-
Notifications
You must be signed in to change notification settings - Fork 37
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Sync with 3.1.1 #93
Sync with 3.1.1 #93
Conversation
LaurenzV
commented
Feb 7, 2024
•
edited
Loading
edited
Status | Commit message | HB Link |
---|---|---|
⚪️ | [doc] Add missing version tags | Link |
⚪️ | Make the code able to compile with MSVC++ from Visual Studio 2015 (Version 14.0.25431.01 | Link |
⚪️ | [repacker] fix heap use after free in repacker. | Link |
⚪️ | Rename HBGlyphID to HBGlyphID16 | Link |
⚪️ | Add stub HBUINT15 | Link |
⚪️ | Make the file UTF-8 again | Link |
⚪️ | [myanmar] Towards separating Myanmar from Indic table | Link |
⚪️ | [khmer] Towards separating Khmer from Indic table | Link |
🟢 | Add support for Pwo Karen / Ason Chin medial la | Link |
⚪️ | [set-digest] Use using instead of typedef | Link |
⚪️ | [subset] add closure_glyphs () method for MATH table | Link |
⚪️ | [subset] fix find_duplicate_features | Link |
⚪️ | [subset] Fix subset_offset_array adding unused space to serializer. | Link |
⚪️ | [subset] subset MATH table | Link |
⚪️ | [atomic] Cannot fail anymore | Link |
⚪️ | [mutex] Add hb_mutex_t std::mutex implementation | Link |
⚪️ | Add a debug message when offset overflow resolution fails. | Link |
⚪️ | [repacker] begin implementing the ability to isolate extension subtables. | Link |
⚪️ | [repacker] extract overflows processing into its own method. | Link |
⚪️ | [repacker] do extension subtable isolation before starting resolution attempts. | Link |
⚪️ | [repacker] recursively duplicate nodes during isolation. | Link |
⚪️ | [repacker] Add repacker test for subgraph isolation. | Link |
⚪️ | [repacker] put each 32 bit subgraph into it's own packing space. | Link |
⚪️ | [repacker] default space to 0. | Link |
⚪️ | [repacker] correctly update incoming_edges in duplicate. | Link |
⚪️ | [repacker] add a couple more complex isolation tests. | Link |
⚪️ | [repacker] add repacker isolation test on real font. | Link |
⚪️ | [repacker] reduce the bits used by order by 2 to give more bits to distance. | Link |
⚪️ | [repacker] handle a couple of duplication edge cases. | Link |
⚪️ | [repacker] s/0/HB_TAG_NONE | Link |
⚪️ | [getn-hb-version.py] Open old output in UTF-8 mode as well | Link |
⚪️ | [repacker] fix bug in subgraph isolation. | Link |
⚪️ | [subset] Fix bug in composite operation counting. | Link |
⚪️ | [repacker] begin storing each nodes parents. | Link |
⚪️ | [repacker] add space assignment based on connected components. | Link |
⚪️ | [repacker] when assigning each connected subgraph a space | Link |
⚪️ | [repacker] restrict 32 bit subgraph connected component search to only nodes reachable via directed links. | Link |
⚪️ | [subset] fuzzer fix: https://oss-fuzz.com/testcase-detail/6616166961905664 | Link |
⚪️ | [repacker] Handle the case where a subgraph root has an incoming 32 and 16 bit edge. | Link |
⚪️ | [repacker] add the ability to move subgraphs from a shared space into their own space. | Link |
🟢 | [Unicode 14] Handle U+180F FVS4 | Link |
🟢 | [USE] Remove overrides for hieroglyphic categories | Link |
🟢 | [USE] Update the data files | Link |
🟢 | [ot-shape] Tweak table order some more | Link |
🟢 | [indic] Apply 'calt' and 'clig' across syllable boundaries | Link |
🟢 | [test] Add test for harfbuzz/harfbuzz#3244 | Link |
🟢 | [test] Add test for harfbuzz/harfbuzz#2757 (comment) | Link |
⚪️ | [repacker] when assigning spaces use post isolation node indices. | Link |
⚪️ | [repacker] add test for splitting spaces. | Link |
⚪️ | [repacker] Add additional splitting spaces test. | Link |
⚪️ | [repacker] Add repacker test that requires space splitting. | Link |
⚪️ | [repacker] refactor resolve overflow tests to use a helper. | Link |
⚪️ | [repacker] add missing fini for parents vector. | Link |
⚪️ | [repacker] remove clone buffer | Link |
⚪️ | [repacker] performance optimizations for topological sorting. | Link |
🟢 | Update comments for U+180F FVS4 | Link |
🟢 | [USE] Only skip default ignorables within CCSes | Link |
🟢 | [USE] Merge the categories S and O | Link |
🟢 | [USE] Document customizations of USE categories | Link |
⚪️ | Merge pull request #3249 from harfbuzz/use-merge-s-o | Link |
🟢 | [USE] Categorize U+11046 BRAHMI VIRAMA as HALANT | Link |
🟢 | [USE] Insert dotted circles before lone viramas | Link |
🟢 | [USE] Categorize characters that only have AJT | Link |
⚪️ | [subset] code cleanup | Link |
⚪️ | [subset] Export header & archive when HB_BUILD_SUBSET | Link |
⚪️ | [subset] Refine the coding style | Link |
⚪️ | Fix Wbitwise-instead-of-logical warnings | Link |
⚪️ | Add a writeup of the overflow resolution algorithm in harfbuzz. | Link |
⚪️ | Err on -Wbitwise-instead-of-logical | Link |
⚪️ | [docs] Rename overflow_resolution to repacker | Link |
⚪️ | [repacker] use possibly updated root idx after isolate_subgraph. | Link |
⚪️ | [head] Add is_expanded() internally | Link |
⚪️ | [style] Use head->macStyle.isExtended in style fallback | Link |
⚪️ | [meson] Add graphite2 option and deprecate graphite | Link |
⚪️ | [subset] update COLRv1 table struct definitions | Link |
⚪️ | [subset] add DeltasetIndexMap struct definition | Link |
⚪️ | [subset] COLRv1: add ClipList/ClipRecord/ClipBox struct definition | Link |
⚪️ | [subset] COLRv1: update subset() method for new strutc ClipList and VarIdxMap | Link |
⚪️ | [subset] Move struct DeltaSetIndexMap into hb-ot-var-common.hh | Link |
⚪️ | Merge pull request #3248 from googlefonts/connected_components | Link |
⚪️ | Revert "[hb-directwrite] Don't load dwrit.dll dynamically" | Link |
⚪️ | Remove direct link to dwrite from meson scripts | Link |
⚪️ | [API] Add hb_buffer_[sg]et_not_found_glyph() and --not-found-glyph | Link |
⚪️ | [ci] Add CIFuzz integration | Link |
⚪️ | [subset] fuzzer fix: https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=40406 | Link |
⚪️ | Document hb_ot_name_t values | Link |
⚪️ | docs: Add a link to the OpenType spec | Link |
⚪️ | Better round positions during scaling | Link |
⚪️ | [algs] Add hb_swap() ala | Link |
⚪️ | [vector] Add initializer_list constructor & tests | Link |
⚪️ | Revert "[vector] Add initializer_list constructor & tests" | Link |
⚪️ | [vector] Add initializer_list constructor & tests | Link |
⚪️ | Add a test font scale rounding behavior | Link |
⚪️ | [vector] Implement move assignment using swap | Link |
⚪️ | [vector] Add initializer from iterable | Link |
⚪️ | [vector] Move test code into separate file | Link |
⚪️ | [vector] Pre-allocate if iterator has random-access length | Link |
⚪️ | [vector] Add more tests | Link |
⚪️ | [set] Add standard the 5 standard methods & tests | Link |
⚪️ | [bit-set-invertible] Implement standard-5 methods & swap | Link |
⚪️ | [bit-set] Implement standard-5 methods plus swap | Link |
⚪️ | [vector] Use default constructor | Link |
⚪️ | [set] More default operators | Link |
⚪️ | Verbiage | Link |
⚪️ | [vector] Add swap() friend to hb_sorted_vector_t | Link |
⚪️ | [vector] Complete sorted_vector 5-methods | Link |
⚪️ | [array] Use default methods | Link |
⚪️ | [bimap] Simplify initialization | Link |
⚪️ | [set] Add successful check in swap() | Link |
⚪️ | [map] Add 5 methods | Link |
⚪️ | [algs] Include <initializer_list> | Link |
⚪️ | [map] Add initializer_list and swap() | Link |
⚪️ | [map] Add iterable constructor | Link |
⚪️ | Simplify swap() | Link |
⚪️ | [map] Try moving items when resizing | Link |
⚪️ | [algs] include | Link |
⚪️ | [meta] Use std::is_trivially_... instead of internal copies | Link |
⚪️ | [meta] Remove unused traits | Link |
⚪️ | [algs] Use std::move instead of hb_move() | Link |
⚪️ | [meta] Use std::forward instead of hb_forward() | Link |
⚪️ | [meta] Use more std type_traits | Link |
⚪️ | [meta] Use std::conditional instead of hb_conditional() | Link |
⚪️ | [meta] Add in-house impl of hb_is_convertible() | Link |
⚪️ | [meta] Remove otherwise-empty test-meta | Link |
⚪️ | Merge pull request #3278 from harfbuzz/more-std | Link |
⚪️ | [map] Fix copy assignment operator | Link |
⚪️ | [repacker] add 'virtual links' to the serializer. | Link |
⚪️ | [repacker] Add test for virtual links in the repacker. | Link |
⚪️ | [repacker] remove VirtualOffset struct. | Link |
⚪️ | [repacker] add more detailed comment about virtual links. | Link |
⚪️ | [subset] During LigatureSubstFormat1 subsetting always place Coverage last. | Link |
⚪️ | [subset] Don't assume FeatureList is sorted | Link |
⚪️ | [subset] fuzzer fix: https://oss-fuzz.com/testcase?key=6254792024915968 | Link |
⚪️ | [subset] update subset goldens to fix tests. | Link |
⚪️ | [subset] use glyphset_gsub instead of glyphset for GDEF | Link |
⚪️ | [subset] Update test goldens for gdef.glyphset. | Link |
⚪️ | [subset] fix bug in (Chain)ContextFormat2 | Link |
⚪️ | [map] Add INVALID_KEY and INVALID_VALUE | Link |
⚪️ | [subset] Fix fuzzer found memory leak. | Link |
⚪️ | [array] Remove defaulted destructor | Link |
⚪️ | 3.1.0 | Link |
⚪️ | [mutex] Try work around GCC cast-align error/warning | Link |
⚪️ | docs: Clarify a doc comment | Link |
⚪️ | Cosmetics: Avoid confusing indentation | Link |
⚪️ | Typo fix | Link |
⚪️ | 3.1.1 | Link |
Sorry for the many uncertainties, but the codebase is so complex, I'm not always able to tell what's relevant and what isn't. :( But I tried my best. Other than that, it's mostly done. |
No problem. I know that it's hard to figure out where harfbuzz code is in rustybuzz. I'm looking for a solution. As for changes, can you reduce the number of changes in
Completely unrelated.
Same. Anything that relates to TrueType parsing is out of scope. That's ttf-parser problem. And unless it affects output (tests are failing) you can ignore it.
Unrelated.
Unrelated. Anything with the
I guess it can be safely ignored. It allows swapping the
We do not support scaling at all. Everything is in font units right now. |
Better now? I mostly copy pasted it from their repo, not sure why their formatting is so weird sometimes. |
I ran my scripts through black, afair. |
All good. Thanks. Note that in case of rustybuzz I do not squash commits, so prefer not to have random/duplicate commits. |