profile
viewpoint
Ryan A. Pavlik rpavlik Collabora http://www.ryanpavlik.com Principal Software Engineer at Collabora, OpenXR Working Group Spec Editor

rpavlik/cmake-modules 674

My collection of CMake modules

rpavlik/cmake-multitool 12

Python modules for parsing and working with CMake, and a utility to automatically clean CMake build scripts

rpavlik/chromium 8

My changes to the Chromium OpenGL system - not the web browser - link goes to the fork I started with

rpavlik/assembler 2

Yasm Assembler rewrite (experimental)

rpavlik/Clang-tutorial 2

A collection of code samples showing usage of clang and llvm as a library

rpavlik/Arduino-Makefile 1

My modifications/improvements to the Arduino-Makefile project.

rpavlik/checklist-specification-language 1

A little Lua tool working with a DSL for specifying broadly-construed "checklists"

rpavlik/clang-meta 1

Scripts to maintain a Git-based snapshot build of Clang and friends

rpavlik/CMake 1

My work based on the upstream CMake git

godbyk/chocolatey-packages 0

Custom packages for use with Chocolatey

Pull request review commentKhronosGroup/OpenXR-SDK-Source

Missing attributes

 def outputLayerCommands(self):                     generated_commands += '#if %s\n' % cur_cmd.protect_string                  prototype = cur_cmd.cdecl.replace(" xr", " ApiDumpLayerXr")-                prototype = prototype.replace(self.genOpts.apicall, "").replace(self.genOpts.apientry, "")+                # prototype = prototype.replace(self.genOpts.apicall, "").replace(self.genOpts.apientry, "")

Yeah should just remove it, good catch.

rpavlik

comment created time in 2 days

pull request commentKhronosGroup/OpenXR-SDK-Source

Work around std::filesystem::canonical bug on UWP

Next patch release at the latest. Should be very soon.

brycehutchings

comment created time in 2 days

push eventrpavlik/cmake-modules

Davy Durham

commit sha 5a0a556ba080cf953d9f259c629ab7ed6102a4a2

moving PUSH below potential return statement where it otherwise wouldn't get POP-ed

view details

Ryan A. Pavlik

commit sha 3afc54cf839c8a7fefa98cdb43ada81b39066109

Merge pull request #64 from ddurham2/main moving PUSH below potential return statement where it otherwise wouldn't get POP-ed

view details

push time in 2 days

push eventrpavlik/cmake-modules

Steven Winston

commit sha cc38559b6fb9bba0bebc8b4a941e9a7814d5c074

Fix to allow discovery of the loader. This will now poll the user-supplied directory correctly for the loader library target.

view details

Ryan A. Pavlik

commit sha 857e72a5c6f3ac248ecd2db0a5b3ebe2f1ff9f4e

Merge pull request #63 from gpx1000/patch-2 Fix to allow discovery of the loader.

view details

push time in 2 days

PR merged rpavlik/cmake-modules

Fix to allow discovery of the loader.

This will now poll the user-supplied directory correctly for the loader library target.

+2 -2

0 comment

1 changed file

gpx1000

pr closed time in 2 days

PR opened adafruit/Adafruit_ICM20X

Update README.md

These chips are by Invensense :) just a readme change.

+1 -1

0 comment

1 changed file

pr created time in 2 days

push eventrpavlik/Adafruit_ICM20X

Ryan A. Pavlik

commit sha ec04a523ad3ed933022e845907a43bc57c101b0b

Update README.md These chips are by Invensense :)

view details

push time in 2 days

push eventrpavlik/Adafruit_ICM20X

Ryan A. Pavlik

commit sha baf0567f4b6a2619c2488d66cdc05dfdfa7fde42

Update README.md These are Invensense chips, not SY

view details

push time in 2 days

push eventrpavlik/OpenXR-SDK-Source

Ryan Pavlik

commit sha b7d6bb41f2740fb1d4b858228a6d1de63b9d6cee

cmake: Fix configure

view details

Ryan Pavlik

commit sha 6cd712bc60552a26d1bf0765d589f876582e0f77

scripts: Disable bogus error messages in clang-format script.

view details

Ryan Pavlik

commit sha 39537f83ccc37ad27f55d553883b6b5a274f3a00

layers: Make sure to apply function attributes

view details

push time in 3 days

push eventrpavlik/OpenXR-SDK-Source

Ryan Pavlik

commit sha 561ad4f77558cf290d0f3079a8bc247d66193108

layers: Make sure to apply function attributes

view details

Ryan Pavlik

commit sha 0ae88558e88134601d89412659df3f0e019dacf8

cmake: Fix configure

view details

Ryan Pavlik

commit sha 0e93ff2b2ce1e179aa73537e3d9e552f032addc1

build: Handle the libc++ filesystem lib as well.

view details

Ryan Pavlik

commit sha 1fbd03e3fa484071a8157e3e04bb5852a4651e34

Porting to Android NDK

view details

Ryan Pavlik

commit sha 3a8b2e4aaac18971dd09786e6322426534ffd17d

Add gradle wrapper DO NOT MERGE, NEEDS LICENSE DATA

view details

Ryan Pavlik

commit sha 52c063671d710d2ca87ab40fd0d4766ac5fc18e6

Add Gradle build for hello_xr

view details

push time in 3 days

push eventrpavlik/OpenXR-SDK-Source

Ryan Pavlik

commit sha 561ad4f77558cf290d0f3079a8bc247d66193108

layers: Make sure to apply function attributes

view details

Ryan Pavlik

commit sha 0ae88558e88134601d89412659df3f0e019dacf8

cmake: Fix configure

view details

push time in 3 days

push eventrpavlik/OpenXR-SDK-Source

Lubosz Sarnecki

commit sha a013fcd3bee509a7d95ce7602a6f53ef91dc3274

hello_xr/vulkan: Fix build with Vulkan headers 1.2.136.

view details

Ryan Pavlik

commit sha adcf4972362f64717403d06280a8b10de833435f

validation layer: Remove unneeded string copies from ExtensionEnabled

view details

Ryan Pavlik

commit sha 89f1b680e7da62f7ec7237573d7d863aedac2ccf

validation layer: Move repeated code out of generated code. Noticed because clang-tidy wanted to modernize the for loop.

view details

Ryan Pavlik

commit sha 066f624ebc4ef117635bdec7764330eaaf386bda

validation: Fix a clang-tidy warning

view details

Ryan Pavlik

commit sha 5f000fb6593517c39adfd9a0b49f001a71794a0a

Add changelog fragment

view details

Bryce Hutchings

commit sha 0c67e110d9c0c195c18c61dac1c697e74435dfa6

Skip arm/arm64 on non-UWP platform

view details

Bryce Hutchings

commit sha df6604b169eefe037ede52360b2d7952c997b99c

test on any branch

view details

Bryce Hutchings

commit sha 079a103cb0a9a0865bf79e642383594cd4bc729b

improve comment

view details

Ryan A. Pavlik

commit sha 9984e3da282eea36fe48b5cefbd8f405184164fd

Merge pull request #181 from rpavlik/vulkan-fix hello_xr/vulkan: Fix build with Vulkan headers 1.2.136.

view details

Ryan A. Pavlik

commit sha 7ab03516e9de9753aa3ae398254425c86e55afca

Merge pull request #183 from brycehutchings/bryceh_fix_pipeline Fix pipeline

view details

Ryan A. Pavlik

commit sha a179a056d0e4c3343ab9557ea7f4612429234967

Merge pull request #176 from rpavlik/layer-improvements Layer improvements

view details

Matthieu Coquet

commit sha a011f9814dbdc7ffb90dcfe6c331f99a6ce67c82

Fix Visual Studio 19 16.6 build

view details

Ryan Pavlik

commit sha 38dca88abef5cfe1eb7957cbaea769221626f3c5

Rename/fix changelog fragments for correct ref parsing.

view details

Matthieu Coquet

commit sha f533f06e1ffaec8381bac26cb989429767c276d5

Add changelog fragment

view details

Ryan A. Pavlik

commit sha 3e40561410c72a56c1b55a5f421a6bd5f9c1b490

Merge pull request #186 from matthieucoquet/fix_vs_16_6 Fix Visual Studio 19 16.6 build

view details

Ryan Pavlik

commit sha 415430354ce50d74c93288993e1fb4d904129164

loader: Add info-level debug messages outputting all places we look for the manifest

view details

Matthieu Coquet

commit sha 5ba16a80d6f06400aba706b70e3d7c7f36d2f83b

Export OpenXRConfig.cmake during install

view details

Matthieu Coquet

commit sha 3ddb95601941fb12a1e2d620f454dda9b33e6913

Rename CMake module export to OpenXR. Add export target that countains only headers

view details

Matthieu Coquet

commit sha 1dd608f91930f32d24496927df5b14a31622bcb2

Add changelog fragment

view details

Matthieu Coquet

commit sha 7c29aafc83cfb95cce38f35f889c84f2db31232e

Fix: export openxr-all-supported where in the same file it's defined

view details

push time in 4 days

push eventrpavlik/OpenXR-SDK-Source

Matthieu Coquet

commit sha 5ba16a80d6f06400aba706b70e3d7c7f36d2f83b

Export OpenXRConfig.cmake during install

view details

Matthieu Coquet

commit sha 3ddb95601941fb12a1e2d620f454dda9b33e6913

Rename CMake module export to OpenXR. Add export target that countains only headers

view details

Matthieu Coquet

commit sha 1dd608f91930f32d24496927df5b14a31622bcb2

Add changelog fragment

view details

Matthieu Coquet

commit sha 7c29aafc83cfb95cce38f35f889c84f2db31232e

Fix: export openxr-all-supported where in the same file it's defined

view details

Matthieu Coquet

commit sha d2e2f74fcd706f4076ad2caca97a37c875a9275d

Fix: cmake export for UWP applications

view details

Ryan Pavlik

commit sha 51cade1763267c750f53cefa9834644c7e7b37f3

OpenXR SDK 1.0.9 (2020-05-29) - Registry - Add an author ID, and reserve a vendor extension for Huawei. (OpenXR-Docs/#46) - Reserve vendor extensions for future LunarG overlay and input focus functionality. (internal MR 1720) - Reserve vendor extensions for Microsoft. (internal MR 1723) - Add XR_EXT_hand_tracking multi-vendor extension. (internal MR 1554, internal issue 1266, internal issue 1267, internal issue 1268, internal issue 1269) - Add XR_HUAWEI_controller_interaction vendor extension. (OpenXR-Docs/#47) - Add XR_MNDX_egl_enable provisional vendor extension. (OpenXR-Docs/#48) - Add XR_MSFT_spatial_graph_bridge vendor extension. (internal MR 1730) - Add XR_MSFT_secondary_view_configuration and XR_MSFT_first_person_observer vendor extensions. (internal MR 1731) - Add XR_MSFT_hand_mesh_tracking vendor extension. (internal MR 1736) - Fix missing space in XML definition of XrSpatialAnchorCreateInfoMSFT. (internal MR 1742, internal issue 1351, OpenXR-SDK-Source/#187) - Update a number of contacts for author/vendor tags. (internal MR 1788, internal issue 1326) - SDK - Replaced usage of the _DEBUG macro with NDEBUG. (internal MR 1756) - Allow disabling of std::filesystem usage via CMake, and detect if it’s available and what its requirements are. (OpenXR-SDK-Source/#192, OpenXR-SDK-Source/#188) - CI: Modifications to Azure DevOps build pipeline. Now builds UWP loader DLLs in addition to Win32 loader DLLs. No longer builds static loader libraries due to linkability concerns. Re-arranged release artifact zip to distinguish architecture from 32-bit or 64-bit. - Loader: Replace global static initializers with functions that return static locals. With this change, code that includes OpenXR doesn’t have to page in this code and initialize these during startup. (OpenXR-SDK-Source/#173) - Loader: Unload runtime when xrCreateInstance fails. (internal MR 1778) - Loader: Add “info”-level debug messages listing all the places that we look for the OpenXR active runtime manifest. (OpenXR-SDK-Source/#190) - Validation Layer: Fix crash in dereferencing a nullptr optional array handle when the count > 0. (internal MR 1709, OpenXR-SDK-Source/#161, internal issue 1322) - Validation Layer: Fix static analysis error and possible loss of validation error. (internal MR 1715, OpenXR-SDK-Source/#160, internal issue 1321) - Validation Layer: Simplify some generated code, and minor performance improvements. (OpenXR-SDK-Source/#176) - API Dump Layer: Fix crash in dereferencing a nullptr while constructing a std::string. (internal MR 1712, OpenXR-SDK-Source/#162, internal issue 1323) - hello_xr: Fix releasing a swapchain image with the incorrect image layout. (internal MR 1755) - hello_xr: Prefer VK_LAYER_KHRONOS_validation over VK_LAYER_LUNARG_standard_validation when available. (internal MR 1755) - hello_xr: Optimizations to D3D12 plugin to avoid GPU pipeline stall. (internal MR 1770) (OpenXR-SDK-Source/#175) - hello_xr: Fix build with Vulkan headers 1.2.136. (OpenXR-SDK-Source/#181, OpenXR-SDK-Source/#180, internal issue 1347) - hello_xr: Fix build with Visual Studio 16.6. (OpenXR-SDK-Source/#186, OpenXR-SDK-Source/#184)

view details

Christoph Haag

commit sha a763a74f3cde98f08bbd790758ab3232b37e668c

hello_xr: Add Valve Index bindings

view details

Christoph Haag

commit sha d3dc016ec1236ea1ddae3b2c383a7e31ac8fd9e8

hello_xr: Use trigger/value for grabAction on vive_controller Squeeze is a physical button on this controller. If possible, grabAction should be a float.

view details

Christoph Haag

commit sha e7fcdd9f2371642df7e63df49c40a778a6a81575

add changelog fragment for Valve Index bindings

view details

Christoph Haag

commit sha 8dd68a2c68caeda243b39eb1b43d95ca04fc6882

hello_xr: Use squeeze/force for Valve Index controllers squeeze/value corresponds to how much of the handle is touched which feels undesirable. squeeze/force corresponds to how much the handle is actually squeezed.

view details

Bryce Hutchings

commit sha dd529745083a378957b426bbb01dfe1a297e484a

squash

view details

Bryce Hutchings

commit sha 38218fb4e2b44f055e03a4bc9db2ed9395711fe9

proclamation file

view details

Bryce Hutchings

commit sha efd47217761395f9f7ba47e86c9707ef9562b811

Merge pull request #196 from brycehutchings/bryceh_nuget_gen_script Script to build NuGet package from OpenXR-SDK Release

view details

Ryan A. Pavlik

commit sha 0d5d0c2575048962b44fa8b268dfc9585e3d3c7d

Merge pull request #163 from ChristophHaag/vive_and_index hello_xr: Add Valve Index bindings

view details

Ryan A. Pavlik

commit sha 40de3ed6be9fff43e347409bc208419a7f20639b

Merge pull request #191 from matthieucoquet/cmake_export Export of CMake config-file package during install

view details

Ryan Pavlik

commit sha 3579c5fcc123524a545e6fd361e76b7f819aa8a3

OpenXR SDK 1.0.10 (2020-07-28) Note the relicensing of the registry XML file and some include files provided by or generated by this repository (first item in each changelog section). Each file’s header, or an adjacent file with .license appended to the filename, is the best reference for its license terms. We are currently working on ensuring all files have an SPDX license identifier tag either in them or in an adjacent file. This is still in progress but mostly complete. - Registry - Relicense registry XML from MIT-like “Khronos Free Use License for Software and Documentation” to, at your option, either the Apache License, Version 2.0, found at http://www.apache.org/licenses/LICENSE-2.0, or the MIT License, found at http://opensource.org/licenses/MIT, for broader license compatibility with downstream projects. (SPDX License Identifier expression “Apache-2.0 OR MIT”) (internal MR 1814, OpenXR-Docs/#3, internal issue 958) - Add XR_MSFT_holographic_window_attachment vendor extension. (internal MR 1833) - Add XR_EXT_hp_mixed_reality_controller multi-vendor extension. (internal MR 1834) - Add XR_EXT_samsung_odyssey_controller multi-vendor extension. (internal MR 1835) - Add XR_VALVE_analog_threshold vendor extension. (internal MR 1859) - Add XR_MND_swapchain_input_attachment vendor extension. (internal MR 1865) - Reserve extension numbers 71 to 78 for Facebook extensions. (internal MR 1839) - Reserve extension numbers 79 to 88 for Valve extensions. (internal MR 1842) - Reserve extension numbers 89 to 92 for Khronos extensions. (internal MR 1844) - Reserve extension numbers 93 to 94 for EXT_unbounded_reference_space and EXT_spatial_anchor. (internal MR 1854) - XR_EPIC_view_configuration_fov: Fix recommendedFov incorrectly being named recommendedMutableFov. This is a SOURCE-INCOMPATIBLE CHANGE to a vendor extension. (internal MR 1812) - schema: Adjust to permit bitmask expansion in extensions, already supported by toolchain thanks to Vulkan. (internal MR 1865) - scripts: Teach xml-consistency to handle bitmask values defined in extensions. (internal MR 1865) - SDK - Relicense generated headers openxr.h, openxr_platform.h, openxr_reflection.h, and static header openxr_platform_defines.h from the Apache License, version 2.0, to, at your option, either the Apache License, Version 2.0, found at http://www.apache.org/licenses/LICENSE-2.0, or the MIT License, found at http://opensource.org/licenses/MIT, for broader license compatibility with downstream projects. (SPDX License Identifier expression “Apache-2.0 OR MIT”) (internal MR 1814, OpenXR-Docs/#3, internal issue 958) - Loader: Fix loading relative runtime libraries on Linux. (internal MR 1817) - Loader: Fix error on xrCreateInstance when explicitly trying to enable an implicit API layer. (internal MR 1858) - Modify Azure DevOps build pipeline to automatically generate a NuGet package. (OpenXR-SDK-Source/#196) - Partially revert build system changes related to detecting Direct3D, to fix builds. (internal MR 1802) - Portability fixes, including checking for timespec_get before enabling XR_USE_TIMESPEC. (internal MR 1804) - cmake: export OpenXRConfig.cmake during install. Two targets can be imported by another CMake application: OpenXR::openxr_loader and OpenXR::headers. (OpenXR-SDK-Source/#191, OpenXR-SDK-Source/#185) - hello_xr: Fix disparity between swapchain and render pass sample count in Vulkan in the case where implementation recommends a value higher than one. (internal MR 1794) - hello_xr: Fix build on a minimal Linux install by ensuring we check for all dependencies we use. We had missed checking for xcb_glx. (internal MR 1799, internal issue 1360) - hello_xr: Fix a Vulkan crash on Windows related to the mirror window. (internal MR 1823) - hello_xr: Use more proper linear formats (internal MR 1840) - hello_xr: Enable use of glslangValidator to compile shaders if shaderc is not available. (internal MR 1857) - hello_xr: Fix verbose per-layer information. (internal MR 1866) - hello_xr: Add Valve Index Controller bindings. Also use trigger value instead of squeeze click for grab action on Vive Wand controller. (OpenXR-SDK-Source/#163) - openxr_reflection: Add XR_LIST_STRUCT_ expansion macros for structure types, as well as XR_LIST_STRUCTURE_TYPES macro associating types with XrStructureType values. (internal MR 1495) - openxr_reflection: Adds XR_LIST_EXTENSIONS() macro, which will call your supplied macro name with the name and extension number of all known extensions. (internal MR 1864)

view details

Ryan Pavlik

commit sha 2e074015742ff2074169ebd454960473354caf47

Fix error in changelog for OpenXR 1.0.10

view details

Ryan Pavlik

commit sha af531e9eb95697b50ec3fbe2d0dd1b351d45ffe9

Small updates post-release.

view details

Bryce Hutchings

commit sha 7fd2486c1e80dc5ab58e653fdd8844c5e09d4f20

Clean up WIN32/WindowsStore in CMake projects. This simplifies things in most places by removing WindowsStore-specific code or fixing areas that assume "Windows" system when they should assume WIN32 (which includes WindowsStore). As a consequence, more projects can be built for UWP. Finally, this fixes build issues when building ARM/ARM64 for Win32 related to bad OpenGL and Vulkan detection in CMake.

view details

Bryce Hutchings

commit sha 5c1cdc3fe45d46cdefd6f10fde8fc8b8df059fae

Add a changelog fragment

view details

push time in 4 days

startedMozillaReality/vrb

started time in 4 days

startedMozillaReality/FirefoxRealityPC

started time in 4 days

issue commentKhronosGroup/OpenXR-SDK-Source

Export of CMake config-file package missing during install

The basics of this were in 1.0.10, thanks for the contribution. I'll write the little helper needed to deal with the packages

matthieucoquet

comment created time in 4 days

pull request commentKhronosGroup/OpenXR-SDK-Source

Keep runtime loaded after xrEnumerateInstanceExtensionProperties call

@brycehutchings I suspect this is already merged from gitlab - am I right?

brycehutchings

comment created time in 4 days

issue closedKhronosGroup/OpenXR-SDK-Source

Segfault in validation layer on DestroySpace

This is after #24 to be able to get this far in execution. Runs fine until the end of my app arrives and it's time to destroy spaces.

image

closed time in 4 days

rpavlik

issue commentKhronosGroup/OpenXR-SDK-Source

Segfault in validation layer on DestroySpace

I'm assuming this is stale/fixed by now, because I haven't seen this in a long time.

rpavlik

comment created time in 4 days

issue commentKhronosGroup/OpenXR-SDK-Source

Methods defined in openxr_platform.h such as XrGraphicsRequirementsOpenGLESKHR have no impemention in openxr_loader.so

So this is by design, that extension functions are not exported from the loader. However, it is confusing since the prototypes are available by default from the header - we'll address this part of it shortly.

shigaojian1027

comment created time in 4 days

issue commentKhronosGroup/OpenXR-Docs

What is the difference between parentstruct and structextends?

Good catch - this is something that is a little confusing, and that phrasing you found in the code is annoying.

I've added some comments to the relax-ng compact schema, and I'll fix those variable names too. For most purposes, you can use the Vulkan schema docs for OpenXR - OpenXR's schema is nearly a strict subset.

expipiplus1

comment created time in 4 days

issue closedKhronosGroup/OpenXR-Docs

GL_SRGBA

Filing an issue to correspond to the pull request #37 that has been here for a while, so that it will get synced over to gitlab.

cc @jherico

The spec specifies that "With an OpenGL-based graphics API, the texture formats correspond to OpenGL internal formats.", but the example used for for GL is GL_SRGBA, which is not an internal format, but corresponds to a typeless format.

closed time in 4 days

rpavlik

issue commentKhronosGroup/OpenXR-Docs

GL_SRGBA

Released in 1.0.10

rpavlik

comment created time in 4 days

push eventrpavlik/jnipp

Ryan Pavlik

commit sha 7876797781fda1df6749097bdefe146666001a79

Clean up markdown-lint issues in README

view details

push time in 4 days

Pull request review commentKhronosGroup/OpenXR-SDK-Source

Work around std::filesystem::canonical bug on UWP

 bool FileSysUtilsGetAbsolutePath(const std::string& path, std::string& absolute) }  bool FileSysUtilsGetCanonicalPath(const std::string& path, std::string& canonical) {+#if defined(XR_USE_PLATFORM_WIN32)+    // std::filesystem::canonical fails on UWP and must be avoided. This alternative will not+    // follow symbolic links but symbolic links are not needed on Windows since the loader uses+    // the registry as a form of indirection instead.+    wchar_t canonical_wide_path[MAX_PATH];+    if (FAILED(PathCchCanonicalize(canonical_wide_path, MAX_PATH, utf8_to_wide(path).c_str()))) {

there, filed an issue on it so we don't forget :)

brycehutchings

comment created time in 4 days

issue openedKhronosGroup/OpenXR-SDK-Source

Banish MAX_PATH limit

We can use a bigger number, but many parts of windows use MAX_PATH so it is a bit of an unfortunate de-facto limit in Windows. Another option is to use PathCchCanonicalizeEx and dynamically allocate the buffer to the length needed. What I did was a bit of a lazy but common pattern in Windows programming.

Originally posted by @brycehutchings in https://github.com/KhronosGroup/OpenXR-SDK-Source/pull/198

created time in 4 days

Pull request review commentKhronosGroup/OpenXR-SDK-Source

Work around std::filesystem::canonical bug on UWP

 bool FileSysUtilsGetAbsolutePath(const std::string& path, std::string& absolute) }  bool FileSysUtilsGetCanonicalPath(const std::string& path, std::string& canonical) {+#if defined(XR_USE_PLATFORM_WIN32)+    // std::filesystem::canonical fails on UWP and must be avoided. This alternative will not+    // follow symbolic links but symbolic links are not needed on Windows since the loader uses+    // the registry as a form of indirection instead.+    wchar_t canonical_wide_path[MAX_PATH];+    if (FAILED(PathCchCanonicalize(canonical_wide_path, MAX_PATH, utf8_to_wide(path).c_str()))) {

ah, ok. That will work until it doesn't. :)

brycehutchings

comment created time in 4 days

PR opened mitchdowd/jnipp

Upstreaming a variety of local changes

Thanks for this library, it's really nice! As I've been using it, I found a few little holes/gaps I wanted to suggest filling. Not sure if you want all the changes in one branch like this, or if I should submit each commit individually - let me know.

brief "why" of various changes:

  • Android.bp - so it can be used in an Android AOSP (ROM) build
  • templated getField - just a missing functionality - could get a field value directly with templated, but for getting the field ID you had to manually do the type/signature conversion.
  • Init free function taking JavaVM directly - sometimes APIs provide the JavaVM instead of the env, and it was simplest just to expose this capability that was already essentially there.
  • makeLocalReference: To be able to return something from a jni-invoked function, the local reference will transfer to the caller.
  • Expose the env() function - sometimes I can't quite get everything I need from jnipp so this is an "escape hatch"

I added a test where I could, but I couldn't find a "always-there" class guaranteed to have a non-static field to add a getField test.

+105 -3

0 comment

6 changed files

pr created time in 4 days

create barnchrpavlik/jnipp

branch : upstreaming

created branch time in 4 days

push eventrpavlik/jnipp

Ryan Pavlik

commit sha f273c50e017639599fd2101334c4b9d5743d5c46

Add test for makeLocalReference

view details

push time in 4 days

push eventrpavlik/jnipp

WebFolder OÜ

commit sha ebfe1caf923acdb180f49aa05a9083fd43767d4b

Added CI build integration (#26)

view details

Mitchell Dowd

commit sha f6fc04184fafc604231358bc7971b585c8ed85d3

Clear exception on failure of `FindClass()`

view details

Ryan Pavlik

commit sha 8f6508880749bc81897b21596ac08f5e8d8ca745

Merge remote-tracking branch 'origin/master' into android

view details

Ryan Pavlik

commit sha 65fe2c871e0ab41ce8f867098bf2f06bea746717

Add definition of JavaVM

view details

Ryan Pavlik

commit sha 2ecac3f17f4ad9424aa08dbf2a2506335eef2968

Makefile: Provide a reasonable default JAVA_HOME for Linux

view details

Ryan Pavlik

commit sha 2e359c1ccbfa6a2f2f1693db38db40f4e0f75aa8

Makefile: Add a "clean" target

view details

Ryan Pavlik

commit sha 352700a916158982fde15637c539586c04dc7e84

Ignore gradle files

view details

push time in 4 days

push eventKhronosGroup/OpenXR-SDK-Source

Bryce Hutchings

commit sha 7fd2486c1e80dc5ab58e653fdd8844c5e09d4f20

Clean up WIN32/WindowsStore in CMake projects. This simplifies things in most places by removing WindowsStore-specific code or fixing areas that assume "Windows" system when they should assume WIN32 (which includes WindowsStore). As a consequence, more projects can be built for UWP. Finally, this fixes build issues when building ARM/ARM64 for Win32 related to bad OpenGL and Vulkan detection in CMake.

view details

Bryce Hutchings

commit sha 5c1cdc3fe45d46cdefd6f10fde8fc8b8df059fae

Add a changelog fragment

view details

Bryce Hutchings

commit sha 896d9651e643e0fe619f3b2482d08ec04e9b4af0

Rename runtime_list to openxr_runtime_list

view details

Ryan A. Pavlik

commit sha 73c4051d969a8f6b1d26887a1daef1dc044725d7

Merge pull request #199 from brycehutchings/build_more_projects_for_uwp Clean up WIN32/WindowsStore in CMake projects.

view details

push time in 4 days

PR merged KhronosGroup/OpenXR-SDK-Source

Clean up WIN32/WindowsStore in CMake projects.

This simplifies things in most places by removing WindowsStore-specific code or fixing areas that assume "Windows" system when they should assume WIN32 (which includes WindowsStore). As a consequence, more projects can be built for UWP. Finally, this fixes build issues when building ARM/ARM64 for Win32 related to bad OpenGL and Vulkan detection in CMake.

+27 -28

1 comment

8 changed files

brycehutchings

pr closed time in 4 days

pull request commentKhronosGroup/OpenXR-SDK-Source

Clean up WIN32/WindowsStore in CMake projects.

awesome, thanks!

brycehutchings

comment created time in 4 days

Pull request review commentKhronosGroup/OpenXR-SDK-Source

Work around std::filesystem::canonical bug on UWP

 bool FileSysUtilsGetAbsolutePath(const std::string& path, std::string& absolute) }  bool FileSysUtilsGetCanonicalPath(const std::string& path, std::string& canonical) {+#if defined(XR_USE_PLATFORM_WIN32)+    // std::filesystem::canonical fails on UWP and must be avoided. This alternative will not+    // follow symbolic links but symbolic links are not needed on Windows since the loader uses+    // the registry as a form of indirection instead.+    wchar_t canonical_wide_path[MAX_PATH];+    if (FAILED(PathCchCanonicalize(canonical_wide_path, MAX_PATH, utf8_to_wide(path).c_str()))) {

isn't max_path an artificial limitation when using wide apis now?

brycehutchings

comment created time in 4 days

issue closedrpavlik/cmake-modules

Option to prevent get_git_head_revision from traversing parent directories

There's a patch here. They mentioned upstreaming it but never got around to it.

To summarize, if you are versioning your home directory with git and unpack a source tarball then the CMake module will still traverse upwards and use the revision of the home directory. Most projects use a CMakeLists.txt in the root directory, so the traversal is not needed, and disabling it solves the issue.

closed time in 4 days

Mathnerd314

startedgregsaun/maker_cheatsheet

started time in 4 days

create barnchOSVR/OSVR-HDK-MCU-Firmware

branch : fixes

created branch time in 5 days

push eventOSVR/OSVR-Docs

Nando

commit sha a1b25b3d8bd103e3586257a8c30ea97fbbbee0da

Update UpdateHDKLinux.md Add example command to generate SHA-256 hash as used in newer firmware versions.

view details

Ryan A. Pavlik

commit sha 33e5c22fb6b6d12d8ca13181b524d5e6f266ceff

Merge pull request #126 from Nandox7/patch-1 Update UpdateHDKLinux.md to include example of generate SHA-256 hash

view details

push time in 5 days

PR merged OSVR/OSVR-Docs

Update UpdateHDKLinux.md to include example of generate SHA-256 hash

Add example command to generate SHA-256 hash as used in newer firmware versions.

+6 -1

0 comment

1 changed file

Nandox7

pr closed time in 5 days

delete branch rpavlik/cmake-modules

delete branch : rework-boost

delete time in 5 days

delete branch rpavlik/cmake-modules

delete branch : launchers-cmake-3

delete time in 5 days

create barnchrpavlik/cmake-modules

branch : main

created branch time in 5 days

issue commentrpavlik/cmake-modules

VS Launcher with different RUNTIME_LIBRARY_DIR's per configuration

Generator expression work has been merged. Not sure if that helps your situation (or if you even still have this problem) - please re-check if you would like.

ulrichard

comment created time in 5 days

push eventrpavlik/cmake-modules

Ryan Pavlik

commit sha 7d8f0314b8229e6383c7b2839cc2fac92e18d3b9

Fix a missed change in CreateLaunchers

view details

Ryan Pavlik

commit sha be8fdce6dbde88b8d46551ae04ffc12ec691b027

Add an update-modules.sh script

view details

push time in 5 days

issue commentcaseymcc/CreateLaunchers

Remote debugging target

pretty cool! You're welcome to submit this "upstream" - I finally merged @caseymcc 's changes. https://github.com/rpavlik/cmake-modules

haferburg

comment created time in 5 days

PR opened caseymcc/CreateLaunchers

Backport from upstream, fix license file

Finally got around to merging your PR, sorry it took so long! "Upstream" I have cleaned up the files a little bit more (cmake-format is a nice thing to have!), so I wanted to submit a PR bringing this repo in sync with https://github.com/rpavlik/cmake-modules

Also, I saw that your LICENSE file did not correctly reflect the license of the modules from my repo. They're BSL-1.0, which (if you can believe it) is an even more permissive license than MIT, though for things like this that don't get compiled, I don't think there's really much effective difference, though I'm not a lawyer.

Thanks again for your contribution!

+115 -65

0 comment

2 changed files

pr created time in 5 days

create barnchrpavlik/CreateLaunchers

branch : backport

created branch time in 5 days

fork rpavlik/CreateLaunchers

CMake module for creating batch/shell scripts and Visual Studio user file for launching your application.

fork in 5 days

push eventrpavlik/cmake-modules

Ryan Pavlik

commit sha 741d7afb51906b9f2b43443abd7a3fc9c8088436

Port back some CreateLaunchers changes from a forked repo. See https://github.com/caseymcc/CreateLaunchers

view details

Ryan Pavlik

commit sha 8a9510d52527afd6813c4cc60bdb83231d25c96e

Update header on CreateLaunchers

view details

push time in 5 days

issue closedrpavlik/cmake-modules

GetGitRevisionDescription: git_local_changes() returns false positive when repo. source file touched

My project contains a source file detailing software version information, which I touch at the start of every build to ensure that it is unconditionally built such that a project build timestamp derived from C __DATE__ and __TIME__ macros is up to date.

The git diff-index --quiet HEAD -- command used in the git_local_changes() function results in the repo. being evaluated as dirty once a source file has been touched, and doesn't come right until some other git operation performs an index refresh.

See https://stackoverflow.com/questions/34807971 for a discussion of git diff-index behaviour.

Seems to me that using git diff --quiet HEAD -- instead results in desired behaviour. Alternately can issue git update-index --refresh before the git diff-index call.

closed time in 5 days

RichardLangFromNZ

issue commentrpavlik/cmake-modules

GetGitRevisionDescription: git_local_changes() returns false positive when repo. source file touched

Hmm. I'm not sure the best way to do this, especially given that generating/modifying a source file in the source directory (instead of the build directory) is not really recommended with CMake. Additionally, reproducible builds work makes use of DATE and TIME something that compilers are warning about these days.

If it still sounds like there's an issue here, you're welcome to submit a pull request. Do note however that this module is by far the most popular in this repo, so changes should be made with care :)

RichardLangFromNZ

comment created time in 5 days

issue closedrpavlik/cmake-modules

Conan module

I drafted up a little conan module to import the CMake scripts:

https://github.com/kheaactua/conan-cmake-modules

If your interested, I could create a pull request to add details into the README of how to include it, and also add a test_package to make it a proper conan recipe.

closed time in 5 days

kheaactua

issue commentrpavlik/cmake-modules

Conan module

If Conan is used by people still then you're welcome to submit one!

kheaactua

comment created time in 5 days

issue commentrpavlik/cmake-modules

Document how to use GetGitRevisionDescription "offline"

Yeah, that would be a good thing. I'm not sure there's a single best way, especially since projects tend to use a git archive as their release tarball, which leaves little room to inject a version file. Probably best approach is to include a version file, but ignore it if you find out there's a git repo?

alerque

comment created time in 5 days

issue closedrpavlik/cmake-modules

FindNSS and FindNSPR?

Hi @rpavlik

Would you be interested in adding FindNSS and FindNSPR to your repo?

This repo helps loads, and it would be great to add scripts we have separately to it - they are both BSD-Clause-3 licensed, originally by Andreas Schneider at RedHat.

closed time in 5 days

shearer12345

issue commentrpavlik/cmake-modules

FindNSS and FindNSPR?

You're welcome to submit these at any time!

shearer12345

comment created time in 5 days

issue closedrpavlik/cmake-modules

CMake Find C++11 Support

Saw this the other day, looks quite useful for finding C++ 11 support:

http://stackoverflow.com/a/13571695

closed time in 5 days

carlsonp

issue commentrpavlik/cmake-modules

CMake Find C++11 Support

Fixed upstream :)

carlsonp

comment created time in 5 days

issue commentrpavlik/cmake-modules

Question on get git revision description

This one is actually easier - you can do something like the following. add_custom_target is considered "always out of date" so it will always run.

set(DESCRIBEFILE ${CMAKE_CURRENT_BINARY_DIRECTORY}/git-describe-results.txt)
add_custom_target(git-describe-target
    ALL
    "${GIT_EXECUTABLE}" describe --dirty > "${DESCRIBEFILE}"
    BYPRODUCTS "${DESCRIBEFILE}"
    WORKING_DIRECTORY "${CMAKE_CURRENT_SOURCE_DIR}"
    COMMENT "Running git describe"
    VERBATIM)

Then you can freely use ${DESCRIBEFILE} though you'll want to add_dependencies() from wherever you use it, to the git-describe-target. Might need to mark DESCRIBEFILE as generated if you use it as a source file, not sure. (You can modify the command line here however you please - I assumed --dirty because the script should get triggered to re-configure if you only care about commits.)

Let me know if this helps!

kobi-ca

comment created time in 5 days

pull request commentldc-developers/ldc

Add option to avoid recursing in get_git_head_revision cmake function

FYI, I think I fixed this upstream in https://github.com/rpavlik/cmake-modules/commit/43a50aab1f2d453900972ef67bf03b25f6acb800 by not going above CMAKE_SOURCE_DIR. See https://github.com/rpavlik/cmake-modules/issues/58

WebDrake

comment created time in 5 days

issue commentrpavlik/cmake-modules

Option to prevent get_git_head_revision from traversing parent directories

I think 43a50aa should fix this, by a slightly different (and backward compatible) method, can you double-check?

Mathnerd314

comment created time in 5 days

IssuesEvent

push eventrpavlik/cmake-modules

Ryan Pavlik

commit sha 43a50aab1f2d453900972ef67bf03b25f6acb800

Don't let get_git_head_revision go above the CMAKE_SOURCE_DIR, for safety. Think this should fix #58

view details

Ryan Pavlik

commit sha 245d153128893655d850052b439cd453b6b9199b

Namespace a function in the git module.

view details

Ryan Pavlik

commit sha ec5ea830592b1cb9caedf9a40564d2d1cf67d7c0

Format and update my most popular module

view details

push time in 5 days

issue closedrpavlik/cmake-modules

Option to prevent get_git_head_revision from traversing parent directories

There's a patch here. They mentioned upstreaming it but never got around to it.

To summarize, if you are versioning your home directory with git and unpack a source tarball then the CMake module will still traverse upwards and use the revision of the home directory. Most projects use a CMakeLists.txt in the root directory, so the traversal is not needed, and disabling it solves the issue.

closed time in 5 days

Mathnerd314

issue closedrpavlik/cmake-modules

CreateLauncher: needs to be ported to generator expressions - else violates policy (use of the LOCATION) -> CMP0026

After recently upgrading to cmake 2.8.12 I get lots of warnings like this:

1> CMake Warning (dev) at App/cmake/CreateLaunchers.cmake:204 (get_target_property): 1> Policy CMP0026 is not set: Disallow use of the LOCATION target property. 1> Run "cmake --help-policy CMP0026" for policy details. Use the cmake_policy 1> command to set the policy and suppress this warning. 1>
1> The LOCATION property should not be read from target "CppExtTest". Use the 1> target name directly with add_custom_command, or use the generator 1> expression $<TARGET_FILE>, as appropriate. 1>
1> Call Stack (most recent call first): 1> App/cmake/CreateLaunchers.cmake:281 (_launcher_produce_vcproj_user) 1> App/CMakeLists.txt:264 (create_target_launcher) 1> App/CppExt/UnitTests/CMakeLists.txt:32 (PL_ADD_UNIT_TEST) 1> This warning is for project developers. Use -Wno-dev to suppress it.

And here is the long description:

cmake --help-policy CMP0026

CMP0026

Disallow use of the LOCATION target property.

CMake 2.8.12 and lower allowed reading the LOCATION target property to determine the eventual location of build targets. This relies on the assumption that all necessary information is available at configure-time to determine the final location and filename of the target. However, this property is not fully determined until later at generate-time. At generate time, the $<TARGET_FILE> generator expression can be used to determine the eventual LOCATION of a target output.

Code which reads the LOCATION target property can be ported to use the $<TARGET_FILE> generator expression together with the file(GENERATE) subcommand to generate a file containing the target location.

The OLD behavior for this policy is to allow reading the LOCATION property from build-targets. The NEW behavior for this policy is to not to allow reading the LOCATION property from build-targets.

This policy was introduced in CMake version 3.0.0. CMake version 2.8.12.20140106-gff6f80 warns when the policy is not set and uses OLD behavior. Use the cmake_policy command to set it to OLD or NEW explicitly.

closed time in 5 days

ulrichard

issue closedrpavlik/cmake-modules

git_describe() and --dirty

It seems that git_describe() can't be used with --dirty - it will always return -128-NOTFOUND. When running the same command from the command line yields this:

$git describe HEAD --dirty
fatal: --dirty is incompatible with commit-ishes

So basically git doesn't like that we're passing both hash and --dirty. I don't see why we have to pass commit hash to git describe in git_describe(), by default git will just assume HEAD.

closed time in 5 days

Zeex

issue commentrpavlik/cmake-modules

git_describe() and --dirty

There is now a git_describe_working_tree that does a git describe --dirty without the fancy commit-ish magic of the normal function. I think this should solve things.

Zeex

comment created time in 5 days

issue closedrpavlik/cmake-modules

get_git_head_revision doesn't work from submodule

Looks like it can't find correct folder, instead of trying to find .git by hand, maybe it's more wise to use git rev-parse --git-dir.

closed time in 5 days

jimon

issue commentrpavlik/cmake-modules

get_git_head_revision doesn't work from submodule

I think this is fixed after merging #61

jimon

comment created time in 5 days

pull request commentrpavlik/cmake-modules

Godbyk changes

Not sure which of these still make sense to merge. If you wish, you can update and submit these, otherwise I'll eventually close this as stale, since it's been like 8 years already :facepalm: Sorry @godbyk

rpavlik

comment created time in 5 days

push eventrpavlik/cmake-modules

dawid-aurobit

commit sha d82da0b9e839aec34d0eba8f903bb426cf537284

git_describe_working_tree function added Returns the results of git describe on the working tree (--dirty option), and adjusting the output so that it tests false if an error occurs.

view details

Ryan A. Pavlik

commit sha fc454f191bebc3678c6ab0eafa10b0e55a9b8a82

Merge pull request #41 from dawid-aurobit/git-describe git_describe_working_tree function added

view details

push time in 5 days

PR merged rpavlik/cmake-modules

git_describe_working_tree function added

Returns the results of git describe on the working tree (--dirty option), and adjusting the output so that it tests false if an error occurs.

+34 -0

1 comment

1 changed file

dawid-aurobit

pr closed time in 5 days

PR closed rpavlik/cmake-modules

Add generator expression support to CreateLauncher merged-after-rebase

This uses generator expression to get rid of the cmake warnings with using the LOCATION property of the target. Also since generator expressions are being decoded in the file(GENERATE ...) call you can feed CreateLauncher generator expressions.

+69 -35

2 comments

3 changed files

caseymcc

pr closed time in 5 days

pull request commentrpavlik/cmake-modules

Add generator expression support to CreateLauncher

I don't concern myself with the whitespace police here - I wasn't very consistent in these files, and there wasn't a tool like cmake-format until recently. I have rebased this work, on top of a normalization of formatting, and merged it! Thank you!

caseymcc

comment created time in 5 days

push eventrpavlik/cmake-modules

Ryan Pavlik

commit sha d32e6c19d08acabb240fba96058476d8bc3053d2

clang-format CreateLaunchers

view details

caseymcc

commit sha a2a0b06c76ed32508def27c8710fa59cf01fd937

add generator expression support v2: Ryan Pavlik rebased and ran cmake-format

view details

Ryan Pavlik

commit sha 8c1ee2b9ff8bc366411fad5deee453f999f4f851

Slight reword

view details

Ryan Pavlik

commit sha 6bc8c2e8b42e57bd4799e7fe1c59750ea0d4b090

Merge branch 'launcher_generator_exp' See PR #53

view details

push time in 5 days

push eventrpavlik/cmake-modules

Matthew Russell

commit sha 1e1cd70abbaf47af9423507d8994b61ddfd12fd5

Added windows version 17134 https://blogs.msdn.microsoft.com/chuckw/2018/05/02/windows-10-april-2018-update-sdk/

view details

Matthew Russell

commit sha babae73177b16234e08a2a1c7ffeefc79e51236b

Added October 2018 update

view details

Ryan A. Pavlik

commit sha a9c1ee19bb06f87e76ebd8bdec8772aa4a8745c1

Merge branch 'master' into master

view details

Ryan A. Pavlik

commit sha 6022a08078de3fd068b00fe769284772480e1e7e

Merge pull request #51 from kheaactua/master Added windows version 17134 and 17763

view details

push time in 5 days

PR merged rpavlik/cmake-modules

Added windows version 17134 and 17763

https://blogs.msdn.microsoft.com/chuckw/2018/05/02/windows-10-april-2018-update-sdk/

+0 -0

3 comments

0 changed file

kheaactua

pr closed time in 5 days

pull request commentrpavlik/cmake-modules

Added windows version 17134 and 17763

Thanks!

kheaactua

comment created time in 5 days

push eventkheaactua/cmake-modules

nwtour

commit sha e9b27d77aa5c0064d9652f9a5e63c27bd8e526ad

Direct hint on CMake command line options

view details

Mal

commit sha e22a3e681438411c3fa86f9958428caf91fe62fd

Add support for git-worktree when finding GIT_DIR

view details

Ryan Pavlik

commit sha c3d6d9a07fd508d3dda26d0bc95aea1a8e7484cd

FindJsonCpp: Fix linking in Debian package building.

view details

Ryan Pavlik

commit sha df13ba813ecdcedf705e70deb38dd65175c693c4

FindHIDAPI: Overhaul for better Linux support

view details

Ryan Pavlik

commit sha e58535ae29a75d3540bbf7712e6bb6fa0fff69af

EnableExtraCompilerWarnings: Import changes made elsewhere

view details

Ryan Pavlik

commit sha ba80b248d7fd4d8d4d523d975e7881afa1048c3d

Add some new modules

view details

Ryan Pavlik

commit sha d9181c3f7d71db95465b39cc6c37151e55b1beb4

cppcheck: fix indent

view details

Ryan Pavlik

commit sha b62ab98e36ed29d4b001e8fde05664a8aff6e40d

Windows SDK: Add more versions

view details

Ryan Pavlik

commit sha dfb1f69530ccc4a486e288ebfba41ab00bb7438b

DebugSymbols: Fix typo

view details

Ryan Pavlik

commit sha 74bb568c1a3dbcf4942c9e9b554f6238b01a3dd9

Add FindRangeV3

view details

Ryan A. Pavlik

commit sha 8d8b4898bf2191e0bc3ba16a683528a90a9a572d

Merge pull request #55 from malicious/worktree-support Add support for git-worktree when finding GIT_DIR

view details

Ryan A. Pavlik

commit sha c97690cc0024a1842cd8de71334a2ee9c5f7e729

Merge pull request #54 from nwtour/master Direct hint on CMake command line options

view details

Caleb Maclennan

commit sha ef17650bafa9bd84d2479d347a1d3851dcf697c2

Allow `git describe` to function without tags Closes #12 Note that --always must be paired with --tags to get the expected output in the most common use case when there are tags. Without it only the short hash is ever returned. Combined with --tags the output is the same as previous when tags are present but will fall back to identifying the commit by its hash if no tags are found.

view details

Florian Fleißner

commit sha d9bcdedfb39e6426e5419215407ac65ba85637aa

Enabled correct handling of worktrees This change adds correct handling of git worktrees.

view details

Ryan Pavlik

commit sha 1c7e5d70f7d4d29fb8df3218b2df3bea6e9503a1

Add cmake-format config. Not used universally (yet)

view details

Ryan Pavlik

commit sha f079726a43da415c37637a20d1c71aa49d54256f

Update FindOpenXR

view details

Ryan Pavlik

commit sha 023f68403409d1cc8a6d6d030770f2dafacc471b

Add FindLib3ds

view details

Ryan Pavlik

commit sha d9826cc35c9d16bdccd617ea48b386be79507c0e

Fix bug in FindOpenXR

view details

Ryan Pavlik

commit sha 49494fec9f6cf58fb36a476c30a8fade97cc3f1d

Update FindHIDAPI, fixing bugs

view details

Ryan Pavlik

commit sha 42b6b68c476a9ab56763bf90de4fd420ab9ef449

Clean up FindLibcheck

view details

push time in 5 days

PR closed rpavlik/cmake-modules

GetGitRevisionDescription: match parent path when .git doesn't have newlines

The parent path of a submodule is not matched if the .git file in the submodule path doesn't have newlines. I removed the newline in the regex and added a strip step to remove newlines (if they are present).

+2 -1

1 comment

1 changed file

Banus

pr closed time in 5 days

pull request commentrpavlik/cmake-modules

GetGitRevisionDescription: match parent path when .git doesn't have newlines

Thanks for submitting this, sorry it took me so long to merge it! I rebased your change on other changes but left your author credit.

Banus

comment created time in 5 days

push eventrpavlik/cmake-modules

Emanuele Plebani

commit sha f7e7f0610b16e9c44f12cd71be3b726594c08e25

match parent path when .git doesn't have newlines The parent path of a submodule is not matched if the .git file in the submodule path doesn't have newlines. Removed the newline in the regex and added a strip step to remove newlines (if they are present). v2: Rebased by Ryan Pavlik - changes applied to revised code.

view details

Ryan Pavlik

commit sha 731fbd93ce96de3f40e493d105bf400c7b00a12e

Merge branch 'fix_submodule'

view details

push time in 5 days

push eventrpavlik/cmake-modules

Caleb Maclennan

commit sha ef17650bafa9bd84d2479d347a1d3851dcf697c2

Allow `git describe` to function without tags Closes #12 Note that --always must be paired with --tags to get the expected output in the most common use case when there are tags. Without it only the short hash is ever returned. Combined with --tags the output is the same as previous when tags are present but will fall back to identifying the commit by its hash if no tags are found.

view details

Ryan A. Pavlik

commit sha 139efb304f5b7ca58b189206835841a6a50a05f1

Merge pull request #56 from alerque/git-always Allow `git describe` to function without tags

view details

push time in 5 days

PR merged rpavlik/cmake-modules

Allow `git describe` to function without tags

Closes #12

Note that --always must be paired with --tags to get the expected output in the most common use case when there are tags. Without it only the short hash is ever returned. Combined with --tags the output is the same as previous when tags are present but will fall back to identifying the commit by its hash if no tags are found.

+1 -1

1 comment

1 changed file

alerque

pr closed time in 5 days

issue closedrpavlik/cmake-modules

Add --always to 'git describe' function in GetGitRevisionDescription.cmake

Otherwise you might get this output: $ git describe fatal: No tags can describe 'ffdc9780376f57d4cff1f07c835f7516c482d28d'. Try --always, or create some tags.

closed time in 5 days

krf

pull request commentrpavlik/cmake-modules

Allow `git describe` to function without tags

Ah, great, thank you!

alerque

comment created time in 5 days

pull request commentrpavlik/cmake-modules

GetGitRevisionDescription: Fix problem with worktrees

Thanks for submitting this! I got two submissions around the same time, and the other one had more documentation so I merged that, but I do appreciate this contribution! Can you check and make sure that the current main branch (after merging #61) solves your problem?

sebwolf-de

comment created time in 5 days

push eventrpavlik/cmake-modules

Florian Fleißner

commit sha d9bcdedfb39e6426e5419215407ac65ba85637aa

Enabled correct handling of worktrees This change adds correct handling of git worktrees.

view details

Ryan A. Pavlik

commit sha 6878d13f10901527bcb6837d5865df18c76f96e7

Merge pull request #61 from almechd/feature/git_worktree_handling Enabled correct handling of worktrees

view details

push time in 5 days

PR merged rpavlik/cmake-modules

Enabled correct handling of worktrees

This change adds more robust handling of git worktrees.

An attempt to handle worktrees had already been submitted as https://github.com/rpavlik/cmake-modules/commit/e22a3e681438411c3fa86f9958428caf91fe62fd but does not work under all circumstances.

Here's why the current worktree handling fails:

The part newly introduced by https://github.com/rpavlik/cmake-modules/commit/e22a3e681438411c3fa86f9958428caf91fe62fd

if(NOT IS_DIRECTORY "${GIT_DIR}")
   file(READ ${GIT_DIR} worktree)
   string(REGEX REPLACE "gitdir: (.*)worktrees(.*)\n$" "\\1" GIT_DIR ${worktree})
endif()

assumes that GIT_DIR is an existing path, either a directory or a file. But in my case the variable GIT_DIR evaluates to something like e.g. C:/some_path/C:/some_path/.git/worktrees/some_repo which is not a valid path. Nonetheless, the expression NOT IS_DIRECTORY "${GIT_DIR}" evaluates to true for anything that is not an exisiting file. CMake thus tries to read the broken GIT_DIR and fails.

+69 -21

1 comment

1 changed file

almechd

pr closed time in 5 days

pull request commentrpavlik/cmake-modules

Enabled correct handling of worktrees

This looks really nice, thank you for submitting this!

almechd

comment created time in 5 days

push eventrpavlik/cmake-modules

Ryan Pavlik

commit sha 213cdaac44c2d3f37fa5e9a08e9d7765147557cb

Add pkg-config support to FindOpenXR

view details

push time in 5 days

push eventrpavlik/cmake-modules

Ryan Pavlik

commit sha 1c7e5d70f7d4d29fb8df3218b2df3bea6e9503a1

Add cmake-format config. Not used universally (yet)

view details

Ryan Pavlik

commit sha f079726a43da415c37637a20d1c71aa49d54256f

Update FindOpenXR

view details

Ryan Pavlik

commit sha 023f68403409d1cc8a6d6d030770f2dafacc471b

Add FindLib3ds

view details

Ryan Pavlik

commit sha d9826cc35c9d16bdccd617ea48b386be79507c0e

Fix bug in FindOpenXR

view details

Ryan Pavlik

commit sha 49494fec9f6cf58fb36a476c30a8fade97cc3f1d

Update FindHIDAPI, fixing bugs

view details

Ryan Pavlik

commit sha 42b6b68c476a9ab56763bf90de4fd420ab9ef449

Clean up FindLibcheck

view details

Ryan Pavlik

commit sha 85b1053d5095c15945f8c6e7b642d874b63a70f9

Clean up FindOpenHMD

view details

Ryan Pavlik

commit sha 1ff28323197d4249101c066b8bc8f2a1f329389a

Some SPDX tags

view details

Ryan Pavlik

commit sha 9ca190f61b7995be8b43838ea9ecf53acf7acb83

Add FindOpenGLES

view details

Ryan Pavlik

commit sha 0d6078c3f33de5caa38f5056a9846c2706607377

Add FindcJSON

view details

push time in 5 days

Pull request review commentKhronosGroup/OpenXR-SDK-Source

Clean up WIN32/WindowsStore in CMake projects.

 if(MSVC) endif()  set_target_properties(runtime_list PROPERTIES FOLDER ${TESTS_FOLDER})-set_target_properties(runtime_list PROPERTIES OUTPUT_NAME openxr_runtime_list)

Ah, weird, go figure. Yep, we can rename the whole thing, that works for me!

brycehutchings

comment created time in 5 days

startedxinghaochen/awesome-hand-pose-estimation

started time in 5 days

Pull request review commentKhronosGroup/OpenXR-SDK-Source

Clean up WIN32/WindowsStore in CMake projects.

 if(MSVC) endif()  set_target_properties(runtime_list PROPERTIES FOLDER ${TESTS_FOLDER})-set_target_properties(runtime_list PROPERTIES OUTPUT_NAME openxr_runtime_list)

Why was this removed?

brycehutchings

comment created time in 5 days

more