Request.host and get_host return the empty string if the header is missing or has invalid characters.
Signed-off-by: Wang Mingyu <wangmy@fujitsu.com>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
Changelog:
===========
- Use dashes instead of underscores for METADATA.toml field names
- Update most test/lint dependencies
- Update mypy to 1.20.0
Signed-off-by: Wang Mingyu <wangmy@fujitsu.com>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
Changelog:
=============
- Add pre_load and post_load parameters to marshmallow.fields.Field for field-level pre- and post-processing
- Typing: improvements to marshmallow.validate
- marshmallow.validate.URL and marshmallow.validate.Email accept Internationalized Domain Names (IDNs)
marshmallow.validate.Email also correctly rejects IDN domains with leading/trailing hyphens.
- Typing: Fix typing of nested in marshmallow.fields.Nested
Signed-off-by: Wang Mingyu <wangmy@fujitsu.com>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
Changelog:
============
- Improved: Fixes Japanese character encoding issues.
- Added: noColor option for configureOutput, which disables
colored output. It is set to False by default.
Signed-off-by: Wang Mingyu <wangmy@fujitsu.com>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
Changelog:
============
* Switch to 3.14 for testing
* Drop 3.9 support
* More visible deprecation
Signed-off-by: Wang Mingyu <wangmy@fujitsu.com>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
Changelog:
===========
- Non-animated images can now be loaded as single frame animations
- Fixed animated WebP frame composition
- Fixed potential buffer overflow in tRNS handling
- Fixed out of bounds read in XCF image loader (CVE-2026-35444)
Signed-off-by: Wang Mingyu <wangmy@fujitsu.com>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
0001-Bug-fix-Faulty-signal-handler-prototype-spoiled-comp.patch
removed since it's included in 1.5.8
Changelog:
=============
* Bug fix: burn_offst_source_new() parameter "size" rolled over at 2 exp 31
* Bug fix: ATA and SATA drives under sg did not register their SCSI address
tuple
* Bug fix: WAVE files with unsuitable audio format were read as raw audio file.
* New API calls use and return off_t sizes instead of int or uint32_t
* Augmented struct burn_toc_entry by new off_t block addresses and counters
* New struct burn_progress_v2 and API call burn_drive_get_status_v2()
* New API call burn_disc_track_lba_nwa_v2()
* New API calls burn_disc_get_sectors_v2(), burn_session_get_sectors_v2(),
burn_track_get_sectors_v2()
* New API call burn_get_read_capacity_v2()
* New API call burn_drive_release_v2()
* New API call burn_write_opts_set_perform_opc_v2
* New cdrskin option --audio_not_raw
* New cdrskin options --perform_opc and --perform_opc_growisofs
Signed-off-by: Wang Mingyu <wangmy@fujitsu.com>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
Changelog:
==========
- Fix: unmap when surface is immediately requested to close after opening
- Fix: time out after 1 second of waiting for configure
- Fix: edge cases around montiros changing while being mapped
- gtk-priv: rename ast.py -> c_ast.py to avoid name collision with std lib
- Compat: bump supported GTK to v3.24.52
Signed-off-by: Wang Mingyu <wangmy@fujitsu.com>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
Changelog:
==========
* Update tests to reflect the %f/%F changes implemented in v3.12
* Disallow %f and %n format specifiers in --action and --info strings;
abort with an error message if those are encountered. Use %F and %N
instead. Rationale: %f and %n do not escape shell-specific syntax and
are thus a security risk when passing untrusted file names to feh. %F and
%N, which have been available since v2.3 (Feb 2012), do escape
shell-specific syntax. Migration path: Replace %f (or '%f') and %n (or
'%n') with %F and %N (without '') in --action and --info commands.
* Fix formatting in man page for groff 1.23.0+
* Do not skip URLs when '--sort mtime' or similar are used
* Define '_GNU_SOURCE' macro for mkstemps usage (if enabled)
Signed-off-by: Wang Mingyu <wangmy@fujitsu.com>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
Fixes
======
- Added missing <cstdint> includes.
- Fixed suppression of empty variadic macro arguments warning on Clang <19.
- Fixed catch_discover_tests failing during PRE_TEST discovery if a target does
not have discoverable tests.
- Fixed build of the main library failing with CATCH_CONFIG_PREFIX_ALL defined.
- JUnit reporter outputs single failed (errored/skipped) assertion per test
case.
Improvements =============
- The default implementation of --list-tags and --list-listeners has a quiet
variant.
- Suppressed the new Clang warning about __COUNTER__ usage.
- Line-wrapping counts utf-8 codepoints instead of bytes.
- Combining character sequences are still miscounted, but Catch2 does not aim
to fully support Unicode.
Signed-off-by: Wang Mingyu <wangmy@fujitsu.com>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
There is a bug (see https://github.com/jqlang/jq/issues/434), which
results in an empty version being used if autoreconf is run on the jq
sources when using a release tar ball. The incorrect assumption is that
autoreconf is only used when fetching the code using Git.
The empty version results in an incorrect libjq.pc file being created
where the version is not set, which results in, e.g.,
`pkgconf --libs 'libjq > 1.6'` failing even if version 1.8.1 of jq is
actually installed.
Switch to fetching the code using Git to workaround the bug.
Signed-off-by: Peter Kjellerstedt <peter.kjellerstedt@axis.com>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
Here is the setting in bcc recipe:
COMPATIBLE_HOST = "(x86_64.*|aarch64.*|powerpc64.*|riscv64.*)-linux"
bcc only support 64 bit arch, so add it into NON_MULTILIB_RECIPES
to avoid expanding lib32- variant
Signed-off-by: Harish Sadineni <Harish.Sadineni@windriver.com>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
Patches have been refreshed to apply to the new version.
Signed-off-by: Pascal Bach <pascal.bach@siemens.com>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
It is a new feature in 1.15.0, enabled upstream by default.
It is part of Foundation library.
So this makes it possible to disable it.
Signed-off-by: Peter Marko <peter.marko@siemens.com>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
Add patch which was also submitted upstream to quill.
Poco should get it when they update quill component.
Signed-off-by: Peter Marko <peter.marko@siemens.com>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
Those systemd services were added in 1.54 upstream
https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/merge_requests/2089
According the comments we can see they are meant for the inird
other than using for rootfs. They will conflict with the main
services and can lead to potentially confusing error messages.
So remove them for now to avoid the following issue.
>$systemd-analyze --man=false verify \
> /lib/systemd/system/NetworkManager-wait-online-initrd.service
Failed to put bus name to hashmap: File exists
NetworkManager-initrd.service: Two services allocated for the \
same bus name org.freedesktop.NetworkManager, refusing operation.
Test:
PASS: bitbake core-image-minimal
PASS: runqemu qemux86-64
PASS: systemd-analyze --man=false verify \
/lib/systemd/system/NetworkManager.service
Signed-off-by: Zhixiong Chi <zhixiong.chi@windriver.com>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
gkt-doc build uses the deprecated gtkdoc-mktmpl tool, which was dropped by upstream
in version 1.26 of gtk-doc and fails if 'api-documentation' is in DISTRO_FEATURES.
Signed-off-by: Markus Volk <f_l_k@t-online.de>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
Change SRC_URI to https://codeberg.org/sanlock/sanlock.git as sanlock source moved to it.
Signed-off-by: Liu Yiding <liuyd.fnst@fujitsu.com>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
Add do_install function to resolve following issue.
python3-legacy-cgi-2.6.4-r0 do_package_qa: QA Issue: /usr/lib/python3.14/site-packages/.pc/0001-cgi.py-fixup-interpreter-according-to-OE.patch/cgi.py contained in package python3-legacy-cgi requires /usr/local/bin/python, but no providers found in RDEPENDS:python3-legacy-cgi? [file-rdeps]
Signed-off-by: Liu Yiding <liuyd.fnst@fujitsu.com>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
Current code fails to parse meta-oe layer for machine with unsupported
architecture.
This is too restrictive as it requires downstream layers to play tricks
with masking the recipe. Usual way is to skip the recipe instead.
Signed-off-by: Peter Marko <peter.marko@siemens.com>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
