Open transparent1998 opened 1 month ago
Correct, s3_url_style
is not currently settable.
As discussed a bit in #197, we might want to move towards a more flexible way of supporting all the various options. For instance it would also be nice to support the scope
option… and there's many others.
I guess this will be fixed by #151, once we implement it.
postgres=# select duckdb.install_extension('httpfs'); install_extension
t (1 row)
postgres=# select from read_csv('s3://jyt/test.csv') as t (a int,b int); WARNING: (PGDuckDB/CreatePlan) Prepared query returned an error: 'HTTP Error: HTTP GET error on 'https://jyt.s3.amazonaws.com/test.csv' (HTTP 403) WARNING: (PGDuckDB/CreatePlan) Prepared query returned an error: 'Conversion Error: Type VARCHAR with value '{}' can't be cast to the destination type LIST LINE 1: ...est.csv'::text, false, true, true, '{}'::text[], 'auto'::character varying, ''... ^ ERROR: Function
read_csv(TEXT)
only works with Duckdb execution. CONTEXT: PL/pgSQL function read_csv(text,boolean,boolean,boolean,text[],character varying,character varying,character varying,character varying,character varying,boolean,text[],boolean,boolean,boolean,bigint,text[],character varying,boolean,boolean,text[],boolean,character varying,bigint,character varying,bigint,character varying,text[],boolean) line 3 at RAISE postgres=# select from read_csv('/tmp/a.csv') as t (a text); a1 (1 row)
It can be accessed normally in duckdb