The current code is inherently racy: if screen update takes
considerable time (e.g. when working over network) and a user
performed a series of resizes the final event might get ignored and
the display will be left in inconsistent state.
Fix the race by unsetting the flag first so if the next WINCH signal
is received while display is resizing it'll be processed on the next
iteration.
* use custom memory descriptor that `abort()`s on `malloc()` failure
* use static log descriptor
* don't always re-create all contexts
* de-duplicate code of `.._connect()` and `.._register()`
Signed-off-by: Steffen Jaeckel <jaeckel-floss@eyet-services.de>
The current code enters an infinite loop if the input string happens
to get an invalid utf-8 sequence somehow. For me it was reproducible
by running profanity in a Screen session and pressing Alt-т (cyrillic
letter).
Fix it the way borrowed from 0501e49623
where mbrlen is used for the same purposes.
When preferences do not specify a program to be used for /editor
command, try getting it from EDITOR (which POSIX.1-2017 calls one of
"variables that are frequently exported by widely used command
interpreters and applications"), fall back to "vim" if not set.
In case the user decides to ignore the validity-state of certificates
we also have to configure libcurl accordingly.
`tls.policy` can be set via
```
/account set <account> tls trust
```
Signed-off-by: Steffen Jaeckel <jaeckel-floss@eyet-services.de>
The profanity-internal mechanism to allow connecting to a server isn't
easily portable to cURL. Therefor introduce a profanity-specific CAfile
which is managed individually and will be configured in libcurl calls.
Signed-off-by: Steffen Jaeckel <jaeckel-floss@eyet-services.de>
Fixes https://github.com/profanity-im/profanity/issues/1649
Type `>` then press tab or shift tab to autocomplete previous messages,
then type your reply and send message.
Newlines are replaced with newline followed by `> `.
A newline is added at the end so that the user can immediately type a
message without manually adding a new line.
Add tab completion for `/cmd`.
Complete the `list` and `exec` subcommands.
Didn't complete the `<jid>` part since I think it's usually used for
components and etc. Things that might not be in the roster.
Fix https://github.com/profanity-im/profanity/issues/1650
* use GLib functions to write&read compose file
* trim trailing new-line characters from compose file
Signed-off-by: Steffen Jaeckel <jaeckel-floss@eyet-services.de>
As all parts of the code invoking the `files_get_account_data_path()`
function did the same afterwards, a function has been added with the same
behavior.
1. create path
2. `mkdir` of that path
3. return final path
Signed-off-by: Steffen Jaeckel <jaeckel-floss@eyet-services.de>
In case we're not connected yet and press Alt+c a segfault occurred
since `conn.xmpp_conn` is dereferenced while it's still `NULL`.
Signed-off-by: Steffen Jaeckel <jaeckel-floss@eyet-services.de>
Output before:
```
11:00:00 - Annonuce OpenPGP Key for OX ~/test/testuser.pub.gpg ...
```
After:
```
11:00:00 - Annonuce OpenPGP Key for OX /home/user/test/testuser.pub.gpg ...
```
Now we expand the path so that we can check for `~` properly.
And test if the file is actually a normal file.
Set this to 0.
We might want to have this configurable later.
For now we fix the valgrind report:
```
Conditional jump or move depends on uninitialised value
xmpp_debug_verbose()
```
Which will be fixed in libstrophe > 0.11.0 by commit
28f3ce19b8
Previously it relied on AX_PYTHON_DEVEL, which in turn executes
python-config to get the build flags. However this does not work while
cross compiling because we can't execute the python-config build for the
target platform. To circumvent this problem the python build flags are
now queried via pkgconfig, which has the drawback of not having some
extra build flags, but they do not seem to be needed.
I tested this patch with the termux build system and it build without
their existing hack of injecting python after the configure step. I also
tested non cross compile build on Arch Linux and it also still works.
Fixes#851
Remove support for libsignal-protocol-c < 2.3.2.
Debian 10 uses 2.3.2, Debian 11 and 12 use 2.3.3.
openSUSE from 15.2 onward uses 2.3.3.
Fedora since 28 uses 2.3.2.
We should be good.
When a `see-other-host` stream-error is received we try to re-connect to
the other host. Erroneously this also started the `reconnect_timer`.
This lead to the behavior that in cases where e.g. the login failed
we try to reconnect instead of bailing out with an error.
This commit fixes the wrong behavior by not starting the `reconnect_timer`.
Fix 0e58509c16
Signed-off-by: Steffen Jaeckel <jaeckel-floss@eyet-services.de>
