search

start-jsk / rtmros_hironx

hironx controller and applications using rtmros packages
http://wiki.ros.org/rtmros_hironx
10 stars 27 forks source link

[Travis] Permit ROS shadow repo failure. #451

Closed 130s closed 8 years ago

130s commented 8 years ago

When packages in ROS shadow-fixed repo are failing, Travis never passes with the current config (e.g. #450). So use the official repo as a default, and shadow repo as an option.

k-okada commented 8 years ago

I believe failing on ros-shadow-fixed will be solve very soon, after that, we do not need to ignore failure.

◉ Kei Okada

On Mon, Apr 11, 2016 at 1:41 AM, Isaac I.Y. Saito notifications@github.com wrote:

When packages in ROS shadow-fixed repo are failing, Travis never passes with the current config (e.g. #450 https://github.com/start-jsk/rtmros_hironx/pull/450). So use the

official repo as a default, and shadow repo as an option.

You can view, comment on, or merge this pull request online at:

https://github.com/start-jsk/rtmros_hironx/pull/451 Commit Summary

  • [Travis] Permit ROS shadow repo failure.

File Changes

Patch Links:

— You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub https://github.com/start-jsk/rtmros_hironx/pull/451

130s commented 8 years ago

上流のエラーによって下流の PR が2ヶ月もレヴュできないとか，ちょっと不都合過ぎるので，一旦この PR をマージして先に進めたいです．

あと一般論として，

k-okada commented 8 years ago
  • Official レポを判定基準に含めるのは悪くないと思いますがどうでしょうか．

これはそのとおりですね．いれましょう．

  • Shadow を判定基準に含めるかどうかは方針次第ということじゃないでしょうか．

今までの基準を緩めている，一時的な対応である，上流のエラーは治せる立場に居る，の理由でこれは良くないと思います．

◉ Kei Okada

k-okada commented 8 years ago

https://github.com/k-okada/rtmros_hironx/commit/3c9cfafece44c0ab8dcc3904eaf16554c2f8d399

たぶんですが，http:の:がyamlのenv:の:と同じ解釈されているので"でかこう必要が有るのが原因でしょうか． よく見ると，最初にexport USE_DEB=..と出てくるところで，export ROS_REPOSITORY_PATH=...がない時点で気づける，というのがポイントですかね．

130s commented 8 years ago

k-okada thank you for taking a look. For Travis failure, I found that the cause was passing env var to sudo shell. It is now already addressed in my latest commit https://github.com/start-jsk/rtmros_hironx/pull/451/commits/b433c6128fb9d8d28ee9d1fe4d6aab9df2db8c4e.

I agree that disregarding the rule that has been in effect for awhile (i.e. using shadow repo as a criteria) isn't good. Commit is updated.

Now, only failing test is the one uses shadow repo, which is expected due to the shadow-repo glitch.

k-okada commented 8 years ago

humm..

https://github.com/k-okada/rtmros_hironx/blob/1e6fcf70957dfc34747ea4660ddae407ba1eddb5/.travis.yml / https://travis-ci.org/k-okada/rtmros_hironx/jobs/122395933#L160 outputs like


Setting environment variables from .travis.yml

$ export ROS_DISTRO=indigo
$ export ROSWS=wstool
$ export BUILDER=catkin
$ export USE_DEB=false

and

0.00s$ env | grep PATH
MANPATH=/usr/local/clang-3.5.0/share/man:/usr/local/man:/usr/local/share/man:/usr/share/man
PATH=/home/travis/bin:/home/travis/.local/bin:/home/travis/.rvm/gems/ruby-2.2.3/bin:/home/travis/.rvm/gems/ruby-2.2.3@global
/bin:/home/travis/.rvm/rubies/ruby-2.2.3/bin:/home/travis/.pyenv/shims:/home/travis/.phpenv/shims:/home/travis/.php-build/bin:/home/travis/.local/bin:/home/travis/.pyenv/shims:/opt/pyenv/bin:/home/travis/.phpenv/shims:/home/travis/perl5/perlbrew/bin:./node_modules/.bin:/usr/local/clang-3.5.0/bin:/home/travis/.local/bin:/home/travis/.pyenv/shims:/opt/pyenv/bin:/home/travis/.phpenv/shims:./node_modules/.bin:/usr/local/clang-3.5.0/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/home/travis/.phpenv/bin:/home/travis/.phpenv/bin:/home/travis/.rvm/bin:/home/travis/.rvm/bin
GEM_PATH=/home/travis/.rvm/gems/ruby-2.2.3:/home/travis/.rvm/gems/ruby-2.2.3@global
CI_SOURCE_PATH=/home/travis/build/k-okada/rtmros_hironx

so i think ROS_REPOSITORY_PATH is not set to the travis shell, but you PR seems working, so it is ok. Sorry for noise.

◉ Kei Okada

130s commented 8 years ago

Thank you for the review. Let me merge to see how this improves tests on other PRs (I'll rebase those).