timsaucer commented on issue #715:
URL:
https://github.com/apache/datafusion-python/issues/715#issuecomment-2131229316
My statement above about testing on rust side is likely incorrect. I ran the
same test above but loading the dataframe from a parquet file instead of
creating in memory and the expected behavior is reproduced.
If you amend these lines to the bottom of the minimal example
```
df.write_parquet("save_out.parquet")
df_reread = ctx.read_parquet("save_out.parquet")
df_reread.show()
df_reread.select(col("a")["outer_1"]["inner_2"]).show()
```
You get the expected result
```
DataFrame()
+-------------------------------------+
| a |
+-------------------------------------+
| {outer_1: {inner_1: 1, inner_2: 2}} |
| {outer_1: {inner_1: 1, inner_2: }} |
| {outer_1: } |
+-------------------------------------+
DataFrame()
+-----------------------------+
| ?table?.a[outer_1][inner_2] |
+-----------------------------+
| 2 |
| |
| |
+-----------------------------+
```
It also shows the original table is reproduced. I'll continue digging but I
no longer am convinced this is a python binding issue.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscr...@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscr...@datafusion.apache.org
For additional commands, e-mail: github-h...@datafusion.apache.org
