NixOS/hydra
NixOS/hydra
1
0
Fork 0
mirror of https://github.com/NixOS/hydra.git synced 2024-10-17 16:37:26 -04:00
Code Activity
hydra/t/Makefile.am

40 lines
1,023 B
Makefile
Raw Normal View History

Fix the test
2011-11-30 12:03:50 -05:00
TESTS_ENVIRONMENT = \
BZR_HOME="$(abs_builddir)/data" \
Run PostgreSQL during "make check"
2015-06-25 09:51:44 -04:00
HYDRA_DBI="dbi:Pg:dbname=hydra-test-suite;port=6433" \
Fix the test
2011-11-30 12:03:50 -05:00
HYDRA_DATA="$(abs_builddir)/data" \
tests: $(topsrc_dir) -> $(top_srcdir).
2011-03-16 12:18:59 -04:00
HYDRA_HOME="$(top_srcdir)/src" \
Clear $HYDRA_CONFIG in the tests Otherwise one might accidentally send out HipChat notifications when running the tests...
2013-07-19 08:42:34 -04:00
HYDRA_CONFIG= \
tests: Use $(builddir) instead of `pwd`.
2011-03-16 12:18:56 -04:00
NIX_REMOTE= \
Tests: Clear $NIX_REMOTE_SYSTEMS
2016-10-06 11:19:16 -04:00
NIX_REMOTE_SYSTEMS= \
2011-03-17 09:25:27 -04:00
NIX_CONF_DIR="$(abs_builddir)/nix/etc/nix" \
nix needs env vars as absolute paths
2011-03-17 03:43:33 -04:00
NIX_STATE_DIR="$(abs_builddir)/nix/var/nix" \
NIX_STORE_DIR="$(abs_builddir)/nix/store" \
NIX_LOG_DIR="$(abs_builddir)/nix/var/log/nix" \
tests: change postgresql socket dir to /tmp In https://github.com/NixOS/nixpkgs/pull/57677 default postgresql socket directory was changed to `/run/postgresql`, which doesn't exist (and can't be created) in Nix build environment. We'll use /tmp as socket dir explicitly then. Fixes build failure https://hydra.nixos.org/build/91221682 Cc @aszlig @edolstra
2019-03-29 17:46:38 -04:00
PGHOST=/tmp \
Fix tests
2017-05-24 10:00:38 -04:00
PERL5LIB="$(srcdir):$(abs_top_srcdir)/src/lib:$$PERL5LIB" \
Fix Mercurial tests In Nixpkgs 20.03, Mercurial fails if PYTHONPATH is set: $ hg Traceback (most recent call last): File "/nix/store/q7s856v6nw4dffdrm9k3w38qs35i8kr3-mercurial-5.2.2/bin/..hg-wrapped-wrapped", line 37, in <module> dispatch.run() File "/nix/store/bffdy7q3wi3qinflnvbdkigqj39zzynd-python3-3.7.6/lib/python3.7/importlib/util.py", line 245, in __getattribute__ self.__spec__.loader.exec_module(self) File "<frozen importlib._bootstrap_external>", line 728, in exec_module File "<frozen importlib._bootstrap>", line 219, in _call_with_frames_removed File "/nix/store/q7s856v6nw4dffdrm9k3w38qs35i8kr3-mercurial-5.2.2/lib/python3.7/site-packages/mercurial/dispatch.py", line 10, in <module> import difflib File "/nix/store/bffdy7q3wi3qinflnvbdkigqj39zzynd-python3-3.7.6/lib/python3.7/difflib.py", line 1084, in <module> import re File "/nix/store/bffdy7q3wi3qinflnvbdkigqj39zzynd-python3-3.7.6/lib/python3.7/re.py", line 143, in <module> class RegexFlag(enum.IntFlag): AttributeError: module 'enum' has no attribute 'IntFlag' (cherry picked from commit 4009d4295e1e3c0e59e716b0b111f48c590a6020)
2020-03-31 07:33:11 -04:00
PYTHONPATH= \
Concurrent hydra-evaluator This rewrites the top-level loop of hydra-evaluator in C++. The Perl stuff is moved into hydra-eval-jobset. (Rewriting the entire evaluator would be nice but is a bit too much work.) The new version has some advantages: * It can run multiple jobset evaluations in parallel. * It uses PostgreSQL notifications so it doesn't have to poll the database. So if a jobset is triggered via the web interface or from a GitHub / Bitbucket webhook, evaluation of the jobset will start almost instantaneously (assuming the evaluator is not at its concurrency limit). * It imposes a timeout on evaluations. So if e.g. hydra-eval-jobset hangs connecting to a Mercurial server, it will eventually be killed.
2016-10-13 09:53:05 -04:00
PATH=$(abs_top_srcdir)/src/hydra-evaluator:$(abs_top_srcdir)/src/script:$(abs_top_srcdir)/src/hydra-eval-jobs:$(abs_top_srcdir)/src/hydra-queue-runner:$$PATH \
Remove trailing whitespace
2013-01-22 08:41:02 -05:00
perl -w
2011-03-16 06:27:32 -04:00
make sure jobs/* and Setup.pm is in tarball
2011-03-17 03:30:19 -04:00
EXTRA_DIST = \
$(wildcard *.pm) \
$(wildcard jobs/*.nix) \
Rename hydra_*.pl to hydra-* The underscores are ugly and the .pl extension is an implementation detail that shouldn't be visible to the outside. Also, get rid of the *.in files. It's not really necessary to generate them. And I was always modifying the wrong file.
2011-11-30 11:32:50 -05:00
$(wildcard jobs/*.sh) \
$(TESTS)
first test, not yet in buildprocess
2011-03-16 04:50:31 -04:00
TESTS = \
Start checking PRs with perlcritic
2021-08-19 16:13:45 -04:00
perlcritic.pl \
tests: delete set-up.pl / tear-down.pl We'll set these up on a per-test basis.
2021-02-22 13:33:25 -05:00
test.pl
add first evaluations tests
2011-03-16 09:18:12 -04:00
Update "make check" for the new queue runner Also, if the machines file contains an entry for localhost, then run "nix-store --serve" directly, without going through SSH.
2015-06-25 09:29:22 -04:00
check_SCRIPTS = repos
first test, not yet in buildprocess
2011-03-16 04:50:31 -04:00
Fix handling of IPC::Run::run exit status Turns out that the exit status is returned in $?, not as the return value of run(). So our error checking was completely bogus.
2013-01-23 09:56:28 -05:00
repos: dirs
set BZR_HOME env var so bzr won't fail
2011-03-17 05:14:48 -04:00
Fix handling of IPC::Run::run exit status Turns out that the exit status is returned in $?, not as the return value of run(). So our error checking was completely bogus.
2013-01-23 09:56:28 -05:00
dirs:
setup environment variables for tests
2011-03-16 05:46:47 -04:00
mkdir -p data
2011-03-16 06:50:50 -04:00
touch data/hydra.conf
2011-03-16 06:27:32 -04:00
mkdir -p nix
2011-03-17 09:25:27 -04:00
mkdir -p nix/etc/nix
2011-03-16 06:27:32 -04:00
mkdir -p nix/store
mkdir -p nix/var
Copy permalink