Implement sort_values for Index/MultiIndex

[itholic]

Implement sort_values for Index/MultiIndex
(https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.Index.sort_values.html#pandas.Index.sort_values)

>>> idx = ks.Index([10, 100, 1, 1000])
>>> idx
Int64Index([10, 100, 1, 1000], dtype='int64')

>>> idx.sort_values()
Int64Index([1, 10, 100, 1000], dtype='int64')

>>> idx.sort_values(ascending=False)
Int64Index([1000, 100, 10, 1], dtype='int64')

Support for MultiIndex.

>>> kidx = ks.MultiIndex.from_tuples([('a', 'x', 1), ('c', 'y', 2), ('b', 'z', 3)])
>>> kidx
MultiIndex([('a', 'x', 1),
            ('c', 'y', 2),
            ('b', 'z', 3)],
           )

>>> kidx.sort_values()
MultiIndex([('a', 'x', 1),
            ('b', 'z', 3),
            ('c', 'y', 2)],
           )

>>> kidx.sort_values(ascending=False)
MultiIndex([('c', 'y', 2),
            ('b', 'z', 3),
            ('a', 'x', 1)],
           )

[codecov-io]

Codecov Report

Merging #1120 into master will increase coverage by <.01%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master    #1120      +/-   ##
==========================================
+ Coverage   95.19%   95.19%   +<.01%     
==========================================
  Files          35       35              
  Lines        7071     7075       +4     
==========================================
+ Hits         6731     6735       +4     
  Misses        340      340

Impacted Files	Coverage Δ
databricks/koalas/missing/indexes.py	100% <ø> (ø)	⬆️
databricks/koalas/indexes.py	96.45% <100%> (+0.07%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 468bf3a...a8dda7d. Read the comment docs.

[ueshin]

Do we need this?

[itholic]

without this, i think we can't keep names when below case.

>>> kidx = ks.MultiIndex.from_tuples([('a', 'x', 1), ('c', 'y', 2), ('b', 'z', 3)])
>>> kidx.names = ['A', 'B', 'C']
>>> kidx.sort_values()
MultiIndex([('a', 'x', 1),
            ('b', 'z', 3),
            ('c', 'y', 2)],
           )

[ueshin]

ah, I see. @names.setter seems wrong.

[itholic]

ah, i got it.

in @names.setter, the new index_map is overwritten to self._kdf._internal, not to self._internal.

like below

self._kdf._internal = internal.copy(index_map=list(zip(internal.index_columns, names)))

at this point, i curious why we overwrite self._kdf._internal rather than simply self._internal?

For now, i've fixed it to the current implementation

[HyukjinKwon]

Looks fine but let me leave it to @ueshin

[softagram-bot]

Softagram Impact Report for pull/1120 (head commit: 8ae7722)

⚠️ Copy paste found

ℹ️ test_indexes.py: Copy paste fragment on line 30 shared with ../test_dataframe.py, ../test_numpy_compat.py:

    @property
    def pdf(self):
        return pd.DataFrame({
            'a': [1, 2, 3, 4, 5, 6, 7, 8, 9],
            'b': [4, 5, 6, 3, 2, 1, ...(truncated 160 chars)

ℹ️ indexes.py: Copy paste fragment inside the same file on lines 720, 1163:

            raise NotImplementedError(
                \"Doesn't support symmetric_difference between Index & MultiIndex for now\")

        sdf_self = self._kdf._s...(truncated 477 chars)

Now that you are on the file, it would be easier to pay back some tech. debt.

⭐ Change Overview

(Open in Softagram Desktop for full details)

📄 Full report

• Permalink: Full report for pull/1120

Impact Report explained. Give feedback on this report to [email protected]

[HyukjinKwon]

@itholic can you resolve conflicts?

[itholic]

@HyukjinKwon resolved :)

[ueshin]

LGTM.