| 1 | // Copyright (C) 2002-2007 Detlev Offenbach <detlev@die-offenbachs.de> |
| 2 | // Copyright (C) 2021 The Qt Company Ltd. |
| 3 | // SPDX-License-Identifier: LicenseRef-Qt-Commercial OR GPL-3.0-only WITH Qt-GPL-exception-1.0 |
| 4 | |
| 5 | #include <translator.h> |
| 6 | #include "lupdate.h" |
| 7 | #include "metastrings.h" |
| 8 | |
| 9 | #include <QtCore/qhash.h> |
| 10 | #include <QtCore/qlist.h> |
| 11 | #include <QtCore/qstring.h> |
| 12 | #include <QtCore/qtextstream.h> |
| 13 | #include <QtCore/qstack.h> |
| 14 | |
| 15 | #include <cctype> |
| 16 | #include <cerrno> |
| 17 | #include <cstdio> |
| 18 | #include <cstring> |
| 19 | |
| 20 | using namespace Qt::StringLiterals; |
| 21 | |
| 22 | QT_BEGIN_NAMESPACE |
| 23 | |
| 24 | class PythonParser |
| 25 | { |
| 26 | |
| 27 | enum Token { |
| 28 | Tok_Eof, |
| 29 | Tok_class, |
| 30 | Tok_def, |
| 31 | Tok_return, |
| 32 | Tok_tr, |
| 33 | Tok_trUtf8, |
| 34 | Tok_translate, |
| 35 | Tok_Ident, |
| 36 | Tok_Dot, |
| 37 | Tok_String, |
| 38 | Tok_LeftParen, |
| 39 | Tok_RightParen, |
| 40 | Tok_Comma, |
| 41 | Tok_None, |
| 42 | Tok_Integer |
| 43 | }; |
| 44 | |
| 45 | enum class StringType { NoString, String, FormatString, RawString }; |
| 46 | |
| 47 | public: |
| 48 | PythonParser(Translator &translator, const QString &fileName, bool &error, ConversionData &cd) |
| 49 | : tor(translator), m_cd(cd) |
| 50 | { |
| 51 | #ifdef Q_CC_MSVC |
| 52 | const auto *fileNameC = reinterpret_cast<const wchar_t *>(fileName.utf16()); |
| 53 | error = _wfopen_s(&yyInFile, fileNameC, L"r" ) != 0; |
| 54 | #else |
| 55 | const QByteArray fileNameC = QFile::encodeName(fileName); |
| 56 | yyInFile = std::fopen(filename: fileNameC.constData(), modes: "r" ); |
| 57 | error = yyInFile == nullptr; |
| 58 | #endif |
| 59 | if (!error) |
| 60 | startTokenizer(fileName); |
| 61 | } |
| 62 | |
| 63 | /* |
| 64 | Accomplishes a very easy task: It finds all strings inside a tr() or translate() |
| 65 | call, and possibly finds out the context of the call. It supports |
| 66 | three cases: |
| 67 | (1) the context is specified, as in FunnyDialog.tr("Hello") or |
| 68 | translate("FunnyDialog", "Hello"); |
| 69 | (2) the call appears within an inlined function; |
| 70 | (3) the call appears within a function defined outside the class definition. |
| 71 | */ |
| 72 | void parse(const QByteArray &initialContext = {}, const QByteArray &defaultContext = {}) |
| 73 | { |
| 74 | QByteArray context; |
| 75 | QByteArray text; |
| 76 | QByteArray ; |
| 77 | QByteArray prefix; |
| 78 | bool utf8 = false; |
| 79 | |
| 80 | yyTok = getToken(); |
| 81 | while (yyTok != Tok_Eof) { |
| 82 | |
| 83 | switch (yyTok) { |
| 84 | case Tok_class: { |
| 85 | if (yyIndentationSize < 0 && yyContinuousSpaceCount > 0) |
| 86 | yyIndentationSize = yyContinuousSpaceCount; // First indented "class" |
| 87 | const int indent = |
| 88 | yyIndentationSize > 0 ? yyContinuousSpaceCount / yyIndentationSize : 0; |
| 89 | while (!yyContextStack.isEmpty() && yyContextStack.top().second >= indent) |
| 90 | yyContextStack.pop(); |
| 91 | yyTok = getToken(); |
| 92 | yyContextStack.push(t: { yyIdent, indent }); |
| 93 | yyTok = getToken(); |
| 94 | } break; |
| 95 | case Tok_def: |
| 96 | if (yyIndentationSize < 0 && yyContinuousSpaceCount > 0) |
| 97 | yyIndentationSize = yyContinuousSpaceCount; // First indented "def" |
| 98 | if (!yyContextStack.isEmpty()) { |
| 99 | // Pop classes if the function is further outdented than the class on the top |
| 100 | // (end of a nested class). |
| 101 | const int classIndent = yyIndentationSize > 0 |
| 102 | ? yyContinuousSpaceCount / yyIndentationSize - 1 |
| 103 | : 0; |
| 104 | while (!yyContextStack.isEmpty() && yyContextStack.top().second > classIndent) |
| 105 | yyContextStack.pop(); |
| 106 | } |
| 107 | yyTok = getToken(); |
| 108 | break; |
| 109 | case Tok_tr: |
| 110 | case Tok_trUtf8: { |
| 111 | utf8 = true; |
| 112 | yyTok = getToken(); |
| 113 | const int lineNo = yyCurLineNo; |
| 114 | if (match(t: Tok_LeftParen) && matchString(s: &text)) { |
| 115 | comment.clear(); |
| 116 | bool plural = false; |
| 117 | |
| 118 | MetaStrings metaBackup = std::move(metaStrings); |
| 119 | |
| 120 | if (match(t: Tok_RightParen)) { |
| 121 | // There is no comment or plural arguments. |
| 122 | } else if (match(t: Tok_Comma) && matchStringOrNone(s: &comment)) { |
| 123 | // There is a comment argument. |
| 124 | if (match(t: Tok_RightParen)) { |
| 125 | // There is no plural argument. |
| 126 | } else if (match(t: Tok_Comma)) { |
| 127 | // There is a plural argument. |
| 128 | plural = true; |
| 129 | } |
| 130 | } |
| 131 | |
| 132 | if (prefix.isEmpty()) |
| 133 | context = defaultContext; |
| 134 | else if (prefix == "self" ) |
| 135 | context = yyContextStack.isEmpty() ? initialContext |
| 136 | : yyContextStack.top().first; |
| 137 | else |
| 138 | context = prefix; |
| 139 | |
| 140 | prefix.clear(); |
| 141 | TranslatorMessage message(QString::fromUtf8(ba: context), QString::fromUtf8(ba: text), |
| 142 | QString::fromUtf8(ba: comment), {}, yyFileName, lineNo, |
| 143 | {}, TranslatorMessage::Unfinished, plural); |
| 144 | setMessageParameters(message: &message, meta: metaBackup); |
| 145 | tor.extend(msg: message, cd&: m_cd); |
| 146 | } |
| 147 | } break; |
| 148 | case Tok_translate: { |
| 149 | bool plural{}; |
| 150 | const int lineNo = yyCurLineNo; |
| 151 | MetaStrings metaBackup = std::move(metaStrings); |
| 152 | if (parseTranslate(text: &text, context: &context, comment: &comment, utf8: &utf8, plural: &plural)) { |
| 153 | TranslatorMessage message(QString::fromUtf8(ba: context), QString::fromUtf8(ba: text), |
| 154 | QString::fromUtf8(ba: comment), {}, yyFileName, lineNo, |
| 155 | {}, TranslatorMessage::Unfinished, plural); |
| 156 | setMessageParameters(message: &message, meta: metaBackup); |
| 157 | tor.extend(msg: message, cd&: m_cd); |
| 158 | } else { |
| 159 | metaStrings = std::move(metaBackup); |
| 160 | } |
| 161 | } break; |
| 162 | case Tok_Ident: |
| 163 | if (!prefix.isEmpty()) |
| 164 | prefix += '.'; |
| 165 | prefix += yyIdent; |
| 166 | yyTok = getToken(); |
| 167 | if (yyTok != Tok_Dot) |
| 168 | prefix.clear(); |
| 169 | break; |
| 170 | default: |
| 171 | yyTok = getToken(); |
| 172 | } |
| 173 | } |
| 174 | |
| 175 | if (yyParenDepth != 0) { |
| 176 | qWarning(msg: "%s: Unbalanced parentheses in Python code" , qPrintable(yyFileName)); |
| 177 | } |
| 178 | } |
| 179 | |
| 180 | ~PythonParser() { std::fclose(stream: yyInFile); } |
| 181 | |
| 182 | private: |
| 183 | QHash<QByteArray, Token> fillTokens() |
| 184 | { |
| 185 | QHash<QByteArray, Token> tokens = { { "None" , Tok_None }, { "class" , Tok_class }, |
| 186 | { "def" , Tok_def }, { "return" , Tok_return }, |
| 187 | { "__tr" , Tok_tr }, // Legacy? |
| 188 | { "__trUtf8" , Tok_trUtf8 } }; |
| 189 | |
| 190 | const auto &nameMap = trFunctionAliasManager.nameToTrFunctionMap(); |
| 191 | for (auto it = nameMap.cbegin(), end = nameMap.cend(); it != end; ++it) { |
| 192 | switch (it.value()) { |
| 193 | case TrFunctionAliasManager::Function_tr: |
| 194 | case TrFunctionAliasManager::Function_QT_TR_NOOP: |
| 195 | tokens.insert(key: it.key().toUtf8(), value: Tok_tr); |
| 196 | break; |
| 197 | case TrFunctionAliasManager::Function_trUtf8: |
| 198 | tokens.insert(key: it.key().toUtf8(), value: Tok_trUtf8); |
| 199 | break; |
| 200 | case TrFunctionAliasManager::Function_translate: |
| 201 | case TrFunctionAliasManager::Function_QT_TRANSLATE_NOOP: |
| 202 | // QTranslator::findMessage() has the same parameters as QApplication::translate(). |
| 203 | case TrFunctionAliasManager::Function_findMessage: |
| 204 | tokens.insert(key: it.key().toUtf8(), value: Tok_translate); |
| 205 | break; |
| 206 | default: |
| 207 | break; |
| 208 | } |
| 209 | } |
| 210 | return tokens; |
| 211 | } |
| 212 | |
| 213 | QHash<QByteArray, Token> &getTokens() |
| 214 | { |
| 215 | static QHash<QByteArray, Token> tokens = fillTokens(); |
| 216 | return tokens; |
| 217 | } |
| 218 | |
| 219 | int getChar() |
| 220 | { |
| 221 | int c; |
| 222 | |
| 223 | if (buf < 0) { |
| 224 | c = getc(stream: yyInFile); |
| 225 | } else { |
| 226 | c = buf; |
| 227 | buf = -1; |
| 228 | } |
| 229 | if (c == '\n') { |
| 230 | yyCurLineNo++; |
| 231 | yyCountingIndentation = true; |
| 232 | yyContinuousSpaceCount = 0; |
| 233 | } else if (yyCountingIndentation && (c == 32 || c == 9)) { |
| 234 | yyContinuousSpaceCount++; |
| 235 | } else { |
| 236 | yyCountingIndentation = false; |
| 237 | } |
| 238 | return c; |
| 239 | } |
| 240 | |
| 241 | int peekChar() |
| 242 | { |
| 243 | int c = getc(stream: yyInFile); |
| 244 | buf = c; |
| 245 | return c; |
| 246 | } |
| 247 | |
| 248 | void startTokenizer(const QString &fileName) |
| 249 | { |
| 250 | yyInPos = 0; |
| 251 | buf = -1; |
| 252 | |
| 253 | yyFileName = fileName; |
| 254 | yyCh = getChar(); |
| 255 | yyParenDepth = 0; |
| 256 | yyCurLineNo = 1; |
| 257 | |
| 258 | yyIndentationSize = -1; |
| 259 | yyContinuousSpaceCount = 0; |
| 260 | yyContextStack.clear(); |
| 261 | } |
| 262 | |
| 263 | bool parseStringEscape(int quoteChar, StringType stringType) |
| 264 | { |
| 265 | static const char tab[] = "abfnrtv" ; |
| 266 | static const char backTab[] = "\a\b\f\n\r\t\v" ; |
| 267 | |
| 268 | yyCh = getChar(); |
| 269 | if (yyCh == EOF) |
| 270 | return false; |
| 271 | |
| 272 | if (stringType == StringType::RawString) { |
| 273 | if (yyCh != quoteChar) // Only quotes can be escaped in raw strings |
| 274 | yyString[yyStringLen++] = '\\'; |
| 275 | yyString[yyStringLen++] = yyCh; |
| 276 | yyCh = getChar(); |
| 277 | return true; |
| 278 | } |
| 279 | |
| 280 | if (yyCh == 'x' || yyCh == 'u' || yyCh == 'U') { |
| 281 | qsizetype maxSize = 2; // \x |
| 282 | if (yyCh == 'u') |
| 283 | maxSize = 4; |
| 284 | else if (yyCh == 'U') |
| 285 | maxSize = 8; |
| 286 | |
| 287 | QByteArray hex; |
| 288 | yyCh = getChar(); |
| 289 | if (yyCh == EOF) |
| 290 | return false; |
| 291 | |
| 292 | while (maxSize-- && std::isxdigit(yyCh)) { |
| 293 | hex += char(yyCh); |
| 294 | yyCh = getChar(); |
| 295 | if (yyCh == EOF) |
| 296 | return false; |
| 297 | } |
| 298 | uint n; |
| 299 | #ifdef Q_CC_MSVC |
| 300 | sscanf_s(hex, "%x" , &n); |
| 301 | #else |
| 302 | std::sscanf(s: hex, format: "%x" , &n); |
| 303 | #endif |
| 304 | |
| 305 | QByteArray hexChar = QString(QChar(n)).toUtf8(); |
| 306 | if (yyStringLen < sizeof(yyString) - hexChar.size()) |
| 307 | for (char c : std::as_const(t&: hexChar)) |
| 308 | yyString[yyStringLen++] = c; |
| 309 | return true; |
| 310 | } |
| 311 | |
| 312 | if (yyCh >= '0' && yyCh < '8') { |
| 313 | QByteArray oct; |
| 314 | int n = 0; |
| 315 | do { |
| 316 | oct += char(yyCh); |
| 317 | ++n; |
| 318 | yyCh = getChar(); |
| 319 | if (yyCh == EOF) |
| 320 | return false; |
| 321 | } while (yyCh >= '0' && yyCh < '8' && n < 3); |
| 322 | #ifdef Q_CC_MSVC |
| 323 | sscanf_s(oct, "%o" , &n); |
| 324 | #else |
| 325 | std::sscanf(s: oct, format: "%o" , &n); |
| 326 | #endif |
| 327 | if (yyStringLen < sizeof(yyString) - 1) |
| 328 | yyString[yyStringLen++] = char(n); |
| 329 | return true; |
| 330 | } |
| 331 | |
| 332 | const char *p = std::strchr(s: tab, c: yyCh); |
| 333 | if (yyStringLen < sizeof(yyString) - 1) { |
| 334 | yyString[yyStringLen++] = p == nullptr ? char(yyCh) : backTab[p - tab]; |
| 335 | } |
| 336 | yyCh = getChar(); |
| 337 | return true; |
| 338 | } |
| 339 | |
| 340 | Token parseString(StringType stringType = StringType::NoString) |
| 341 | { |
| 342 | int quoteChar = yyCh; |
| 343 | bool tripleQuote = false; |
| 344 | bool singleQuote = true; |
| 345 | bool in = false; |
| 346 | |
| 347 | yyCh = getChar(); |
| 348 | |
| 349 | while (yyCh != EOF) { |
| 350 | if (singleQuote && (yyCh == '\n' || (in && yyCh == quoteChar))) |
| 351 | break; |
| 352 | |
| 353 | if (yyCh == quoteChar) { |
| 354 | if (peekChar() == quoteChar) { |
| 355 | yyCh = getChar(); |
| 356 | if (!tripleQuote) { |
| 357 | tripleQuote = true; |
| 358 | singleQuote = false; |
| 359 | in = true; |
| 360 | yyCh = getChar(); |
| 361 | } else { |
| 362 | yyCh = getChar(); |
| 363 | if (yyCh == quoteChar) { |
| 364 | tripleQuote = false; |
| 365 | break; |
| 366 | } |
| 367 | } |
| 368 | } else if (tripleQuote) { |
| 369 | if (yyStringLen < sizeof(yyString) - 1) |
| 370 | yyString[yyStringLen++] = char(yyCh); |
| 371 | yyCh = getChar(); |
| 372 | continue; |
| 373 | } else { |
| 374 | break; |
| 375 | } |
| 376 | } else { |
| 377 | in = true; |
| 378 | } |
| 379 | |
| 380 | if (yyCh == '\\') { |
| 381 | if (!parseStringEscape(quoteChar, stringType)) |
| 382 | return Tok_Eof; |
| 383 | } else { |
| 384 | char *yStart = yyString + yyStringLen; |
| 385 | char *yp = yStart; |
| 386 | while (yyCh != EOF && (tripleQuote || yyCh != '\n') && yyCh != quoteChar |
| 387 | && yyCh != '\\') { |
| 388 | *yp++ = char(yyCh); |
| 389 | yyCh = getChar(); |
| 390 | } |
| 391 | yyStringLen += yp - yStart; |
| 392 | } |
| 393 | } |
| 394 | yyString[yyStringLen] = '\0'; |
| 395 | |
| 396 | if (yyCh != quoteChar) { |
| 397 | printf(format: "%c\n" , yyCh); |
| 398 | |
| 399 | qWarning(msg: "%s:%d: Unterminated string" , qPrintable(yyFileName), yyLineNo); |
| 400 | } |
| 401 | |
| 402 | if (yyCh == EOF) |
| 403 | return Tok_Eof; |
| 404 | yyCh = getChar(); |
| 405 | return Tok_String; |
| 406 | } |
| 407 | |
| 408 | QByteArray readLine() |
| 409 | { |
| 410 | QByteArray result; |
| 411 | while (true) { |
| 412 | yyCh = getChar(); |
| 413 | if (yyCh == EOF || yyCh == '\n') |
| 414 | break; |
| 415 | result.append(c: char(yyCh)); |
| 416 | } |
| 417 | return result; |
| 418 | } |
| 419 | |
| 420 | Token getToken(StringType stringType = StringType::NoString) |
| 421 | { |
| 422 | yyIdent.clear(); |
| 423 | yyStringLen = 0; |
| 424 | while (yyCh != EOF) { |
| 425 | yyLineNo = yyCurLineNo; |
| 426 | |
| 427 | if (std::isalpha(yyCh) || yyCh == '_') { |
| 428 | do { |
| 429 | yyIdent.append(c: char(yyCh)); |
| 430 | yyCh = getChar(); |
| 431 | } while (std::isalnum(yyCh) || yyCh == '_'); |
| 432 | |
| 433 | return getTokens().value(key: yyIdent, defaultValue: Tok_Ident); |
| 434 | } |
| 435 | switch (yyCh) { |
| 436 | case '#': { |
| 437 | auto = QString::fromUtf8(ba: readLine()); |
| 438 | if (!metaStrings.parse(string&: comment)) { |
| 439 | qWarning() << qPrintable(yyFileName) << ':' << yyLineNo << ": " |
| 440 | << metaStrings.popError().toStdString(); |
| 441 | break; |
| 442 | } |
| 443 | if (metaStrings.magicComment()) { |
| 444 | auto [context, comment] = *metaStrings.magicComment(); |
| 445 | TranslatorMessage msg(ParserTool::transcode(str: context), QString(), |
| 446 | ParserTool::transcode(str: comment), QString(), yyFileName, |
| 447 | yyCurLineNo, QStringList(), TranslatorMessage::Finished, |
| 448 | false); |
| 449 | msg.setExtraComment( |
| 450 | ParserTool::transcode(str: metaStrings.extracomment().simplified())); |
| 451 | tor.append(msg); |
| 452 | tor.setExtras(metaStrings.extra()); |
| 453 | metaStrings.clear(); |
| 454 | } |
| 455 | break; |
| 456 | } |
| 457 | case '"': |
| 458 | case '\'': |
| 459 | return parseString(stringType); |
| 460 | case '(': |
| 461 | yyParenDepth++; |
| 462 | yyCh = getChar(); |
| 463 | return Tok_LeftParen; |
| 464 | case ')': |
| 465 | yyParenDepth--; |
| 466 | yyCh = getChar(); |
| 467 | return Tok_RightParen; |
| 468 | case ',': |
| 469 | yyCh = getChar(); |
| 470 | return Tok_Comma; |
| 471 | case '.': |
| 472 | yyCh = getChar(); |
| 473 | return Tok_Dot; |
| 474 | case '0': |
| 475 | case '1': |
| 476 | case '2': |
| 477 | case '3': |
| 478 | case '4': |
| 479 | case '5': |
| 480 | case '6': |
| 481 | case '7': |
| 482 | case '8': |
| 483 | case '9': { |
| 484 | QByteArray ba; |
| 485 | ba += char(yyCh); |
| 486 | yyCh = getChar(); |
| 487 | const bool hex = yyCh == 'x'; |
| 488 | if (hex) { |
| 489 | ba += char(yyCh); |
| 490 | yyCh = getChar(); |
| 491 | } |
| 492 | while ((hex ? std::isxdigit(yyCh) : std::isdigit(yyCh))) { |
| 493 | ba += char(yyCh); |
| 494 | yyCh = getChar(); |
| 495 | } |
| 496 | bool ok; |
| 497 | auto v = ba.toLongLong(ok: &ok); |
| 498 | Q_UNUSED(v); |
| 499 | if (ok) |
| 500 | return Tok_Integer; |
| 501 | break; |
| 502 | } |
| 503 | default: |
| 504 | yyCh = getChar(); |
| 505 | } |
| 506 | } |
| 507 | return Tok_Eof; |
| 508 | } |
| 509 | |
| 510 | bool match(Token t) |
| 511 | { |
| 512 | const bool matches = (yyTok == t); |
| 513 | if (matches) |
| 514 | yyTok = getToken(); |
| 515 | return matches; |
| 516 | } |
| 517 | |
| 518 | bool matchStringStart() |
| 519 | { |
| 520 | if (yyTok == Tok_String) |
| 521 | return true; |
| 522 | // Check for f"bla{var}" and raw strings r"bla". |
| 523 | if (yyTok == Tok_Ident && yyIdent.size() == 1) { |
| 524 | switch (yyIdent.at(i: 0)) { |
| 525 | case 'r': |
| 526 | yyTok = getToken(stringType: StringType::RawString); |
| 527 | return yyTok == Tok_String; |
| 528 | case 'f': |
| 529 | yyTok = getToken(stringType: StringType::FormatString); |
| 530 | return yyTok == Tok_String; |
| 531 | } |
| 532 | } |
| 533 | return false; |
| 534 | } |
| 535 | |
| 536 | bool matchString(QByteArray *s) |
| 537 | { |
| 538 | s->clear(); |
| 539 | bool ok = false; |
| 540 | while (matchStringStart()) { |
| 541 | *s += yyString; |
| 542 | yyTok = getToken(); |
| 543 | ok = true; |
| 544 | } |
| 545 | return ok; |
| 546 | } |
| 547 | |
| 548 | bool matchEncoding(bool *utf8) |
| 549 | { |
| 550 | // Remove any leading module paths. |
| 551 | if (yyTok == Tok_Ident && std::strcmp(s1: yyIdent, s2: "PySide6" ) == 0) { |
| 552 | yyTok = getToken(); |
| 553 | |
| 554 | if (yyTok != Tok_Dot) |
| 555 | return false; |
| 556 | |
| 557 | yyTok = getToken(); |
| 558 | } |
| 559 | |
| 560 | if (yyTok == Tok_Ident |
| 561 | && (std::strcmp(s1: yyIdent, s2: "QtGui" ) == 0 || std::strcmp(s1: yyIdent, s2: "QtCore" ) == 0)) { |
| 562 | yyTok = getToken(); |
| 563 | |
| 564 | if (yyTok != Tok_Dot) |
| 565 | return false; |
| 566 | |
| 567 | yyTok = getToken(); |
| 568 | } |
| 569 | |
| 570 | if (yyTok == Tok_Ident) { |
| 571 | if (std::strcmp(s1: yyIdent, s2: "QApplication" ) == 0 |
| 572 | || std::strcmp(s1: yyIdent, s2: "QGuiApplication" ) == 0 |
| 573 | || std::strcmp(s1: yyIdent, s2: "QCoreApplication" ) == 0) { |
| 574 | yyTok = getToken(); |
| 575 | |
| 576 | if (yyTok == Tok_Dot) |
| 577 | yyTok = getToken(); |
| 578 | } |
| 579 | |
| 580 | *utf8 = QByteArray(yyIdent).endsWith(bv: "UTF8" ); |
| 581 | yyTok = getToken(); |
| 582 | return true; |
| 583 | } |
| 584 | return false; |
| 585 | } |
| 586 | |
| 587 | bool matchStringOrNone(QByteArray *s) |
| 588 | { |
| 589 | bool matches = matchString(s); |
| 590 | |
| 591 | if (!matches) |
| 592 | matches = match(t: Tok_None); |
| 593 | |
| 594 | return matches; |
| 595 | } |
| 596 | |
| 597 | /* |
| 598 | * match any expression that can return a number, which can be |
| 599 | * 1. Literal number (e.g. '11') |
| 600 | * 2. simple identifier (e.g. 'm_count') |
| 601 | * 3. simple function call (e.g. 'size()') |
| 602 | * 4. function call on an object (e.g. 'list.size()') |
| 603 | * * Other cases: |
| 604 | * size(2,4) |
| 605 | * list().size() |
| 606 | * list(a,b).size(2,4) |
| 607 | * etc... |
| 608 | */ |
| 609 | bool matchExpression() |
| 610 | { |
| 611 | if (match(t: Tok_Integer)) |
| 612 | return true; |
| 613 | |
| 614 | int parenlevel = 0; |
| 615 | while (match(t: Tok_Ident) || parenlevel > 0) { |
| 616 | if (yyTok == Tok_RightParen) { |
| 617 | if (parenlevel == 0) |
| 618 | break; |
| 619 | --parenlevel; |
| 620 | yyTok = getToken(); |
| 621 | } else if (yyTok == Tok_LeftParen) { |
| 622 | yyTok = getToken(); |
| 623 | if (yyTok == Tok_RightParen) { |
| 624 | yyTok = getToken(); |
| 625 | } else { |
| 626 | ++parenlevel; |
| 627 | } |
| 628 | } else if (yyTok == Tok_Ident) { |
| 629 | continue; |
| 630 | } else if (parenlevel == 0) { |
| 631 | return false; |
| 632 | } |
| 633 | } |
| 634 | return true; |
| 635 | } |
| 636 | |
| 637 | bool parseTranslate(QByteArray *text, QByteArray *context, QByteArray *, bool *utf8, |
| 638 | bool *plural) |
| 639 | { |
| 640 | text->clear(); |
| 641 | context->clear(); |
| 642 | comment->clear(); |
| 643 | *utf8 = false; |
| 644 | *plural = false; |
| 645 | |
| 646 | yyTok = getToken(); |
| 647 | if (!match(t: Tok_LeftParen) || !matchString(s: context) || !match(t: Tok_Comma) |
| 648 | || !matchString(s: text)) { |
| 649 | return false; |
| 650 | } |
| 651 | |
| 652 | if (match(t: Tok_RightParen)) |
| 653 | return true; |
| 654 | |
| 655 | // not a comma or a right paren, illegal syntax |
| 656 | if (!match(t: Tok_Comma)) |
| 657 | return false; |
| 658 | |
| 659 | // python accepts trailing commas within parenthesis, so allow a comma with nothing after |
| 660 | if (match(t: Tok_RightParen)) |
| 661 | return true; |
| 662 | |
| 663 | // check for comment |
| 664 | if (!matchStringOrNone(s: comment)) |
| 665 | return false; // not a comment, or a trailing comma... something is wrong |
| 666 | |
| 667 | if (match(t: Tok_RightParen)) |
| 668 | return true; |
| 669 | |
| 670 | // not a comma or a right paren, illegal syntax |
| 671 | if (!match(t: Tok_Comma)) |
| 672 | return false; |
| 673 | |
| 674 | // python accepts trailing commas within parenthesis, so allow a comma with nothing after |
| 675 | if (match(t: Tok_RightParen)) |
| 676 | return true; |
| 677 | |
| 678 | // look for optional encoding information |
| 679 | if (matchEncoding(utf8)) { |
| 680 | if (match(t: Tok_RightParen)) |
| 681 | return true; |
| 682 | |
| 683 | // not a comma or a right paren, illegal syntax |
| 684 | if (!match(t: Tok_Comma)) |
| 685 | return false; |
| 686 | |
| 687 | // python accepts trailing commas within parenthesis, so allow a comma with nothing |
| 688 | // after |
| 689 | if (match(t: Tok_RightParen)) |
| 690 | return true; |
| 691 | } |
| 692 | |
| 693 | // Must be a plural expression |
| 694 | if (!matchExpression()) |
| 695 | return false; |
| 696 | |
| 697 | *plural = true; |
| 698 | |
| 699 | // Ignore any trailing comma here |
| 700 | match(t: Tok_Comma); |
| 701 | |
| 702 | // This must be the end, or there are too many parameters |
| 703 | if (match(t: Tok_RightParen)) |
| 704 | return true; |
| 705 | |
| 706 | return false; |
| 707 | } |
| 708 | |
| 709 | void setMessageParameters(TranslatorMessage *message, const MetaStrings &meta) |
| 710 | { |
| 711 | // PYSIDE-2863: parseTranslate() can read past the message |
| 712 | // and capture extraComments intended for the next message. |
| 713 | // Use only extraComments for the current message. |
| 714 | |
| 715 | message->setExtraComment(ParserTool::transcode(str: meta.extracomment().simplified())); |
| 716 | message->setId(meta.msgid()); |
| 717 | message->setExtras(meta.extra()); |
| 718 | if (!meta.label().isEmpty() && meta.msgid().isEmpty()) |
| 719 | m_cd.appendError(error: "%1:%2: labels cannot be used with text-based translation. " |
| 720 | "Ignoring\n"_L1 .arg(args&: yyFileName) |
| 721 | .arg(a: yyLineNo)); |
| 722 | else |
| 723 | message->setLabel(meta.label()); |
| 724 | } |
| 725 | |
| 726 | QString yyFileName; |
| 727 | Token yyTok{}; |
| 728 | int yyCh{}; |
| 729 | QByteArray yyIdent; |
| 730 | char yyString[65536]; |
| 731 | size_t yyStringLen{}; |
| 732 | int yyParenDepth{}; |
| 733 | int yyLineNo = 1; |
| 734 | int yyCurLineNo{}; |
| 735 | // the file to read from (if reading from a file) |
| 736 | FILE *yyInFile; |
| 737 | // the string to read from and current position in the string (otherwise) |
| 738 | int yyInPos{}; |
| 739 | int buf{}; |
| 740 | int yyIndentationSize{}; |
| 741 | int yyContinuousSpaceCount{}; |
| 742 | bool yyCountingIndentation = false; |
| 743 | // (Context, indentation level) pair. |
| 744 | using ContextPair = QPair<QByteArray, int>; |
| 745 | // Stack of (Context, indentation level) pairs. |
| 746 | using ContextStack = QStack<ContextPair>; |
| 747 | ContextStack yyContextStack; |
| 748 | MetaStrings metaStrings; |
| 749 | Translator &tor; |
| 750 | ConversionData &m_cd; |
| 751 | }; |
| 752 | |
| 753 | bool loadPython(Translator &translator, const QString &fileName, ConversionData &cd) |
| 754 | { |
| 755 | |
| 756 | bool error = false; |
| 757 | PythonParser parser(translator, fileName, error, cd); |
| 758 | if (error) { |
| 759 | cd.appendError(QStringLiteral("Cannot open %1" ).arg(a: fileName)); |
| 760 | return false; |
| 761 | } |
| 762 | |
| 763 | parser.parse(); |
| 764 | return true; |
| 765 | } |
| 766 | |
| 767 | QT_END_NAMESPACE |
| 768 | |