1 | /****************************************************************************
|
---|
2 | **
|
---|
3 | ** Copyright (C) 2010 Nokia Corporation and/or its subsidiary(-ies).
|
---|
4 | ** All rights reserved.
|
---|
5 | ** Contact: Nokia Corporation (qt-info@nokia.com)
|
---|
6 | **
|
---|
7 | ** This file is part of the Qt Linguist of the Qt Toolkit.
|
---|
8 | **
|
---|
9 | ** $QT_BEGIN_LICENSE:LGPL$
|
---|
10 | ** Commercial Usage
|
---|
11 | ** Licensees holding valid Qt Commercial licenses may use this file in
|
---|
12 | ** accordance with the Qt Commercial License Agreement provided with the
|
---|
13 | ** Software or, alternatively, in accordance with the terms contained in
|
---|
14 | ** a written agreement between you and Nokia.
|
---|
15 | **
|
---|
16 | ** GNU Lesser General Public License Usage
|
---|
17 | ** Alternatively, this file may be used under the terms of the GNU Lesser
|
---|
18 | ** General Public License version 2.1 as published by the Free Software
|
---|
19 | ** Foundation and appearing in the file LICENSE.LGPL included in the
|
---|
20 | ** packaging of this file. Please review the following information to
|
---|
21 | ** ensure the GNU Lesser General Public License version 2.1 requirements
|
---|
22 | ** will be met: http://www.gnu.org/licenses/old-licenses/lgpl-2.1.html.
|
---|
23 | **
|
---|
24 | ** In addition, as a special exception, Nokia gives you certain additional
|
---|
25 | ** rights. These rights are described in the Nokia Qt LGPL Exception
|
---|
26 | ** version 1.1, included in the file LGPL_EXCEPTION.txt in this package.
|
---|
27 | **
|
---|
28 | ** GNU General Public License Usage
|
---|
29 | ** Alternatively, this file may be used under the terms of the GNU
|
---|
30 | ** General Public License version 3.0 as published by the Free Software
|
---|
31 | ** Foundation and appearing in the file LICENSE.GPL included in the
|
---|
32 | ** packaging of this file. Please review the following information to
|
---|
33 | ** ensure the GNU General Public License version 3.0 requirements will be
|
---|
34 | ** met: http://www.gnu.org/copyleft/gpl.html.
|
---|
35 | **
|
---|
36 | ** If you have questions regarding the use of this file, please contact
|
---|
37 | ** Nokia at qt-info@nokia.com.
|
---|
38 | ** $QT_END_LICENSE$
|
---|
39 | **
|
---|
40 | ****************************************************************************/
|
---|
41 |
|
---|
42 | #include "translator.h"
|
---|
43 |
|
---|
44 | #ifndef QT_BOOTSTRAPPED
|
---|
45 | #include <QtCore/QCoreApplication>
|
---|
46 | #endif
|
---|
47 | #include <QtCore/QDebug>
|
---|
48 | #include <QtCore/QDir>
|
---|
49 | #include <QtCore/QFile>
|
---|
50 | #include <QtCore/QFileInfo>
|
---|
51 | #include <QtCore/QMap>
|
---|
52 | #include <QtCore/QString>
|
---|
53 | #include <QtCore/QTextCodec>
|
---|
54 |
|
---|
55 | QT_BEGIN_NAMESPACE
|
---|
56 |
|
---|
57 | // magic number for the file
|
---|
58 | static const int MagicLength = 16;
|
---|
59 | static const uchar magic[MagicLength] = {
|
---|
60 | 0x3c, 0xb8, 0x64, 0x18, 0xca, 0xef, 0x9c, 0x95,
|
---|
61 | 0xcd, 0x21, 0x1c, 0xbf, 0x60, 0xa1, 0xbd, 0xdd
|
---|
62 | };
|
---|
63 |
|
---|
64 |
|
---|
65 | namespace {
|
---|
66 |
|
---|
67 | enum Tag {
|
---|
68 | Tag_End = 1,
|
---|
69 | Tag_SourceText16 = 2,
|
---|
70 | Tag_Translation = 3,
|
---|
71 | Tag_Context16 = 4,
|
---|
72 | Tag_Obsolete1 = 5,
|
---|
73 | Tag_SourceText = 6,
|
---|
74 | Tag_Context = 7,
|
---|
75 | Tag_Comment = 8,
|
---|
76 | Tag_Obsolete2 = 9
|
---|
77 | };
|
---|
78 |
|
---|
79 | enum Prefix {
|
---|
80 | NoPrefix,
|
---|
81 | Hash,
|
---|
82 | HashContext,
|
---|
83 | HashContextSourceText,
|
---|
84 | HashContextSourceTextComment
|
---|
85 | };
|
---|
86 |
|
---|
87 | } // namespace anon
|
---|
88 |
|
---|
89 | static uint elfHash(const QByteArray &ba)
|
---|
90 | {
|
---|
91 | const uchar *k = (const uchar *)ba.data();
|
---|
92 | uint h = 0;
|
---|
93 | uint g;
|
---|
94 |
|
---|
95 | if (k) {
|
---|
96 | while (*k) {
|
---|
97 | h = (h << 4) + *k++;
|
---|
98 | if ((g = (h & 0xf0000000)) != 0)
|
---|
99 | h ^= g >> 24;
|
---|
100 | h &= ~g;
|
---|
101 | }
|
---|
102 | }
|
---|
103 | if (!h)
|
---|
104 | h = 1;
|
---|
105 | return h;
|
---|
106 | }
|
---|
107 |
|
---|
108 | class ByteTranslatorMessage
|
---|
109 | {
|
---|
110 | public:
|
---|
111 | ByteTranslatorMessage(
|
---|
112 | const QByteArray &context,
|
---|
113 | const QByteArray &sourceText,
|
---|
114 | const QByteArray &comment,
|
---|
115 | const QStringList &translations) :
|
---|
116 | m_context(context),
|
---|
117 | m_sourcetext(sourceText),
|
---|
118 | m_comment(comment),
|
---|
119 | m_translations(translations)
|
---|
120 | {}
|
---|
121 | const QByteArray &context() const { return m_context; }
|
---|
122 | const QByteArray &sourceText() const { return m_sourcetext; }
|
---|
123 | const QByteArray &comment() const { return m_comment; }
|
---|
124 | const QStringList &translations() const { return m_translations; }
|
---|
125 | bool operator<(const ByteTranslatorMessage& m) const;
|
---|
126 |
|
---|
127 | private:
|
---|
128 | QByteArray m_context;
|
---|
129 | QByteArray m_sourcetext;
|
---|
130 | QByteArray m_comment;
|
---|
131 | QStringList m_translations;
|
---|
132 | };
|
---|
133 |
|
---|
134 | Q_DECLARE_TYPEINFO(ByteTranslatorMessage, Q_MOVABLE_TYPE);
|
---|
135 |
|
---|
136 | bool ByteTranslatorMessage::operator<(const ByteTranslatorMessage& m) const
|
---|
137 | {
|
---|
138 | if (m_context != m.m_context)
|
---|
139 | return m_context < m.m_context;
|
---|
140 | if (m_sourcetext != m.m_sourcetext)
|
---|
141 | return m_sourcetext < m.m_sourcetext;
|
---|
142 | return m_comment < m.m_comment;
|
---|
143 | }
|
---|
144 |
|
---|
145 | class Releaser
|
---|
146 | {
|
---|
147 | public:
|
---|
148 | struct Offset {
|
---|
149 | Offset()
|
---|
150 | : h(0), o(0)
|
---|
151 | {}
|
---|
152 | Offset(uint hash, uint offset)
|
---|
153 | : h(hash), o(offset)
|
---|
154 | {}
|
---|
155 |
|
---|
156 | bool operator<(const Offset &other) const {
|
---|
157 | return (h != other.h) ? h < other.h : o < other.o;
|
---|
158 | }
|
---|
159 | bool operator==(const Offset &other) const {
|
---|
160 | return h == other.h && o == other.o;
|
---|
161 | }
|
---|
162 | uint h;
|
---|
163 | uint o;
|
---|
164 | };
|
---|
165 |
|
---|
166 | enum { Contexts = 0x2f, Hashes = 0x42, Messages = 0x69, NumerusRules = 0x88 };
|
---|
167 |
|
---|
168 | Releaser() : m_codec(0) {}
|
---|
169 |
|
---|
170 | void setCodecName(const QByteArray &codecName)
|
---|
171 | {
|
---|
172 | m_codec = QTextCodec::codecForName(codecName);
|
---|
173 | }
|
---|
174 |
|
---|
175 | bool save(QIODevice *iod);
|
---|
176 |
|
---|
177 | void insert(const TranslatorMessage &msg, const QStringList &tlns, bool forceComment);
|
---|
178 | void insertIdBased(const TranslatorMessage &message, const QStringList &tlns);
|
---|
179 |
|
---|
180 | void squeeze(TranslatorSaveMode mode);
|
---|
181 |
|
---|
182 | void setNumerusRules(const QByteArray &rules);
|
---|
183 |
|
---|
184 | private:
|
---|
185 | Q_DISABLE_COPY(Releaser)
|
---|
186 |
|
---|
187 | // This should reproduce the byte array fetched from the source file, which
|
---|
188 | // on turn should be the same as passed to the actual tr(...) calls
|
---|
189 | QByteArray originalBytes(const QString &str, bool isUtf8) const;
|
---|
190 |
|
---|
191 | void insertInternal(const TranslatorMessage &message, const QStringList &tlns,
|
---|
192 | bool forceComment, bool isUtf8);
|
---|
193 |
|
---|
194 | static Prefix commonPrefix(const ByteTranslatorMessage &m1, const ByteTranslatorMessage &m2);
|
---|
195 |
|
---|
196 | static uint msgHash(const ByteTranslatorMessage &msg);
|
---|
197 |
|
---|
198 | void writeMessage(const ByteTranslatorMessage & msg, QDataStream & stream,
|
---|
199 | TranslatorSaveMode strip, Prefix prefix) const;
|
---|
200 |
|
---|
201 | // for squeezed but non-file data, this is what needs to be deleted
|
---|
202 | QByteArray m_messageArray;
|
---|
203 | QByteArray m_offsetArray;
|
---|
204 | QByteArray m_contextArray;
|
---|
205 | QMap<ByteTranslatorMessage, void *> m_messages;
|
---|
206 | QByteArray m_numerusRules;
|
---|
207 |
|
---|
208 | // Used to reproduce the original bytes
|
---|
209 | QTextCodec *m_codec;
|
---|
210 | };
|
---|
211 |
|
---|
212 | QByteArray Releaser::originalBytes(const QString &str, bool isUtf8) const
|
---|
213 | {
|
---|
214 | if (str.isEmpty()) {
|
---|
215 | // Do not use QByteArray() here as the result of the serialization
|
---|
216 | // will be different.
|
---|
217 | return QByteArray("");
|
---|
218 | }
|
---|
219 | if (isUtf8)
|
---|
220 | return str.toUtf8();
|
---|
221 | return m_codec ? m_codec->fromUnicode(str) : str.toLatin1();
|
---|
222 | }
|
---|
223 |
|
---|
224 | uint Releaser::msgHash(const ByteTranslatorMessage &msg)
|
---|
225 | {
|
---|
226 | return elfHash(msg.sourceText() + msg.comment());
|
---|
227 | }
|
---|
228 |
|
---|
229 | Prefix Releaser::commonPrefix(const ByteTranslatorMessage &m1, const ByteTranslatorMessage &m2)
|
---|
230 | {
|
---|
231 | if (msgHash(m1) != msgHash(m2))
|
---|
232 | return NoPrefix;
|
---|
233 | if (m1.context() != m2.context())
|
---|
234 | return Hash;
|
---|
235 | if (m1.sourceText() != m2.sourceText())
|
---|
236 | return HashContext;
|
---|
237 | if (m1.comment() != m2.comment())
|
---|
238 | return HashContextSourceText;
|
---|
239 | return HashContextSourceTextComment;
|
---|
240 | }
|
---|
241 |
|
---|
242 | void Releaser::writeMessage(const ByteTranslatorMessage &msg, QDataStream &stream,
|
---|
243 | TranslatorSaveMode mode, Prefix prefix) const
|
---|
244 | {
|
---|
245 | for (int i = 0; i < msg.translations().count(); ++i)
|
---|
246 | stream << quint8(Tag_Translation) << msg.translations().at(i);
|
---|
247 |
|
---|
248 | if (mode == SaveEverything)
|
---|
249 | prefix = HashContextSourceTextComment;
|
---|
250 |
|
---|
251 | // lrelease produces "wrong" QM files for QByteArrays that are .isNull().
|
---|
252 | switch (prefix) {
|
---|
253 | default:
|
---|
254 | case HashContextSourceTextComment:
|
---|
255 | stream << quint8(Tag_Comment) << msg.comment();
|
---|
256 | // fall through
|
---|
257 | case HashContextSourceText:
|
---|
258 | stream << quint8(Tag_SourceText) << msg.sourceText();
|
---|
259 | // fall through
|
---|
260 | case HashContext:
|
---|
261 | stream << quint8(Tag_Context) << msg.context();
|
---|
262 | break;
|
---|
263 | }
|
---|
264 |
|
---|
265 | stream << quint8(Tag_End);
|
---|
266 | }
|
---|
267 |
|
---|
268 |
|
---|
269 | bool Releaser::save(QIODevice *iod)
|
---|
270 | {
|
---|
271 | QDataStream s(iod);
|
---|
272 | s.writeRawData((const char *)magic, MagicLength);
|
---|
273 |
|
---|
274 | if (!m_offsetArray.isEmpty()) {
|
---|
275 | quint32 oas = quint32(m_offsetArray.size());
|
---|
276 | s << quint8(Hashes) << oas;
|
---|
277 | s.writeRawData(m_offsetArray.constData(), oas);
|
---|
278 | }
|
---|
279 | if (!m_messageArray.isEmpty()) {
|
---|
280 | quint32 mas = quint32(m_messageArray.size());
|
---|
281 | s << quint8(Messages) << mas;
|
---|
282 | s.writeRawData(m_messageArray.constData(), mas);
|
---|
283 | }
|
---|
284 | if (!m_contextArray.isEmpty()) {
|
---|
285 | quint32 cas = quint32(m_contextArray.size());
|
---|
286 | s << quint8(Contexts) << cas;
|
---|
287 | s.writeRawData(m_contextArray.constData(), cas);
|
---|
288 | }
|
---|
289 | if (!m_numerusRules.isEmpty()) {
|
---|
290 | quint32 nrs = m_numerusRules.size();
|
---|
291 | s << quint8(NumerusRules) << nrs;
|
---|
292 | s.writeRawData(m_numerusRules.constData(), nrs);
|
---|
293 | }
|
---|
294 | return true;
|
---|
295 | }
|
---|
296 |
|
---|
297 | void Releaser::squeeze(TranslatorSaveMode mode)
|
---|
298 | {
|
---|
299 | if (m_messages.isEmpty() && mode == SaveEverything)
|
---|
300 | return;
|
---|
301 |
|
---|
302 | QMap<ByteTranslatorMessage, void *> messages = m_messages;
|
---|
303 |
|
---|
304 | // re-build contents
|
---|
305 | m_messageArray.clear();
|
---|
306 | m_offsetArray.clear();
|
---|
307 | m_contextArray.clear();
|
---|
308 | m_messages.clear();
|
---|
309 |
|
---|
310 | QMap<Offset, void *> offsets;
|
---|
311 |
|
---|
312 | QDataStream ms(&m_messageArray, QIODevice::WriteOnly);
|
---|
313 | QMap<ByteTranslatorMessage, void *>::const_iterator it, next;
|
---|
314 | int cpPrev = 0, cpNext = 0;
|
---|
315 | for (it = messages.constBegin(); it != messages.constEnd(); ++it) {
|
---|
316 | cpPrev = cpNext;
|
---|
317 | next = it;
|
---|
318 | ++next;
|
---|
319 | if (next == messages.constEnd())
|
---|
320 | cpNext = 0;
|
---|
321 | else
|
---|
322 | cpNext = commonPrefix(it.key(), next.key());
|
---|
323 | offsets.insert(Offset(msgHash(it.key()), ms.device()->pos()), (void *)0);
|
---|
324 | writeMessage(it.key(), ms, mode, Prefix(qMax(cpPrev, cpNext + 1)));
|
---|
325 | }
|
---|
326 |
|
---|
327 | QMap<Offset, void *>::Iterator offset;
|
---|
328 | offset = offsets.begin();
|
---|
329 | QDataStream ds(&m_offsetArray, QIODevice::WriteOnly);
|
---|
330 | while (offset != offsets.end()) {
|
---|
331 | Offset k = offset.key();
|
---|
332 | ++offset;
|
---|
333 | ds << quint32(k.h) << quint32(k.o);
|
---|
334 | }
|
---|
335 |
|
---|
336 | if (mode == SaveStripped) {
|
---|
337 | QMap<QByteArray, int> contextSet;
|
---|
338 | for (it = messages.constBegin(); it != messages.constEnd(); ++it)
|
---|
339 | ++contextSet[it.key().context()];
|
---|
340 |
|
---|
341 | quint16 hTableSize;
|
---|
342 | if (contextSet.size() < 200)
|
---|
343 | hTableSize = (contextSet.size() < 60) ? 151 : 503;
|
---|
344 | else if (contextSet.size() < 2500)
|
---|
345 | hTableSize = (contextSet.size() < 750) ? 1511 : 5003;
|
---|
346 | else
|
---|
347 | hTableSize = (contextSet.size() < 10000) ? 15013 : 3 * contextSet.size() / 2;
|
---|
348 |
|
---|
349 | QMultiMap<int, QByteArray> hashMap;
|
---|
350 | QMap<QByteArray, int>::const_iterator c;
|
---|
351 | for (c = contextSet.constBegin(); c != contextSet.constEnd(); ++c)
|
---|
352 | hashMap.insert(elfHash(c.key()) % hTableSize, c.key());
|
---|
353 |
|
---|
354 | /*
|
---|
355 | The contexts found in this translator are stored in a hash
|
---|
356 | table to provide fast lookup. The context array has the
|
---|
357 | following format:
|
---|
358 |
|
---|
359 | quint16 hTableSize;
|
---|
360 | quint16 hTable[hTableSize];
|
---|
361 | quint8 contextPool[...];
|
---|
362 |
|
---|
363 | The context pool stores the contexts as Pascal strings:
|
---|
364 |
|
---|
365 | quint8 len;
|
---|
366 | quint8 data[len];
|
---|
367 |
|
---|
368 | Let's consider the look-up of context "FunnyDialog". A
|
---|
369 | hash value between 0 and hTableSize - 1 is computed, say h.
|
---|
370 | If hTable[h] is 0, "FunnyDialog" is not covered by this
|
---|
371 | translator. Else, we check in the contextPool at offset
|
---|
372 | 2 * hTable[h] to see if "FunnyDialog" is one of the
|
---|
373 | contexts stored there, until we find it or we meet the
|
---|
374 | empty string.
|
---|
375 | */
|
---|
376 | m_contextArray.resize(2 + (hTableSize << 1));
|
---|
377 | QDataStream t(&m_contextArray, QIODevice::WriteOnly);
|
---|
378 |
|
---|
379 | quint16 *hTable = new quint16[hTableSize];
|
---|
380 | memset(hTable, 0, hTableSize * sizeof(quint16));
|
---|
381 |
|
---|
382 | t << hTableSize;
|
---|
383 | t.device()->seek(2 + (hTableSize << 1));
|
---|
384 | t << quint16(0); // the entry at offset 0 cannot be used
|
---|
385 | uint upto = 2;
|
---|
386 |
|
---|
387 | QMap<int, QByteArray>::const_iterator entry = hashMap.constBegin();
|
---|
388 | while (entry != hashMap.constEnd()) {
|
---|
389 | int i = entry.key();
|
---|
390 | hTable[i] = quint16(upto >> 1);
|
---|
391 |
|
---|
392 | do {
|
---|
393 | const char *con = entry.value().constData();
|
---|
394 | uint len = uint(entry.value().length());
|
---|
395 | len = qMin(len, 255u);
|
---|
396 | t << quint8(len);
|
---|
397 | t.writeRawData(con, len);
|
---|
398 | upto += 1 + len;
|
---|
399 | ++entry;
|
---|
400 | } while (entry != hashMap.constEnd() && entry.key() == i);
|
---|
401 | if (upto & 0x1) {
|
---|
402 | // offsets have to be even
|
---|
403 | t << quint8(0); // empty string
|
---|
404 | ++upto;
|
---|
405 | }
|
---|
406 | }
|
---|
407 | t.device()->seek(2);
|
---|
408 | for (int j = 0; j < hTableSize; j++)
|
---|
409 | t << hTable[j];
|
---|
410 | delete [] hTable;
|
---|
411 |
|
---|
412 | if (upto > 131072) {
|
---|
413 | qWarning("Releaser::squeeze: Too many contexts");
|
---|
414 | m_contextArray.clear();
|
---|
415 | }
|
---|
416 | }
|
---|
417 | }
|
---|
418 |
|
---|
419 | void Releaser::insertInternal(const TranslatorMessage &message, const QStringList &tlns,
|
---|
420 | bool forceComment, bool isUtf8)
|
---|
421 | {
|
---|
422 | ByteTranslatorMessage bmsg(originalBytes(message.context(), isUtf8),
|
---|
423 | originalBytes(message.sourceText(), isUtf8),
|
---|
424 | originalBytes(message.comment(), isUtf8),
|
---|
425 | tlns);
|
---|
426 | if (!forceComment) {
|
---|
427 | ByteTranslatorMessage bmsg2(
|
---|
428 | bmsg.context(), bmsg.sourceText(), QByteArray(""), bmsg.translations());
|
---|
429 | if (!m_messages.contains(bmsg2)) {
|
---|
430 | m_messages.insert(bmsg2, 0);
|
---|
431 | return;
|
---|
432 | }
|
---|
433 | }
|
---|
434 | m_messages.insert(bmsg, 0);
|
---|
435 | }
|
---|
436 |
|
---|
437 | void Releaser::insert(const TranslatorMessage &message, const QStringList &tlns, bool forceComment)
|
---|
438 | {
|
---|
439 | insertInternal(message, tlns, forceComment, message.isUtf8());
|
---|
440 | if (message.isUtf8() && message.isNonUtf8())
|
---|
441 | insertInternal(message, tlns, forceComment, false);
|
---|
442 | }
|
---|
443 |
|
---|
444 | void Releaser::insertIdBased(const TranslatorMessage &message, const QStringList &tlns)
|
---|
445 | {
|
---|
446 | ByteTranslatorMessage bmsg("", originalBytes(message.id(), false), "", tlns);
|
---|
447 | m_messages.insert(bmsg, 0);
|
---|
448 | }
|
---|
449 |
|
---|
450 | void Releaser::setNumerusRules(const QByteArray &rules)
|
---|
451 | {
|
---|
452 | m_numerusRules = rules;
|
---|
453 | }
|
---|
454 |
|
---|
455 | static quint8 read8(const uchar *data)
|
---|
456 | {
|
---|
457 | return *data;
|
---|
458 | }
|
---|
459 |
|
---|
460 | static quint32 read32(const uchar *data)
|
---|
461 | {
|
---|
462 | return (data[0] << 24) | (data[1] << 16) | (data[2] << 8) | (data[3]);
|
---|
463 | }
|
---|
464 |
|
---|
465 | static void fromBytes(const char *str, int len, QTextCodec *codec, QTextCodec *utf8Codec,
|
---|
466 | QString *out, QString *utf8Out,
|
---|
467 | bool *isSystem, bool *isUtf8, bool *needs8Bit)
|
---|
468 | {
|
---|
469 | for (int i = 0; i < len; ++i)
|
---|
470 | if (str[i] & 0x80) {
|
---|
471 | if (utf8Codec) {
|
---|
472 | QTextCodec::ConverterState cvtState;
|
---|
473 | *utf8Out = utf8Codec->toUnicode(str, len, &cvtState);
|
---|
474 | *isUtf8 = !cvtState.invalidChars;
|
---|
475 | }
|
---|
476 | QTextCodec::ConverterState cvtState;
|
---|
477 | *out = codec->toUnicode(str, len, &cvtState);
|
---|
478 | *isSystem = !cvtState.invalidChars;
|
---|
479 | *needs8Bit = true;
|
---|
480 | return;
|
---|
481 | }
|
---|
482 | *out = QString::fromLatin1(str, len);
|
---|
483 | *isSystem = true;
|
---|
484 | if (utf8Codec) {
|
---|
485 | *utf8Out = *out;
|
---|
486 | *isUtf8 = true;
|
---|
487 | }
|
---|
488 | *needs8Bit = false;
|
---|
489 | }
|
---|
490 |
|
---|
491 | bool loadQM(Translator &translator, QIODevice &dev, ConversionData &cd)
|
---|
492 | {
|
---|
493 | QByteArray ba = dev.readAll();
|
---|
494 | const uchar *data = (uchar*)ba.data();
|
---|
495 | int len = ba.size();
|
---|
496 | if (len < MagicLength || memcmp(data, magic, MagicLength) != 0) {
|
---|
497 | cd.appendError(QLatin1String("QM-Format error: magic marker missing"));
|
---|
498 | return false;
|
---|
499 | }
|
---|
500 |
|
---|
501 | enum { Contexts = 0x2f, Hashes = 0x42, Messages = 0x69, NumerusRules = 0x88 };
|
---|
502 |
|
---|
503 | // for squeezed but non-file data, this is what needs to be deleted
|
---|
504 | const uchar *messageArray = 0;
|
---|
505 | const uchar *offsetArray = 0;
|
---|
506 | const uchar *contextArray = 0;
|
---|
507 | const uchar *numerusRulesArray = 0;
|
---|
508 | uint messageLength = 0;
|
---|
509 | uint offsetLength = 0;
|
---|
510 | uint contextLength = 0;
|
---|
511 | uint numerusRulesLength = 0;
|
---|
512 |
|
---|
513 | bool ok = true;
|
---|
514 | const uchar *end = data + len;
|
---|
515 |
|
---|
516 | data += MagicLength;
|
---|
517 |
|
---|
518 | while (data < end - 4) {
|
---|
519 | quint8 tag = read8(data++);
|
---|
520 | quint32 blockLen = read32(data);
|
---|
521 | //qDebug() << "TAG:" << tag << "BLOCKLEN:" << blockLen;
|
---|
522 | data += 4;
|
---|
523 | if (!tag || !blockLen)
|
---|
524 | break;
|
---|
525 | if (data + blockLen > end) {
|
---|
526 | ok = false;
|
---|
527 | break;
|
---|
528 | }
|
---|
529 |
|
---|
530 | if (tag == Contexts) {
|
---|
531 | contextArray = data;
|
---|
532 | contextLength = blockLen;
|
---|
533 | //qDebug() << "CONTEXTS: " << contextLength << QByteArray((const char *)contextArray, contextLength).toHex();
|
---|
534 | } else if (tag == Hashes) {
|
---|
535 | offsetArray = data;
|
---|
536 | offsetLength = blockLen;
|
---|
537 | //qDebug() << "HASHES: " << offsetLength << QByteArray((const char *)offsetArray, offsetLength).toHex();
|
---|
538 | } else if (tag == Messages) {
|
---|
539 | messageArray = data;
|
---|
540 | messageLength = blockLen;
|
---|
541 | //qDebug() << "MESSAGES: " << messageLength << QByteArray((const char *)messageArray, messageLength).toHex();
|
---|
542 | } else if (tag == NumerusRules) {
|
---|
543 | numerusRulesArray = data;
|
---|
544 | numerusRulesLength = blockLen;
|
---|
545 | //qDebug() << "NUMERUSRULES: " << numerusRulesLength << QByteArray((const char *)numerusRulesArray, numerusRulesLength).toHex();
|
---|
546 | }
|
---|
547 |
|
---|
548 | data += blockLen;
|
---|
549 | }
|
---|
550 |
|
---|
551 |
|
---|
552 | size_t numItems = offsetLength / (2 * sizeof(quint32));
|
---|
553 | //qDebug() << "NUMITEMS: " << numItems;
|
---|
554 |
|
---|
555 | QTextCodec *codec = QTextCodec::codecForName(
|
---|
556 | cd.m_codecForSource.isEmpty() ? QByteArray("Latin1") : cd.m_codecForSource);
|
---|
557 | QTextCodec *utf8Codec = 0;
|
---|
558 | if (codec->name() != "UTF-8")
|
---|
559 | utf8Codec = QTextCodec::codecForName("UTF-8");
|
---|
560 |
|
---|
561 | QString strProN = QLatin1String("%n");
|
---|
562 | QLocale::Language l;
|
---|
563 | QLocale::Country c;
|
---|
564 | Translator::languageAndCountry(translator.languageCode(), &l, &c);
|
---|
565 | QStringList numerusForms;
|
---|
566 | bool guessPlurals = true;
|
---|
567 | if (getNumerusInfo(l, c, 0, &numerusForms))
|
---|
568 | guessPlurals = (numerusForms.count() == 1);
|
---|
569 |
|
---|
570 | QString context, contextUtf8;
|
---|
571 | bool contextIsSystem, contextIsUtf8, contextNeeds8Bit;
|
---|
572 | QString sourcetext, sourcetextUtf8;
|
---|
573 | bool sourcetextIsSystem, sourcetextIsUtf8, sourcetextNeeds8Bit;
|
---|
574 | QString comment, commentUtf8;
|
---|
575 | bool commentIsSystem, commentIsUtf8, commentNeeds8Bit;
|
---|
576 | QStringList translations;
|
---|
577 |
|
---|
578 | for (const uchar *start = offsetArray; start != offsetArray + (numItems << 3); start += 8) {
|
---|
579 | //quint32 hash = read32(start);
|
---|
580 | quint32 ro = read32(start + 4);
|
---|
581 | //qDebug() << "\nHASH:" << hash;
|
---|
582 | const uchar *m = messageArray + ro;
|
---|
583 |
|
---|
584 | for (;;) {
|
---|
585 | uchar tag = read8(m++);
|
---|
586 | //qDebug() << "Tag:" << tag << " ADDR: " << m;
|
---|
587 | switch(tag) {
|
---|
588 | case Tag_End:
|
---|
589 | goto end;
|
---|
590 | case Tag_Translation: {
|
---|
591 | int len = read32(m);
|
---|
592 | if (len % 1) {
|
---|
593 | cd.appendError(QLatin1String("QM-Format error"));
|
---|
594 | return false;
|
---|
595 | }
|
---|
596 | m += 4;
|
---|
597 | QString str = QString((const QChar *)m, len/2);
|
---|
598 | if (QSysInfo::ByteOrder == QSysInfo::LittleEndian) {
|
---|
599 | for (int i = 0; i < str.length(); ++i)
|
---|
600 | str[i] = QChar((str.at(i).unicode() >> 8) +
|
---|
601 | ((str.at(i).unicode() << 8) & 0xff00));
|
---|
602 | }
|
---|
603 | translations << str;
|
---|
604 | m += len;
|
---|
605 | break;
|
---|
606 | }
|
---|
607 | case Tag_Obsolete1:
|
---|
608 | m += 4;
|
---|
609 | //qDebug() << "OBSOLETE";
|
---|
610 | break;
|
---|
611 | case Tag_SourceText: {
|
---|
612 | quint32 len = read32(m);
|
---|
613 | m += 4;
|
---|
614 | //qDebug() << "SOURCE LEN: " << len;
|
---|
615 | //qDebug() << "SOURCE: " << QByteArray((const char*)m, len);
|
---|
616 | fromBytes((const char*)m, len, codec, utf8Codec,
|
---|
617 | &sourcetext, &sourcetextUtf8,
|
---|
618 | &sourcetextIsSystem, &sourcetextIsUtf8, &sourcetextNeeds8Bit);
|
---|
619 | m += len;
|
---|
620 | break;
|
---|
621 | }
|
---|
622 | case Tag_Context: {
|
---|
623 | quint32 len = read32(m);
|
---|
624 | m += 4;
|
---|
625 | //qDebug() << "CONTEXT LEN: " << len;
|
---|
626 | //qDebug() << "CONTEXT: " << QByteArray((const char*)m, len);
|
---|
627 | fromBytes((const char*)m, len, codec, utf8Codec,
|
---|
628 | &context, &contextUtf8,
|
---|
629 | &contextIsSystem, &contextIsUtf8, &contextNeeds8Bit);
|
---|
630 | m += len;
|
---|
631 | break;
|
---|
632 | }
|
---|
633 | case Tag_Comment: {
|
---|
634 | quint32 len = read32(m);
|
---|
635 | m += 4;
|
---|
636 | //qDebug() << "COMMENT LEN: " << len;
|
---|
637 | //qDebug() << "COMMENT: " << QByteArray((const char*)m, len);
|
---|
638 | fromBytes((const char*)m, len, codec, utf8Codec,
|
---|
639 | &comment, &commentUtf8,
|
---|
640 | &commentIsSystem, &commentIsUtf8, &commentNeeds8Bit);
|
---|
641 | m += len;
|
---|
642 | break;
|
---|
643 | }
|
---|
644 | default:
|
---|
645 | //qDebug() << "UNKNOWN TAG" << tag;
|
---|
646 | break;
|
---|
647 | }
|
---|
648 | }
|
---|
649 | end:;
|
---|
650 | TranslatorMessage msg;
|
---|
651 | msg.setType(TranslatorMessage::Finished);
|
---|
652 | if (translations.count() > 1) {
|
---|
653 | // If guessPlurals is not false here, plural form discard messages
|
---|
654 | // will be spewn out later.
|
---|
655 | msg.setPlural(true);
|
---|
656 | } else if (guessPlurals) {
|
---|
657 | // This might cause false positives, so it is a fallback only.
|
---|
658 | if (sourcetext.contains(strProN))
|
---|
659 | msg.setPlural(true);
|
---|
660 | }
|
---|
661 | msg.setTranslations(translations);
|
---|
662 | translations.clear();
|
---|
663 | if (contextNeeds8Bit || sourcetextNeeds8Bit || commentNeeds8Bit) {
|
---|
664 | if (utf8Codec && contextIsUtf8 && sourcetextIsUtf8 && commentIsUtf8) {
|
---|
665 | // The message is utf-8, but file is not.
|
---|
666 | msg.setUtf8(true);
|
---|
667 | msg.setContext(contextUtf8);
|
---|
668 | msg.setSourceText(sourcetextUtf8);
|
---|
669 | msg.setComment(commentUtf8);
|
---|
670 | translator.append(msg);
|
---|
671 | continue;
|
---|
672 | }
|
---|
673 | if (!(contextIsSystem && sourcetextIsSystem && commentIsSystem)) {
|
---|
674 | cd.appendError(QLatin1String(
|
---|
675 | "Cannot read file with specified input codec"));
|
---|
676 | return false;
|
---|
677 | }
|
---|
678 | // The message is 8-bit in the file's encoding (utf-8 or not).
|
---|
679 | }
|
---|
680 | msg.setContext(context);
|
---|
681 | msg.setSourceText(sourcetext);
|
---|
682 | msg.setComment(comment);
|
---|
683 | translator.append(msg);
|
---|
684 | }
|
---|
685 | return ok;
|
---|
686 | }
|
---|
687 |
|
---|
688 |
|
---|
689 |
|
---|
690 | static bool containsStripped(const Translator &translator, const TranslatorMessage &msg)
|
---|
691 | {
|
---|
692 | foreach (const TranslatorMessage &tmsg, translator.messages())
|
---|
693 | if (tmsg.sourceText() == msg.sourceText()
|
---|
694 | && tmsg.context() == msg.context()
|
---|
695 | && tmsg.comment().isEmpty())
|
---|
696 | return true;
|
---|
697 | return false;
|
---|
698 | }
|
---|
699 |
|
---|
700 | static bool saveQM(const Translator &translator, QIODevice &dev, ConversionData &cd)
|
---|
701 | {
|
---|
702 | Releaser releaser;
|
---|
703 | QLocale::Language l;
|
---|
704 | QLocale::Country c;
|
---|
705 | Translator::languageAndCountry(translator.languageCode(), &l, &c);
|
---|
706 | QByteArray rules;
|
---|
707 | if (getNumerusInfo(l, c, &rules, 0))
|
---|
708 | releaser.setNumerusRules(rules);
|
---|
709 | releaser.setCodecName(translator.codecName());
|
---|
710 |
|
---|
711 | int finished = 0;
|
---|
712 | int unfinished = 0;
|
---|
713 | int untranslated = 0;
|
---|
714 | int missingIds = 0;
|
---|
715 | int droppedData = 0;
|
---|
716 |
|
---|
717 | for (int i = 0; i != translator.messageCount(); ++i) {
|
---|
718 | const TranslatorMessage &msg = translator.message(i);
|
---|
719 | TranslatorMessage::Type typ = msg.type();
|
---|
720 | if (typ != TranslatorMessage::Obsolete) {
|
---|
721 | if (cd.m_idBased && msg.id().isEmpty()) {
|
---|
722 | ++missingIds;
|
---|
723 | continue;
|
---|
724 | }
|
---|
725 | if (typ == TranslatorMessage::Unfinished) {
|
---|
726 | if (!cd.m_idBased && msg.translation().isEmpty()) {
|
---|
727 | ++untranslated;
|
---|
728 | continue;
|
---|
729 | } else {
|
---|
730 | if (cd.ignoreUnfinished())
|
---|
731 | continue;
|
---|
732 | ++unfinished;
|
---|
733 | }
|
---|
734 | } else {
|
---|
735 | ++finished;
|
---|
736 | }
|
---|
737 | QStringList tlns = msg.translations();
|
---|
738 | if (msg.type() == TranslatorMessage::Unfinished
|
---|
739 | && (cd.m_idBased || !cd.m_unTrPrefix.isEmpty()))
|
---|
740 | for (int j = 0; j < tlns.size(); ++j)
|
---|
741 | if (tlns.at(j).isEmpty())
|
---|
742 | tlns[j] = cd.m_unTrPrefix + msg.sourceText();
|
---|
743 | if (cd.m_idBased) {
|
---|
744 | if (!msg.context().isEmpty() || !msg.comment().isEmpty())
|
---|
745 | ++droppedData;
|
---|
746 | releaser.insertIdBased(msg, tlns);
|
---|
747 | } else {
|
---|
748 | // Drop the comment in (context, sourceText, comment),
|
---|
749 | // unless the context is empty,
|
---|
750 | // unless (context, sourceText, "") already exists or
|
---|
751 | // unless we already dropped the comment of (context,
|
---|
752 | // sourceText, comment0).
|
---|
753 | bool forceComment =
|
---|
754 | msg.comment().isEmpty()
|
---|
755 | || msg.context().isEmpty()
|
---|
756 | || containsStripped(translator, msg);
|
---|
757 | releaser.insert(msg, tlns, forceComment);
|
---|
758 | }
|
---|
759 | }
|
---|
760 | }
|
---|
761 |
|
---|
762 | if (missingIds)
|
---|
763 | cd.appendError(QCoreApplication::translate("LRelease",
|
---|
764 | "Dropped %n message(s) which had no ID.", 0,
|
---|
765 | QCoreApplication::CodecForTr, missingIds));
|
---|
766 | if (droppedData)
|
---|
767 | cd.appendError(QCoreApplication::translate("LRelease",
|
---|
768 | "Excess context/disambiguation dropped from %n message(s).", 0,
|
---|
769 | QCoreApplication::CodecForTr, droppedData));
|
---|
770 |
|
---|
771 | releaser.squeeze(cd.m_saveMode);
|
---|
772 | bool saved = releaser.save(&dev);
|
---|
773 | if (saved && cd.isVerbose()) {
|
---|
774 | int generatedCount = finished + unfinished;
|
---|
775 | cd.appendError(QCoreApplication::translate("LRelease",
|
---|
776 | " Generated %n translation(s) (%1 finished and %2 unfinished)\n", 0,
|
---|
777 | QCoreApplication::CodecForTr, generatedCount).arg(finished).arg(unfinished));
|
---|
778 | if (untranslated)
|
---|
779 | cd.appendError(QCoreApplication::translate("LRelease",
|
---|
780 | " Ignored %n untranslated source text(s)\n", 0,
|
---|
781 | QCoreApplication::CodecForTr, untranslated));
|
---|
782 | }
|
---|
783 | return saved;
|
---|
784 | }
|
---|
785 |
|
---|
786 | int initQM()
|
---|
787 | {
|
---|
788 | Translator::FileFormat format;
|
---|
789 |
|
---|
790 | format.extension = QLatin1String("qm");
|
---|
791 | format.description = QObject::tr("Compiled Qt translations");
|
---|
792 | format.fileType = Translator::FileFormat::TranslationBinary;
|
---|
793 | format.priority = 0;
|
---|
794 | format.loader = &loadQM;
|
---|
795 | format.saver = &saveQM;
|
---|
796 | Translator::registerFileFormat(format);
|
---|
797 |
|
---|
798 | return 1;
|
---|
799 | }
|
---|
800 |
|
---|
801 | Q_CONSTRUCTOR_FUNCTION(initQM)
|
---|
802 |
|
---|
803 | QT_END_NAMESPACE
|
---|