Closed quintenpalmer closed 5 years ago
sorry, meant to push this against my fork, if you (upstream) are happy with this change, then would be nice to have this accepted to keep errcheck (https://github.com/kisielk/errcheck) happy.
If you're not interested in this change I can happily close it or handle the errors differently, definitely open to discussion.
Hey,
I'm a fan of static analysis, so I definitely don't have a thing against errcheck. In fact, we should probably add it to the Travis tests.
All these assignments to _
do feel silly, but I guess that's the price we have to pay for it.
Would you mind hooking errcheck
into the Travis config?
Awesome, glad to hear that you're up for satisfying errcheck
. I'm not too familiar with Travis, I'll try to find some time soon to figure out how to add errcheck
and run it as part of the Travis checks.
Shouldn't be too hard, it's just a YAML-shaped shell script, so you'll probably just need to add a go get
line and a line to run it.
sounds good, I'll give that a shot soon, thank you!
@rubenv sorry for the delay, let me know if these last two commits look good. There was one last error not being handled that I caught and I introduced the Travis config changes to run this. I verified that the errcheck check is running and is passing, you can see an example here: https://travis-ci.org/rubenv/sql-migrate/jobs/467150613
Perfect, thanks!
Not sure if upstream is using errcheck, so likely will not publish this change to them, unless they are welcome to it. Publishing as its own PR to keep that bookkeeping separate.
There is nothing really to do with these errors.