VIS: Fix DataFrame.plot() produces incorrect legend markers when plotting multiple series on the same axis

[charlesdong1991]

I found out that the reason this happens is because subplots is not updated accordingly when plt.subplots() is used and assign to ax, e.g. fig, ax = plt.subplots(nrows=1, ncols=3). I tested my solution and things all get solved, and figure looks very correct like below:

BTW, i am not very sure how to properly test this, and would be nice to know!

• closes DataFrame.plot() produces incorrect legend labels when plotting multiple series on the same axis #18222
• tests added / passed
• passes black pandas
• passes git diff upstream/master -u -- "*.py" | flake8 --diff
• whatsnew entry

[TomAugspurger]

A test would be nice. Does the issue post have one?

[charlesdong1991]

@TomAugspurger thanks very much for your quick review! Just curious about what you mean by Does the issue post have one? this issue does provide a fail case, but no test is given.
I do see some tests for plotting, but since this issue is mainly about missing markers in legend, i haven't found a proper solution to test if this marker/color is printed correctly. Would be very nice if you could give some hint! ^^

[TomAugspurger]

Ideally we would make a plot and verify that the legend labels and markers are matched up. I can never remember the matplotlib API for that though.

…

On Wed, Aug 7, 2019 at 4:16 PM Kaiqi Dong ***@***.***> wrote: @TomAugspurger <https://github.com/TomAugspurger> thanks very much for your quick review! Just curious about what you mean by Does the issue post have one? this issue does provide a fail case, but no test is given. I do see some tests for plotting, but since this failure is mainly caused by missing markers in legend, haven't found a proper solution to test if this marker/color is printed correctly. Would be very nice if you could give some hint! ^^ — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#27808?email_source=notifications&email_token=AAKAOITMP33G25W33WD6V2TQDM3STA5CNFSM4IKD3SXKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD3ZXOFA#issuecomment-519272212>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAKAOITIAQNWQETQ7ESO74LQDM3STANCNFSM4IKD3SXA> .

[charlesdong1991]

Thanks for your quick reply, I will check it up @TomAugspurger

[pep8speaks]

Hello @charlesdong1991! Thanks for updating this PR. We checked the lines you've touched for PEP 8 issues, and found:

There are currently no PEP 8 issues detected in this Pull Request. Cheers! 🍻

Comment last updated at 2019-08-14 20:47:39 UTC

[charlesdong1991]

This is kind of new working solution with a test case that is passing all tests on my local computer, not sure if it can also pass all tests here. But in the meantime pls feel free to take a look and provide reviews if you have some time. @TomAugspurger

[charlesdong1991]

any follow-up reviews? all are much appreciated! ^^

[TomAugspurger]

Looks good. I'll restart the windows azure CI failure.

[charlesdong1991]

ping

[TomAugspurger]

Thanks!

[comready]

If you need another example of this problem let me know