Debian Patches
Status for syncevolution/2.0.0-4
| Patch | Description | Author | Forwarded | Bugs | Origin | Last update |
|---|---|---|---|---|---|---|
| 020240920~9d9535b.patch | src/dbus/interfaces/syncevo-*-full.xml: Fix annotations <annotation> need to be outside of <arg> to be recognised by qdbusxml2cpp, the In*|Out* of their name maps them to the corresponding <arg>. |
OPNA2608 <opna2608@protonmail.com> | no | 2026-08-06 | ||
| 020250404~2aed588.patch | syncevo: explicitly set which FDs will be passed to children Starting from GLib 2.75.4, FD returned from g_file_open_tmp() now have O_CLOEXEC set, which breaks our interprocess communication. . But instead of just un-setting O_CLOEXEC on the FDs, let's use the new API introduced in GLib 2.68 and tell it explicitly which FD will be passed to children. For older GLib, still manually unset O_CLOEXEC on the FDs specified to be passed. |
Ratchanan Srirattanamet <ratchanan@ubports.com> | no | 2026-08-06 | ||
| 020250917~f4620c2.patch | evolution: Adapt to EVCard API changes in 3.59.1 The (currently) upcoming evolution-data-server 3.59.1 contains API changes in the libebook-contacts' EVCard and EContact. This change handles it in a way that allows to compile with the old and with the new evolution-data-server. |
Milan Crha <mcrha@redhat.com> | no | 2026-08-06 | ||
| 020250917~f5d29d8.patch | configure.ac: Move declaration of IT_PROG_INTLTOOL Without this, I get a build error during autoreconf call: . autoreconf: running: intltoolize --copy --force ERROR: 'IT_PROG_INTLTOOL' must appear in configure.ac for intltool to work. autoreconf: error: intltoolize failed with exit status: 1 |
Milan Crha <mcrha@redhat.com> | no | 2026-08-06 | ||
| 020251112~8db2564.patch | CurlTransportAgent: fix passing lambda to curl_easy_setopt() curl_easy_setopt()'s third argument is a C-style variadic arguement, which allows it to accept arbitrary-type arguments. In this case, my understanding is that C++ doesn't know that it has to convert lambda to plain C-style function pointer. So when curl_easy_setopt() read the third parameter from va_args, it receives NULL instead of a function pointer. . When this happens, cURL will invoke fallback functions instead, but the subsequent call to set "userdata" doesn't fail. So what happens is that the fallback functions are called with incorrect "userdata", which inevitably leads to crash. . To fix this, explicitly cast the lambdas to the correct types before passing to curl_easy_setopt(). . Fixes: 623397c6 ("C++: lambdas instead of static methods") |
Ratchanan Srirattanamet <ratchanan@ubports.com> | no | 2026-08-06 | ||
| 020260507~dac2d9e.patch | source2html: Remove lineanchors attribute Per Pygments documentation (https://pygments.org/docs/formatters/): . lineanchors If set to a nonempty string, e.g. ``foo``, the formatter will wrap each output line in an anchor tag with an ``id`` (and `name`) of ``foo-linenumber``. This allows easy linking to certain lines. . lineanchors is not supposed to be boolean and it wasn't ever taken into account. The underlying code in recent Pygments changed to always expect a string, causing syncevolution to fall into the except block. . It fixes compatibility with Pygments 2.20+ |
Karolina Surma <ksurma@redhat.com> | no | 2026-08-06 | ||
| 1001_SOCK_CLOEXEC.patch | Fix FTBFS on kfreebsd due to missing SOCK_CLOEXEC Work around missing SOCK_CLOEXEC on kfreebsd by setting FD_CLOEXEC afterwards. |
Tino Mettler <tino+debian@tikei.de> | no | 2021-09-29 | ||
| 1003_64time_t.patch | Fix FTBFS on 32-bit with 64-bit time_t | Adrian Bunk <bunk@debian.org> | no | debian | 2026-08-06 | |
| 1004_reproducible_build.patch | do not update the datestamps of .rst files | Vagrant Cascadian <vagrant@reproducible-builds.org> | no | debian | 2026-08-06 |