[ZEPPELIN-1255] Add cast to string in z.show() for Pandas DataFrame

[bustios]

What is this PR for?

Casting data types in Pandas DataFrame to string in z.show()

What type of PR is it?

Bug Fix

What is the Jira issue?

ZEPPELIN-1255

How should this be tested?

%python
import pandas as pd
df = pd.read_csv('https://archive.ics.uci.edu/ml/machine-learning-databases/iris/iris.data', header=None)
df.columns=[1, 2, 3, 'PetalWidth', 'Name']
z.show(df)
%python.sql
SELECT * FROM df LIMIT 10

Questions:

• Does the licenses files need update? No
• Is there breaking changes for older versions? No
• Does this needs documentation? No

[felixcheung]

looks good!

[felixcheung]

could you add tests for this?

[bzz]

Looks great to me, @bustios could you add a simple test case that reproduce this case to the python/src/test/java/org/apache/zeppelin/python/PythonInterpreterPandasSqlTest.java ?

[bustios]

Tranks for the revision @felixcheung, @bzz. I have already added the test case.

I think it would be nice to show a column with the row number or index if the DataFrame had one.

[bzz]

Looks great to me.

I think it would be nice to show a column with the row number or index if the DataFrame had one.

@bustios This makes sense, please feel free to open a separate JIRA issue for such improvement. We just need to make sure that it's clearly distinguished and can not be confused by the user as a part of the dataset.

CI failure is due to flaky R integration tests, which are not related

Results :
Failed tests: ZeppelinSparkClusterTest.sparkRTest:116 expected:<[[1] 3]> but was:<[<pre><code>Error in getSparkSession(): SparkSession not initialized
</code></pre>
<pre><code>Error in (function (classes, fdef, mtable) : unable to find an inherited method for function 'count' for signature '&quot;function&quot;'
</code></pre>]>
Tests run: 65, Failures: 1, Errors: 0, Skipped: 0

Merging if there is no further discussion.