Changes In Branch spellsugg Excluding Merge-Ins
This is equivalent to a diff from 8ddc15536e to 6c5050fe91
2017-11-21
| ||
16:57 | [core] better suggestion engine Closed-Leaf check-in: 6c5050fe91 user: olr tags: core, spellsugg | |
2017-11-09
| ||
11:56 | [core] ibdawg: suggestion mechanism > reduce 1toX replacements overload (much, much faster) check-in: 767e396f2d user: olr tags: core, spellsugg | |
2017-11-08
| ||
15:57 | [fr] màj: écriture dystypographique check-in: 0bd6a29f03 user: olr tags: trunk, fr | |
2017-11-07
| ||
11:56 | [core] ibdawg: another suggestion method check-in: 1edae62ad8 user: olr tags: core, spellsugg | |
2017-11-06
| ||
17:21 | [core][py] timer for testing check-in: 8ddc15536e user: olr tags: trunk, core | |
14:21 | [core][py] ibdawg: debugging check-in: cbe1422160 user: olr tags: trunk, core | |
Modified cli.py from [ed7458a4d3] to [68793c6981].
︙ | |||
206 207 208 209 210 211 212 213 214 215 216 217 218 219 | 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 | + | echo("* " + sWord) for sMorph in oDict.getMorph(sWord): echo(" {:<32} {}".format(sMorph, oLexGraphe.formatTags(sMorph))) elif sText.startswith("!"): for sWord in sText[1:].strip().split(): if sWord: echo(" | ".join(oDict.suggest(sWord))) #echo(" | ".join(oDict.suggest2(sWord))) elif sText.startswith(">"): oDict.drawPath(sText[1:].strip()) elif sText.startswith("="): for sRes in oDict.select(sText[1:].strip()): echo(sRes) elif sText.startswith("/+ "): gce.setOptions({ opt:True for opt in sText[3:].strip().split() if opt in gce.getOptions() }) |
︙ |
Modified gc_core/js/char_player.js from [f5b2267e7d] to [0547b59e35].
︙ | |||
22 23 24 25 26 27 28 | 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 | - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - + - - - - - - - + - - + - - - - - - - - - - - - - - - - - - - - | let sRes = ""; for (let c of sWord) { sRes += this._dTransChars.gl_get(c, c); } return sRes.replace("eau", "o").replace("au", "o"); }, |
︙ | |||
223 224 225 226 227 228 229 | 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 | - - + + - - | d1toX: new Map([ ["æ", ["ae",]], ["Æ", ["AE",]], ["b", ["bb",]], ["B", ["BB",]], ["c", ["cc", "ss", "qu", "ch"]], ["C", ["CC", "SS", "QU", "CH"]], |
︙ | |||
262 263 264 265 266 267 268 269 270 271 272 273 274 275 | 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 | + + + + + + + | ["t", ["tt", "th"]], ["T", ["TT", "TH"]], ["x", ["cc", "ct", "xx"]], ["X", ["CC", "CT", "XX"]], ["z", ["ss", "zh"]], ["Z", ["SS", "ZH"]], ]), get1toXReplacement: function (cPrev, cCur, cNext) { if (this.aConsonant.has(cCur) && (this.aConsonant.has(cPrev) || this.aConsonant.has(cNext))) { return []; } return this.d1toX.gl_get(cCur, []); }, d2toX: new Map([ ["an", ["en",]], ["AN", ["EN",]], ["au", ["eau", "o", "ô"]], ["AU", ["EAU", "O", "Ô"]], ["en", ["an",]], |
︙ |
Modified gc_core/js/ibdawg.js from [c87880e4ad] to [b7bc0fa0c9].
︙ | |||
13 14 15 16 17 18 19 20 21 22 23 24 25 26 | 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 | + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + | // Don’t remove <string>. Necessary in TB. ${string} ${map} ${set} class SuggResult { // Structure for storing, classifying and filtering suggestions constructor (sWord, nDistLimit=-1) { this.sWord = sWord; this.sCleanWord = char_player.cleanWord(sWord); this.nDistLimit = (nDistLimit >= 0) ? nDistLimit : Math.floor(sWord.length / 3) + 1; this.nMinDist = 1000; this.aSugg = new Set(); this.dSugg = new Map([ [0, []], [1, []] ]); } addSugg (sSugg, nDeep=0) { // add a suggestion if (!this.aSugg.has(sSugg)) { let nDist = str_transform.distanceDamerauLevenshtein(this.sCleanWord, char_player.cleanWord(sSugg)); if (nDist <= this.nDistLimit) { if (!this.dSugg.has(nDist)) { this.dSugg.set(nDist, []); } this.dSugg.get(nDist).push(sSugg); this.aSugg.add(sSugg); if (nDist < this.nMinDist) { this.nMinDist = nDist; } this.nDistLimit = Math.min(this.nDistLimit, this.nMinDist+2); } } } getSuggestions (nSuggLimit=10, nDistLimit=-1) { // return a list of suggestions let lRes = []; if (this.dSugg.get(0).length) { // we sort the better results with the original word let dDistTemp = new Map(); lRes.forEach((sSugg) => { dDistTemp.set(sSugg, str_transform.distanceDamerauLevenshtein(this.sWord, sSugg)); }); lRes = lRes.sort((sA, sB) => { return dDistTemp.get(sA) - dDistTemp.get(sB); }); dDistTemp.clear(); } for (let lSugg of this.dSugg.values()) { for (let sSugg of lSugg) { lRes.push(sSugg); } if (lRes.length > nSuggLimit) { break; } } lRes = char_player.filterSugg(lRes); if (this.sWord.gl_isTitle()) { lRes = lRes.map((sSugg) => { return sSugg.gl_toCapitalize(); }); } else if (this.sWord.gl_isUpperCase()) { lRes = lRes.map((sSugg) => { return sSugg.toUpperCase(); }); } return lRes.slice(0, nSuggLimit); } reset () { this.aSugg.clear(); this.dSugg.clear(); } } class IBDAWG { // INDEXABLE BINARY DIRECT ACYCLIC WORD GRAPH constructor (sDicName, sPath="") { try { let sURL = (sPath !== "") ? sPath + "/" + sDicName : "resource://grammalecte/_dictionaries/"+sDicName; |
︙ | |||
189 190 191 192 193 194 195 196 197 | 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 | + + - + - + - + - - + - - - - - - - - - - + - - + - + - + - - - + + + + + + + + + + + + - - - - - - + + + + + + - - - - + + - - - - - - - - - - - - + + + + + + + + + + + + + + + + - - - - + + - + - + - + - + + - - - - + + + + + + + + + + + | } suggest (sWord, nMaxSugg=10) { // returns a array of suggestions for <sWord> let sPfx = ""; let sSfx = ""; [sPfx, sWord, sSfx] = char_player.cut(sWord); let nMaxSwitch = Math.max(Math.floor(sWord.length / 3), 1); let nMaxDel = Math.floor(sWord.length / 5); let nMaxHardRepl = Math.max(Math.floor((sWord.length - 5) / 4), 1); let oSuggResult = new SuggResult(sWord); |
︙ |
Modified gc_core/js/str_transform.js from [6745507121] to [e70f6ede55].
1 2 3 4 5 6 7 8 9 10 11 12 13 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 | + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + | //// STRING TRANSFORMATION /*jslint esversion: 6*/ // Note: 48 is the ASCII code for "0" var str_transform = { distanceDamerauLevenshtein: function (s1, s2) { // distance of Damerau-Levenshtein between <s1> and <s2> // https://fr.wikipedia.org/wiki/Distance_de_Damerau-Levenshtein try { let nLen1 = s1.length; let nLen2 = s2.length; let matrix = []; for (let i = 0; i <= nLen1; i++) { matrix[i] = new Array(nLen2 + 1); } for (let i = 0; i <= nLen1; i++) { matrix[i][0] = i; } for (let j = 0; j <= nLen2; j++) { matrix[0][j] = j; } for (let i = 1; i <= nLen1; i++) { for (let j = 1; j <= nLen2; j++) { let nCost = (s1[i] === s2[j]) ? 0 : 1; matrix[i][j] = Math.min( matrix[i-1][j] + 1, // Deletion matrix[i][j-1] + 1, // Insertion matrix[i-1][j-1] + nCost // Substitution ); if (i > 1 && j > 1 && s1[i] == s2[j-1] && s1[i-1] == s2[j]) { matrix[i][j] = Math.min(matrix[i][j], matrix[i-2][j-2] + nCost); // Transposition } } } //console.log(s2 + ": " + matrix[nLen1][nLen2]); return matrix[nLen1][nLen2]; } catch (e) { helpers.logerror(e); } }, showDistance (s1, s2) { console.log(`Distance: ${s1} / ${s2} = ${this.distanceDamerauLevenshtein(s1, s2)})`); }, getStemFromSuffixCode: function (sFlex, sSfxCode) { // Suffix only if (sSfxCode == "0") { return sFlex; } return sSfxCode[0] == '0' ? sFlex + sSfxCode.slice(1) : sFlex.slice(0, -(sSfxCode.charCodeAt(0)-48)) + sSfxCode.slice(1); }, |
︙ |
Modified gc_core/py/char_player.py from [97f69bd70d] to [b0152aab01].
︙ | |||
15 16 17 18 19 20 21 | 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 | - - - - - - - - - - - - - - - - - - - - - - - + + - - - - - - + - - - - - - - - - - - - - - - - - - - - - - | }) def cleanWord (sWord): "word simplication before calculating distance between words" return sWord.lower().translate(_xTransChars).replace("eau", "o").replace("au", "o") |
︙ | |||
201 202 203 204 205 206 207 | 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 | - - + + - - | d1toX = { "æ": ("ae",), "Æ": ("AE",), "b": ("bb",), "B": ("BB",), "c": ("cc", "ss", "qu", "ch"), "C": ("CC", "SS", "QU", "CH"), |
︙ | |||
240 241 242 243 244 245 246 247 248 249 250 251 252 253 | 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 | + + + + + + + | "t": ("tt", "th"), "T": ("TT", "TH"), "x": ("cc", "ct", "xx"), "X": ("CC", "CT", "XX"), "z": ("ss", "zh"), "Z": ("SS", "ZH"), } def get1toXReplacement (cPrev, cCur, cNext): if cCur in aConsonant and (cPrev in aConsonant or cNext in aConsonant): return () return d1toX.get(cCur, ()) d2toX = { "an": ("en",), "AN": ("EN",), "au": ("eau", "o", "ô"), "AU": ("EAU", "O", "Ô"), "en": ("an",), |
︙ |
Modified gc_core/py/ibdawg.py from [a0b40e49a1] to [dcdab2db72].
1 2 3 4 5 6 7 8 9 10 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 | - - + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + | #!python3 import os import traceback import pkgutil import re from itertools import chain from functools import wraps import time |
︙ | |||
200 201 202 203 204 205 206 207 208 | 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 | + + - + - + - - - - - + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + - + - - - - - + + + + + + - - - + + - - - - - + + + + - - - - - - - + - - - - - - - - - - - - - - + + + - - - - - - - - - - - - + + + + + + + + - + - - + + - - - - - - + + + + + | l.extend(self.morph(sWord.capitalize())) return l @timethis def suggest (self, sWord, nMaxSugg=10): "returns a set of suggestions for <sWord>" sPfx, sWord, sSfx = cp.cut(sWord) nMaxSwitch = max(len(sWord) // 3, 1) nMaxDel = len(sWord) // 5 nMaxHardRepl = max((len(sWord) - 5) // 4, 1) oSuggResult = SuggResult(sWord) |
︙ |
Modified gc_core/py/str_transform.py from [23bc31f0e4] to [3d25d1270b].
1 2 3 4 5 6 7 8 9 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 | + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + | #!python3 #### DISTANCE CALCULATIONS def longestCommonSubstring (s1, s2): # http://en.wikipedia.org/wiki/Longest_common_substring_problem # http://en.wikibooks.org/wiki/Algorithm_implementation/Strings/Longest_common_substring M = [ [0]*(1+len(s2)) for i in range(1+len(s1)) ] longest, x_longest = 0, 0 for x in range(1, 1+len(s1)): for y in range(1, 1+len(s2)): if s1[x-1] == s2[y-1]: M[x][y] = M[x-1][y-1] + 1 if M[x][y] > longest: longest = M[x][y] x_longest = x else: M[x][y] = 0 return s1[x_longest-longest : x_longest] def distanceDamerauLevenshtein (s1, s2): "distance of Damerau-Levenshtein between <s1> and <s2>" # https://fr.wikipedia.org/wiki/Distance_de_Damerau-Levenshtein d = {} nLen1 = len(s1) nLen2 = len(s2) for i in range(-1, nLen1+1): d[i, -1] = i + 1 for j in range(-1, nLen2+1): d[-1, j] = j + 1 for i in range(nLen1): for j in range(nLen2): nCost = 0 if s1[i] == s2[j] else 1 d[i, j] = min( d[i-1, j] + 1, # Deletion d[i, j-1] + 1, # Insertion d[i-1, j-1] + nCost, # Substitution ) if i and j and s1[i] == s2[j-1] and s1[i-1] == s2[j]: d[i, j] = min(d[i, j], d[i-2, j-2] + nCost) # Transposition return d[nLen1-1, nLen2-1] def distanceSift4 (s1, s2, nMaxOffset=5): "implementation of general Sift4." # https://siderite.blogspot.com/2014/11/super-fast-and-accurate-string-distance.html if not s1: return len(s2) if not s2: return len(s1) nLen1, nLen2 = len(s1), len(s2) i1, i2 = 0, 0 # Cursors for each string nLargestCS = 0 # Largest common substring nLocalCS = 0 # Local common substring nTrans = 0 # Number of transpositions ('ab' vs 'ba') lOffset = [] # Offset pair array, for computing the transpositions while i1 < nLen1 and i2 < nLen2: if s1[i1] == s2[i2]: nLocalCS += 1 # Check if current match is a transposition bTrans = False i = 0 while i < len(lOffset): t = lOffset[i] if i1 <= t[0] or i2 <= t[1]: bTrans = abs(i2-i1) >= abs(t[1] - t[0]) if bTrans: nTrans += 1 elif not t[2]: t[2] = True nTrans += 1 break elif i1 > t[1] and i2 > t[0]: del lOffset[i] else: i += 1 lOffset.append([i1, i2, bTrans]) else: nLargestCS += nLocalCS nLocalCS = 0 if i1 != i2: i1 = i2 = min(i1, i2) for i in range(nMaxOffset): if i1 + i >= nLen1 and i2 + i >= nLen2: break elif i1 + i < nLen1 and s1[i1+i] == s2[i2]: i1 += i - 1 i2 -= 1 break elif i2 + i < nLen2 and s1[i1] == s2[i2+i]: i2 += i - 1 i1 -= 1 break i1 += 1 i2 += 1 if i1 >= nLen1 or i2 >= nLen2: nLargestCS += nLocalCS nLocalCS = 0 i1 = i2 = min(i1, i2) nLargestCS += nLocalCS return round(max(nLen1, nLen2) - nLargestCS + nTrans) def showDistance (s1, s2): print(f"Damerau-Levenshtein: {s1} / {s2} = {distanceDamerauLevenshtein(s1, s2)}") print(f"Sift4: {s1} / {s2} = {distanceSift4(s1, s2)}") #### STEMMING OPERATIONS ## No stemming def noStemming (sFlex, sStem): return sStem def rebuildWord (sFlex, cmd1, cmd2): if cmd1 == "_": |
︙ | |||
36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 | 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 | + + | return "0" jSfx = 0 for i in range(min(len(sFlex), len(sStem))): if sFlex[i] != sStem[i]: break jSfx += 1 return chr(len(sFlex)-jSfx+48) + sStem[jSfx:] def changeWordWithSuffixCode (sWord, sSfxCode): if sSfxCode == "0": return sWord return sWord[:-(ord(sSfxCode[0])-48)] + sSfxCode[1:] if sSfxCode[0] != '0' else sWord + sSfxCode[1:] # Prefix and suffix def defineAffixCode (sFlex, sStem): """ Returns a string defining how to get stem from flexion. Examples: "0" if stem = flexion "stem" if no common substring "n(pfx)/m(sfx)" with n and m: chars with numeric meaning, "0" = 0, "1" = 1, ... ":" = 10, etc. (See ASCII table.) Says how many letters to strip from flexion. pfx [optional]: string to add before the flexion |
︙ | |||
72 73 74 75 76 77 78 | 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 | - - - - - - - - - - - - - - - | n = sFlex.find(sSubs) m = len(sFlex) - (len(sSubs)+n) sAff = "{}/".format(chr(n+48)) if not sPfx else "{}{}/".format(chr(n+48), sPfx) sAff += chr(m+48) if not sSfx else "{}{}".format(chr(m+48), sSfx) return sAff return sStem |
Modified gc_lang/fr/config.ini from [c76d410d86] to [bd9c32e343].
︙ | |||
29 30 31 32 33 34 35 | 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 | - + | oxt_identifier = French.linguistic.resources.from.Dicollecte.by.OlivierR # Firefox fx_identifier = French-GC@grammalecte.net fx_name = Grammalecte [fr] win_fx_dev_path = C:\Program Files\Firefox Developer Edition\firefox.exe |
︙ |