Reduce the amount of logging messages that get too frequently printed during debug#1721
Conversation
amarkpayne
left a comment
There was a problem hiding this comment.
I talked with @mliu49 and we both agree that we can just get rid of this code instead of commenting it out. There are a couple of changes that need to be made due to the failing unit tests (at least one of them needs to be modified because it expects to see a logging statement that we are now getting rid of.
f546992 to
d123f68
Compare
Codecov Report
@@ Coverage Diff @@
## master #1721 +/- ##
==========================================
- Coverage 32.6% 32.57% -0.03%
==========================================
Files 87 87
Lines 26115 26090 -25
Branches 6874 6866 -8
==========================================
- Hits 8514 8500 -14
+ Misses 16640 16616 -24
- Partials 961 974 +13
Continue to review full report at Codecov.
|
|
@amarkpayne, this PR should be ready for review now |
amarkpayne
left a comment
There was a problem hiding this comment.
This all looks good, thanks for making these changes!
Motivation or Problem
When debugging RMG or a software that depends on RMG the amount of debugging statements that get printed jam the buffer, not allowing the developer to conveniently see the actual original error massage.
Description of Changes
Some logging statements, most of which are debugging level, were commented out. I only marked those which get printed too frequently in my experience. The purpose here was to flag them in the code for a discussion of how to best deal with them - leave commented out (so can be conveniently brought back to life if needed) or completely deleting them.
I think we can start the discussion even though we won't merge until the Py3 transition, since the same logging statements will be relevant after transitioning.