louiz’
fdf336afa0
Use ensure_future if available, otherwise use asyncio.async
2016-10-31 01:39:35 +01:00
louiz’
1d6aa89ab9
Test self disco
2016-10-31 01:39:35 +01:00
louiz’
ec555d05df
e2e: add the possibility to launch a specific list of scenarios only
2016-10-31 01:39:35 +01:00
louiz’
1ff71ca79e
Test raw messages
2016-10-31 01:39:35 +01:00
louiz’
3e7c8ab4bc
Trivial cleanup
2016-10-31 01:39:35 +01:00
louiz’
911258bc65
Make AddrinfoDeleter a class
2016-10-31 01:39:35 +01:00
louiz’
0d8a2bfd13
Rename a variable that shadows a class member
2016-10-31 01:39:35 +01:00
louiz’
f7f4e3f638
Do the previous commit, but correctly
2016-10-28 19:07:12 +02:00
louiz’
8c5c217a9e
Default to make -j1 if nproc doesn’t exist
2016-10-28 18:55:10 +02:00
louiz’
ac61450184
Refactor remove_invalid_xml_chars to use correct types directly
2016-10-27 01:37:55 +02:00
louiz’
6d65e9527e
Remove calls to INFO() in catch, they are useless
2016-10-27 01:16:04 +02:00
louiz’
6d5d7eff68
Directly use Botan::byte instead of char, to avoid an unnecessary cast
2016-10-27 01:15:26 +02:00
louiz’
021f025cb0
Refactor the sha1 digest into its own function, and do not use sprintf
2016-10-26 21:19:53 +02:00
louiz’
3fdae6b624
Add a new badge to our collection! (Sonarqube)
2016-10-27 00:19:56 +02:00
louiz’
5ffd976d7f
CONTRIBUTING.rst formatting and a typo
2016-10-25 01:17:58 +02:00
louiz
e5aaf7f8f4
Merge branch 'fix_tests_encoding' into 'master'
...
Fix test encoding by setting LANG and LC_ALL to C.UTF-8
Because we don’t want to rely on each projet setting their own variable LC_ALL and LANG in gitlab’s CI preferences.
See merge request !4
2016-10-24 11:46:49 +02:00
louiz’
b141f3ddee
Explicitely use the docker runner on some builds
2016-10-24 11:39:28 +02:00
louiz’
4359680f9d
Fix test encoding by setting LANG and LC_ALL to C.UTF-8
2016-10-24 11:21:12 +02:00
louiz’
fed21b7e9f
ci: Rename the test archive
2016-10-24 10:23:09 +02:00
louiz’
00a2e8dd70
Suggest adding tests when contributing
2016-10-24 19:22:46 +02:00
louiz’
7b2215b882
Add a CONTRIBUTING file
2016-10-24 19:01:52 +02:00
louiz’
5990a8bf8a
Correctly handle the nick change inside the virtual channel
2016-10-24 00:19:30 +02:00
louiz’
c3bb9fe2e2
Handle forced-join by just sending an invitation
...
fix #3116
2016-10-23 23:59:21 +02:00
louiz’
00eb18bae8
Refactor channel->parting to a new location
2016-10-23 23:58:03 +02:00
louiz’
8b8bbceeb3
Revert "Use ensure_future instead of async"
...
This reverts commit a4d67ce041 .
2016-10-21 19:12:25 +02:00
louiz’
ac32d25738
Fix the broken commit 4388b9c
2016-10-21 18:50:04 +02:00
louiz’
82a6fa6d09
e2e: Fix some logic in check_list_of_xpath
2016-10-21 01:05:38 +02:00
louiz’
a4d67ce041
Use ensure_future instead of async
2016-10-21 01:05:15 +02:00
louiz’
4388b9c311
Remove an unused variable in e2e
2016-10-21 01:00:52 +02:00
louiz’
479708c83e
Add an other kewl badge
2016-10-20 18:55:18 +02:00
louiz’
52f0aca246
Coverity upload is allowed to fail and is manual. Also name the artifacts
2016-10-21 00:05:21 +02:00
louiz’
aa4255224e
Optimize tcp_socket::on_send by using vector::erase() only once per call
2016-10-20 20:08:29 +02:00
louiz’
ce06c25e93
Very little optimization by using a simpler scope_guard when possible
...
The version with the vector, that can be disabled etc, is “very” slow, so we
use unique_ptr when we don’t need to disable it, and when it only contains
one function
2016-10-20 19:32:20 +02:00
louiz’
6b4d2e8e3e
Use expect_unordered in a few more places
2016-10-17 19:49:45 +02:00
louiz’
2a0f5f1b6a
Add tests for the nick change, and the nick conflict
2016-10-17 19:48:51 +02:00
louiz’
ef36a769a5
Fix the indent in the coverage.cmake file
...
[skip-ci]
2016-10-15 01:55:39 +02:00
louiz’
501a6f2661
Add two e2e tests on adhoc commands
2016-10-14 23:51:04 +02:00
louiz’
7dfc08ca9b
Use g++ for the openbsd tests
2016-10-13 10:22:37 +02:00
louiz’
34bd8b93da
Disable the output in the config unit test
2016-10-12 22:18:53 +02:00
louiz’
86c927ae17
ci: Add a test on openbsd
2016-10-12 22:18:16 +02:00
louiz’
44f35a256d
e2e: test the private and no-copy thingy
2016-10-12 21:00:05 +02:00
louiz’
c3a7cd3627
e2e: the port to use should be an int, not a string
2016-10-12 21:00:05 +02:00
louiz’
827a1eedf8
On EINPROGRESS, we need to also check for read events (because openBSD lies)
...
“It is possible to select(2) or poll(2) for completion by selecting the
socket for writing”
Yeah, sure, “writing”…
2016-10-12 21:00:05 +02:00
louiz’
5f2e4820df
Fix an off-by-one issue in the POLL code
2016-10-12 21:00:05 +02:00
louiz’
c54f28d29d
Conditionally use strptime if we don’t have std::get_time
2016-10-11 21:00:56 +02:00
louiz’
dfc0793ef2
Include a private and no-copy nodes in private <message/> to avoid carbon duplication
2016-10-11 00:43:46 +02:00
louiz’
548e4ad473
Parse the timezone myself, instead of using the broken strptime
...
See
https://lab.louiz.org/louiz/biboumi/issues/3215
https://github.com/andikleen/glibc/blob/master/time/strptime_l.c#L746-L747
for why strptime() sucks
We use std::get_time now, to parse the date and time. And we parse the
timezone by hand.
fix #3215
2016-10-11 00:20:43 +02:00
louiz’
116472920c
Fix the muc#traffic response
...
Was completely broken, and the test was just useless
2016-10-07 23:29:14 +02:00
louiz’
8ac8d2b242
Correctly set status="110" in the presence for the target of a kick
2016-10-07 23:28:40 +02:00
louiz’
8cf292fa44
e2e: add expec_unordered to be able to test things even if the arrive in a different order
...
Also, some tests that were raising some exception and thus were not ran at all
fix #3213
2016-10-07 23:26:44 +02:00