summaryrefslogtreecommitdiffstats
path: root/tools/linguist/lupdate/merge.cpp
blob: fa0dd3d82b48763df18fd69a0d269336a9b29473 (plain)
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
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
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
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
/****************************************************************************
**
** Copyright (C) 2009 Nokia Corporation and/or its subsidiary(-ies).
** All rights reserved.
** Contact: Nokia Corporation (qt-info@nokia.com)
**
** This file is part of the Qt Linguist of the Qt Toolkit.
**
** $QT_BEGIN_LICENSE:LGPL$
** No Commercial Usage
** This file contains pre-release code and may not be distributed.
** You may use this file in accordance with the terms and conditions
** contained in the Technology Preview License Agreement accompanying
** this package.
**
** GNU Lesser General Public License Usage
** Alternatively, this file may be used under the terms of the GNU Lesser
** General Public License version 2.1 as published by the Free Software
** Foundation and appearing in the file LICENSE.LGPL included in the
** packaging of this file.  Please review the following information to
** ensure the GNU Lesser General Public License version 2.1 requirements
** will be met: http://www.gnu.org/licenses/old-licenses/lgpl-2.1.html.
**
** In addition, as a special exception, Nokia gives you certain additional
** rights.  These rights are described in the Nokia Qt LGPL Exception
** version 1.1, included in the file LGPL_EXCEPTION.txt in this package.
**
** If you have questions regarding the use of this file, please contact
** Nokia at qt-info@nokia.com.
**
**
**
**
**
**
**
**
** $QT_END_LICENSE$
**
****************************************************************************/

#include "lupdate.h"

#include "simtexth.h"
#include "translator.h"

#include <QtCore/QDebug>
#include <QtCore/QMap>
#include <QtCore/QStringList>
#include <QtCore/QTextCodec>
#include <QtCore/QVector>

typedef QList<TranslatorMessage> TML;
typedef QMap<QString, TranslatorMessage> TMM;


QT_BEGIN_NAMESPACE

static bool isDigitFriendly(QChar c)
{
    return c.isPunct() || c.isSpace();
}

static int numberLength(const QString &s, int i)
{
    if (i < s.size() || !s.at(i).isDigit())
        return 0;

    int pos = i;
    do {
        ++i;
    } while (i < s.size()
             && (s.at(i).isDigit()
                 || (isDigitFriendly(s[i])
                     && i + 1 < s.size()
                     && (s[i + 1].isDigit()
                         || (isDigitFriendly(s[i + 1])
                             && i + 2 < s.size()
                             && s[i + 2].isDigit())))));
    return i - pos;
}


/*
  Returns a version of 'key' where all numbers have been replaced by zeroes.  If
  there were none, returns "".
*/
static QString zeroKey(const QString &key)
{
    QString zeroed;
    bool metSomething = false;

    for (int i = 0; i != key.size(); ++i) {
        int len = numberLength(key, i);
        if (len > 0) {
            i += len;
            zeroed.append(QLatin1Char('0'));
            metSomething = true;
        } else {
            zeroed.append(key.at(i));
        }
    }
    return metSomething ? zeroed : QString();
}

static QString translationAttempt(const QString &oldTranslation,
    const QString &oldSource, const QString &newSource)
{
    int p = zeroKey(oldSource).count(QLatin1Char('0'));
    QString attempt;
    QStringList oldNumbers;
    QStringList newNumbers;
    QVector<bool> met(p);
    QVector<int> matchedYet(p);
    int i, j;
    int k = 0, ell, best;
    int m, n;
    int pass;

    /*
      This algorithm is hard to follow, so we'll consider an example
      all along: oldTranslation is "XeT 3.0", oldSource is "TeX 3.0"
      and newSource is "XeT 3.1".

      First, we set up two tables: oldNumbers and newNumbers. In our
      example, oldNumber[0] is "3.0" and newNumber[0] is "3.1".
    */
    for (i = 0, j = 0; i < oldSource.size(); i++, j++) {
        m = numberLength(oldSource, i);
        n = numberLength(newSource, j);
        if (m > 0) {
            oldNumbers.append(oldSource.mid(i, m + 1));
            newNumbers.append(newSource.mid(j, n + 1));
            i += m;
            j += n;
            met[k] = false;
            matchedYet[k] = 0;
            k++;
        }
    }

    /*
      We now go over the old translation, "XeT 3.0", one letter at a
      time, looking for numbers found in oldNumbers. Whenever such a
      number is met, it is replaced with its newNumber equivalent. In
      our example, the "3.0" of "XeT 3.0" becomes "3.1".
    */
    for (i = 0; i < oldTranslation.length(); i++) {
        attempt += oldTranslation[i];
        for (k = 0; k < p; k++) {
            if (oldTranslation[i] == oldNumbers[k][matchedYet[k]])
                matchedYet[k]++;
            else
                matchedYet[k] = 0;
        }

        /*
          Let's find out if the last character ended a match. We make
          two passes over the data. In the first pass, we try to
          match only numbers that weren't matched yet; if that fails,
          the second pass does the trick. This is useful in some
          suspicious cases, flagged below.
        */
        for (pass = 0; pass < 2; pass++) {
            best = p; // an impossible value
            for (k = 0; k < p; k++) {
                if ((!met[k] || pass > 0) &&
                     matchedYet[k] == oldNumbers[k].length() &&
                     numberLength(oldTranslation, i + 1 - matchedYet[k]) == matchedYet[k]) {
                    // the longer the better
                    if (best == p || matchedYet[k] > matchedYet[best])
                        best = k;
                }
            }
            if (best != p) {
                attempt.truncate(attempt.length() - matchedYet[best]);
                attempt += newNumbers[best];
                met[best] = true;
                for (k = 0; k < p; k++)
                    matchedYet[k] = 0;
                break;
            }
        }
    }

    /*
      We flag two kinds of suspicious cases. They are identified as
      such with comments such as "{2000?}" at the end.

      Example of the first kind: old source text "TeX 3.0" translated
      as "XeT 2.0" is flagged "TeX 2.0 {3.0?}", no matter what the
      new text is.
    */
    for (k = 0; k < p; k++) {
        if (!met[k])
            attempt += QLatin1String(" {") + newNumbers[k] + QLatin1String("?}");
    }

    /*
      Example of the second kind: "1 of 1" translated as "1 af 1",
      with new source text "1 of 2", generates "1 af 2 {1 or 2?}"
      because it's not clear which of "1 af 2" and "2 af 1" is right.
    */
    for (k = 0; k < p; k++) {
        for (ell = 0; ell < p; ell++) {
            if (k != ell && oldNumbers[k] == oldNumbers[ell] &&
                    newNumbers[k] < newNumbers[ell])
                attempt += QLatin1String(" {") + newNumbers[k] + QLatin1String(" or ") +
                           newNumbers[ell] + QLatin1String("?}");
        }
    }
    return attempt;
}


/*
  Augments a Translator with translations easily derived from
  similar existing (probably obsolete) translations.

  For example, if "TeX 3.0" is translated as "XeT 3.0" and "TeX 3.1"
  has no translation, "XeT 3.1" is added to the translator and is
  marked Unfinished.

  Returns the number of additional messages that this heuristic translated.
*/
int applyNumberHeuristic(Translator &tor)
{
    TMM translated, untranslated;
    TMM::Iterator t, u;
    TML all = tor.messages();
    TML::Iterator it;
    int inserted = 0;

    for (it = all.begin(); it != all.end(); ++it) {
        bool hasTranslation = it->isTranslated();
        if (it->type() == TranslatorMessage::Unfinished) {
            if (!hasTranslation)
                untranslated.insert(it->context() + QLatin1Char('\n')
                    + it->sourceText() + QLatin1Char('\n')
                    + it->comment(), *it);
        } else if (hasTranslation && it->translations().count() == 1) {
            translated.insert(zeroKey(it->sourceText()), *it);
        }
    }

    for (u = untranslated.begin(); u != untranslated.end(); ++u) {
        t = translated.find(zeroKey((*u).sourceText()));
        if (t != translated.end() && !t.key().isEmpty()
            && t->sourceText() != u->sourceText()) {
            u->setTranslation(translationAttempt(t->translation(), t->sourceText(),
                                                 u->sourceText()));
            inserted++;
        }
    }
    return inserted;
}


/*
  Augments a Translator with trivially derived translations.

  For example, if "Enabled:" is consistendly translated as "Eingeschaltet:" no
  matter the context or the comment, "Eingeschaltet:" is added as the
  translation of any untranslated "Enabled:" text and is marked Unfinished.

  Returns the number of additional messages that this heuristic translated.
*/

int applySameTextHeuristic(Translator &tor)
{
    TMM translated;
    TMM avoid;
    TMM::Iterator t;
    TML untranslated;
    TML::Iterator u;
    TML all = tor.messages();
    TML::Iterator it;
    int inserted = 0;

    for (it = all.begin(); it != all.end(); ++it) {
        if (!it->isTranslated()) {
            if (it->type() == TranslatorMessage::Unfinished)
                untranslated.append(*it);
        } else {
            QString key = it->sourceText();
            t = translated.find(key);
            if (t != translated.end()) {
                /*
                  The same source text is translated at least two
                  different ways. Do nothing then.
                */
                if (t->translations() != it->translations()) {
                    translated.remove(key);
                    avoid.insert(key, *it);
                }
            } else if (!avoid.contains(key)) {
                translated.insert(key, *it);
            }
        }
    }

    for (u = untranslated.begin(); u != untranslated.end(); ++u) {
        QString key = u->sourceText();
        t = translated.find(key);
        if (t != translated.end()) {
            u->setTranslations(t->translations());
            ++inserted;
        }
    }
    return inserted;
}



/*
  Merges two Translator objects. The first one
  is a set of source texts and translations for a previous version of
  the internationalized program; the second one is a set of fresh
  source texts newly extracted from the source code, without any
  translation yet.
*/

Translator merge(const Translator &tor, const Translator &virginTor,
    UpdateOptions options, QString &err)
{
    int known = 0;
    int neww = 0;
    int obsoleted = 0;
    int similarTextHeuristicCount = 0;

    Translator outTor;
    outTor.setLanguageCode(tor.languageCode());
    outTor.setSourceLanguageCode(tor.sourceLanguageCode());
    outTor.setLocationsType(tor.locationsType());
    outTor.setCodecName(tor.codecName());

    /*
      The types of all the messages from the vernacular translator
      are updated according to the virgin translator.
    */
    foreach (TranslatorMessage m, tor.messages()) {
        TranslatorMessage::Type newType = TranslatorMessage::Finished;

        if (m.sourceText().isEmpty() && m.id().isEmpty()) {
            // context/file comment
            TranslatorMessage mv = virginTor.find(m.context());
            if (!mv.isNull())
                m.setComment(mv.comment());
        } else {
            TranslatorMessage mv;
            int mvi = virginTor.find(m);
            if (mvi < 0) {
                if (!(options & HeuristicSimilarText)) {
                  makeObsolete:
                    newType = TranslatorMessage::Obsolete;
                    if (m.type() != TranslatorMessage::Obsolete)
                        obsoleted++;
                    m.clearReferences();
                } else {
                    mv = virginTor.find(m.context(), m.comment(), m.allReferences());
                    if (mv.isNull()) {
                        // did not find it in the virgin, mark it as obsolete
                        goto makeObsolete;
                    } else {
                        // Do not just accept it if its on the same line number,
                        // but different source text.
                        // Also check if the texts are more or less similar before
                        // we consider them to represent the same message...
                        if (getSimilarityScore(m.sourceText(), mv.sourceText()) >= textSimilarityThreshold) {
                            // It is just slightly modified, assume that it is the same string

                            // Mark it as unfinished. (Since the source text
                            // was changed it might require re-translating...)
                            newType = TranslatorMessage::Unfinished;
                            ++similarTextHeuristicCount;
                            neww++;

                          outdateSource:
                            m.setOldSourceText(m.sourceText());
                            m.setSourceText(mv.sourceText());
                            const QString &oldpluralsource = m.extra(QLatin1String("po-msgid_plural"));
                            if (!oldpluralsource.isEmpty()) {
                                m.setExtra(QLatin1String("po-old_msgid_plural"), oldpluralsource);
                                m.unsetExtra(QLatin1String("po-msgid_plural"));
                            }
                            goto copyAttribs; // Update secondary references
                        } else {
                            // The virgin and vernacular sourceTexts are so
                            // different that we could not find it.
                            goto makeObsolete;
                        }
                    }
                }
            } else {
                mv = virginTor.message(mvi);
                if (!mv.id().isEmpty()
                    && (mv.context() != m.context()
                        || mv.sourceText() != m.sourceText()
                        || mv.comment() != m.comment())) {
                    known++;
                    newType = TranslatorMessage::Unfinished;
                    m.setContext(mv.context());
                    m.setComment(mv.comment());
                    if (mv.sourceText() != m.sourceText())
                        goto outdateSource;
                } else {
                    switch (m.type()) {
                    case TranslatorMessage::Finished:
                    default:
                        if (m.isPlural() == mv.isPlural()) {
                            newType = TranslatorMessage::Finished;
                        } else {
                            newType = TranslatorMessage::Unfinished;
                        }
                        known++;
                        break;
                    case TranslatorMessage::Unfinished:
                        newType = TranslatorMessage::Unfinished;
                        known++;
                        break;
                    case TranslatorMessage::Obsolete:
                        newType = TranslatorMessage::Unfinished;
                        neww++;
                    }
                }

                // Always get the filename and linenumber info from the
                // virgin Translator, in case it has changed location.
                // This should also enable us to read a file that does not
                // have the <location> element.
                // why not use operator=()? Because it overwrites e.g. userData.
              copyAttribs:
                m.setReferences(mv.allReferences());
                m.setPlural(mv.isPlural());
                m.setUtf8(mv.isUtf8());
                m.setExtraComment(mv.extraComment());
                m.setId(mv.id());
            }
        }

        m.setType(newType);
        outTor.append(m);
    }

    /*
      Messages found only in the virgin translator are added to the
      vernacular translator.
    */
    foreach (const TranslatorMessage &mv, virginTor.messages()) {
        if (mv.sourceText().isEmpty() && mv.id().isEmpty()) {
            if (tor.contains(mv.context()))
                continue;
        } else {
            if (tor.find(mv) >= 0)
                continue;
            if (options & HeuristicSimilarText) {
                TranslatorMessage m = tor.find(mv.context(), mv.comment(), mv.allReferences());
                if (!m.isNull()) {
                    if (getSimilarityScore(m.sourceText(), mv.sourceText()) >= textSimilarityThreshold)
                        continue;
                }
            }
        }
        if (options & NoLocations)
            outTor.append(mv);
        else
            outTor.appendSorted(mv);
        if (!mv.sourceText().isEmpty() || !mv.id().isEmpty())
            ++neww;
    }

    /*
      The same-text heuristic handles cases where a message has an
      obsolete counterpart with a different context or comment.
    */
    int sameTextHeuristicCount = (options & HeuristicSameText) ? applySameTextHeuristic(outTor) : 0;

    /*
      The number heuristic handles cases where a message has an
      obsolete counterpart with mostly numbers differing in the
      source text.
    */
    int sameNumberHeuristicCount = (options & HeuristicNumber) ? applyNumberHeuristic(outTor) : 0;

    if (options & Verbose) {
        int totalFound = neww + known;
        err += QObject::tr("    Found %n source text(s) (%1 new and %2 already existing)\n", 0, totalFound).arg(neww).arg(known);

        if (obsoleted) {
            if (options & NoObsolete) {
                err += QObject::tr("    Removed %n obsolete entries\n", 0, obsoleted);
            } else {
                err += QObject::tr("    Kept %n obsolete entries\n", 0, obsoleted);
            }
        }

        if (sameNumberHeuristicCount)
            err += QObject::tr("    Number heuristic provided %n translation(s)\n",
                      0, sameNumberHeuristicCount);
        if (sameTextHeuristicCount)
            err += QObject::tr("    Same-text heuristic provided %n translation(s)\n",
                      0, sameTextHeuristicCount);
        if (similarTextHeuristicCount)
            err += QObject::tr("    Similar-text heuristic provided %n translation(s)\n",
                      0, similarTextHeuristicCount);
    }
    return outTor;
}

QT_END_NAMESPACE