Re: KDE Gear projects with failing CI (master) (4 June 2024)

2024-06-14 Thread Thiago Macieira
On Thursday 13 June 2024 14:08:43 GMT-7 Elvis Angelaccio wrote:
>   auto tempDir = new QTemporaryDir(QDir::currentPath() +
> "/.test-qtempdir-");

Indeed, that works, as does trying QFileInfo or QDir on the returned path and 
then using absolutePath.

> I'd expect at least the QTemporaryDir documentation to get updated with a
> warning about this new behavior. 

True, but I think it's unnecessary. That doesn't fix existing code. For anyone 
writing new code, they'll realise the problem right away, so they'll make the 
path absolute or deal with the relative path.

> Currently it still claims that
> "Auto-remove mode will automatically delete the directory from disk upon
> destruction.", but this is no longer true if you happened to create the
> QTemporaryDir with a relative template and then change the current
> directory in the process before deleting the QTemporaryDir instance.

True.

But there's also a converse case: if you renamed the current directory, the 
old code had failed, but now it works.
-- 
Thiago Macieira - thiago (AT) macieira.info - thiago (AT) kde.org
  Principal Engineer - Intel DCAI Fleet Systems Engineering





KDE Gear 24.08 release schedule

2024-06-14 Thread Albert Astals Cid
Hi folks, 

This is the release schedule the release team agreed on

  https://community.kde.org/Schedules/KDE_Gear_24.08_Schedule

Dependency freeze is in around 4 weeks (July 18) and feature freeze one 
after that. Get your stuff ready!

Cheers,
  Albert




[ANNOUNCE] CMake 3.30.0-rc3 is ready for testing

2024-06-14 Thread John Parent
I am proud to announce the third CMake 3.30 release candidate.
  https://cmake.org/download/

Documentation is available at:
  https://cmake.org/cmake/help/v3.30

Release notes appear below and are also published at
  https://cmake.org/cmake/help/v3.30/release/3.30.html

Release milestone is available at:
  https://gitlab.kitware.com/cmake/cmake/-/milestones/142

Some of the most significant changes in CMake 3.30 are:

* "cmake-presets(7)" files now support schema version "9". "include"
  fields now expand all macros except "$env{}" and preset-specific
  macros, i.e., those derived from the fields inside a preset's
  definition.

* The "Compile Features" functionality now implements support for the
  "cxx_std_26" and "cuda_std_26" meta-features to indicate that the
  compiler mode must be at least C++26. These meta-features were first
  documented by CMake 3.25, but were not fully implemented.

* The "CMAKE__STANDARD_LATEST" variable was added to describe
  the latest "" language standard CMake supports for the
  selected compiler.

* The "CMAKE_TLS_VERSION" variable and "CMAKE_TLS_VERSION" environment
  variable were added to specify a default minimum TLS version for
  connections to "https://; URLs by the "file(DOWNLOAD)" and
  "file(UPLOAD)" commands.

* The "GENERATED" source file property is now visible in all
  directories.  See policy "CMP0163".  Policy "CMP0118"'s
  documentation has been revised to describe its actual effects.

* The "FindPython", "FindPython2", and "FindPython3" modules, on
  Windows, now offer better support for the Python debug variant.

* The "TARGET_PROPERTY" generator expression learned to evaluate
  custom transitive properties defined by new
  "TRANSITIVE_COMPILE_PROPERTIES" and "TRANSITIVE_LINK_PROPERTIES"
  target properties.

* The "CPack WIX Generator" gained support for WiX Toolset v4. See the
  "CPACK_WIX_VERSION" variable.

* The "FindBoost" module has been removed by policy "CMP0167". Port
  projects to upstream Boost's "BoostConfig.cmake" package
  configuration file, for which "find_package(Boost)" now searches.

* The precompiled Windows ".msi" installers provided on cmake.org,
  when performing a fresh installation, now modify the system-wide
  "PATH" by default.



CMake 3.30 Release Notes


Changes made since CMake 3.29 include the following.


New Features



Presets
---

* "cmake-presets(7)" files now support schema version "9". "include"
  fields now expand all macros except "$env{}" and preset-specific
  macros, i.e., those derived from the fields inside a preset's
  definition.


File-Based API
--

* The "cmake-file-api(7)" "cmakeFiles" version 1 object's "version"
  field has been updated to 1.1.  It gained a "globsDependent" field
  to report "file(GLOB)" calls using "CONFIGURE_DEPENDS".


Generators
--

* Visual Studio Generators now add "UseDebugLibraries" indicators to
  ".vcxproj" files to denote which configurations are debug
  configurations. See policy "CMP0162".


Languages
-

* The "Compile Features" functionality now implements support for the
  "cxx_std_26" and "cuda_std_26" meta-features to indicate that the
  compiler mode must be at least C++26. These meta-features were first
  documented by CMake 3.25, but were not fully implemented.


Commands


* The "add_library()" command, on platforms that do not support shared
  libraries, now rejects creation of shared libraries instead of
  automatically converting them to static libraries.  See policy
  "CMP0164".

* The "enable_language()" command now fails with an error if it is
  called before the first "project()" call. See policy "CMP0165".

* The "file(DOWNLOAD)" and "file(UPLOAD)" commands gained a
  "TLS_VERSION " option to specify the minimum TLS version for
  connections to "https://; URLs.


Variables
-

* The "CMAKE__STANDARD_LATEST" variable was added to describe
  the latest "" language standard CMake supports for the
  selected compiler.

* The "CMAKE_TLS_VERIFY" environment variable was added as a fallback
  to the existing "CMAKE_TLS_VERIFY" variable.  It specifies whether
  to verify the server certificate for "https://; URLs by default.

* The "CMAKE_TLS_VERSION" variable and "CMAKE_TLS_VERSION" environment
  variable were added to specify a default minimum TLS version for
  connections to "https://; URLs by the "file(DOWNLOAD)" and
  "file(UPLOAD)" commands.

* The "CMAKE_VS_USE_DEBUG_LIBRARIES" variable and corresponding
  "VS_USE_DEBUG_LIBRARIES" target property were added to explicitly
  control "UseDebugLibraries" indicators in ".vcxproj" files.


Properties
--

* The "GENERATED" source file property is now visible in all
  directories.  See policy "CMP0163".  Policy "CMP0118"'s
  documentation has been revised to describe its actual effects.

* The "PROPAGATE_TOP_LEVEL_INCLUDES_TO_TRY_COMPILE" global property
  can be used to propagate "CMAKE_PROJECT_TOP_LEVEL_INCLUDES" into
  

Re: KDE Gear projects with failing CI (release/24.05) (11 June 2024)

2024-06-14 Thread Ben Cooksley
On Fri, Jun 14, 2024 at 10:21 AM Albert Astals Cid  wrote:

> El dijous, 13 de juny del 2024, a les 22:37:04 (CEST), Ingo Klöcker va
> escriure:
> > On Mittwoch, 12. Juni 2024 00:22:35 CEST Albert Astals Cid wrote:
> > > kwalletmanager - 3rd week
> > >
> > >  * https://invent.kde.org/utilities/kwalletmanager/-/pipelines/711185
> > >
> > >   * appstream test fails because  https://apps.kde.org/kwalletmanager5
> > >
> > > doesn't exist
> >
> > This will be fixed by
> > https://invent.kde.org/websites/apps-kde-org/-/merge_requests/94
> > Proof:
> > https://invent.kde.org/kloecker/apps-kde-org/-/jobs/1897729/artifacts/
> > external_file/public/kwalletmanager5/index.html
>
> Can we have someone that knows about that code review it or should i just
> press the bottom and YOLO it? [ the html_test on CI fails, is that
> expected
> when run on a repo fork? ]
>

Only phunh or Carl would have the necessary knowledge to review this i'm
afraid, so this is probably going to have to be a case of verify as best we
can and then merge if we can't get one of them to review.


>
> Cheers,
>   Albert
>

Cheers,
Ben


>
> >
> > Regards,
> > Ingo
>
>
>
>
>


Re: KDE Gear projects with failing CI (master) (4 June 2024)

2024-06-14 Thread Ingo Klöcker
On Donnerstag, 13. Juni 2024 23:08:43 CEST Elvis Angelaccio wrote:
> On 09/06/24 17:16, Thiago Macieira wrote:
> > The path is correct at the point it is issued, referring to the same
> > directory. The fact that it is in a different form was not expected to be
> > a
> > breaking change.
> 
> I had a look at the 6.7 release notes and I've found the commit 935562a77b
> which seems to be what changed this behavior.
> 
> Indeed, if I apply the same fix done in that commit to the test-case I
> attached before, i.e. if I replace
> 
>   auto tempDir = new QTemporaryDir(".test-qtempdir-");
> 
> with
>   auto tempDir = new QTemporaryDir(QDir::currentPath() + "/.test-
qtempdir-");
> 
> then the returned QTemporaryDir::path() goes back to being absolute.
> 
> I'd expect at least the QTemporaryDir documentation to get updated with a
> warning about this new behavior. Currently it still claims that
> "Auto-remove mode will automatically delete the directory from disk upon
> destruction.", but this is no longer true if you happened to create the
> QTemporaryDir with a relative template and then change the current
> directory in the process before deleting the QTemporaryDir instance.

That looks like a serious unforeseen regression and was surely not intended. 
Best provide a simple test case for QTemporaryDir for this.

Regards,
Ingo

signature.asc
Description: This is a digitally signed message part.


Re: KDE Gear projects with failing CI (release/24.05) (11 June 2024)

2024-06-14 Thread Ingo Klöcker
On Freitag, 14. Juni 2024 00:21:41 CEST Albert Astals Cid wrote:
> El dijous, 13 de juny del 2024, a les 22:37:04 (CEST), Ingo Klöcker va
> 
> escriure:
> > On Mittwoch, 12. Juni 2024 00:22:35 CEST Albert Astals Cid wrote:
> > > kwalletmanager - 3rd week
> > > 
> > >  * https://invent.kde.org/utilities/kwalletmanager/-/pipelines/711185
> > >  
> > >   * appstream test fails because  https://apps.kde.org/kwalletmanager5
> > > 
> > > doesn't exist
> > 
> > This will be fixed by
> > https://invent.kde.org/websites/apps-kde-org/-/merge_requests/94
> > Proof:
> > https://invent.kde.org/kloecker/apps-kde-org/-/jobs/1897729/artifacts/
> > external_file/public/kwalletmanager5/index.html
> 
> Can we have someone that knows about that code review it or should i just
> press the bottom and YOLO it? [ the html_test on CI fails, is that expected
> when run on a repo fork? ]

The html_test always fails for MRs. It checks URLs like
https://kloecker.local-kde.org/-/apps-kde-org/-/jobs/1897729/artifacts/public/
bovo/
which will never work because unlike on the real page with a real web server 
there are no redirects to or automatic loading of
https://kloecker.local-kde.org/-/apps-kde-org/-/jobs/1897729/artifacts/public/
bovo/index.html

Fixed with the following MR which disables the test for all non-protected 
branches where it will anyway fail
https://invent.kde.org/websites/apps-kde-org/-/merge_requests/95

Regards,
Ingo


signature.asc
Description: This is a digitally signed message part.