UI.table from_pandas method

[zilch42]

Adds from_pandas method to ui.table based on the similar method in ui.aggrid.

Addresses #1972, including conversion of the non-serializable datatypes identified in #1698

I think I have updated all of the documentation and tests but let me know if I have missed anything.

[falkoschindler]

Thanks a lot for this wonderful pull request, @zilch42!
Let's merge it right away 🚀

[rohitsathish]

@falkoschindler, as far as I can see, this problem still persists in 1.4.12.
I get the following error when using ui.table.from_pandas():

Type is not JSON serializable: Timestamp
TypeError: Object of type Timestamp is not JSON serializable

[falkoschindler]

@rohitsathish Can you share a minimal reproducible code example? The following code seems to be working:

df = pd.DataFrame({'Name': ['John'], 'DOB': ['1/1/2000']})
df['DOB'] = pd.to_datetime(df['DOB'])
ui.table.from_pandas(df)

[rohitsathish]

@falkoschindler, looks like the time component is causing the error.

df = pd.DataFrame({'Name': ['John'], 'DOB': ['2021-06-25 03:00:00+00:00']})
df['DOB'] = pd.to_datetime(df['DOB'])
ui.table.from_pandas(df)

[falkoschindler]

@rohitsathish Looks like the timezone is causing the error. 🤦‍♂️
This works:

df = pd.DataFrame({'Name': ['John'], 'DOB': ['2021-06-25 03:00:00']})
df['DOB'] = pd.to_datetime(df['DOB'])
ui.table.from_pandas(df)

I'll create a new issue.