summaryrefslogtreecommitdiffstats
path: root/tools/linguist/shared
Commit message (Collapse)AuthorAgeFilesLines
* Fix locale encoding in XLIFF filesAndy Shaw2009-09-301-4/+11
| | | | | | xlf uses xx-YY notation instead of xx_YY Reviewed-by: ossi
* Update license headers again.Jason McDonald2009-09-0828-112/+112
| | | | Reviewed-by: Trust Me
* Add missing license headersJason McDonald2009-09-031-0/+40
| | | | Reviewed-by: Trust Me
* Update license headers.Jason McDonald2009-09-021-3/+0
| | | | Reviewed-by: Trust Me
* Update tech preview license header.Jason McDonald2009-08-3127-351/+351
| | | | Reviewed-by: Trust Me
* remove withdrawn attributes (and fix syntax)Oswald Buddenhagen2009-08-141-9/+0
|
* Update license headers.Jason McDonald2009-08-1127-27/+27
| | | | Reviewed-by: Trust Me
* Fix compilation with xlC 7: operands to ?: must match.Thiago Macieira2009-08-071-2/+2
| | | | | | | See 3ae2cab9c8bd1790a00da2755ac036143a3a35f4 for another similar fix. Reviewed-by: Trust Me (cherry picked from commit 18fbfdf0f774198e2e1277e064cc3a8eb9dbb29d)
* Fix compilation with xlC 7: operands to ?: must match.Thiago Macieira2009-08-071-1/+1
| | | | | | | "../shared/qm.cpp", line 556.45: 1540-0207 (S) No common type found for operands with type "const char [7]" and "QByteArray". Reviewed-by: Trust Me (cherry picked from commit 3ae2cab9c8bd1790a00da2755ac036143a3a35f4)
* don't rely on system codec when handling PO filesOswald Buddenhagen2009-07-173-4/+7
| | | | | - make -input-codec affect PO files, default to UTF-8 - add -output-codec for PO files, same default
* reorganize numerus form count normalizationOswald Buddenhagen2009-07-175-29/+39
| | | | | | replace implicit normalization of individual messages on file writeout with global normalization which is called by the command line tools. this should a) be faster and b) cover the most critical case: lrelease.
* implement proper vpath handlingOswald Buddenhagen2009-07-134-113/+92
| | | | | | | | | | | | this also removes the bogus special casing of various filename-containing variables inside the pro parser. while this is a feature as such, it restores backwards compatibility without damaging the correct behavior again, so it qualifies for 4.5. based on a2f40fce2a1cf3c19a13fa27eea08192493ab76e from creator Task-number: 254098
* factor out linguist-specific partOswald Buddenhagen2009-07-135-66/+166
| | | | it is simpler to keep it in sync with creator then.
* Fix memory leak in $$system() calls from .pro filesOswald Buddenhagen2009-07-101-0/+4
| | | | cherry-picked 07730341bd739aac823ac9b4336d8294510a35e2 from creator
* remove totally pointless conditionalOswald Buddenhagen2009-07-101-1/+1
| | | | cherry-picked 167a09b20614d282ec898f69fc2a3f0bafa11229 from creator
* remove dead codeOswald Buddenhagen2009-07-101-10/+0
| | | | cherry-picked d8b1cc5f0ecbb8de734d241d72a05b325c2bbb2c from creator
* mark plural messages as such in the qm readerOswald Buddenhagen2009-06-231-0/+18
| | | | | | this contains both a "clean" code path based on translation count and a fallback path based on looking for "%n" (for languages with only one form).
* add -input-codec option for qm readerOswald Buddenhagen2009-06-231-3/+2
| | | | | basing the input codec on the locale was broken by design and didn't really work anyway. so adding this option really is a bugfix, kind of.
* Update license headers as requested by the marketing department.Jason McDonald2009-06-1625-50/+50
| | | | Reviewed-by: Trust Me
* no point in ProFile being a QObjectOswald Buddenhagen2009-06-151-4/+1
|
* remove unused includesOswald Buddenhagen2009-06-152-2/+0
|
* fix arabic pluralsOswald Buddenhagen2009-06-111-2/+2
|
* add plural rules for urduOswald Buddenhagen2009-06-111-0/+1
|
* don't let comments weaved into strings confuse usOswald Buddenhagen2009-06-101-1/+3
| | | | | | Reviewed-by: TrustMe Task-number: 254260 Testcase: 323072@p4
* explicitly handle windows and mac9 line endingsOswald Buddenhagen2009-06-051-5/+20
| | | | | | in practice, this matters only for backslashed line continuations Task-number: 255336
* fix catalan plural rulesOswald Buddenhagen2009-06-031-2/+7
|
* rename some plural formsOswald Buddenhagen2009-06-031-4/+3
|
* drop traditional czech plural rules in favor of today's slovak-like rulesOswald Buddenhagen2009-06-031-8/+2
|
* fix turkish plural formsOswald Buddenhagen2009-06-031-1/+1
|
* fix tagalog plural formsOswald Buddenhagen2009-06-031-2/+8
|
* fix icelandic plural formsOswald Buddenhagen2009-06-031-1/+5
|
* fix lithuanian plural rulesOswald Buddenhagen2009-06-031-2/+2
|
* Fix parsing method calls with null arguments in Java codeEskil Abrahamsen Blomfeldt2009-05-201-5/+11
| | | | | | | | | | | In Java, "null" is represented as a keyword, not as the integer 0. The old code assumed the latter. Code such as translate("fooBar", "fooBar", null); would thus not be detected by lupdate when parsing Java files. Reviewed-by: ossi
* protect against circular inclusion of pro/pri filesOswald Buddenhagen2009-04-151-1/+7
| | | | Task-number: 250574
* do not crash on lines starting with an equal signOswald Buddenhagen2009-04-151-0/+3
|
* fix uninitialized variable readOswald Buddenhagen2009-04-151-4/+12
|
* add some commentsOswald Buddenhagen2009-04-151-4/+4
|
* compile with aCCOswald Buddenhagen2009-04-071-2/+2
|
* re-initialize some variables for each fileOswald Buddenhagen2009-04-021-5/+4
|
* fix evaluation of CONFIG() and contains()Oswald Buddenhagen2009-04-021-3/+4
| | | | cherry-picked from creator's b3ec859c80bd0656cfc48cfe6514ec4f631b8206
* fix wrong evaluation of arguments to qmake functionsOswald Buddenhagen2009-04-021-1/+2
| | | | cherry-pick 28dacdfdf3eed04ec47a1e8eb206bd3ffb979c08 from creator
* duplicate message handling improvementsOswald Buddenhagen2009-04-025-50/+67
| | | | | | | | | | | | | | | - lrelease will not abort on duplicates any more - lconvert now gets noisy, but only if messages within one file are duplicated (combining files with identical messages is ok) - lupdate stays silent, but will eliminate duplicates again - consolidate handling of dual-encoded and duplicate messages - for performance - instead of the format loader, now the app is responsible for calling the duplicate handler. this allows for the fine-grained control necessary for optimal performance. Task-number: 247738
* optimize duplicate resolutionOswald Buddenhagen2009-04-021-5/+26
| | | | use pointers in the index hash to avoid needless allocs and copies
* properly process backslash line continuationsOswald Buddenhagen2009-03-271-6/+13
| | | | | | this is a backport of be5a9587865f9f042cac8feb5296b71b11a1a80f Task-number: 249633
* static is even better than constOswald Buddenhagen2009-03-271-4/+4
|
* declare type info next to the type, not "somewhere"Oswald Buddenhagen2009-03-272-2/+2
|
* properly deal with messages which appear in multiple encodingsOswald Buddenhagen2009-03-276-220/+299
| | | | | | | | | | | | | in ts 1.1 and qm files, messages appear in their native encoding. that means that a message can appear multiple times - once in utf8 and once in the codecForTr. however, in ts 2.0 files, everything is utf8 and messages can have a utf8 flag for the later transformation into qm. unfortunately, there was no flag to mark that the message is needed in *both* encodings, and the respective case was completely ignored when reading ts 1.1 and qm files (causing error messages). Task-number: 249022 AutoTest: 322690
* micro-optimization: don't evaluate same condition twiceOswald Buddenhagen2009-03-271-2/+2
|
* actually ignore untranslated messages as it claimsOswald Buddenhagen2009-03-271-2/+4
|
* avoid empty codec nameOswald Buddenhagen2009-03-272-2/+4
|