summaryrefslogtreecommitdiff
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
* Minor cosmetic fixesMarti Maria2022-01-072-1/+5
| | | | | | - Left typos - meson build by #292 - Fix valgrind warning on writing uninitialized memory
* Final changes for 2.13Marti Maria2022-01-0610-20/+20
| | | | | | - typos - changelog - readme.1st
* Make sure context mutex critical section is initializedMarti Maria2022-01-061-26/+57
| | | | | AppVerifier complains in some cases if the critical section is not initialized. Should address #289
* better utf16 to wchar_tMarti Maria2022-01-051-0/+47
| | | | Try to promote correctly to utf8 to wchar_t when unicode is 32 bits
* first 2.13 betaMarti Maria2021-12-223-8/+8
| | | | Making things ready for 2.13 release
* Premultiplied alpha works on tifficcMarti Maria2021-12-221-16/+18
| | | | Some tested formats are now working
* tifficc premultiplied alphaMarti Maria2021-12-211-26/+26
| | | | | First attempt to add premultiplied alpha support Other cosmetic changes
* fix an overflow conditionMarti Maria2021-12-121-1/+1
| | | | on ill-formed CGATS file
* Missing part from last commitMarti Maria2021-12-101-0/+1
| | | | My fault on merging, sorry
* update CGATS parser to deal with very long valuesMarti Maria2021-12-101-34/+108
| | | | Allows larger strings on datasets
* More work on premultiplied alphaMarti Maria2021-12-101-168/+211
| | | | Still not tested, nor covered by testbed
* more work on pre-multiplied alphaMarti Maria2021-12-032-2/+192
| | | | Ongoing effort
* Fix parametric curves 4 and 5 inversion safeguardMarti Maria2021-11-301-29/+35
| | | | | | A bug introduced circa 2017 in a try of preventing security risks. Parametric curve inversions needs divisions, the change was trying to prevent a division by zero. Unfortunately it was discarding legit cases as well. Now it seems to be done correctly.
* Safeguard against infinite .include recursionMarti Maria2021-11-231-19/+38
| | | | Detected by oss-fuzz
* Recreate autotools filesmm22021-11-061-2/+2
| | | | fix a typo in pointer casts
* Fix for #287Marti Maria2021-10-241-28/+42
| | | | | Fix for #286 Multiprocessor elements sampled curves missing node
* cmsReadTag memory raceSergey Bylokhov2021-09-051-1/+2
|
* Unbounded Transform for integer-to-float formatsMarti Maria2021-09-012-2/+189
| | | | Color transforms going from integer types to floating point works now in unbounded mode. Thanks to John Papandriopoulos for discovering the issue.
* Prefer to use gmtime_s insteadKleis Auke Wolthuizen2021-07-141-3/+3
| | | | | | | | | MinGW-w64 defines _gmtime64_s on both i686 and x86_64 targets. Prefer to use the gmtime_s macro to avoid that _gmtime64_s is being called on Windows 32-bit, which will not work. This commit also generally ensures that a threadsafe alternative is provided for toolchains targeting Windows i686.
* Remove all autotools filesmm22021-06-241-0/+1
|
* Add automatic linear space detectionMarti Maria2021-06-204-19/+85
| | | | | - Added a function to estimate gamma space of RGB profiles (only a subset). This function is still undocumented. - Lcms now automatically turns off optimizations when 16-bit on RGB and input linear space is found.
* Provide a thread-safe get time functionMarti Maria2021-06-093-16/+33
| | | | Use the context pool mutex in the case of none of the re-entrant alternatives to gmtime() is available
* Ensure threadsafe alternative of gmtime is usedKleis Auke Wolthuizen2021-06-041-1/+18
|
* add cmsGetStageContextIDMarti Maria2021-04-132-0/+6
| | | | cmsGetStageContextID function per issue #253
* Fixes from Artifexmm22021-02-222-4/+21
| | | | Thanks to Robin Watts for pointing out
* Merge branch 'master' of https://github.com/mm2/Little-CMSMarti Maria2021-01-223-7/+7
|\
| * fix some compiler warningsmm22021-01-223-7/+7
| | | | | | | | Thanks to Noel Carboni to report them
* | prevent segfaults when NULL country or language codeMarti Maria2021-01-201-3/+7
|/ | | | Add codeguard. Thanks to Christian Schmitz for pointing it out
* Prevent segfault on atoi()Marti Maria2021-01-181-6/+14
| | | | | Prevent segfaults on atoi when input buffer is NULL (harmless, non-exploitable)
* fix possible out-ob-bounds readMarti Maria2021-01-051-3/+4
| | | | Thanks to tsdgeos for finding the bug
* Clean the _cmsInterpFn16 messMarti Maria2020-12-204-18/+12
| | | | A cleanup of a function pointer being mixed in several places
* Update cmsopt.cMarti Maria2020-12-181-1/+1
| | | | fix const mismatch
* revert a wrong fixupMarti Maria2020-12-181-1/+1
| | | | Ops, I did too much fixing
* More warnings removedMarti Maria2020-12-181-9/+11
| | | | regarding different pointer types
* remove warnings & add updated docMarti Maria2020-12-181-2/+2
| | | | | remove pointer conversion warnings updated plugin documentation
* Black preservation on multiprofile transformsMarti Maria2020-12-121-14/+79
| | | | Black preservation now works on multiprofile transforms that have CMYK->CMYK devicelinks at the end. This improves scenarios where black preservation AND ink-limiting needs to be used together.
* Solve Incompatible function pointer type casts #145Marti Maria2020-12-083-3/+3
| | | | Thanks to vlad902 for discovering the issue
* Autotools for latest changesmm22020-12-031-2/+3
|
* Add Lab -> RGB/CMYK/Gray/Lab kernel to fast float pluginMarti Maria2020-12-023-8/+44
| | | | | | First try, although it seems to work. Is is more accurate close to neutral axis and less accurate on self-flourescent hyper-saturated colors, which are rare. configure have to be regenerated (it will be wiped out soon)
* Plugin supports RGB->LabMarti Maria2020-11-221-27/+24
| | | | | | Add support for "RGB to any" in the fast float plug-in cleanup of interpolation (cosmetics) ongoing "lab to any" support in the fast float plug-in
* Update cmsgamma.cMarti Maria2020-11-141-3/+10
| | | | add ability to bypass checks
* First try for fast float plugin automakemm22020-11-024-7/+7
| | | | | | Early 2.12 alpha includes dynamic linking with fast float plugin Next steps in removing generated files from repository
* fast plugin update to development sourcesMarti Maria2020-10-021-0/+6
| | | | | | | Fixes several issues on C++ (functions not found due to mangling) Converts all optimizers to version 2 Adds complete alpha handling Improves speed on certain cases
* cosmetic changesMarti Maria2020-10-014-3/+4
| | | | to make codeQL happy
* Update cmsopt.cMarti Maria2020-09-291-3/+3
| | | | remove unnecessary asignments
* added interpolation kernels for more input channelsMarti Maria2020-09-031-341/+139
| | | | Thanks to Carles Llopis for the idean and the code
* Fix matrix write on LUT8Marti Maria2020-08-041-7/+5
| | | | LUT16 was fixed but LUT8 not, so fix it as well.
* Two checks on null pointersMarti Maria2020-07-262-0/+16
| | | | Both are on malloc for few bytes, so arent't necessary at all.
* some fixes for XCodemm22020-07-151-1/+1
| | | | XCode found some issues
* setlocale damage controlMarti Maria2020-07-071-0/+8
| | | | setlocale can mess out things, take drastic action
View Lorry Controller Status