*** caphrim007_ has joined #openstack-jjb | 00:28 | |
*** caphrim007_ has quit IRC | 00:29 | |
*** caphrim007_ has joined #openstack-jjb | 00:29 | |
*** caphrim00_ has joined #openstack-jjb | 00:31 | |
*** caphrim007 has quit IRC | 00:31 | |
*** caphrim007_ has quit IRC | 00:34 | |
*** caphrim00_ has quit IRC | 00:36 | |
*** ckonstanski has quit IRC | 01:52 | |
openstackgerrit | Anil Belur proposed openstack-infra/jenkins-job-builder master: Add support for fallback Job Filters https://review.openstack.org/549091 | 02:45 |
---|---|---|
openstackgerrit | Anil Belur proposed openstack-infra/jenkins-job-builder master: Add support for job status job filter https://review.openstack.org/549092 | 02:45 |
*** ckonstanski has joined #openstack-jjb | 03:03 | |
*** zxiiro has quit IRC | 04:10 | |
*** zaro_ has quit IRC | 04:11 | |
*** abelur has quit IRC | 04:11 | |
*** jplf has quit IRC | 04:11 | |
*** larainema has quit IRC | 04:11 | |
*** larainema_ has joined #openstack-jjb | 04:12 | |
*** zxiiro has joined #openstack-jjb | 04:12 | |
*** abelur has joined #openstack-jjb | 04:17 | |
*** zaro_ has joined #openstack-jjb | 04:17 | |
*** jplf has joined #openstack-jjb | 04:29 | |
*** caphrim007 has joined #openstack-jjb | 07:30 | |
*** caphrim007 has quit IRC | 07:35 | |
*** larainema_ has quit IRC | 11:17 | |
openstackgerrit | Sorin Sbarnea proposed openstack-infra/jenkins-job-builder master: Add multibranch project-type https://review.openstack.org/403940 | 11:56 |
*** hashar has joined #openstack-jjb | 12:06 | |
ssbarnea | does anyone know a nice way to display a diff from python? I want to make test failures more friendly on jjb | 13:52 |
ssbarnea | to look more like "git diff" instead of printing both expected vs actual xml, which can be huge. | 13:52 |
ssbarnea | i know that I could save files to tmp and use git as an alternative, but maybe i can do it easier with pure python / extra module. | 13:53 |
zxiiro | personally I kinda like that it prints the whole thing with the diff inside (it is diff already isn't it?) | 14:01 |
zxiiro | I think not everyone knows how to read a diff but having it print the full xml with the +/- where it is different helps. | 14:02 |
ssbarnea | zxiiro: not sure where it prints the diffs, but it was not printing diffs in my tests. i fixed by using git diff, and pleased about result. | 14:42 |
ssbarnea | i think that's due to the outdated implementation in my patch. i need to look how to use same behavior. | 14:50 |
zxiiro | ssbarnea: take one of the existing plugin tests and change the expected result. You should see a diff on failure | 14:55 |
ssbarnea | zxiiro: yep, fixing now, already found. | 14:55 |
ssbarnea | i am starting to believe that it could have being easier to rewrite multibranch from scratch instead of fixing it. | 14:56 |
ssbarnea | the patch was too old, too many things changed (improved) in between. | 14:56 |
ssbarnea | zxiiro: do you think it would be preferred to split the test files to one for each scm supported instead of one for each project type? | 15:03 |
zxiiro | ssbarnea: the way I see it, it's one per plugin and the multibranch is a project that uses many plugins | 15:04 |
zxiiro | ssbarnea: I also have been trying to push for 2 types of tests for every plugin. A "full" and a "minimal" test. The minimal one provides the absolute smallest yaml that get's the plugin to work and relies on defaults to fill out all the blanks, and the full one configures every possible configuration in the non-default value. | 15:05 |
ssbarnea | zxiiro: yep, was thinking the same, in fact in my special case i have only two tests: one minimal and one maximal, so very close to what you describe. | 15:06 |
ssbarnea | i prefer to include the minimal in documentation, or it could be overwhelming for the user. | 15:07 |
zxiiro | great! please use the "minimal" and "full" language. We want to try to keep the wording consistent and I believe that language is the current most used one. | 15:07 |
zxiiro | ssbarnea: the way I do it is I include the minimal first in documentation and full test last. that way the first thing the user sees is the minimal configuration but if they are interested they can look at the full configuration at the bottom. | 15:08 |
zxiiro | ssbarnea: I've found both is good because some people are visual and like to see exactly how to set something. | 15:08 |
*** caphrim007 has joined #openstack-jjb | 16:33 | |
openstackgerrit | Major Hayden proposed openstack-infra/jenkins-job-builder master: [Docs] Ensure commands are monospaced https://review.openstack.org/549284 | 16:41 |
openstackgerrit | Merged openstack-infra/jenkins-job-builder master: [Docs] Ensure commands are monospaced https://review.openstack.org/549284 | 17:07 |
openstackgerrit | Sorin Sbarnea proposed openstack-infra/jenkins-job-builder master: Add multibranch project-type https://review.openstack.org/403940 | 17:08 |
ssbarnea | zxiiro: finally, now with multiple smaller tests that are using same | 17:54 |
zxiiro | ssbarnea: awesome | 17:55 |
zxiiro | ssbarnea: great this patch looks a lot easier to read | 17:56 |
zxiiro | ssbarnea: if you can get electrofelix's +2 I think we can merge this. | 17:56 |
*** hashar is now known as hasharAway | 18:27 | |
*** hasharAway has quit IRC | 23:45 | |
*** caphrim007_ has joined #openstack-jjb | 23:59 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!