DOC: Fix flake8 issue in whatsnew files

[saurav2608]

In some of the whatsnew files we are using implicit imports. For example in "whatsnew/v0.11.0.rst"
we import pandas from pandas import * implicitly. We instantiate using the implicitly imported class DataFrame later in the code.

df1 = DataFrame(randn(8, 1), columns = ['A'], dtype = 'float32')
df1

The preferred approach is to use explicit import as import pandas as pd and replace the call to DataFrame with pd.DateFrame.

Note:

1. import pandas as pd is part of the header, so it is imported by default. In that case just replacing DataFrame with pd.DateFrame and removing from pandas import * should be enough.
2. Many whatnew files are using the preferred approach of explicit calling pd.*. However, the import statement of from pandas import * is included by default. In such case, just remove the import statement.

cc: @datapythonista

Status of whatsnew files

File Name	Implicit Imports	Flake8-rst
v0.24.0.rst	Pass	Pass: #24217
v0.23.4.rst	Pass	Pass
v0.23.3.rst	Pass	Pass #24273
v0.23.2.rst	Pass	Pass
v0.23.1.rst	Pass: #24199	Pass #24273
v0.23.0.rst	Pass	Pass: #24217
v0.22.0.rst	Pass	Pass: #24217
v0.21.1.rst	Pass	Pass
v0.21.0.rst	Pass	Pass: #24217
v0.20.3.rst	Pass	Pass
v0.20.2.rst	Pass	Pass
v0.20.0.rst	Pass	Pass: #24236
v0.19.2.rst	Pass	Pass
v0.19.1.rst	Pass	Pass
v0.19.0.rst	Pass #24236	Pass #24236
v0.18.1.rst	Pass	Pass
v0.18.0.rst	Pass #24233	Pass #24233
v0.17.1.rst	Pass #24235	Pass #24235
v0.17.0.rst	Pass #24235	Pass #24235
v0.16.2.rst	Pass #24338	Pass #24338
v0.16.1.rst	Pass #24338	Pass #24338
v0.16.0.rst	Pass #24338	Pass #24338
v0.15.2.rst	Pass #24635	Pass #24635
v0.15.1.rst	Pass #24635	Pass #24635
v0.15.0.rst	Pass #24635	Pass #24635
v0.14.1.rst	Pass #24253	Pass #24253
v0.14.0.rst	Pass #24253	Pass #24253
v0.13.1.rst	Pass #24322	Pass #24322
v0.13.0.rst	Pass #24322	Pass #24322
v0.12.0.rst	Pass #24309	Pass #24309
v0.11.0.rst	Pass #24277	Pass #24277
v0.10.1.rst	Pass #24277	Pass #24277
v0.10.0.rst	Pass #24277	Pass #24277
v0.7*-v0.9*	Pass #24273	Pass #24273
v0.4*-v0.6*	Pass	Pass

[datapythonista]

Thanks for creating the issue @saurav2608

Did you check if the changes required are small to have a single PR for all files? I'm guessing if it'd make sense to create an issue to fix all the flake8-rst errors for each version (all the 0.23.* in one issue). I see all the whatsnew files are skipped right now.

If you want to take a look yourself, may be fix some cases, see what's reasonable to do (in terms of having issues that are not too big), and create the issues, that would be really great. Otherwise I'll take a look myself.

[saurav2608]

@datapythonista :
There are 1015 flake8-rst errors in all the whatsnew files. Of that 200 have F405 type errors. F405 is due to import * but could have other reasons as well. My suggestion is to fix all the F405 errors in one PR. I will take this up.
For the rest, I can create one issue per whatsnew. And one master issue tracking them all. (similar to how you have organized for the rest of .rst files)

[datapythonista]

Sounds good. I'd start by the newest whatsnew files, and if you see there are too many F405 errors for one PR, just open one half way and continue in another. Huge PRs make reviews very inefficient.

[saurav2608]

I have created a PR for errors in recent whatsnew ( v20.0 and above). I am creating issues for the flake-8 error. One for each whatsnew.

[datapythonista]

I changed the formats in the table to be the same as the issue for the other .rst files, I hope you don't mind. Makes it easier to read having the same.

For the Pass or OK, I consider them OK when we don't exclude them anymore in in setup.cfg. Otherwise, we don't guarantee that a new change does not break them again (in this case there shouldn't be changes to the old whatsnew, but still, better to be in the safe side).

Thanks a lot for taking care of this.

[thoo]

@saurav2608 Can you update the table: status of whatsnew files for my PR? This PR is related to v0.23.0.rst and v0.22.0.rst?

[datapythonista]

updated

[thoo]

PR #24217 now covers #24200 (v0.24.0.rst) too.

[saurav2608]

@jreback - this issue should remain open. This is a master issue with reference to all the issues in whatsnew.

[jreback]

@saurav2608 see my comment

don’t put ‘closes’ in the PR, instead use xref

[saurav2608]

@jreback I see what you mean. This issue changed over time. Going forward it will be streamlined.

[saurav2608]

@datapythonista : can you have a quick look at the issue I pointed to in the PR #24236. I am cleaning up the above table need clarifications.

[datapythonista]

@saurav2608 I couldn't find the issue, can you send me the link to the exact comment, or tag me on it?

[saurav2608]

@saurav2608 I couldn't find the issue, can you send me the link to the exact comment, or tag me on it?

@datapythonista : I tagged you on that. I am not able to get the link to the exact comment (it is a review ). I have added a new comment tagging you. Line 1036.

[datapythonista]

I can't find anything in #24236 and didn't received any notification, no idea what's going on, but no idea what's the issue, sorry.

[saurav2608]

@datapythonista : I am replicating the comment here. The issue is near line 1030.

I know this is closed and merged. But I noticed an exception while building docs on my local. And traced it back to here. This 2 space indentation is probably wrong. In my local build, I am getting an error

Exception in .../pandas-saurav/doc/source/whatsnew/v0.20.0.rst at block ending on line 1043
Specify :okexcept: as an option in the ipython:: block to suppress this message
  File "<ipython-input-114-e2b4e607b6f6>", line 1
    .. _whatsnew_0200.api_breaking.hdfstore_where:
    ^
SyntaxError: invalid syntax

The indentation puts this line of code inside .. code-block:: ipython
@FHaase , retag @datapythonista

[datapythonista]

ok, I see, can you open a PR to remove the indentation please?

[jreback]

these are left (aside from open PR to fix 0.15.x)

    doc/source/basics.rst
    doc/source/contributing_docstring.rst
    doc/source/enhancingperf.rst

[saurav2608]

There are no issues created for one and two above. I can create those and if reasonable effort finish them off.

[saurav2608]

@jreback / @datapythonista - this issue is now closed. All what whatsnew files are off the exception list.

[datapythonista]

thanks a lot for this @saurav2608