Concise dates test #16215
Concise dates test #16215
Conversation
| ], | ||
| [datetime.timedelta(seconds=40), | ||
| ['', '05', '10', '15', '20', '25', '30', '35', '40'] | ||
| ], |
There was a problem hiding this comment.
OK, these results are a little confusing, I suspect because you are starting 1 Jan? Does it make sense to start on a different date and a bit later in the day?
There was a problem hiding this comment.
I made the last element in zero_formats "%S.%f' in stead of ".%f" to make the expected string '00' instead of the empty string above.
There was a problem hiding this comment.
I think its a bit confusing to have the zero-format for seconds be seconds, but I guess this indeed tests the functionality.
lib/matplotlib/tests/test_dates.py
Outdated
| 'day: 22', | ||
| '03/1997', | ||
| 'day: 22', | ||
| '04/1997', |
There was a problem hiding this comment.
I don't know that this should be laid out with so many carriage returns. I appreciate it may make it a bit easier to debug, but it makes it harder for other to scroll through the file.
lib/matplotlib/tests/test_dates.py
Outdated
| return sts | ||
|
|
||
| d1 = datetime.datetime(1997, 1, 1) | ||
| results = ([datetime.timedelta(weeks=52 * 200), |
There was a problem hiding this comment.
The indentation here is strange. I'm suprised it's not being picked up by the pep8 checker. Can you check this is pep8 compatible?
lib/matplotlib/tests/test_dates.py
Outdated
| ax.set_ylim(date1, date2) | ||
| fig.canvas.draw() | ||
| sts = [] | ||
| for st in ax.get_yticklabels(): |
There was a problem hiding this comment.
The following would probably be faster and more readable:
sts = [st.get_text() for st in ax.get_yticklabels()]| [datetime.timedelta(weeks=52 * 200), [str(t) for t in range(1980, | ||
| 2201, 20)]], | ||
| [datetime.timedelta(weeks=52), [ | ||
| '1997', '02/1997', '03/1997', '04/1997', '05/1997', '06/1997', |
There was a problem hiding this comment.
| '1997', '02/1997', '03/1997', '04/1997', '05/1997', '06/1997', | |
| '01/1997', '02/1997', '03/1997', '04/1997', '05/1997', '06/1997', |
This bug is not failing the test. Why? Something is fishy here.
There was a problem hiding this comment.
The output you suggest is what you get if you set,
formats = ['%m/%Y', '%m/%Y', 'day: %d', '%H hr %M min', '%H hr %M min', '%S.%f sec']
however, the original is indeed what you get with formats set the way it is in line 512.
PR Summary
PR fixes #15182. Two test cases added to
test_dates.pyboth to testConciseDateFormatterparametersformatsandzero_formatsrespectively. Expected outputs are compared with generated outputs for different time intervals. The file was also formatted with autopep8, hence some minor edits elsewhere in the file--I can revert the formatting easily if requested.PR Checklist