summaryrefslogtreecommitdiff
path: root/include
Commit message (Collapse)AuthorAgeFilesLines
* Misc. typosluz.paz2018-05-211-1/+1
| | | Found via `codespell -q 3`
* adjusting rc3Marti Maria2017-10-231-1/+1
|
* Revert "revert "-fvisibility=hidden""Thomas Weber2017-10-211-1/+6
| | | | | | | This reverts commit 4ce2fee79c35465b9f182a2e55b7b703a5e60594. Further work on the ax_gcc_func_attribute.m4 will allow this to work on OSX.
* revert "-fvisibility=hidden"Marti Maria2017-10-061-6/+1
| | | | It does not work on OSX
* Revert last tryMarti Maria2017-09-281-1/+1
|
* fixed visibility on OSXMarti Maria2017-09-281-1/+1
| | | | Try to fix visibility
* upgrade autotools to 2.9Marti Maria2017-09-271-6/+24
| | | | Matches release candidate
* Get rid of warnings when compiling as C++ with -WallMarti Maria2017-08-021-12/+12
| | | | | Thanks to Noel Carboni for suggesting & providing a proof that it could be done
* Prepare for 2.9 releaseMarti Maria2017-07-142-9/+9
| | | | | Set copyright notices and version stamp, Still there are missing files, like documentation
* Fix windows static buildsİsmail Dönmez2017-02-031-0/+3
| | | After the latest changes CMSEXPORT and CMSAPI were undefined for Windows static (non-DLL) builds.
* Explicitly mark symbols meant for export under unixoid systemsThomas Weber2017-02-021-7/+9
| | | | | | | The compiler has to support this, but most semi-modern compilers do. This brings these builds closer to the Windows builds. If the compiler does not support hiding of symbols, all symbols will be exported, just as before.
* spelling fixesklemens2017-01-222-5/+5
|
* fix IT8 parsing glitchMarti Maria2016-12-041-1/+1
| | | | | Fixed a bug when trying to parse junk as IT8 file. Added the corresponding test case
* Better critical section initializationMarti2016-08-231-0/+3
| | | | | | | Contributed patch from Robin Watts (Artifex). Prevents ApplicationVerifer to complain about the static initialization of critical section. Applies to windows only.
* Revert "better critical section initialization"Marti2016-08-231-3/+0
| | | | This reverts commit c964a251268e95c40d170c64c6eb84b594d91de0.
* better critical section initializationMarti2016-08-231-0/+3
| | | | | | Contributed patch from Robin Watts (Artifex). Prevents ApplicationVerifier to complain about the use of static intialization of critical section. Applies only to windows.
* Fix some warnings found by coverityMarti2016-07-141-1/+1
|
* downgrade to automake 1.14.1 which is stableMarti2016-05-181-18/+6
|
* Fix some typos (found by codespell)Stefan Weil2016-05-112-3/+3
| | | | Signed-off-by: Stefan Weil <sw@weilnetz.de>
* Change copyright dateMarti2016-03-292-2/+2
|
* Added support for Argyll's ARTS tagMarti2016-01-301-1/+2
|
* Upgrade to modern autotoolsMarti2015-11-151-50/+155
| | | | Tested on a mac. Partial work to 2.8 release
* Moved a fixup on wrong XYZ valuesMarti Maria2015-08-012-3/+3
|
* Removed some wrong line endingsMarti Maria2015-06-191-46/+11
|
* Big drop sponsorized by AlienSkin SoftwareMarti Maria2015-06-182-10/+50
|
* Simplify/Improve endian decisions in lcms2.hChris Liddell2015-05-211-34/+35
| | | | | | | | | | | | | | Firstly, protect the endian decisions in lcms2.h so that settings given on the compiler command line are not silently overridden. Thus being set explicitly gets the highest priority. Secondly, use the endianness detected by the configure script, and remove the complex and error prone stuff for PPC platforms from lcms2.h. Thus the endianess from configure gets second highest priority. Thirdly, if neither of the above are set, fall back to some simple (and long standing) platform/compiler specific tests. These should rarely, if ever, come into play in a "normal" build for a "normal" Unix-like system.
* Added a version retrieval functionlcms2.7rc3lcms2-2.7Marti Maria2015-03-171-1/+5
|
* lcms2-2.7 rc2 bumpMarti Maria2015-02-211-1/+1
|
* Merge from private branch in preparation of 2.7 rc1Marti Maria2014-12-172-5/+12
|
* Added a function to retrieve the iohandler of a given profile objectMarti Maria2014-08-121-0/+1
|
* Fix (again) endianess detection on PowerPCMarti Maria2014-04-171-15/+13
|
* Documentation for 2.6Marti Maria2014-03-161-2/+2
|
* Some copyright updatesMarti Maria2014-03-071-6/+2
|
* Added pthread dependency, and multithreading supportMarti Maria2014-02-162-2/+16
|
* First attempt to pthread integrationMarti Maria2014-02-142-1/+5
|
* Fix for endianess detectionlcms2.6rc1Marti Maria2014-02-141-0/+3
|
* Fixed context guessingMarti Maria2014-02-141-5/+1
|
* Merge from Artifex branchlcms2.6rc0Marti Maria2014-02-122-18/+85
|
* Fixing endianess detection and memory leaksMarti Maria2013-10-301-7/+18
|
* Added a function for IOhandler writeMarti Maria2013-07-191-0/+1
|
* Final changes to 2.5lcms2.5rc2Marti Maria2013-06-171-1/+1
|
* Added access to MLU translation codesMarti Maria2013-05-101-0/+7
|
* Memory squeezing fix: lcms2 cmsPipeline constructionRobin Watts2013-03-291-1/+1
| | | | | | | | | | | When creating a new pipeline, lcms would often try to allocate a stage and pass it to cmsPipelineInsertStage without checking whether the allocation succeeded. cmsPipelineInsertStage would then assert (or crash) if it had not. The fix here is to change cmsPipelineInsertStage to check and return an error value. All calling code is then checked to test this return value and cope.
* >Changes from Richard HughesMarti Maria2013-03-261-0/+2
|
* smooth tone curve fixesMarti Maria2013-02-281-2/+2
|
* Fix LCMS threading issue with plugin registration.Robin Watts2013-02-151-0/+1
| | | | | | | | | | When registering a plugin, allocations can be done. This can result in calls back to the user supplied malloc functions, which assume that a cmsContext will be supplied. In the current code there is no way to pass a cmsContext into the plugin registration. We therefore add a new function cmsPluginTHR that does take such a context.
* Updated BPC algorithmMarti Maria2013-01-281-1/+1
|
* Upgrade autotoolsMarti Maria2012-12-031-2/+27
|
* lcms2-2.4 rc2 is hereMarti Maria2012-08-312-2/+19
|
* Bugs and modifications for 2.4rc2Marti Maria2012-08-141-0/+3
|
View Lorry Controller Status