Grammalecte  Check-in [5a15898d11]

Overview
Comment:[core] rename parameter tags -> aTags
Downloads: Tarball | ZIP archive | SQL archive
Timelines: family | ancestors | descendants | both | core | rg
Files: files | file ages | folders
SHA3-256: 5a15898d113cdcbf3ade04bd4ea0e563ad34b463fb19dfd68272ba386229449e
User & Date: olr on 2018-09-12 10:38:22
Other Links: branch diff | manifest | tags
Context
2018-09-12
17:21
[core][js] gc engine: fix several issues check-in: 900ff7dc44 user: olr tags: core, rg
10:38
[core] rename parameter tags -> aTags check-in: 5a15898d11 user: olr tags: core, rg
09:49
[core][js] small code cleaning check-in: 05090f91ec user: olr tags: core, rg
Changes

Modified gc_core/js/lang_core/gc_engine.js from [225ec72077] to [eabdda0297].

183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
...
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
...
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
...
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
....
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
        s += "sentence: " + this.sSentence0 + "\n";
        s += "now:      " + this.sSentence  + "\n";
        for (let dToken of this.lToken) {
            s += `#${dToken["i"]}\t${dToken["nStart"]}:${dToken["nEnd"]}\t${dToken["sValue"]}\t${dToken["sType"]}`;
            if (dToken.hasOwnProperty("lMorph")) {
                s += "\t" + dToken["lMorph"].toString();
            }
            if (dToken.hasOwnProperty("tags")) {
                s += "\t" + dToken["tags"].toString();
            }
            s += "\n";
        }
        return s;
    }

    parse (sCountry="${country_default}", bDebug=false, dOptions=null, bContext=false) {
................................................................................
        // update <sSentence> and retokenize
        this.sSentence = sSentence;
        let lNewToken = Array.from(_oTokenizer.genTokens(sSentence, true));
        for (let dToken of lNewToken) {
            if (this.dTokenPos.gl_get(dToken["nStart"], {}).hasOwnProperty("lMorph")) {
                dToken["lMorph"] = this.dTokenPos.get(dToken["nStart"])["lMorph"];
            }
            if (this.dTokenPos.gl_get(dToken["nStart"], {}).hasOwnProperty("tags")) {
                dToken["tags"] = this.dTokenPos.get(dToken["nStart"])["tags"];
            }
        }
        this.lToken = lNewToken;
        this.dTokenPos.clear();
        for (let dToken of this.lToken) {
            if (dToken["sType"] != "INFO") {
                this.dTokenPos.set(dToken["nStart"], dToken);
................................................................................
                                }
                            }
                        }
                    }
                }
            }
            // token tags
            if (dToken.hasOwnProperty("tags") && dNode.hasOwnProperty("<tags>")) {
                for (let sTag in dToken["tags"]) {
                    if (dNode["<tags>"].hasOwnProperty(sTag)) {
                        if (bDebug) {
                            console.log("  MATCH: /" + sTag);
                        }
                        yield { "iNode1": iNode1, "dNode": dGraph[dNode["<tags>"][sTag]] };
                        bTokenFound = true;
                    }
................................................................................
                                }
                            }
                            else if (cActionType == "/") {
                                // Tag
                                let nTokenStart = (eAct[0] > 0) ? nTokenOffset + eAct[0] : nLastToken + eAct[0];
                                let nTokenEnd = (eAct[1] > 0) ? nTokenOffset + eAct[1] : nLastToken + eAct[1];
                                for (let i = nTokenStart; i <= nTokenEnd; i++) {
                                    if (this.lToken[i].hasOwnProperty("tags")) {
                                        this.lToken[i]["tags"].add(...sWhat.split("|"))
                                    } else {
                                        this.lToken[i]["tags"] = new Set(sWhat.split("|"));
                                    }
                                }
                                if (bDebug) {
                                    console.log(`    TAG:  ${sWhat} > [${this.lToken[nTokenStart]["sValue"]}:${this.lToken[nTokenEnd]["sValue"]}]`);
                                }
                                if (!this.dTags.has(sWhat)) {
                                    this.dTags.set(sWhat, [nTokenStart, nTokenStart]);
................................................................................
    if (dToken["i"] < dTags.get(sTag)[1]) {
        return true;
    }
    return false;
}

function g_tag (dToken, sTag) {
    return dToken.hasOwnProperty("tags") && dToken["tags"].has(sTag);
}

function g_space_between_tokens (dToken1, dToken2, nMin, nMax=null) {
    let nSpace = dToken2["nStart"] - dToken1["nEnd"]
    if (nSpace < nMin) {
        return false;
    }







|
|







 







|
|







 







|
|







 







|
|

|







 







|







183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
...
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
...
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
...
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
....
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
        s += "sentence: " + this.sSentence0 + "\n";
        s += "now:      " + this.sSentence  + "\n";
        for (let dToken of this.lToken) {
            s += `#${dToken["i"]}\t${dToken["nStart"]}:${dToken["nEnd"]}\t${dToken["sValue"]}\t${dToken["sType"]}`;
            if (dToken.hasOwnProperty("lMorph")) {
                s += "\t" + dToken["lMorph"].toString();
            }
            if (dToken.hasOwnProperty("aTags")) {
                s += "\t" + dToken["aTags"].toString();
            }
            s += "\n";
        }
        return s;
    }

    parse (sCountry="${country_default}", bDebug=false, dOptions=null, bContext=false) {
................................................................................
        // update <sSentence> and retokenize
        this.sSentence = sSentence;
        let lNewToken = Array.from(_oTokenizer.genTokens(sSentence, true));
        for (let dToken of lNewToken) {
            if (this.dTokenPos.gl_get(dToken["nStart"], {}).hasOwnProperty("lMorph")) {
                dToken["lMorph"] = this.dTokenPos.get(dToken["nStart"])["lMorph"];
            }
            if (this.dTokenPos.gl_get(dToken["nStart"], {}).hasOwnProperty("aTags")) {
                dToken["aTags"] = this.dTokenPos.get(dToken["nStart"])["aTags"];
            }
        }
        this.lToken = lNewToken;
        this.dTokenPos.clear();
        for (let dToken of this.lToken) {
            if (dToken["sType"] != "INFO") {
                this.dTokenPos.set(dToken["nStart"], dToken);
................................................................................
                                }
                            }
                        }
                    }
                }
            }
            // token tags
            if (dToken.hasOwnProperty("aTags") && dNode.hasOwnProperty("<tags>")) {
                for (let sTag in dToken["aTags"]) {
                    if (dNode["<tags>"].hasOwnProperty(sTag)) {
                        if (bDebug) {
                            console.log("  MATCH: /" + sTag);
                        }
                        yield { "iNode1": iNode1, "dNode": dGraph[dNode["<tags>"][sTag]] };
                        bTokenFound = true;
                    }
................................................................................
                                }
                            }
                            else if (cActionType == "/") {
                                // Tag
                                let nTokenStart = (eAct[0] > 0) ? nTokenOffset + eAct[0] : nLastToken + eAct[0];
                                let nTokenEnd = (eAct[1] > 0) ? nTokenOffset + eAct[1] : nLastToken + eAct[1];
                                for (let i = nTokenStart; i <= nTokenEnd; i++) {
                                    if (this.lToken[i].hasOwnProperty("aTags")) {
                                        this.lToken[i]["aTags"].add(...sWhat.split("|"))
                                    } else {
                                        this.lToken[i]["aTags"] = new Set(sWhat.split("|"));
                                    }
                                }
                                if (bDebug) {
                                    console.log(`    TAG:  ${sWhat} > [${this.lToken[nTokenStart]["sValue"]}:${this.lToken[nTokenEnd]["sValue"]}]`);
                                }
                                if (!this.dTags.has(sWhat)) {
                                    this.dTags.set(sWhat, [nTokenStart, nTokenStart]);
................................................................................
    if (dToken["i"] < dTags.get(sTag)[1]) {
        return true;
    }
    return false;
}

function g_tag (dToken, sTag) {
    return dToken.hasOwnProperty("aTags") && dToken["aTags"].has(sTag);
}

function g_space_between_tokens (dToken1, dToken2, nMin, nMax=null) {
    let nSpace = dToken2["nStart"] - dToken1["nEnd"]
    if (nSpace < nMin) {
        return false;
    }

Modified gc_core/py/lang_core/gc_engine.py from [a6af5aa7ed] to [b63b69316e].

229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
...
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
...
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
...
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
....
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
        s = "===== TEXT =====\n"
        s += "sentence: " + self.sSentence0 + "\n"
        s += "now:      " + self.sSentence  + "\n"
        for dToken in self.lToken:
            s += '#{i}\t{nStart}:{nEnd}\t{sValue}\t{sType}'.format(**dToken)
            if "lMorph" in dToken:
                s += "\t" + str(dToken["lMorph"])
            if "tags" in dToken:
                s += "\t" + str(dToken["tags"])
            s += "\n"
        #for nPos, dToken in self.dTokenPos.items():
        #    s += "{}\t{}\n".format(nPos, dToken)
        return s

    def parse (self, sCountry="${country_default}", bDebug=False, dOptions=None, bContext=False):
        "analyses the paragraph sText and returns list of errors"
................................................................................
    def update (self, sSentence, bDebug=False):
        "update <sSentence> and retokenize"
        self.sSentence = sSentence
        lNewToken = list(_oTokenizer.genTokens(sSentence, True))
        for dToken in lNewToken:
            if "lMorph" in self.dTokenPos.get(dToken["nStart"], {}):
                dToken["lMorph"] = self.dTokenPos[dToken["nStart"]]["lMorph"]
            if "tags" in self.dTokenPos.get(dToken["nStart"], {}):
                dToken["tags"] = self.dTokenPos[dToken["nStart"]]["tags"]
        self.lToken = lNewToken
        self.dTokenPos = { dToken["nStart"]: dToken  for dToken in self.lToken  if dToken["sType"] != "INFO" }
        if bDebug:
            echo("UPDATE:")
            echo(self)

    def _getNextPointers (self, dToken, dGraph, dPointer, bDebug=False):
................................................................................
                                continue
                            if not sPattern or any(re.search(sPattern, sMorph)  for sMorph in lMorph):
                                if bDebug:
                                    echo("  MATCH: @" + sRegex)
                                yield { "iNode1": iNode1, "dNode": dGraph[dNode["<re_morph>"][sRegex]] }
                                bTokenFound = True
        # token tags
        if "tags" in dToken and "<tags>" in dNode:
            for sTag in dToken["tags"]:
                if sTag in dNode["<tags>"]:
                    if bDebug:
                        echo("  MATCH: /" + sTag)
                    yield { "iNode1": iNode1, "dNode": dGraph[dNode["<tags>"][sTag]] }
                    bTokenFound = True
        # meta arc (for token type)
        if "<meta>" in dNode:
................................................................................
                                    echo("    COND_OK")
                                pass
                            elif cActionType == "/":
                                # Tag
                                nTokenStart = nTokenOffset + eAct[0]  if eAct[0] > 0  else nLastToken + eAct[0]
                                nTokenEnd = nTokenOffset + eAct[1]  if eAct[1] > 0  else nLastToken + eAct[1]
                                for i in range(nTokenStart, nTokenEnd+1):
                                    if "tags" in self.lToken[i]:
                                        self.lToken[i]["tags"].update(sWhat.split("|"))
                                    else:
                                        self.lToken[i]["tags"] = set(sWhat.split("|"))
                                if bDebug:
                                    echo("    TAG: {} >  [{}:{}]".format(sWhat, self.lToken[nTokenStart]["sValue"], self.lToken[nTokenEnd]["sValue"]))
                                if sWhat not in self.dTags:
                                    self.dTags[sWhat] = [nTokenStart, nTokenStart]
                                else:
                                    self.dTags[sWhat][0] = min(nTokenStart, self.dTags[sWhat][0])
                                    self.dTags[sWhat][1] = max(nTokenEnd, self.dTags[sWhat][1])
................................................................................
        return False
    if dToken["i"] < dTags[sTag][1]:
        return True
    return False


def g_tag (dToken, sTag):
    return "tags" in dToken and sTag in dToken["tags"]


def g_space_between_tokens (dToken1, dToken2, nMin, nMax=None):
    nSpace = dToken2["nStart"] - dToken1["nEnd"]
    if nSpace < nMin:
        return False
    if nMax is not None and nSpace > nMax:







|
|







 







|
|







 







|
|







 







|
|

|







 







|







229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
...
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
...
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
...
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
....
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
        s = "===== TEXT =====\n"
        s += "sentence: " + self.sSentence0 + "\n"
        s += "now:      " + self.sSentence  + "\n"
        for dToken in self.lToken:
            s += '#{i}\t{nStart}:{nEnd}\t{sValue}\t{sType}'.format(**dToken)
            if "lMorph" in dToken:
                s += "\t" + str(dToken["lMorph"])
            if "aTags" in dToken:
                s += "\t" + str(dToken["aTags"])
            s += "\n"
        #for nPos, dToken in self.dTokenPos.items():
        #    s += "{}\t{}\n".format(nPos, dToken)
        return s

    def parse (self, sCountry="${country_default}", bDebug=False, dOptions=None, bContext=False):
        "analyses the paragraph sText and returns list of errors"
................................................................................
    def update (self, sSentence, bDebug=False):
        "update <sSentence> and retokenize"
        self.sSentence = sSentence
        lNewToken = list(_oTokenizer.genTokens(sSentence, True))
        for dToken in lNewToken:
            if "lMorph" in self.dTokenPos.get(dToken["nStart"], {}):
                dToken["lMorph"] = self.dTokenPos[dToken["nStart"]]["lMorph"]
            if "aTags" in self.dTokenPos.get(dToken["nStart"], {}):
                dToken["aTags"] = self.dTokenPos[dToken["nStart"]]["aTags"]
        self.lToken = lNewToken
        self.dTokenPos = { dToken["nStart"]: dToken  for dToken in self.lToken  if dToken["sType"] != "INFO" }
        if bDebug:
            echo("UPDATE:")
            echo(self)

    def _getNextPointers (self, dToken, dGraph, dPointer, bDebug=False):
................................................................................
                                continue
                            if not sPattern or any(re.search(sPattern, sMorph)  for sMorph in lMorph):
                                if bDebug:
                                    echo("  MATCH: @" + sRegex)
                                yield { "iNode1": iNode1, "dNode": dGraph[dNode["<re_morph>"][sRegex]] }
                                bTokenFound = True
        # token tags
        if "aTags" in dToken and "<tags>" in dNode:
            for sTag in dToken["aTags"]:
                if sTag in dNode["<tags>"]:
                    if bDebug:
                        echo("  MATCH: /" + sTag)
                    yield { "iNode1": iNode1, "dNode": dGraph[dNode["<tags>"][sTag]] }
                    bTokenFound = True
        # meta arc (for token type)
        if "<meta>" in dNode:
................................................................................
                                    echo("    COND_OK")
                                pass
                            elif cActionType == "/":
                                # Tag
                                nTokenStart = nTokenOffset + eAct[0]  if eAct[0] > 0  else nLastToken + eAct[0]
                                nTokenEnd = nTokenOffset + eAct[1]  if eAct[1] > 0  else nLastToken + eAct[1]
                                for i in range(nTokenStart, nTokenEnd+1):
                                    if "aTags" in self.lToken[i]:
                                        self.lToken[i]["aTags"].update(sWhat.split("|"))
                                    else:
                                        self.lToken[i]["aTags"] = set(sWhat.split("|"))
                                if bDebug:
                                    echo("    TAG: {} >  [{}:{}]".format(sWhat, self.lToken[nTokenStart]["sValue"], self.lToken[nTokenEnd]["sValue"]))
                                if sWhat not in self.dTags:
                                    self.dTags[sWhat] = [nTokenStart, nTokenStart]
                                else:
                                    self.dTags[sWhat][0] = min(nTokenStart, self.dTags[sWhat][0])
                                    self.dTags[sWhat][1] = max(nTokenEnd, self.dTags[sWhat][1])
................................................................................
        return False
    if dToken["i"] < dTags[sTag][1]:
        return True
    return False


def g_tag (dToken, sTag):
    return "aTags" in dToken and sTag in dToken["aTags"]


def g_space_between_tokens (dToken1, dToken2, nMin, nMax=None):
    nSpace = dToken2["nStart"] - dToken1["nEnd"]
    if nSpace < nMin:
        return False
    if nMax is not None and nSpace > nMax: