Query.wildcard.LEADING) && (clause.term.charAt(0) != lunr.Query.wildcard)) {
|
|
|
2910 |
clause.term = "*" + clause.term
|
|
|
2911 |
}
|
|
|
2912 |
|
|
|
2913 |
if ((clause.wildcard & lunr.Query.wildcard.TRAILING) && (clause.term.slice(-1) != lunr.Query.wildcard)) {
|
|
|
2914 |
clause.term = "" + clause.term + "*"
|
|
|
2915 |
}
|
|
|
2916 |
|
|
|
2917 |
if (!('presence' in clause)) {
|
|
|
2918 |
clause.presence = lunr.Query.presence.OPTIONAL
|
|
|
2919 |
}
|
|
|
2920 |
|
|
|
2921 |
this.clauses.push(clause)
|
|
|
2922 |
|
|
|
2923 |
return this
|
|
|
2924 |
}
|
|
|
2925 |
|
|
|
2926 |
/**
|
|
|
2927 |
* A negated query is one in which every clause has a presence of
|
|
|
2928 |
* prohibited. These queries require some special processing to return
|
|
|
2929 |
* the expected results.
|
|
|
2930 |
*
|
|
|
2931 |
* @returns boolean
|
|
|
2932 |
*/
|
|
|
2933 |
lunr.Query.prototype.isNegated = function () {
|
|
|
2934 |
for (var i = 0; i < this.clauses.length; i++) {
|
|
|
2935 |
if (this.clauses[i].presence != lunr.Query.presence.PROHIBITED) {
|
|
|
2936 |
return false
|
|
|
2937 |
}
|
|
|
2938 |
}
|
|
|
2939 |
|
|
|
2940 |
return true
|
|
|
2941 |
}
|
|
|
2942 |
|
|
|
2943 |
/**
|
|
|
2944 |
* Adds a term to the current query, under the covers this will create a {@link lunr.Query~Clause}
|
|
|
2945 |
* to the list of clauses that make up this query.
|
|
|
2946 |
*
|
|
|
2947 |
* The term is used as is, i.e. no tokenization will be performed by this method. Instead conversion
|
|
|
2948 |
* to a token or token-like string should be done before calling this method.
|
|
|
2949 |
*
|
|
|
2950 |
* The term will be converted to a string by calling `toString`. Multiple terms can be passed as an
|
|
|
2951 |
* array, each term in the array will share the same options.
|
|
|
2952 |
*
|
|
|
2953 |
* @param {object|object[]} term - The term(s) to add to the query.
|
|
|
2954 |
* @param {object} [options] - Any additional properties to add to the query clause.
|
|
|
2955 |
* @returns {lunr.Query}
|
|
|
2956 |
* @see lunr.Query#clause
|
|
|
2957 |
* @see lunr.Query~Clause
|
|
|
2958 |
* @example <caption>adding a single term to a query</caption>
|
|
|
2959 |
* query.term("foo")
|
|
|
2960 |
* @example <caption>adding a single term to a query and specifying search fields, term boost and automatic trailing wildcard</caption>
|
|
|
2961 |
* query.term("foo", {
|
|
|
2962 |
* fields: ["title"],
|
|
|
2963 |
* boost: 10,
|
|
|
2964 |
* wildcard: lunr.Query.wildcard.TRAILING
|
|
|
2965 |
* })
|
|
|
2966 |
* @example <caption>using lunr.tokenizer to convert a string to tokens before using them as terms</caption>
|
|
|
2967 |
* query.term(lunr.tokenizer("foo bar"))
|
|
|
2968 |
*/
|
|
|
2969 |
lunr.Query.prototype.term = function (term, options) {
|
|
|
2970 |
if (Array.isArray(term)) {
|
|
|
2971 |
term.forEach(function (t) { this.term(t, lunr.utils.clone(options)) }, this)
|
|
|
2972 |
return this
|
|
|
2973 |
}
|
|
|
2974 |
|
|
|
2975 |
var clause = options || {}
|
|
|
2976 |
clause.term = term.toString()
|
|
|
2977 |
|
|
|
2978 |
this.clause(clause)
|
|
|
2979 |
|
|
|
2980 |
return this
|
|
|
2981 |
}
|
|
|
2982 |
lunr.QueryParseError = function (message, start, end) {
|
|
|
2983 |
this.name = "QueryParseError"
|
|
|
2984 |
this.message = message
|
|
|
2985 |
this.start = start
|
|
|
2986 |
this.end = end
|
|
|
2987 |
}
|
|
|
2988 |
|
|
|
2989 |
lunr.QueryParseError.prototype = new Error
|
|
|
2990 |
lunr.QueryLexer = function (str) {
|
|
|
2991 |
this.lexemes = []
|
|
|
2992 |
this.str = str
|
|
|
2993 |
this.length = str.length
|
|
|
2994 |
this.pos = 0
|
|
|
2995 |
this.start = 0
|
|
|
2996 |
this.escapeCharPositions = []
|
|
|
2997 |
}
|
|
|
2998 |
|
|
|
2999 |
lunr.QueryLexer.prototype.run = function () {
|
|
|
3000 |
var state = lunr.QueryLexer.lexText
|
|
|
3001 |
|
|
|
3002 |
while (state) {
|
|
|
3003 |
state = state(this)
|
|
|
3004 |
}
|
|
|
3005 |
}
|
|
|
3006 |
|
|
|
3007 |
lunr.QueryLexer.prototype.sliceString = function () {
|
|
|
3008 |
var subSlices = [],
|
|
|
3009 |
sliceStart = this.start,
|
|
|
3010 |
sliceEnd = this.pos
|
|
|
3011 |
|
|
|
3012 |
for (var i = 0; i < this.escapeCharPositions.length; i++) {
|
|
|
3013 |
sliceEnd = this.escapeCharPositions[i]
|
|
|
3014 |
subSlices.push(this.str.slice(sliceStart, sliceEnd))
|
|
|
3015 |
sliceStart = sliceEnd + 1
|
|
|
3016 |
}
|
|
|
3017 |
|
|
|
3018 |
subSlices.push(this.str.slice(sliceStart, this.pos))
|
|
|
3019 |
this.escapeCharPositions.length = 0
|
|
|
3020 |
|
|
|
3021 |
return subSlices.join('')
|
|
|
3022 |
}
|
|
|
3023 |
|
|
|
3024 |
lunr.QueryLexer.prototype.emit = function (type) {
|
|
|
3025 |
this.lexemes.push({
|
|
|
3026 |
type: type,
|
|
|
3027 |
str: this.sliceString(),
|
|
|
3028 |
start: this.start,
|
|
|
3029 |
end: this.pos
|
|
|
3030 |
})
|
|
|
3031 |
|
|
|
3032 |
this.start = this.pos
|
|
|
3033 |
}
|
|
|
3034 |
|
|
|
3035 |
lunr.QueryLexer.prototype.escapeCharacter = function () {
|
|
|
3036 |
this.escapeCharPositions.push(this.pos - 1)
|
|
|
3037 |
this.pos += 1
|
|
|
3038 |
}
|
|
|
3039 |
|
|
|
3040 |
lunr.QueryLexer.prototype.next = function () {
|
|
|
3041 |
if (this.pos >= this.length) {
|
|
|
3042 |
return lunr.QueryLexer.EOS
|
|
|
3043 |
}
|
|
|
3044 |
|
|
|
3045 |
var char = this.str.charAt(this.pos)
|
|
|
3046 |
this.pos += 1
|
|
|
3047 |
return char
|
|
|
3048 |
}
|
|
|
3049 |
|
|
|
3050 |
lunr.QueryLexer.prototype.width = function () {
|
|
|
3051 |
return this.pos - this.start
|
|
|
3052 |
}
|
|
|
3053 |
|
|
|
3054 |
lunr.QueryLexer.prototype.ignore = function () {
|
|
|
3055 |
if (this.start == this.pos) {
|
|
|
3056 |
this.pos += 1
|
|
|
3057 |
}
|
|
|
3058 |
|
|
|
3059 |
this.start = this.pos
|
|
|
3060 |
}
|
|
|
3061 |
|
|
|
3062 |
lunr.QueryLexer.prototype.backup = function () {
|
|
|
3063 |
this.pos -= 1
|
|
|
3064 |
}
|
|
|
3065 |
|
|
|
3066 |
lunr.QueryLexer.prototype.acceptDigitRun = function () {
|
|
|
3067 |
var char, charCode
|
|
|
3068 |
|
|
|
3069 |
do {
|
|
|
3070 |
char = this.next()
|
|
|
3071 |
charCode = char.charCodeAt(0)
|
|
|
3072 |
} while (charCode > 47 && charCode < 58)
|
|
|
3073 |
|
|
|
3074 |
if (char != lunr.QueryLexer.EOS) {
|
|
|
3075 |
this.backup()
|
|
|
3076 |
}
|
|
|
3077 |
}
|
|
|
3078 |
|
|
|
3079 |
lunr.QueryLexer.prototype.more = function () {
|
|
|
3080 |
return this.pos < this.length
|
|
|
3081 |
}
|
|
|
3082 |
|
|
|
3083 |
lunr.QueryLexer.EOS = 'EOS'
|
|
|
3084 |
lunr.QueryLexer.FIELD = 'FIELD'
|
|
|
3085 |
lunr.QueryLexer.TERM = 'TERM'
|
|
|
3086 |
lunr.QueryLexer.EDIT_DISTANCE = 'EDIT_DISTANCE'
|
|
|
3087 |
lunr.QueryLexer.BOOST = 'BOOST'
|
|
|
3088 |
lunr.QueryLexer.PRESENCE = 'PRESENCE'
|
|
|
3089 |
|
|
|
3090 |
lunr.QueryLexer.lexField = function (lexer) {
|
|
|
3091 |
lexer.backup()
|
|
|
3092 |
lexer.emit(lunr.QueryLexer.FIELD)
|
|
|
3093 |
lexer.ignore()
|
|
|
3094 |
return lunr.QueryLexer.lexText
|
|
|
3095 |
}
|
|
|
3096 |
|
|
|
3097 |
lunr.QueryLexer.lexTerm = function (lexer) {
|
|
|
3098 |
if (lexer.width() > 1) {
|
|
|
3099 |
lexer.backup()
|
|
|
3100 |
lexer.emit(lunr.QueryLexer.TERM)
|
|
|
3101 |
}
|
|
|
3102 |
|
|
|
3103 |
lexer.ignore()
|
|
|
3104 |
|
|
|
3105 |
if (lexer.more()) {
|
|
|
3106 |
return lunr.QueryLexer.lexText
|
|
|
3107 |
}
|
|
|
3108 |
}
|
|
|
3109 |
|
|
|
3110 |
lunr.QueryLexer.lexEditDistance = function (lexer) {
|
|
|
3111 |
lexer.ignore()
|
|
|
3112 |
lexer.acceptDigitRun()
|
|
|
3113 |
lexer.emit(lunr.QueryLexer.EDIT_DISTANCE)
|
|
|
3114 |
return lunr.QueryLexer.lexText
|
|
|
3115 |
}
|
|
|
3116 |
|
|
|
3117 |
lunr.QueryLexer.lexBoost = function (lexer) {
|
|
|
3118 |
lexer.ignore()
|
|
|
3119 |
lexer.acceptDigitRun()
|
|
|
3120 |
lexer.emit(lunr.QueryLexer.BOOST)
|
|
|
3121 |
return lunr.QueryLexer.lexText
|
|
|
3122 |
}
|
|
|
3123 |
|
|
|
3124 |
lunr.QueryLexer.lexEOS = function (lexer) {
|
|
|
3125 |
if (lexer.width() > 0) {
|
|
|
3126 |
lexer.emit(lunr.QueryLexer.TERM)
|
|
|
3127 |
}
|
|
|
3128 |
}
|
|
|
3129 |
|
|
|
3130 |
// This matches the separator used when tokenising fields
|
|
|
3131 |
// within a document. These should match otherwise it is
|
|
|
3132 |
// not possible to search for some tokens within a document.
|
|
|
3133 |
//
|
|
|
3134 |
// It is possible for the user to change the separator on the
|
|
|
3135 |
// tokenizer so it _might_ clash with any other of the special
|
|
|
3136 |
// characters already used within the search string, e.g. :.
|
|
|
3137 |
//
|
|
|
3138 |
// This means that it is possible to change the separator in
|
|
|
3139 |
// such a way that makes some words unsearchable using a search
|
|
|
3140 |
// string.
|
|
|
3141 |
lunr.QueryLexer.termSeparator = lunr.tokenizer.separator
|
|
|
3142 |
|
|
|
3143 |
lunr.QueryLexer.lexText = function (lexer) {
|
|
|
3144 |
while (true) {
|
|
|
3145 |
var char = lexer.next()
|
|
|
3146 |
|
|
|
3147 |
if (char == lunr.QueryLexer.EOS) {
|
|
|
3148 |
return lunr.QueryLexer.lexEOS
|
|
|
3149 |
}
|
|
|
3150 |
|
|
|
3151 |
// Escape character is '\'
|
|
|
3152 |
if (char.charCodeAt(0) == 92) {
|
|
|
3153 |
lexer.escapeCharacter()
|
|
|
3154 |
continue
|
|
|
3155 |
}
|
|
|
3156 |
|
|
|
3157 |
if (char == ":") {
|
|
|
3158 |
return lunr.QueryLexer.lexField
|
|
|
3159 |
}
|
|
|
3160 |
|
|
|
3161 |
if (char == "~") {
|
|
|
3162 |
lexer.backup()
|
|
|
3163 |
if (lexer.width() > 0) {
|
|
|
3164 |
lexer.emit(lunr.QueryLexer.TERM)
|
|
|
3165 |
}
|
|
|
3166 |
return lunr.QueryLexer.lexEditDistance
|
|
|
3167 |
}
|
|
|
3168 |
|
|
|
3169 |
if (char == "^") {
|
|
|
3170 |
lexer.backup()
|
|
|
3171 |
if (lexer.width() > 0) {
|
|
|
3172 |
lexer.emit(lunr.QueryLexer.TERM)
|
|
|
3173 |
}
|
|
|
3174 |
return lunr.QueryLexer.lexBoost
|
|
|
3175 |
}
|
|
|
3176 |
|
|
|
3177 |
// "+" indicates term presence is required
|
|
|
3178 |
// checking for length to ensure that only
|
|
|
3179 |
// leading "+" are considered
|
|
|
3180 |
if (char == "+" && lexer.width() === 1) {
|
|
|
3181 |
lexer.emit(lunr.QueryLexer.PRESENCE)
|
|
|
3182 |
return lunr.QueryLexer.lexText
|
|
|
3183 |
}
|
|
|
3184 |
|
|
|
3185 |
// "-" indicates term presence is prohibited
|
|
|
3186 |
// checking for length to ensure that only
|
|
|
3187 |
// leading "-" are considered
|
|
|
3188 |
if (char == "-" && lexer.width() === 1) {
|
|
|
3189 |
lexer.emit(lunr.QueryLexer.PRESENCE)
|
|
|
3190 |
return lunr.QueryLexer.lexText
|
|
|
3191 |
}
|
|
|
3192 |
|
|
|
3193 |
if (char.match(lunr.QueryLexer.termSeparator)) {
|
|
|
3194 |
return lunr.QueryLexer.lexTerm
|
|
|
3195 |
}
|
|
|
3196 |
}
|
|
|
3197 |
}
|
|
|
3198 |
|
|
|
3199 |
lunr.QueryParser = function (str, query) {
|
|
|
3200 |
this.lexer = new lunr.QueryLexer (str)
|
|
|
3201 |
this.query = query
|
|
|
3202 |
this.currentClause = {}
|
|
|
3203 |
this.lexemeIdx = 0
|
|
|
3204 |
}
|
|
|
3205 |
|
|
|
3206 |
lunr.QueryParser.prototype.parse = function () {
|
|
|
3207 |
this.lexer.run()
|
|
|
3208 |
this.lexemes = this.lexer.lexemes
|
|
|
3209 |
|
|
|
3210 |
var state = lunr.QueryParser.parseClause
|
|
|
3211 |
|
|
|
3212 |
while (state) {
|
|
|
3213 |
state = state(this)
|
|
|
3214 |
}
|
|
|
3215 |
|
|
|
3216 |
return this.query
|
|
|
3217 |
}
|
|
|
3218 |
|
|
|
3219 |
lunr.QueryParser.prototype.peekLexeme = function () {
|
|
|
3220 |
return this.lexemes[this.lexemeIdx]
|
|
|
3221 |
}
|
|
|
3222 |
|
|
|
3223 |
lunr.QueryParser.prototype.consumeLexeme = function () {
|
|
|
3224 |
var lexeme = this.peekLexeme()
|
|
|
3225 |
this.lexemeIdx += 1
|
|
|
3226 |
return lexeme
|
|
|
3227 |
}
|
|
|
3228 |
|
|
|
3229 |
lunr.QueryParser.prototype.nextClause = function () {
|
|
|
3230 |
var completedClause = this.currentClause
|
|
|
3231 |
this.query.clause(completedClause)
|
|
|
3232 |
this.currentClause = {}
|
|
|
3233 |
}
|
|
|
3234 |
|
|
|
3235 |
lunr.QueryParser.parseClause = function (parser) {
|
|
|
3236 |
var lexeme = parser.peekLexeme()
|
|
|
3237 |
|
|
|
3238 |
if (lexeme == undefined) {
|
|
|
3239 |
return
|
|
|
3240 |
}
|
|
|
3241 |
|
|
|
3242 |
switch (lexeme.type) {
|
|
|
3243 |
case lunr.QueryLexer.PRESENCE:
|
|
|
3244 |
return lunr.QueryParser.parsePresence
|
|
|
3245 |
case lunr.QueryLexer.FIELD:
|
|
|
3246 |
return lunr.QueryParser.parseField
|
|
|
3247 |
case lunr.QueryLexer.TERM:
|
|
|
3248 |
return lunr.QueryParser.parseTerm
|
|
|
3249 |
default:
|
|
|
3250 |
var errorMessage = "expected either a field or a term, found " + lexeme.type
|
|
|
3251 |
|
|
|
3252 |
if (lexeme.str.length >= 1) {
|
|
|
3253 |
errorMessage += " with value '" + lexeme.str + "'"
|
|
|
3254 |
}
|
|
|
3255 |
|
|
|
3256 |
throw new lunr.QueryParseError (errorMessage, lexeme.start, lexeme.end)
|
|
|
3257 |
}
|
|
|
3258 |
}
|
|
|
3259 |
|
|
|
3260 |
lunr.QueryParser.parsePresence = function (parser) {
|
|
|
3261 |
var lexeme = parser.consumeLexeme()
|
|
|
3262 |
|
|
|
3263 |
if (lexeme == undefined) {
|
|
|
3264 |
return
|
|
|
3265 |
}
|
|
|
3266 |
|
|
|
3267 |
switch (lexeme.str) {
|
|
|
3268 |
case "-":
|
|
|
3269 |
parser.currentClause.presence = lunr.Query.presence.PROHIBITED
|
|
|
3270 |
break
|
|
|
3271 |
case "+":
|
|
|
3272 |
parser.currentClause.presence = lunr.Query.presence.REQUIRED
|
|
|
3273 |
break
|
|
|
3274 |
default:
|
|
|
3275 |
var errorMessage = "unrecognised presence operator'" + lexeme.str + "'"
|
|
|
3276 |
throw new lunr.QueryParseError (errorMessage, lexeme.start, lexeme.end)
|
|
|
3277 |
}
|
|
|
3278 |
|
|
|
3279 |
var nextLexeme = parser.peekLexeme()
|
|
|
3280 |
|
|
|
3281 |
if (nextLexeme == undefined) {
|
|
|
3282 |
var errorMessage = "expecting term or field, found nothing"
|
|
|
3283 |
throw new lunr.QueryParseError (errorMessage, lexeme.start, lexeme.end)
|
|
|
3284 |
}
|
|
|
3285 |
|
|
|
3286 |
switch (nextLexeme.type) {
|
|
|
3287 |
case lunr.QueryLexer.FIELD:
|
|
|
3288 |
return lunr.QueryParser.parseField
|
|
|
3289 |
case lunr.QueryLexer.TERM:
|
|
|
3290 |
return lunr.QueryParser.parseTerm
|
|
|
3291 |
default:
|
|
|
3292 |
var errorMessage = "expecting term or field, found '" + nextLexeme.type + "'"
|
|
|
3293 |
throw new lunr.QueryParseError (errorMessage, nextLexeme.start, nextLexeme.end)
|
|
|
3294 |
}
|
|
|
3295 |
}
|
|
|
3296 |
|
|
|
3297 |
lunr.QueryParser.parseField = function (parser) {
|
|
|
3298 |
var lexeme = parser.consumeLexeme()
|
|
|
3299 |
|
|
|
3300 |
if (lexeme == undefined) {
|
|
|
3301 |
return
|
|
|
3302 |
}
|
|
|
3303 |
|
|
|
3304 |
if (parser.query.allFields.indexOf(lexeme.str) == -1) {
|
|
|
3305 |
var possibleFields = parser.query.allFields.map(function (f) { return "'" + f + "'" }).join(', '),
|
|
|
3306 |
errorMessage = "unrecognised field '" + lexeme.str + "', possible fields: " + possibleFields
|
|
|
3307 |
|
|
|
3308 |
throw new lunr.QueryParseError (errorMessage, lexeme.start, lexeme.end)
|
|
|
3309 |
}
|
|
|
3310 |
|
|
|
3311 |
parser.currentClause.fields = [lexeme.str]
|
|
|
3312 |
|
|
|
3313 |
var nextLexeme = parser.peekLexeme()
|
|
|
3314 |
|
|
|
3315 |
if (nextLexeme == undefined) {
|
|
|
3316 |
var errorMessage = "expecting term, found nothing"
|
|
|
3317 |
throw new lunr.QueryParseError (errorMessage, lexeme.start, lexeme.end)
|
|
|
3318 |
}
|
|
|
3319 |
|
|
|
3320 |
switch (nextLexeme.type) {
|
|
|
3321 |
case lunr.QueryLexer.TERM:
|
|
|
3322 |
return lunr.QueryParser.parseTerm
|
|
|
3323 |
default:
|
|
|
3324 |
var errorMessage = "expecting term, found '" + nextLexeme.type + "'"
|
|
|
3325 |
throw new lunr.QueryParseError (errorMessage, nextLexeme.start, nextLexeme.end)
|
|
|
3326 |
}
|
|
|
3327 |
}
|
|
|
3328 |
|
|
|
3329 |
lunr.QueryParser.parseTerm = function (parser) {
|
|
|
3330 |
var lexeme = parser.consumeLexeme()
|
|
|
3331 |
|
|
|
3332 |
if (lexeme == undefined) {
|
|
|
3333 |
return
|
|
|
3334 |
}
|
|
|
3335 |
|
|
|
3336 |
parser.currentClause.term = lexeme.str.toLowerCase()
|
|
|
3337 |
|
|
|
3338 |
if (lexeme.str.indexOf("*") != -1) {
|
|
|
3339 |
parser.currentClause.usePipeline = false
|
|
|
3340 |
}
|
|
|
3341 |
|
|
|
3342 |
var nextLexeme = parser.peekLexeme()
|
|
|
3343 |
|
|
|
3344 |
if (nextLexeme == undefined) {
|
|
|
3345 |
parser.nextClause()
|
|
|
3346 |
return
|
|
|
3347 |
}
|
|
|
3348 |
|
|
|
3349 |
switch (nextLexeme.type) {
|
|
|
3350 |
case lunr.QueryLexer.TERM:
|
|
|
3351 |
parser.nextClause()
|
|
|
3352 |
return lunr.QueryParser.parseTerm
|
|
|
3353 |
case lunr.QueryLexer.FIELD:
|
|
|
3354 |
parser.nextClause()
|
|
|
3355 |
return lunr.QueryParser.parseField
|
|
|
3356 |
case lunr.QueryLexer.EDIT_DISTANCE:
|
|
|
3357 |
return lunr.QueryParser.parseEditDistance
|
|
|
3358 |
case lunr.QueryLexer.BOOST:
|
|
|
3359 |
return lunr.QueryParser.parseBoost
|
|
|
3360 |
case lunr.QueryLexer.PRESENCE:
|
|
|
3361 |
parser.nextClause()
|
|
|
3362 |
return lunr.QueryParser.parsePresence
|
|
|
3363 |
default:
|
|
|
3364 |
var errorMessage = "Unexpected lexeme type '" + nextLexeme.type + "'"
|
|
|
3365 |
throw new lunr.QueryParseError (errorMessage, nextLexeme.start, nextLexeme.end)
|
|
|
3366 |
}
|
|
|
3367 |
}
|
|
|
3368 |
|
|
|
3369 |
lunr.QueryParser.parseEditDistance = function (parser) {
|
|
|
3370 |
var lexeme = parser.consumeLexeme()
|
|
|
3371 |
|
|
|
3372 |
if (lexeme == undefined) {
|
|
|
3373 |
return
|
|
|
3374 |
}
|
|
|
3375 |
|
|
|
3376 |
var editDistance = parseInt(lexeme.str, 10)
|
|
|
3377 |
|
|
|
3378 |
if (isNaN(editDistance)) {
|
|
|
3379 |
var errorMessage = "edit distance must be numeric"
|
|
|
3380 |
throw new lunr.QueryParseError (errorMessage, lexeme.start, lexeme.end)
|
|
|
3381 |
}
|
|
|
3382 |
|
|
|
3383 |
parser.currentClause.editDistance = editDistance
|
|
|
3384 |
|
|
|
3385 |
var nextLexeme = parser.peekLexeme()
|
|
|
3386 |
|
|
|
3387 |
if (nextLexeme == undefined) {
|
|
|
3388 |
parser.nextClause()
|
|
|
3389 |
return
|
|
|
3390 |
}
|
|
|
3391 |
|
|
|
3392 |
switch (nextLexeme.type) {
|
|
|
3393 |
case lunr.QueryLexer.TERM:
|
|
|
3394 |
parser.nextClause()
|
|
|
3395 |
return lunr.QueryParser.parseTerm
|
|
|
3396 |
case lunr.QueryLexer.FIELD:
|
|
|
3397 |
parser.nextClause()
|
|
|
3398 |
return lunr.QueryParser.parseField
|
|
|
3399 |
case lunr.QueryLexer.EDIT_DISTANCE:
|
|
|
3400 |
return lunr.QueryParser.parseEditDistance
|
|
|
3401 |
case lunr.QueryLexer.BOOST:
|
|
|
3402 |
return lunr.QueryParser.parseBoost
|
|
|
3403 |
case lunr.QueryLexer.PRESENCE:
|
|
|
3404 |
parser.nextClause()
|
|
|
3405 |
return lunr.QueryParser.parsePresence
|
|
|
3406 |
default:
|
|
|
3407 |
var errorMessage = "Unexpected lexeme type '" + nextLexeme.type + "'"
|
|
|
3408 |
throw new lunr.QueryParseError (errorMessage, nextLexeme.start, nextLexeme.end)
|
|
|
3409 |
}
|
|
|
3410 |
}
|
|
|
3411 |
|
|
|
3412 |
lunr.QueryParser.parseBoost = function (parser) {
|
|
|
3413 |
var lexeme = parser.consumeLexeme()
|
|
|
3414 |
|
|
|
3415 |
if (lexeme == undefined) {
|
|
|
3416 |
return
|
|
|
3417 |
}
|
|
|
3418 |
|
|
|
3419 |
var boost = parseInt(lexeme.str, 10)
|
|
|
3420 |
|
|
|
3421 |
if (isNaN(boost)) {
|
|
|
3422 |
var errorMessage = "boost must be numeric"
|
|
|
3423 |
throw new lunr.QueryParseError (errorMessage, lexeme.start, lexeme.end)
|
|
|
3424 |
}
|
|
|
3425 |
|
|
|
3426 |
parser.currentClause.boost = boost
|
|
|
3427 |
|
|
|
3428 |
var nextLexeme = parser.peekLexeme()
|
|
|
3429 |
|
|
|
3430 |
if (nextLexeme == undefined) {
|
|
|
3431 |
parser.nextClause()
|
|
|
3432 |
return
|
|
|
3433 |
}
|
|
|
3434 |
|
|
|
3435 |
switch (nextLexeme.type) {
|
|
|
3436 |
case lunr.QueryLexer.TERM:
|
|
|
3437 |
parser.nextClause()
|
|
|
3438 |
return lunr.QueryParser.parseTerm
|
|
|
3439 |
case lunr.QueryLexer.FIELD:
|
|
|
3440 |
parser.nextClause()
|
|
|
3441 |
return lunr.QueryParser.parseField
|
|
|
3442 |
case lunr.QueryLexer.EDIT_DISTANCE:
|
|
|
3443 |
return lunr.QueryParser.parseEditDistance
|
|
|
3444 |
case lunr.QueryLexer.BOOST:
|
|
|
3445 |
return lunr.QueryParser.parseBoost
|
|
|
3446 |
case lunr.QueryLexer.PRESENCE:
|
|
|
3447 |
parser.nextClause()
|
|
|
3448 |
return lunr.QueryParser.parsePresence
|
|
|
3449 |
default:
|
|
|
3450 |
var errorMessage = "Unexpected lexeme type '" + nextLexeme.type + "'"
|
|
|
3451 |
throw new lunr.QueryParseError (errorMessage, nextLexeme.start, nextLexeme.end)
|
|
|
3452 |
}
|
|
|
3453 |
}
|
|
|
3454 |
|
|
|
3455 |
/**
|
|
|
3456 |
* export the module via AMD, CommonJS or as a browser global
|
|
|
3457 |
* Export code from https://github.com/umdjs/umd/blob/master/returnExports.js
|
|
|
3458 |
*/
|
|
|
3459 |
;(function (root, factory) {
|
|
|
3460 |
if (typeof define === 'function' && define.amd) {
|
|
|
3461 |
// AMD. Register as an anonymous module.
|
|
|
3462 |
define(factory)
|
|
|
3463 |
} else if (typeof exports === 'object') {
|
|
|
3464 |
/**
|
|
|
3465 |
* Node. Does not work with strict CommonJS, but
|
|
|
3466 |
* only CommonJS-like enviroments that support module.exports,
|
|
|
3467 |
* like Node.
|
|
|
3468 |
*/
|
|
|
3469 |
module.exports = factory()
|
|
|
3470 |
} else {
|
|
|
3471 |
// Browser globals (root is window)
|
|
|
3472 |
root.lunr = factory()
|
|
|
3473 |
}
|
|
|
3474 |
}(this, function () {
|
|
|
3475 |
/**
|
|
|
3476 |
* Just return a value to define the module export.
|
|
|
3477 |
* This example returns an object, but the module
|
|
|
3478 |
* can return a function as the exported value.
|
|
|
3479 |
*/
|
|
|
3480 |
return lunr
|
|
|
3481 |
}))
|
|
|
3482 |
})();
|