Time zone changed to local TZ in Jupyter Lab

[rzzldzzl]

Hi -

When a pandas dataframe is created with a datetime64[ns, UTC] column, pygwalker converts the time to the local time zone. How do I keep this from happening?

sql = '''select now() time, 1 count'''
cursor = conn.cursor()
cursor.execute(sql)
df = pd.DataFrame(cursor.fetchall())
print(df)

                              time  count
0 2024-01-16 13:09:34.944647+00:00      1

df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 1 entries, 0 to 0
Data columns (total 2 columns):
 #   Column  Non-Null Count  Dtype              
---  ------  --------------  -----              
 0   time    1 non-null      datetime64[ns, UTC]
 1   count   1 non-null      int64              
dtypes: datetime64[ns, UTC](1), int64(1)
memory usage: 148.0 bytes

walker=pyg.walk(df)

Thanks, Joe

[longxiaofei]

Hi @rzzldzzl , There is currently no way to fix it, unless you add a time zone offset to the original time field(this is obviously unreasonable).

Would it be better if users could configure the time zone offset themselves in the UI?

This feature will be launched this week.

[rzzldzzl]

@longxiaofei -

Would it be better if users could configure the time zone offset themselves in the UI?

Yes!

This feature will be launched this week.

Looking forward to it!

Thanks!

[longxiaofei]

A pre-release version of pygwalker was released: pip install pygwalker==0.4.3a0.

It includes the feature to modify time zones, but it may be some flaws(only works on the visualization tab).

details refer it: https://github.com/Kanaries/graphic-walker/pull/297.

[rzzldzzl]

Nice! Works as expected. Thanks!