Update our Logger to use python's logging package, revised #763
Conversation
Codecov Report
@@ Coverage Diff @@
## master #763 +/- ##
==========================================
+ Coverage 99.42% 99.51% +0.08%
==========================================
Files 150 150
Lines 5709 5718 +9
==========================================
+ Hits 5676 5690 +14
+ Misses 33 28 -5
Continue to review full report at Codecov.
|
| assert "EvalML version:" in caplog.text | ||
| assert "EvalML installation directory:" in caplog.text | ||
| assert "SYSTEM INFO" in caplog.text | ||
| assert "INSTALLED VERSIONS" in caplog.text |
There was a problem hiding this comment.
Shouldn't this still use capsys, since this is testing a CLI?
There was a problem hiding this comment.
Seems like caplog grabs the output of the logger--which has a handler to stdout and hence prints to stdout. Similar to the other comment about capsys/caplog
| def test_logger_critical(caplog): | ||
| logger.critical("Test critical") | ||
| assert "Test critical" in caplog.text | ||
| assert "CRITICAL" in caplog.text |
There was a problem hiding this comment.
@angela97lin these tests are helpful! Love it.
I suggest you also add capsys to them, and add another similar test for DEBUG-level. That way, we can ensure everything goes into the log file, and that INFO and above make it into stdout while DEBUG doesn't.
There was a problem hiding this comment.
@dsherry I thought about this, but I don't think I'm able to add capsys fixture for logging. I can share screenshots of what I've tried:
So oddly, it's printed out in the "Captured stdout call" but out returns an empty string. I can keep poking around though!
There was a problem hiding this comment.
Ah, got it. No problem! That's so weird.
There was a problem hiding this comment.
Yeah... I saw a potentially related issue: pytest-dev/pytest#5997
I manually tested by removing one handler and checking that tests still passed but 🤷♀️
There was a problem hiding this comment.
@angela97lin , this rocks!! 😁 So happy this is coming in time for the release too. Great stuff. I left one testing request and some other minor comments. Go ahead and merge once you feel those are addressed!
|
@angela97lin : I responded to a few of your comments. This is still ready to go from my perspective! |
| elif self.max_pipelines: | ||
| if num_pipelines >= self.max_pipelines: | ||
| return False | ||
| elif self.max_time and elapsed >= self.max_time: | ||
| logger.log("\n\nMax time elapsed. Stopping search early.") |
There was a problem hiding this comment.
Removed the elif and cleaned this up because I don't this this was reachable code (would have hit the first if statement)
There was a problem hiding this comment.
Nice! I saw this in codecov a couple weeks ago and tried to clean it up and it kept failing unit tests. Cool that you got it passing
Closes #690, continuation of work in #694
Note:
log_subtitleandlog_titlewill log using name=logger.pysince that's where the call to logger.info is.