Client removes columns from output if they are not aliased about vertica-python HOT 4 CLOSED

Your second code snippet returns [OrderedDict([('a', 'hello'), ('b', 33)])], so your first code snippet should return something like [OrderedDict([('?column?', 'hello'), ('?column?', 33)])]. However, those two columns have the same name '?column?' (i.e. key of the OrderedDict) and there is no schema name or table name to differentiate them, so only one value for that key.

[OrderedDict([('?column?', 'hello')]), OrderedDict([('?column?', 33)])] means you have one column (named '?column?') with two rows.

from vertica-python.

correct.
Do u think the client Can detect it and return something like:
[OrderedDict([('?column0?', 'hello')]), OrderedDict([('?column1?', 33)])]

from vertica-python.

I don't think that kind of change make sense. What's your use cases? Is there any blocker to use cursor_type="list" (a list of cell values) and Cursor.description (retrieve each column name)?

from vertica-python.

Actually, it's perfect! Did not think about it, Thanks :)

from vertica-python.