No upload to coveralls.io

[ujh]

Hey there,

sorry if this is the wrong place but somehow the upload to coveralls.io through Travis fails. It used to work maybe a month ago and then I just realised that the tests were failing. That was easy enough to fix. Now I get the following output: https://travis-ci.org/ujh/iomrascalai/jobs/139514349#L1425 (I hope this opens it on the correct line of the log). It seems that kcov is called with wrong arguments if I read the output correctly.

This is the .travis.yml I'm using: https://github.com/ujh/iomrascalai/blob/master/.travis.yml

As I'm the first reporting this I assume I've done something wrong, but I just can't figure out what I'm doing wrong.

[hexjelly]

I have the same output, and it also stopped working for me about 3-4 weeks ago.

[ujh]

I see. Now the question is if this is an issue with travis-cargo, kcov, or coveralls.io. :) The output from Travis suggests that kcov isn't even called correctly. But I also don't know how the interaction between kcov and coveralls.io is configured. Do I need to do something? Maybe I even did something back in the day to get this to work but I don't even remember and the documentatio of travis-cargo doesn't even mention anything.

[ujh]

As I don't have a Linux box I have a hard time debugging this, but I managed to get it to work by manually calling kcov. This means that it's an issue with the arguments that travis-cargo passes to kcov. My after_success now looks like this:

after_success:
- if [[ "$TRAVIS_OS_NAME" == "linux" ]]; then travis-cargo coveralls --no-sudo --verify; fi
- if [[ "$TRAVIS_OS_NAME" == "linux" ]]; then ./kcov/build/src/kcov --verify --coveralls-id=$TRAVIS_JOB_ID --exclude-pattern=/.cargo target/kcov target/debug/iomrascalai-*; fi

Now stuff is getting uploaded again.

[zetok]

Huh, weird, it worked for me yesterday, but stopped working today when I pushed to master...

Thanks @ujh for posting what made things work again! :)

[cholcombe973]

Mine is also failing with a usage message. I'll try @ujh 's fix.

[Hoverbear]

Related to #55.

[Rafagd]

Is this still broken?

[dpzmick]

yes, mine is failing with a usage message

[gnzlbg]

This is still broke for me.