louiz’
0f14fe83ef
Release 3.0
2016-08-04 11:37:36 +02:00
louiz’
0cf7fac202
Be verbose with make check
2016-08-03 15:17:14 +02:00
louiz’
ec0e87a965
Test the version request, in many ways
2016-08-03 14:28:29 +02:00
louiz’
f0a25ccda4
Lower case the nick, when forwarding a version or ping request
2016-08-03 11:46:13 +02:00
louiz’
3fbe01bb30
Test the kick on a multisession nick
2016-08-02 11:11:27 +02:00
louiz’
58000c36a6
Rephrase an error message
2016-08-01 09:48:58 +02:00
louiz’
f89361c370
Provide a --help option
...
fix #3183
2016-07-31 18:16:29 +02:00
louiz’
4790295313
test kick
2016-07-31 18:16:29 +02:00
louiz’
37800c804e
Actually it works when we surrount the WHOLE string with quotes
...
Not just part of it
2016-07-29 11:00:30 +02:00
louiz’
1dd324aef3
lol k gitlab-ci. It fails to parse a string that contains :
2016-07-29 10:49:31 +02:00
louiz’
b66c006a4b
Trivial micro fix
2016-07-29 10:40:00 +02:00
louiz’
6ca689650a
ci: echo the cmake command line arguments
2016-07-29 10:39:40 +02:00
louiz’
4463e9f5e3
Document the multi-nick session thing
2016-07-28 16:05:36 +02:00
louiz’
085a48ea48
Do not use the logger before the configuration has been loaded first
2016-07-28 11:56:33 +02:00
louiz’
cf9104e3e2
Update the changelog by adding (all?) the changes in 3.0
2016-07-28 11:46:24 +02:00
louiz’
d46e0a9196
Mark some more cmake variables as advanced
2016-07-28 11:12:27 +02:00
louiz’
0d1e0629e7
Fix the timeout test, now that we don't wait 1ms too much everytime
2016-07-28 10:47:44 +02:00
louiz’
5c78692fcd
Do not add 1ms to the timeout of our poller
...
Can’t remember why I did this, but that must be a stupid reason. Everything
must work even with a timeout of 0.
2016-07-28 10:35:12 +02:00
louiz’
11a1c0cc99
Always remove all the DNS sockets on an c-ares event
...
Because c-ares may close one of its socket, even if it’s not the one that
has the event. Otherwise we may not know when a socket has been removed from
our poller (automatically, when close()ed) and this leads to bugs.
2016-07-28 10:34:01 +02:00
louiz’
de0eff6e94
Revert "Don’t use unique_ptr to store dns socket handlers"
...
This reverts commit 5328d0806f .
2016-07-27 10:04:15 +02:00
louiz’
c7e4fc1386
Test the resolving of multiple hostnames at the same time
2016-07-25 14:59:25 +02:00
louiz’
db503b23e8
Use log_error instead of cerr
2016-07-24 20:03:34 +02:00
louiz’
7cc05ab496
Refactor the CI file, build many configurations
2016-07-23 11:29:40 +02:00
louiz’
24824a5015
In reload.cpp, only build the database things if litesql is used
2016-07-18 09:53:15 +02:00
louiz’
c6cfa69cdc
Always build the artifacts
2016-07-18 09:53:01 +02:00
louiz’
8ded8cd33e
Properly use the docstring variables
2016-07-12 11:59:27 +02:00
louiz’
2307f94e5b
Clean the list of options visible in non-advanced ccmake
2016-07-12 11:57:01 +02:00
louiz’
f212b47830
Use an ugly way, because SOURCES property does not work in cmake 3.0
2016-07-12 11:39:27 +02:00
louiz’
fa42d0c178
Bring back DNSSocketHandler's destructor
2016-07-12 11:25:38 +02:00
louiz’
b0d60c0c1a
Set the required cmake version to 3.0
2016-07-12 11:22:01 +02:00
louiz’
4f730f93fa
Don't use target_sources() in cmake because it's >=3.1 only
2016-07-12 11:21:43 +02:00
louiz’
5328d0806f
Don’t use unique_ptr to store dns socket handlers
2016-07-12 11:15:42 +02:00
louiz’
9fb2e116c4
Move reload.*pp from louloulibs to src
2016-07-12 01:15:34 +02:00
louiz’
4c1b9abe7e
Properly catch and handle database errors
...
Do not use a singleton for the database.
fix #3203
2016-07-12 00:31:57 +02:00
louiz’
03feb403f8
Send the iq requests to one random resource instead of the bare JID
2016-07-04 18:49:00 +02:00
louiz’
81f8f45b37
Replace all include guards by #pragma once
...
It’s $CURRENT_YEAR
2016-07-04 17:53:53 +02:00
louiz’
5321d29cbd
List of channels are saved per-request and not globally
...
The list would keep the previous results in memory, forever, and the list
would grow each time a new request was made (even with results from
unrelated servers)…
2016-07-04 17:00:05 +02:00
louiz’
964784497a
Remove forgotten comment
2016-07-04 10:20:16 +02:00
louiz’
0ce75ab521
Properly remove the resource from the server when we leave the last channel
2016-07-03 16:01:15 +02:00
louiz’
dadfc349d9
Use latest git of botan in the dockerfiles
2016-07-03 14:36:38 +02:00
louiz’
29567eefcf
Add https to the unit file DOC thing
2016-07-02 15:10:30 +02:00
louiz’
8cef730318
Add a valgrind suppression file
2016-07-02 15:10:23 +02:00
louiz’
79002aa5c9
Run e2e through with valgrind if BIBOUMI_E2E_VALGRIND is set in the env
2016-06-29 19:53:37 +02:00
louiz’
2fe8878169
Add a e2e test that checks self-ping with multiple resources behind one nick
2016-06-24 11:43:13 +02:00
louiz’
7ae29890c3
Add a e2e for self-ping, with a single channel resource
2016-06-24 11:25:04 +02:00
louiz’
fcb2681b7b
Log a warning when we receive an iq without a from
2016-06-24 11:24:40 +02:00
louiz’
7d2a2dc8cc
Forward ping requests from IRC to XMPP, to one single resource
2016-06-24 11:23:51 +02:00
louiz’
b2e7edeea8
Properly set the “from” of the ping results to the correct full JID
2016-06-24 11:23:01 +02:00
louiz’
4e959a3869
Add a simple e2e test that joins a channel with xep0106-encoded name
2016-06-23 22:50:05 +02:00
louiz’
9206f4723b
Fix a bug in cut()
2016-06-23 22:40:53 +02:00