DataFrame.to_records

[shril]

@rxin, @ueshin Please have a look at my PR. #169
Happy to help regarding my changes.

Reference Pandas Doc: https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.to_records.html

[codecov-io]

Codecov Report

Merging #298 into master will increase coverage by 0.02%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master     #298      +/-   ##
==========================================
+ Coverage   94.31%   94.33%   +0.02%     
==========================================
  Files          35       35              
  Lines        3551     3564      +13     
==========================================
+ Hits         3349     3362      +13     
  Misses        202      202

Impacted Files	Coverage Δ
databricks/koalas/missing/frame.py	100% <ø> (ø)	⬆️
databricks/koalas/testing/utils.py	74.05% <100%> (+0.33%)	⬆️
...tabricks/koalas/tests/test_dataframe_conversion.py	100% <100%> (ø)	⬆️
databricks/koalas/frame.py	95.46% <100%> (+0.03%)	⬆️

---

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 926400f...edf4408. Read the comment docs.

[shril]

@rxin I am getting this exception for Pandas 0.23.4

UNEXPECTED EXCEPTION: TypeError("The pandas version [0.23.4] available does not support parameter 'column_dtypes' for function 'to_records'.",)

Should I skip doctest for it.

[shril]

@rxin, @ueshin I think I need some help here.

[garawalid]

@shril I suggest that you delete the parameter column_dtypes from args if the pandas version is less than 0.23.4

if LooseVersion(pd.__version__) < LooseVersion("0.24.0"):
    del args['column_dtypes'] 

[shril]

@shril I suggest that you delete the parameter column_dtypes from args if the pandas version is less than 0.23.4

if LooseVersion(pd.__version__) < LooseVersion("0.24.0"):
    del args['column_dtypes'] 

Thanks again @garawalid

[rxin]

@shril this doesn't merge cleanly anymore. Can you merge master into your branch so this can merge cleanly?

[HyukjinKwon]

Yea, many PRs are rapidly being merged. I will take a look as soon as it's merged to master.

[shril]

@rxin @HyukjinKwon I will look into this today.
Sorry for the delay.

[shril]

@rxin, @HyukjinKwon I have made the required changes.
I apologize for the delay. Please have a look.
Happy to address any issues regarding my changes.

PR #169.

[HyukjinKwon]

LGTM otherwise.

[HyukjinKwon]

Are we checking different version of doc? Seems not matched - https://github.com/pandas-dev/pandas/blob/v0.24.2/pandas/core/frame.py#L1558-L1741

[shril]

@HyukjinKwon, I am having this issue that dtypes were added in Pandas 0.24.0. So the doc-tests are failing for the lower version of Pandas. So I am skipping the tests for them although I have pushed unit tests for every variation of the function.

[HyukjinKwon]

Thanks, @shril

[shril]

You are most welcome @HyukjinKwon.