DOC: Fix examples in documentation

[ShaharNaveh]

• closes #xxxx
• tests added / passed
• passes black pandas
• passes git diff upstream/master -u -- "*.py" | flake8 --diff
• whatsnew entry

[pep8speaks]

Hello @MomIsBestFriend! 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 2020-02-27 17:35:24 UTC

[ShaharNaveh]

Because of lines like these (and basically every other output line, that's received by df.to_json()), I think it's a good idea that we include a "pprint" example under each one, so it will look somewhat like this:

def example():
    """
    Examples
    --------
    Encoding with table schema

    >>> df = pd.DataFrame(
    ...     [["a", "b"], ["c", "d"]],
    ...     index=["row 1", "row 2"],
    ...     columns=["col 1", "col 2"],
    ... )

    >>> df.to_json(orient='table')
    '{"schema":{"fields":[{"name":"index","type":"string"},{"name":"col 1","type":"string"},{"name":"col 2","type":"string"}],"primaryKey":["index"],"pandas_version":"0.20.0"},"data":[{"index":"row 1","col 1":"a","col 2":"b"},{"index":"row 2","col 1":"c","col 2":"d"}]}'


    Pretty print version:

    >>> import json
    >>> result = df.to_json(orient="table")
    >>> parsed = json.loads(result)
    >>> json.dumps(parsed, indent=4)
    {
        "schema": {
            "fields": [
                {
                    "name": "index",
                    "type": "string"
                },
                {
                    "name": "col 1",
                    "type": "string"
                },
                {
                    "name": "col 2",
                    "type": "string"
                }
            ],
            "primaryKey": [
                "index"
            ],
            "pandas_version": "0.20.0"
        },
        "data": [
            {
                "index": "row 1",
                "col 1": "a",
                "col 2": "b"
            },
            {
                "index": "row 2",
                "col 1": "c",
                "col 2": "d"
            }
        ]
    }
    """

[alimcmaster1]

Flake8 errors in CI:

##[error]./pandas/core/generic.py:2217:89:E501:line too long (94 > 88 characters)

[jreback]

also pls merge master

[jreback]

can you use ... instead here?

[jreback]

can you use ... here? (else could do a json prettify, e.g. json.dump(...., indent=4)

[jreback]

are these on the doctest list that we check?

[ShaharNaveh]

Since 4d66fa8 they are.

[alimcmaster1]

How come you are adding doctest skips opposed to our current pytest -k approach? Think we should be consistent.

[ShaharNaveh]

The problem with describe, is that the output can be random, If you know how to skip a specific line in the output, it would be great!

[ShaharNaveh]

Restarting azure

[datapythonista]

A bit unsure about this, but I think it's reasonable. At least we test all what we can test.

You have conflicts to fix. Also, for inline comments (with #), please leave two spaces instead of one before the hash. That's PEP-8, not sure why the PEP-8 validation is not complaining, I guess we don't have it active in the CI because there are failing cases.

Also, why describe output is not deterministic? I think it should.

[datapythonista]

Thanks for the fixes @MomIsBestFriend. About the describe, why do we need to skip the tests? Isn't the output deterministic?

@jreback, if you want to have another look and see if your comments were addressed...

[ShaharNaveh]

Thanks for the fixes @MomIsBestFriend. About the describe, why do we need to skip the tests? Isn't the output deterministic?

I haven't got it to work without the skip maybe one of the core developers knows something?

[datapythonista]

I haven't got it to work without the skip maybe one of the core developers knows something?

Do you know what was the error?

[datapythonista]

Thanks for fixing those @MomIsBestFriend.

Do you mind opening an issue for the describe? I think the output should be deterministic and we shouldn't need the SKIP there. Would be good to have a look and know what's wrong.