| Age | Commit message (Collapse) | Author |
|---|
|
|
|
This isn't quite Travis, but close enough for local testing.
|
|
* Use fully static build for the bootstrap daemon.
* Store a sha256sum of the binary in the repo.
* Updated documentation for it.
* Add support for fully static build in cmake.
* Enable the docker build on every PR, so we catch changes to the
checksum. I realise this is adding toil, but having the checksum is
valuable for security of released binaries.
|
|
This test was fixed by @robinlinden, but not enabled in all builds.
|
|
|
|
We already have one on Cirrus, and it works fine.
|
|
|
|
|
|
|
|
|
|
It's a maintenance burden. Nobody uses this. It depends on an ancient
version of opencv that less and less systems actually have.
|
|
Removed av_test, because it depends on an ancient opencv that starts to
really not exist on modern systems anymore.
|
|
We should avoid recursion, as it makes reasoning about stack growth
harder. This tool shows (currently) 4 (non-tail) recursive functions, at
least 2 of which are easy to fix.
|
|
|
|
|
|
Limited to 4GiB. That ought to be enough for any save file.
|
|
|
|
This avoids its dependency on FreeBSD.
|
|
Also, avoid the need for putting `_XOPEN_SOURCE` in every test file.
|
|
|
|
|
|
Have one script per build. This means more duplication between the
scripts, but it's much easier to understand and to run locally.
|
|
The new clang version makes gcov segfault. Also, coveralls needs another
package that it doesn't install the right version of by default.
|
|
apidsl is in feature freeze. We can rely on the web service, instead.
|
|
We're not gaining much from this library, and it's a burden, especially
for windows development.
|
|
Also reduce number of people in conference to 5, because on Circle CI the
test times out trying to connect more than 6 or 7 people. The persistent
conferences PR will improve this so we can set it much higher then.
|
|
Since trusty has more up-to-date packages, we can remove some of the
custom install code. Also, we're not using hstox at the moment, so there
is no need to slow down the builds for now.
|
|
|
|
We will only build this during the nightly build. It's very rare for the
autotools build to break when the cmake build does not.
|
|
https://www.freebsd.org/cgi/man.cgi?alloca
If stdlib.h does not define alloca, and we're using GCC (or Clang), we
define the macro ourselves in terms of a GCC builtin.
|
|
|
|
Still run all the jobs during cron and regular push to branch. We
disabled build for push to branch, so cron is the only place where all
builds are run. This also means we need to worry less about spending time
in nightly builds, because they occur only once a day.
See https://github.com/travis-ci/travis-ci/issues/9071
|
|
|
|
Travis has upadted their VM image and by default the latest is being used.
This has caused our Windows builds to fail due to them exceeding Travis's log
limit of 4mb. It appears to be due to the new Docker version the new Trusty
image has, it somehow changed network behaviour in a way that makes toxcore
network module output a lot more of log messages than it used to.
|
|
|
|
|
|
|
|
We can run the spectest from within cmake, so we should exercise that.
|
|
|
|
This reverts commit 5ff099763b1f56414572e1c12eb2f003117db5a0.
|
|
|
|
|
|
This one is not allowed to fail and ensures that toxcore can at least be
built for OSX.
|
|
|
|
|
|
Given that it fails about 80-90% of the time, it's not worth requiring
it to pass. Instead, we'll need to manually look at the osx build to see
in what way it failed.
Also, set `fast_finish` to true, since OSX regularly has multi-hour (up
to half a day) outages that can't block us for trivial code changes or
non-code changes.
|
|
Tests for Windows are disabled for now, until we figure out which tests
can successfully run on wine and select only those.
|
|
|
|
This is the first step towards unifying travis configs for toxcore,
hstox, and qtox.
|
|
It has not been built in a while. We do want to keep this one working (or at
least compiling).
|
