Make is_monotonic~ work properly for index

[itholic]

Resolve #931 (The contents below is same as the contents of this issue)

When we use is_monotonic_decreasing or is_monotonic_increasing for index,

pandas works like below:

>>> s = pd.Series([7, 6, 5, 4, 3, 2, 1], index=[7, 6, 5, 4, 3, 2, 1])
>>> s.index.is_monotonic_decreasing
True

Since the index order is literally motononic decreasing, they return True

But our case,

>>> s = ks.Series([7, 6, 5, 4, 3, 2, 1], index=[7, 6, 5, 4, 3, 2, 1])
>>> s.index.is_monotonic_decreasing
False

as seen above, it returns False.

because our existing logic always order by index column before calculate the each rows increasing or decreasing.

So they always calculate result based on data column, not index column. (index column is always sorted since order by)

So i think maybe it is better to fix this logic to follow behavior of pandas one for index.

[ueshin]

@itholic Good catch!
btw, the proposed fix seems still something missing:

>>> pser
0.097069    0.097069
0.674804    0.674804
0.417065    0.417065
dtype: float64
>>> pser.is_monotonic
False
>>> pser.index.is_monotonic
False

>>> ks.from_pandas(pser).is_monotonic
True
>>> ks.from_pandas(pser).index.is_monotonic
False

maybe we need to apply the fix for Index to Series' case as well.

[codecov-io]

Codecov Report

Merging #930 into master will increase coverage by <.01%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master     #930      +/-   ##
==========================================
+ Coverage   94.46%   94.47%   +<.01%     
==========================================
  Files          34       34              
  Lines        6382     6393      +11     
==========================================
+ Hits         6029     6040      +11     
  Misses        353      353

Impacted Files	Coverage Δ
databricks/koalas/base.py	94.83% <100%> (+0.03%)	⬆️
databricks/koalas/missing/frame.py	100% <0%> (ø)	⬆️
databricks/koalas/missing/indexes.py	100% <0%> (ø)	⬆️
databricks/koalas/missing/series.py	100% <0%> (ø)	⬆️
databricks/koalas/frame.py	96.02% <0%> (ø)	⬆️
databricks/koalas/series.py	95.6% <0%> (+0.01%)	⬆️
databricks/koalas/indexes.py	97.38% <0%> (+0.18%)	⬆️

---

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 465861c...af68952. Read the comment docs.

[itholic]

@ueshin , ah i missed it. just fixed it, Thanks for commenting :)

[ueshin]

Let me retrigger the build and let's see.

[itholic]

@ueshin oops my comment removed by mistake; yeah let's see and maybe hyukjin is something doing for speed up build process. 😃

[HyukjinKwon]

LGTM otherwise

[softagram-bot]

Softagram Impact Report for pull/930 (head commit: af68952)

⭐ Change Overview

(Open in Softagram Desktop for full details)

📄 Full report

• Permalink: Full report for pull/930

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