| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
| |
Shellcheck doesn't like it when you do `rm -r "$A/$B"` : it worries
that the path might accidentally expand to /.
That shouldn't actually be possible in this case, but let's avoid
being the people who accidentally remove somebody's filesystem.
|
| |
|
|
|
| |
This was added in !210; looks like it got removed by mistake with
!221.
|
| |
|
|
|
| |
This makes it easier to see when they're changing, and will make
future patches easier to review.
|
| |\
| |
| |
| |
| |
| |
| | |
Add test coverage tracking with CI/CD
Closes #250
See merge request tpo/core/arti!221
|
| | |
| |
| |
| | |
Follows up on changes from https://gitlab.torproject.org/trinity-1686a/arti/-/merge_requests/2/diffs#bc2cc73a8917fa6fed183148df69119665eb4e55
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
There's some kind of bug in nightly that (for me) keeps grcov from
generating correct results. Instead, it misses some crates entirely.
No trouble, though: I can just use an older nightly until they get
this issue fixed!
This patch makes using a different version of nightly possible with
a new RUST_COVERAGE_TOOLCHAIN environment variable.
|
| | |
| |
| |
| | |
found by shellcheck.
|
| | | |
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
`coverage.sh` is now a more elaborate script to before, thanks to
moving from tarpaulin to grcov via `with_coverage.sh`.
This commit also adds support for the minimalistic
chutney+curl-based integration test from our gitlab-ci.yml file.
See the message from "coverage.sh -h" for information about running
these tests.
Closes #248.
|
| | | |
|
| | |
| |
| |
| |
| | |
(We don't put Tor Project employees in the acknowledgments lists.
And we probably shouldn't put bots either.)
|
| |/
|
|
|
|
|
|
|
|
| |
Previously this code was in .gitlab-ci.yml, but for the purposes of
testing my fix for #275, it turned out to be much better to have it
in a script of its own.
Also, we now need to update our "quote" second-order dependency,
since otherwise our minimal first-order dependencies won't build
correctly.
|
| | |
|
| |
|
|
|
| |
This program is itself for testing; we don't strongly care how much
coverage it has on its own.
|
| |
|
|
|
| |
Fix a bug in postprocess_coverage.py that stuck a big row of '
characters in its output.
|
| |
|
|
|
|
|
|
|
|
|
|
| |
This change jettisons the awk and ed dependencies and instead uses a
real HTML parser, via the BeautifulSoup library in python.
Using BeautifulSoup lets us do trickier stuff, like actually
extracting the coverage totals and adding our own table, with
per-crate coverage.
The script only does this post-processing when it finds python3; the
script exits with an error if BeautifulSoup isn't installed.
|
| | |
|
| |\
| |
| |
| |
| | |
UX improvements to with_coverage script
See merge request tpo/core/arti!143
|
| | |
| |
| |
| | |
The syntax is kinda minimal, but it seems to work okay.
|
| | |
| |
| |
| | |
Applies a suggestion from @trinity-1686a
|
| | |
| |
| |
| |
| | |
This is a slight kludge, but it helps a lot to keep track of what
exactly was tested for coverage.
|
| | |
| |
| |
| |
| |
| |
| |
| | |
If the -c flag is provided, we _continue_ with the previous
consensus_meta data (without deleting it).
If the -s flag is provided, we _skip_ generating the final report
(since we're going to be doing more profiling.)
|
| | |
| |
| |
| |
| | |
If the directories are absent, then deleting them gives you a line
in stderr, even if you do "rm -f foo || true".
|
| | |
| |
| |
| |
| |
| |
| |
| | |
Now the shell is controlled by the presence of an -i flag, and can
be combined with a command.
This commit also adds usage messages and better reports for a couple
of kinds of setup errors.
|
| | |
| |
| |
| | |
Also improve comments and stdout slightly...
|
| |/ |
|
| | |
|
| |
|
|
|
| |
usage: with_coverage <command> [args...]: run <command> with [args...]
with_coverage : run bash
|
| |
|
|
|
|
|
| |
It formats them in a nice little json object, which it writes to
stdout.
Part of arti#172
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
We need this for clock_gettime_nsec_np().
The source is not canonical, but neither was the one we were using
before. To mitigate this, I've added a digest check on the SDK after
downloading it.
A 100% real canonical reproducible build process will probably
require Apple hardware anyway.
Closes #221
|
| |
|
|
|
|
| |
This doesn't actually do the right thing: it will make somebody get
ignored completely for a release if they _only_ use their
alternative name.
|
| | |
|
| | |
|
| |\ |
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Disclaimer: I have no idea what I'm doing when it comes to shell
scripts.
maint/thanks.sh autogenerates a list of contributors since a revision
(that you pass it on the command line), intended to make the work of
going through the shortlog a bit less manual. It can exclude
contributors' names given in maint/exclude_contributors.txt, in order to
filter out people who work for Tor, or who commit under more than one
name.
|
| |/ |
|
| | |
|
| | |
|
| | |
|
| |
|
|
| |
Also, tell the "typos" tool to ignore Cargo.lock.
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Thanks to the chrono update, we no longer include an
obsolete/vulnerable version of the `time` crate. Unfortunately, it
turns out that chrono has the same trouble as `time`: it, too, looks
at the environment via localtime_r, and the environment isn't
threadsafe.
One step forward, one step back. At least the underlying issue is
one that lots of people seem to care about; let's hope they come up
with a solution.
|
| | |
|
| | |
|
| | |
|
| | |
|
