Console View
|
|
|
|||
|
Sune Vuorela
sune@vuorela.dk |
|
||
|
Debug feature for cryptographic signatures While debugging, it is nice to be able to extract both signatures and the data the signature is supposed to be over. |
|||
|
Sune Vuorela
sune@vuorela.dk |
|
||
|
Better signing error handling Just having a bool for many different failure modes is a bit hard to work with, so now we have an enum for the new API, so let's extend it with some of the error possibilities |
|||
|
Sune Vuorela
sune@vuorela.dk |
|
||
| createGfx: use unique pointer for creation | |||
|
Sune Vuorela
sune@vuorela.dk |
|
||
| Signature GpgME backend: Add Kyber to switch | |||
|
Sune Vuorela
sune@vuorela.dk |
|
||
| GfxFont: memory cleanup | |||
|
Albert Astals Cid
aacid@kde.org |
|
||
|
Simplify getting the Font dict We want to both get the Dict object and if it is a Ref, get its ref because we use it later. Previously we did lookup the Font Object from the dict without fetching (NF), and if it was a Ref, we had to fetch to get the actual Dict, and if it was a Dict then all was good. New code uses the "new" lookup function that returns the Ref as an output parameter, so we are doing exactly the same but the code is much simpler |
|||
|
Albert Astals Cid
aacid@kde.org |
|
||
| Update (C) | |||
|
Oliver Sander
oliver.sander@tu-dresden.de |
|
||
| GfxFontDict: Constify a pointer argument | |||
|
Oliver Sander
oliver.sander@tu-dresden.de |
|
||
| Gfx.cc: Reduce scope of local variable | |||
|
Oliver Sander
oliver.sander@tu-dresden.de |
|
||
| GfxResources: Store 'fonts' in a std::unique_ptr | |||
|
Nicolas Fella
nicolas.fella@kde.org |
|
||
|
Handle font size 0 in AnnotAppearanceBuilder::drawSignatureFieldText 0 means 'choose the size according to available space', like we do elsewhere |
|||
|
Nicolas Fella
nicolas.fella@kdab.com |
|
||
|
Set border on signature annotation before generating appearance The appearance depends on the border |
|||
|
Albert Astals Cid
aacid@kde.org |
|
||
| Update (C) | |||
|
G B
glen.browman@veeva.com |
|
||
| Override encoding for Zapf Dingbats | |||
|
Albert Astals Cid
aacid@kde.org |
|
||
| Fix build with in-tree build | |||
|
Albert Astals Cid
aacid@kde.org |
|
||
| Update (C) | |||
|
Albert Astals Cid
aacid@kde.org |
|
||
| export StructTreeRoot so we make it clear it is used/usable | |||
|
Albert Astals Cid
aacid@kde.org |
|
||
|
Install StructTreeRoot.h again I didn't realize it was usable without being exported |
|||
|
Albert Astals Cid
aacid@kde.org |
|
||
| Add a few exports more to classes that can be used from other exported classes | |||
|
Albert Astals Cid
aacid@kde.org |
|
||
|
Install SplashFont again I failed realizing it was usable |
|||
|
Albert Astals Cid
aacid@kde.org |
|
||
| Update (C) | |||
|
Albert Astals Cid
aacid@kde.org |
|
||
|
Add export to ViewerPreferences There's people sadly using this class on the wild so make it clear it's used otherwise we will make it not installed again in the future |
|||
|
Albert Astals Cid
aacid@kde.org |
|
||
|
Install ViewerPreferences again I failed realizing it was usable given it's all inline functions |
|||
|
Nelson Benítez León
nbenitezl@gmail.com |
|
||
|
cairo: Fix JPX transparency regression Commit 2dcfefb43c from https://bugs.freedesktop.org/93468 fixed images 2) and 3) of file[1] (which are jpeg images with transparency) from showing garbage to show fine but with a solid background instead of transparency. But this fix interferes with our previous commit where we are adding support to Cairo backend to correctly show transparency in jpeg files that have it. So we need to partially revert commit 2dcfefb43c but only for the Cairo backend, because the Splash backend still uses this fix to improve some jpeg transparent images by at least showing them with a solid background. So we added a new method JPXStream::setSupportJPXtransparency() so a JPXStream can be informed that the OutputDevice is capable of handling jpx transparency and so to not perform the downgrade fix to solid background. The OutputDevice itself can be marked as handling jpx transparency by new method OutputDev::supportJPXtransparency() which the Cairo backend implements as TRUE. So now file[1] does not regress in Cairo backend, although display changes from showing with solid background to show with proper transparency as CairoOutputDev now supports that. For the Splash backend no display regressions should happen at all, because any display changes are now guard by OutpuDev::supportJPXtransparency() which the Splash backend returns FALSE at this moment. [1] https://bugs.freedesktop.org/attachment.cgi?id=120624 Part of #1486 |
|||
|
Nelson Benítez León
nbenitezl@gmail.com |
|
||
|
cairo: Fix rendering of JPX images with transparency Handle transparent JPX images, they may contain RGBA data when no ColorSpace pdf dict is defined or when SMaskInData is non-zero. PDF files posted in below issues are fixed by this commit: https://gitlab.freedesktop.org/poppler/poppler/-/issues/1486 https://github.com/mozilla/pdf.js/issues/16782 https://github.com/mozilla/pdf.js/issues/11306 https://github.com/mozilla/pdf.js/issues/17416 Inspired by related fix in pdf.js: https://github.com/mozilla/pdf.js/pull/18204 While working on this commit we also succesfully addressed regressions that emerged for the following files: https://github.com/mozilla/pdf.js/issues/18896 https://bugs.freedesktop.org/attachment.cgi?id=49294 Issue #1486 |
|||
|
lbaudin1
lucas.baudin@ensae.fr |
|
||
| glib: Add free text annotations support | |||
|
Albert Astals Cid
aacid@kde.org |
|
||
| poppler 24.11.0 | |||
|
Albert Astals Cid
aacid@kde.org |
|
||
| CI: clang-format 19 | |||
|
Albert Astals Cid
aacid@kde.org |
|
||
| Update (C) | |||
|
Albert Astals Cid
aacid@kde.org |
|
||
| CI: Fedora mingw 40 -> 41 | |||
|
Nicolas Fella
nicolas.fella@kdab.com |
|
||
|
Increase refcount for signature annotation when creating Otherwise it's deleted too early |
|||
|
Albert Astals Cid
aacid@kde.org |
|
||
| qt: respect non-ascii encodings in Launch target file names | |||
|
v4hn
me@v4hn.de |
|
||
|
respect non-ascii encodings in Launch target file names PDF2.0 (and also 1.7) - section 7.11.3 specifies the UF field as a "text string" in the pdf sense, which might be encoded in one of three encodings UTF16. The encoding was ignored here before resulting in broken links when (1) the UF field was actually used and (2) the contents was a marked UTF16 string that would be interpreted as UTF8. I considered to adjust the parameters too, but these are not officially defined as text -either "undefined" or partly as "byte string"- and have been deprecated in PDF2.0. |
|||
|
Sune Vuorela
sune@vuorela.dk |
|
||
| Cleanups after 9d43c190017e621f83cd | |||
|
Sune Vuorela
sune@vuorela.dk |
|
||
|
CharCodeToUnicode memory handling cleanup Rather than do manual ref/deref and delete this, leverage smartpointers instead. |
|||
|
Sune Vuorela
sune@vuorela.dk |
|
||
| Do a little less manual memory handling in splash | |||
|
Albert Astals Cid
aacid@kde.org |
|
||
| Update (C) | |||
|
Albert Astals Cid
aacid@kde.org |
|
||
|
Force passing faceIndex to all functions that may need it I'm relatively sure that for embedded fonts it always will be 0, but since we have the variable use it Fixes issue #1512 in Splash and PS output devices |
|||
|
Albert Astals Cid
aacid@kde.org |
|
||
| Remove the path functions from GfxFontLoc | |||
|
Albert Astals Cid
aacid@kde.org |
|
||
| Return optional string instead of GooString * in some GlobalParam functions | |||
