judahrand commented on issue #2191:
URL: https://github.com/apache/arrow-adbc/issues/2191#issuecomment-2377240249
> Should have done this a moment ago, but I did just check on `main`:
>
> ```
> >>> from adbc_driver_postgresql import dbapi
> >>> postgres =
dbapi.connect("postgresql://localhost:5432/postgres?user=postgres&password=password")
> >>> with postgres.cursor() as cur:
> ... cur.execute('SELECT $1', ([1, 2, 3],))
> ... cur.fetch_arrow_table()
> ...
> pyarrow.Table
> ?column?: list<item: int64>
> child 0, item: int64
> ----
> ?column?: [[[1,2,3]]]
> ```
This should almost sort out my usecase and is precisely the expected
behavior! 🥳 Perhaps worth a separate issue for
```python
>>> from adbc_driver_postgresql import dbapi
>>> postgres =
dbapi.connect("postgresql://localhost:5432/postgres?user=postgres&password=password")
>>> with postgres.cursor() as cur:
... cur.execute('SELECT $1', (pyarrow.array([1, 2, 3]),))
... cur.fetch_arrow_table()
...
pyarrow.Table
?column?: list<item: int64>
child 0, item: int64
----
?column?: [[[1,2,3]]]
```
working because I can see an argument for that being incorrect behavior
since a RecordBatch or Table is arguably just a sequence of Arrays so maybe odd
to have two different behaviors for passing `table` and `tuple(table.columns)`?
🤷♂️
--
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: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
