Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add coverage badge to README #11286

Merged
merged 1 commit into from
May 15, 2015
Merged

Add coverage badge to README #11286

merged 1 commit into from
May 15, 2015

Conversation

kshyatt
Copy link
Contributor

@kshyatt kshyatt commented May 15, 2015

As suggested in #9493.

IainNZ added a commit that referenced this pull request May 15, 2015
@IainNZ IainNZ merged commit 5aeb499 into master May 15, 2015
@kshyatt
Copy link
Contributor Author

kshyatt commented May 15, 2015

Sorry, I forgot to put [ci skip] on this!

@IainNZ
Copy link
Member

IainNZ commented May 15, 2015

No problem, I was thinking about the CI too!

@kshyatt kshyatt deleted the ksh/readme branch May 15, 2015 17:58
@dhoegh
Copy link
Contributor

dhoegh commented May 15, 2015

The badge do not lead to the same site as: https://coveralls.io/r/JuliaLang/julia. Which then do show a lower and older coverage number than Julia and @kshyatt work deserves.

@ViralBShah
Copy link
Member

I verified - the link is not doing the right thing. It takes us to some older commits. How to fix this?

@kshyatt
Copy link
Contributor Author

kshyatt commented May 15, 2015

I did exactly what Coveralls told me to! Grrrrr. I think there's a larger issue where the Repo Stats part of our Coveralls.io page is doing something funky.

E: This is related to this issue I opened at CoverageBase. The correct commit information is not making it to Coveralls from the buildbot and so the repo stats aren't being updated. Interestingly, when I run the same script at julia.mit.edu, everything works fine. @staticfloat and @tkelman, do you have any thoughts?

@staticfloat
Copy link
Member

Yes, @kshyatt is right. It's because the commit information isn't surviving all the way through to Coveralls. I remember having some serious headaches trying to figure out how to bully Coveralls into accepting my git commit information, looks like it's time to revisit it and figure out what's broken.

The OpenStack cluster at MIT had an emergency poweroff this morning due to coolant systems failing at CSAIL; so I'll take a look at this once the buildbots are back up.

@tkelman
Copy link
Contributor

tkelman commented May 16, 2015

I don't have anything to add to this issue except to wholeheartedly second the sentiment behind

lower and older coverage number than Julia and @kshyatt work deserves.

@staticfloat
Copy link
Member

@kshyatt can you give me the exact command sequence you're using to upload to your coveralls stats?

@tkelman
Copy link
Contributor

tkelman commented Jun 4, 2015

Is it expected that in the "Gather inlined test results" and "Gather non-inlined test results" steps of http://buildbot.e.ip.saba.us:8010/builders/coverage_ubuntu14.04-x64/builds/544, that all of the .cov files are being Skipped?

@IainNZ
Copy link
Member

IainNZ commented Jun 4, 2015

Yep, its just because Coverage.jl does a silly thing where it walks through every file in the folder and looks for file.cov - so it works for the .jl files, but when it hits a .cov file it can't find a .cov.cov so it says "skipped".

@kshyatt
Copy link
Contributor Author

kshyatt commented Jun 13, 2015

Just a quick update, thanks to the amazing work of @staticfloat, the coverage badge is now correct!

@tkelman
Copy link
Contributor

tkelman commented Jun 14, 2015

Nice work. I bet running the pkg test during the coverage run would make the numbers a bit better.

@staticfloat
Copy link
Member

Yes, I briefly discussed the best way to do that with Tim Holy here, I'll write something up and see if we can't get it running tonight. :D

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

6 participants