summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2016-10-20Optimize tcp_socket::on_send by using vector::erase() only once per calllouiz’
2016-10-20Very little optimization by using a simpler scope_guard when possiblelouiz’
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
2016-10-17Use expect_unordered in a few more placeslouiz’
2016-10-17Add tests for the nick change, and the nick conflictlouiz’
2016-10-15Fix the indent in the coverage.cmake filelouiz’
[skip-ci]
2016-10-14Add two e2e tests on adhoc commandslouiz’
2016-10-13Use g++ for the openbsd testslouiz’
2016-10-12Disable the output in the config unit testlouiz’
2016-10-12ci: Add a test on openbsdlouiz’
2016-10-12e2e: test the private and no-copy thingylouiz’
2016-10-12e2e: the port to use should be an int, not a stringlouiz’
2016-10-12On EINPROGRESS, we need to also check for read events (because openBSD lies)louiz’
“It is possible to select(2) or poll(2) for completion by selecting the socket for writing” Yeah, sure, “writing”…
2016-10-12Fix an off-by-one issue in the POLL codelouiz’
2016-10-11Conditionally use strptime if we don’t have std::get_timelouiz’
2016-10-11Include a private and no-copy nodes in private <message/> to avoid carbon ↵louiz’
duplication
2016-10-11Parse the timezone myself, instead of using the broken strptimelouiz’
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
2016-10-07Fix the muc#traffic responselouiz’
Was completely broken, and the test was just useless
2016-10-07Correctly set status="110" in the presence for the target of a kicklouiz’
2016-10-07e2e: add expec_unordered to be able to test things even if the arrive in a ↵louiz’
different order Also, some tests that were raising some exception and thus were not ran at all fix #3213
2016-10-07Cleanup __main__.py a little bitlouiz’
2016-10-06Add a coverity badge to the readmelouiz’
2016-10-06Remove a potential nullptr dereference, on mam queryidlouiz’
fix coverity CID 153376
2016-10-06Add a coverity job that runs cov-build and submit the archive to coverity.comlouiz’
2016-10-06Fix the argument of strerror after bind()louiz’
fix coverity CID 134470
2016-10-06Remove a branch that execute identical code in both caseslouiz’
fix coverity CID 134469
2016-10-05Avoid sending PART command for unjoined channelslouiz’
fix #3205
2016-10-05Update the changeloglouiz’
2016-10-04Respond to muc#traffic requestslouiz’
fix #3069
2016-10-03Make version requests work with global user JIDs as welllouiz’
fix #3210
2016-10-03Avoid an exception due to some bad logic in the DNS resolution mechaniclouiz’
fix #3207
2016-10-03Fix some compilation warning/errors that appear on FreeBSDlouiz’
2016-10-03Fix parse_datetime by always using a 'z' as the timezonelouiz’
Because some plateform accept Z and z, but some only accept z…
2016-10-03Remove build/* from the coverage, because that code isn’t part of biboumilouiz’
2016-10-03Add missing include filelouiz’
2016-10-03Modify the charybdis conf to disable the LIST throttlinglouiz’
2016-10-03Implement result-set-management for LIST querieslouiz’
ref #2948
2016-10-02Use LIST * instead of just LIST, because some servers don’t accept itlouiz’
Also add a e2e test for the list query
2016-09-29Re-add the ad-hoc command the was removed by mistake in the previous commitlouiz’
Thank you, e2e tests
2016-09-29Add AdhocCommandHandlers::add_command to simplify the usage of this classlouiz’
And make things a little bit clearer
2016-09-29Look for uuid/uuid.h instead of just uuid.hlouiz’
Avoids a conflict between /usr/include/uuid.h and /usr/local/include/uuid/uuid.h on freebsd
2016-09-28Add trendy badges to the READMElouiz
2016-09-28Remove unused code in sha1 modulelouiz’
2016-09-28Add coverage_e2e, which runs e2e, and regenerate a global coverage reportlouiz’
This coverage is probably more useful than the other one, because it’s not really reasonable to try to cover everything with simple units test (e.g. on network code etc). However, we should strive to have 100% coverage with our e2e tests.
2016-09-22Respond to disco#info requests on IRC server JIDslouiz’
This makes it possible to execute an ad-hoc command on a server, with clients like Gajim, for example.
2016-09-22Add the missing & for 3 std::string function argumentslouiz’
2016-09-21Some little style/typo fixeslouiz’
2016-09-20Make history messages work with fixed_irc_serverlouiz’
fix #3209
2016-09-17make dist creates the archive with the current time, not the git timelouiz’
2016-09-17Fix the parse_datetime by using %Z instead of %zlouiz’
If anybody knows why fedora accepts both, but it only works with %z on debian, please tell me.
2016-09-16In the debian docker, install aiodns from source instead of piplouiz’
Because, I don’t know why but, this doesn’t work.