Confusing error message when SQLAlchemy is not installed

[fredrikw]

Trying to use read_sql_table on an instance without SQLAlchemy installed gives a confusing error message:

NotImplementedError: read_sql_table only supported for SQLAlchemy connectable.

The issue is that _is_sqlalchemy_connectable returns false if SQLAlchemy is not installed. My suggestion is to raise an error stating that SQLAlchemy is required by trying to import it directly in read_sql_table or extending the error message to say that SQLAlchemy is required. I don't know the internals and what might go wrong, but one way would be to not catch the ImportError...

pandas/pandas/io/sql.py

Lines 308 to 311 in 49c7791

	if not _is_sqlalchemy_connectable(con):
	raise NotImplementedError(
	"read_sql_table only supported for SQLAlchemy connectable."
	)

[phofl]

Could you provide something reproducable? If you put a non alchemy connectable in, the Message makes sense...

[fredrikw]

I understand that the message makes sense in that case, but IMHO not in this case.

I prepapred a Dockerfile in the following gist: https://gist.github.com/fredrikw/930e388da9c6f37c2b24544b77a6ba89

Building and running that will give the NotImplementedError.

File "/usr/src/app/testscript.py", line 3, in <module>
print(pd.read_sql_table("test", saconnstring))
File "/usr/local/lib/python3.9/site-packages/pandas/io/sql.py", line 272, in read_sql_table
raise NotImplementedError(
NotImplementedError: read_sql_table only supported for SQLAlchemy connectable.

Adding sqlalchemy to the install line will instead give the expected error that table test doesn't exist.

File "/usr/local/lib/python3.9/site-packages/pandas/io/sql.py", line 280, in read_sql_table
meta.reflect(only=[table_name], views=True)
File "/usr/local/lib/python3.9/site-packages/sqlalchemy/sql/schema.py", line 4703, in reflect
raise exc.InvalidRequestError(
sqlalchemy.exc.InvalidRequestError: Could not reflect: requested table(s) not available in Engine(sqlite:///test.db): (test)

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File "/usr/src/app/testscript.py", line 3, in <module>
print(pd.read_sql_table("test", saconnstring))
File "/usr/local/lib/python3.9/site-packages/pandas/io/sql.py", line 282, in read_sql_table
raise ValueError(f"Table {table_name} not found") from err
ValueError: Table test not found