Lately we had a few times that people pushed to their PR branch
a few times to make small changes. Sadly, this triggers all CIs
every time, which takes ~20 minutes. As we are limited in the
amount of runners we get assigned to us, this means all other CI,
even for other repositories within OpenTTD, are delayed too.
We can avoid this by simply cancelling old runs when a new PR is
pushed. There is a downside: sometimes people already push a new
commit, but still want to know if the old one passed. That will
no longer be possible with this change.
This requires the use of WinHTTP (for Windows) or libcurl (for all
others except Emscripten). Emscripten does not support http(s)
calls currently.
On Linux it requires ca-certificates to be installed, so the HTTPS
certificate can be validated. It is really likely this is installed
on any modern machine, as most connections these days are HTTPS.
(On MacOS and Windows the certificate store is filled by default)
Reminder: in case the http(s):// connection cannot be established,
OpenTTD falls back to a custom TCP-based connection to fetch the
content from the content-service. Emscripten will always do this.
Currently they had a name that the rest of our system cannot
deal with correctly. "cert.pfx" is also not very descriptive from
a system as a whole.
As such, we now name it like any other file, so it can be published
safely to the CDN.
It turns out that having "-g" in the compile-statement causes
Emscripten to pick -g3, which makes for very big binaries. This
is very likely not your intention when building Emscripten, as
smaller really is better.
For comparison, with RelWithDebInfo the binary is ~80MB. With
Release it is ~7.4MB.
With std::variant all memory can be figured out at compile time, so the compiler needs to keep track of fewer elements. It also saves out a unique_ptr and its memory management, over a slight impact for resolving a setting.
The 6th is "is-stable-tag", but it is currently broken in meaning.
Betas and RCs are considered "stable", but final releases are not.
This is the reason it was working for RC1, but not for the final
release.
nielsmh nicely correct us in #8641, pointing out the old comment
is not telling a complete truth. The result is the same, but it
is better to not mislead future-us.
These bundles can be opened on any "modern" Linux machine with
a driver that SDL2 supports.
Machines needs at least glibc 2.15, which was released 10 years ago.
It is build with CentOS 7 as base, and only assumes the following
libraries are available on the system:
- libc
- libdl
- libgcc_s
- libpthread
- librt
- libstdc++
All other libraries the game depends on are bundled together with
the game, so users don't need any library installed to use this
bundle. The downside of course is that this increases the binary
size a bit: 30 MiB of libraries are in this bundle.
RPATH is used to make ld-linux find the folder libraries are
stored in; however, system libraries are always used before these,
in the assumption libraries on the user system are more up-to-date.
Using -DOPTION_PACKAGE_DEPENDENCIES=ON switches on packaging
of libraries in the "lib" folder. This requires CMake 3.16 to
be installed; otherwise it will fail.
Because certain files are read-only, the "restore-cache" step
fails, as Windows tar refuses to overwrite those files (even
if they are identical). Instead of hoping upstream fixes
"restore-cache", just remove the read-only flag on these files
instead.
We use CMakeListsTxtAdvanced, and as such, we have to do this our
self via "-DCMAKE_BUILD_TYPE=RelWithDebInfo". Otherwise we are
producing Debug builds instead of Release builds. Oops.
This has several ways of being triggered:
- When creating a new release via the GitHub interface. Fully
automated that will produce new binaries, upload them, and it
will even update the website to tell about the new version.
- When triggered in an automated way from OpenTTD/workflows to
start a nightly.
- Manually via the Release workflow, which accepts branches,
Pull Requests and tags to build.
Rerunning a job is safe and should be without issues. Everything
retriggers and updates what-ever might have been broken. In fact,
except for dates, it should produce identical results.
Co-authored-by: Charles Pigott <charlespigott@googlemail.com>
When a developer attaches the "preview" label, a build is created
and published on https://preview.openttd.org/. After that, new
pushes to the PR are automatically build (as long as the "preview"
label exists).
If a non-developer attaches the "preview" label, it will be
removed.
Emscripten compiles to WASM, which can be loaded via
HTML / JavaScript. This allows you to play OpenTTD inside a
browser.
Co-authored-by: milek7 <me@milek7.pl>
Also, while at it, make it more like the upcoming release-workflow,
so they look a lot more similar.
Functional it should be the same, except that Windows should
now also output when a test fails.
Stale bot checks if an issue hasn't had any attention in the last
60 days, or if a pull request hasn't had any attention in the last
30 days. If so, it marks it as stale. After 7 more days, it closes
the issue/pull request. Any activity on the issue/pull request
resets the timer.
This helps keeping the counter low, and the work manageable. There
is a lot of reasoning behind this idea, and other projects show it
really helps to keep the issue/pull request count low.
More information here: https://github.com/probot/stale
Rubidium