User guide is incorrect regarding using CLI to register CSV files using schema inference

[andygrove]

Describe the bug

The user guide page at https://arrow.apache.org/datafusion/cli/index.html states that "It is necessary to provide schema information for CSV files since DataFusion does not automatically infer the schema when using SQL to query CSV files." but this is not true, as demonstrated below:

DataFusion CLI v10.0.0
❯ create external table a stored as csv with header row location '/tmp/a.csv';
0 rows in set. Query took 0.017 seconds.
❯ select * from a;
+---+---+---+---+
| a | b | c | d |
+---+---+---+---+
| 1 | 2 | 3 | 4 |
+---+---+---+---+
1 row in set. Query took 0.011 seconds.
❯ describe a;
+-------------+-----------+-------------+
| column_name | data_type | is_nullable |
+-------------+-----------+-------------+
| a           | Int64     | NO          |
| b           | Int64     | NO          |
| c           | Int64     | NO          |
| d           | Int64     | NO          |
+-------------+-----------+-------------+
4 rows in set. Query took 0.017 seconds.

To Reproduce
See above.

Expected behavior
We should update the user guide to state that specifying a schema is optional.

Additional context
None

[kmitchener]

I have a PR coming that will fix this and a bunch of other issues with the docs

[retikulum]

Hi. I think this issue is fixed (I guess on #3171) but not understand why it is still open? https://arrow.apache.org/datafusion/user-guide/sql/ddl.html

[retikulum]

Did you have a chance to take a look? @andygrove

[Dandandan]

Thanks @retikulum for noticing