bots-for-humanity / trio-gitter-bot

Gitter bot for the Trio project

Apache License 2.0

2 stars 2 forks source link

Print that the job is starting. #5

Closed Mariatta closed 5 years ago

codecov-io commented 5 years ago

Codecov Report

Merging #5 into master will not change coverage. The diff coverage is n/a.

@@         Coverage Diff          @@
##           master    #5   +/-   ##
====================================
  Coverage      65%   65%           
====================================
  Files           2     2           
  Lines          20    20           
  Branches        2     2           
====================================
  Hits           13    13           
  Misses          6     6           
  Partials        1     1

Continue to review full report at Codecov.

Legend - Click here to learn more Δ = absolute <relative> (impact), ø = not affected, ? = missing data Powered by Codecov. Last update 1512001...ace7691. Read the comment docs.

njsmith commented 5 years ago

Do we want to put in any countermeasures against stdout buffering, just in case?

Mariatta commented 5 years ago

ok, so use the -u option in Procfile?

njsmith commented 5 years ago

That's a good simple way yeah

Mariatta commented 5 years ago

What would be a better way? 😬

njsmith commented 5 years ago

I think all the ways I know of are:

pass -u on the command line
set PYTHONUNBUFFERED=1 in the environment
(py37+ only) call sys.stdout.reconfigure(line_buffering=True)
weird awkward hacks that have a similar effect and work on earlier versions but are weird awkward hacks

They're all kind of ... just fine? I don't think there's any major reason to prefer one over another.

Mariatta commented 5 years ago

Any harm in doing all of them? 😛 I've done all below ✅ added sys.stdout.reconfigure(line_buffering=True) ✅ set the PYTHONUNBUFFERED environment (in Heroku) ✅ add -u in the Procfile

njsmith commented 5 years ago

belt AND suspenders

AND umm I dunno glue or something?

njsmith commented 5 years ago

(I'm pretty sure they are all literally exactly equivalent, so no, no harm in doing all of them :-))

njsmith commented 5 years ago

anyway LGTM

looks like we're double-testing PRs from branches in the main repo, which I think there's some way to disable but is mostly harmless, just slows down travis slightly

njsmith commented 5 years ago

Ohand we don't even have any tests anyway :-) I guess I don't need to wait for the tests to pass then...