Since commit 87e19535fe, the ETag check when writing out a file through
a symlink (following the symlink) has been incorrectly using the ETag
value of the symlink, rather than the target file. This is incorrect
because the ETag should represent the file content, not its metadata or
links to it.
Fix that, and add a unit test.
Signed-off-by: Philip Withnall <pwithnall@endlessos.org>
Fixes: #2417
This adds g_tls_connection_get_protocol_version(),
g_tls_connection_get_ciphersuite_name(), and DTLS variants. This will
allow populating TLS connection information in the WebKit web inspector.
This is WIP because we found it's not quite possibly to implement
correctly with GnuTLS. See glib-networking!151.
This follows on from !1719. It drops volatile qualifiers on internal
functions in `gbitlock.c`, and casts volatile public arguments to
non-volatile versions to avoid the `g_atomic_*()` macros from
propagating the volatile qualifier.
We can’t drop the `volatile` qualifier from the public API in
`gbitlock.h`, as that would unfortunately be an API break.
Update the documentation in `gbitlock` to mention that users of the API
should not use `volatile`. See http://c.isvolatileusefulwiththreads.com/
Signed-off-by: Philip Withnall <pwithnall@endlessos.org>
Fixes: #2418
This is the result of checking each `Returns:` line in these files. I’ve
only considered nullability and not other (potentially missing or
incorrect) annotations.
Including suggestions by Simon McVittie.
Signed-off-by: Philip Withnall <pwithnall@endlessos.org>
Helps: #2227
Fix a possibility of returning a relative path, according to the
doc, the function should return an absolute path.
Signed-off-by: Luke Yue <lukedyue@gmail.com>
In a PKCS#11 operation there are multiple types of PINs possibly
needed and these flags add a way to expose them to the user.
This design exactly matches gnutls' gnutls_pin_flag_t API.
This changeset exposes
* `not-valid-before`
* `not-valid-after`
* `subject-name`
* `issuer-name`
on GTlsCertificate provided by the underlying TLS Backend.
In order to make use of these changes,
see the related [glib-networking MR][glib-networking].
This change aims to help populate more of the [`Certificate`][wk-cert]
info in the WebKit Inspector Protocol on Linux.
This changeset stems from work in Microsoft Playwright to [add more info
into its HAR capture][pw] generated from the Inspector Protocol events
and will bring feature parity across WebKit platforms.
[wk-cert]: 8afe31a018/Source/JavaScriptCore/inspector/protocol/Security.json
[pw]: https://github.com/microsoft/playwright/pull/6631
[glib-networking]: https://gitlab.gnome.org/GNOME/glib-networking/-/merge_requests/156
Because sometimes you don't want a lone "%s", and you don't
want the compiler yelling at you about format strings that
don't have any format in them.
Closes#663
Specifically, for uclibc at the moment. Other implementations may need
locking, but I haven’t checked any aside from uclibc-ng and glibc.
POSIX.1-2001 specifies that `dlerror()` is not thread-safe, but the
glibc (and likely other libc) implementation is.
Issue #399 was originally reported about eglibc, but that project has
since died and been merged back into glibc. So that’s one less thing to
worry about.
Signed-off-by: Philip Withnall <pwithnall@endlessos.org>
Fixes: #399
gjs has some situations where it's not always aware of the @data that
was passed into g_object_add_toggle_ref, so allow passing %NULL to
just match on @notify.
Rebased and updated by Nitin Wartkar
Closes#817
It is cleaner to define glib_typeof() in a header included after
gversionmacros.h so we can use GLIB_VERSION_MIN_REQUIRED directly
instead of doing it everywhere glib_typeof() is used.
The `GApplication` must be registered before calling
`g_application_mark_busy()`. Document that, and add a guard.
The same is true for `g_application_unmark_busy()`, but the existing
documentation and guard for `busy_count > 0` are enough.
