Tagged release is +2 years old, fix-android-threads patch got upstreamed so it
is no longer necessary here. Also, includes a small patch which makes it
build on uwp.
* [angle] Update to chromium/5249
- Refactor build system based on WebKit's CMake build system for libANGLE
- Add maintainer-notes.md
* [angle] Fetch & generate more files in portfile
* [angle] Move maintainer-notes.md
* [angle] Port changes from #27701
* Merge install tweak from #28547
* [angle] Tweak PlatformLinux
* Run x-add-version
* [angle] Update to chromium/5414
* Run x-add-version
* Fix version database.
* Fix version database
---------
Co-authored-by: Billy Robert O'Neal III <bion@microsoft.com>
Co-authored-by: Jonliu1993 <13720414433@163.com>
* [openturns] add port
* v db
* fix missing boost dep.
* fix static stuff and others.
* v db
* reorder mpc and mpfr for linux
* v db
* add openturns:x64-osx=fail to ci baseline because vm uses old bison.
* move baseline fail above comment
* Add ci.baseline.txt note describing why it doesn't work.
* Add usage text.
* fix version
* v db
---------
Co-authored-by: Billy Robert O'Neal III <bion@microsoft.com>
* Bump to 2023-02-03
* update version database
* Remove unnecessary entries from `ci.baseline.txt`.
---------
Co-authored-by: Billy Robert O'Neal III <bion@microsoft.com>
In https://github.com/microsoft/vcpkg/pull/29267 we asked the user to remove a ci.baseline.txt entry for libcopp , but this failed in the most recent full build https://dev.azure.com/vcpkg/public/_build/results?buildId=84590 . It looks like the port doesn't have the plumbing to hook up ARM or ARM64 compilers on Windows, so this isn't a machine configuration / ci.baseline.txt issue.
There does appear to be some code in upstream trying to select armasm rather than masm to use the arm assembler but the port doesn't know how to hook that up.
Added the ability for the 'default' port version to get a value in case they all change at once (as almost happened due to https://github.com/microsoft/vcpkg/issues/29288)
https://github.com/microsoft/vcpkg/pull/28356/ did not apply post-source stubs for url and locale. I suspect they were just forgotten to be checked in?
https://github.com/microsoft/vcpkg/pull/28617/ did not record the port version change in generate-ports.ps1.
Also fixed all files to have LF line endings. (since changing boost-vcpkg-helper version also changed all the other ports anyway)
Also changed "When changing this file also update and run scripts/boost/generate-ports.ps1" to "Automatically generated by scripts/boost/generate-ports.ps1" because I think that better conveys the consequences (your changes will get nuked).
* Bumped version to 3.0.0 release.
* Update ports/onednn/vcpkg.json
Co-authored-by: Lily Wang <94091114+LilyWangLL@users.noreply.github.com>
* Update ports/onednn/portfile.cmake
Co-authored-by: Lily Wang <94091114+LilyWangLL@users.noreply.github.com>
* Applied changes requests from @LilyWangLL.
* Final database update.
* Fixed inability to download hidden by asset caching.
---------
Co-authored-by: Lily Wang <94091114+LilyWangLL@users.noreply.github.com>
Co-authored-by: Billy Robert O'Neal III <bion@microsoft.com>
* fix gn path error on windows
* add quotes
* vcpkg cmake guidelines
* Fix version database.
* Use Z_VCPKG_BACKCOMPAT_MESSAGE_LEVEL instead of WARNING
* Quotes 🤦
Co-authored-by: Billy Robert O'Neal III <bion@microsoft.com>
* [vcpkg toolchain] Fix CMAKE_CROSSCOMPILING when building on UWP
* [kfr] Set cpu arch to generic to avoid call try_run
* version
* [kfr] Re-fix cross configure issue
* version
* Revert changes about kfr, add kfr:x64-uwp to ci.baseline instead
* version
* Remove VCPKG_LINKER_FLAGS_RELEASE from debug builds, and apply other changes from windows.cmake for consistency.
Co-authored-by: JackBoosY <yuzaiyang@beyondsoft.com>
Co-authored-by: Billy Robert O'Neal III <bion@microsoft.com>
* Print stack traces for failed unit tests
* Add test for vcpkg_fixup_pkgconfig.cmake
* Check for 'optimized' and 'debug' in pc files
* Check for 'NOTFOUND' and for 'ns::target' in pc files
* Check for, and fix, line continuations
* Test file path substitutions
* Move contents processing into separate function
* Update and leverage line break normalization
* Pass prefix to data processing
* Validate only the collapsed Libs
* Test removal of '<field>.private'
* Declare unit test license (same as vcpkg)
* Replace ';' with ' ' in 'Libs:'
* Disambiguate parameter variable names
* Test quoting of variables
* Fix quoting of variables
* Quote whole parameters
* Process and quote libs item-wise
* Resolve keywords 'optimized', 'debug', 'debug'
* Consistency
* CI [skip actions]
* CI [skip actions]
* Don't fail on detected errors
* Disable unit-testing for fatal errors
Co-authored-by: Billy O'Neal <bion@microsoft.com>
* [cgns] update to 4.2.0
* [vtk-m] update to 1.6.0
* [vtk] Update to 9.0.1-pv5.9.1
- VTK update to match release version of ParaView
* [ParaView] Update to 5.9.1
* version stuff
* fix automatic svg linkage in static builds
* update qt5-svg
* revisit vtk
- fix build with vtk-m 1.6
- remove unused patches
- remove comments
* bump version
* remove two commented lines
* version stuff
* remove old PR version
* commit stuff
* commit stuff
* update notes
* wip commit
* update paraview
* fix hdf5 patch
* fix patch
* fix patch
* fix portfile
* add mpi feature
* format manifest
* use relativ path
* trace
* retry
* move qml files again
* try with host triplet
* - use VCPKG_CRT_LINKAGE
- use METIS not metis
* fix casing Zlib
* hopefully more fixes on other platforms.
* fix patch
* fix missing ws
* fix typos
* scale down to warning
* revert jsoncpp change
fix zoltan on osx?
* move trace option to failing port
* move trace option to failing port
* Revert "trilinos vcpkg.json changes"
* verdict fix osx build ?
* fix patch
* fix internal find_package call
* also apply correct case for metis
* disable METIS on osx
* vtk-m don't support arm.
* force more features in Ci
* fix missing ,
* more fixes
* more vtk python try and error.
* - paraview python setup same as vtk
- remove openvr from platforms not supporting it
- trace openimageio
* fix case and remove fmt for openimageio.
* trying to fix stuff
* set hash to 0
* [skip actions] add missing module
* WebCore module moved to paraview&python
* fixes
* remove patch in vtk-dicom
* add var to maybe unused
* use correct 5.11 ref instead of the one in release
* Patch cleanup round 1
* remove empty patch parts
* fix vtkm configs
* reintegrate jsoncpp.patch
* fix patch
* add single seacas/kokkos/zoltan ports
* throw out trilinos, zoltan, kokkos
replace trilinos with seacas
make seacas work for dynamic windows
* fix mpi feature
* fix dependency
* retry dep fix
* rename target
* remove double line
* fix target
* fix patch
* retry
* remove cgns config
* fix cgns
* no seacas trace
* lowercase hdf5 target
* fix case of metis
* disable Fortran
* add iotr patch
* add additional config folder
* handel case sensitive Seacas again
* get trace again and force jsoncpp
* update cmake to 3.25.0
* bump port-version
* add patch from vtk upstream
* cleanup, format-manifest
* fix port version
* v db
* opencv4 always inject deps.
* fix test port to not require stuff which is not supported on x86
* Fix HDF5 lookup in opencv4
* only delete bin folder in static builds
* add license field
* v db
* add missing "
* v db
* trace again
* fix rtabmap python dep
* fix openimageio by deleting the find module
* v db
* cr cleanup
* v db
* opencv4 v db due to merge with master. Had to bump port version
* v db
* v db
* v db
* CR comments
* v db for cr comments
* trace theia output
* revert theia trace
* v db
* try without patch
* revert rtabmap changes
* v db after merge
* v db for all
* [mesa] supports: !staticcrt
* [mesa] Install copyright from license.rst
* [mesa] Revise internal dependencies
* [mesa] Revise external dependencies
* [mesa] Move windows opengl32 lib to manual-link
* [opengl] Fix case of include names
* [opengl] Refactor for simplicity and mingw support
* [opengl] Stop copying out of windows SDK [skip actions]
* Update versions
* Revert "[opengl] Stop copying out of windows SDK"
* [mesa] Align with port opengl