pg.ddx.io pgsql-bugs@postgresql.org mailing list archive
help / color / mirror / Atom feedBUG #17634: Inconsistent view_definition in information_schema.views
3+ messages / 3 participants
[nested] [flat]
* BUG #17634: Inconsistent view_definition in information_schema.views
@ 2022-10-11 13:32 PG Bug reporting form <noreply@postgresql.org>
0 siblings, 1 reply; 3+ messages in thread
From: PG Bug reporting form @ 2022-10-11 13:32 UTC (permalink / raw)
To: pgsql-bugs@lists.postgresql.org; +Cc: kimber7@illinois.edu
The following bug has been logged on the website:
Bug reference: 17634
Logged by: Kimberly Blum
Email address: kimber7@illinois.edu
PostgreSQL version: 14.5
Operating system: RHEL 8
Description:
I use information_schema.views (the view_definition column) in code that
provides dependency information for functions and procedures. All objects in
my code are schema qualified, and when I look for dependencies, I rely on
that qualification.
If I run my query against information_schema.views while my session is
active in a schema that is referenced in my query, information_schema.views
strips out that schema name. All other objects remain schema qualified.
My example query:
SELECT view_definition FROM information_schema.views WHERE table_schema =
'appcode' AND table_name = 'cfg_msg_v';
The results from first query, which was run from schema api (which is not
referenced in the code), and all schema qualification shows up. Note all the
areas where appcfg.<obj> is shown.
query: select view_definition from information_schema.views where
SELECT cfg_msg.id_msg,
cfg_msg.id_src,
cfg_src.src_name,
cfg_msg.msg_type,
cfg_msg.msg_version,
cfg_msg.id_class,
cfg_cls class_name,
cfg_msg.descript
FROM ((appcfg.cfg_msg cfg_msg
JOIN appcfg.cfg_data_src cfg_src ON (((cfg_msg.id_src)::integer =
(cfg_src.id_src)::integer)))
JOIN cmdb.cfg_class cfg_cls ON (((cfg_msg.id_class)::integer =
(cfg_cls.id_class)::integer)));
If I switch my session to the appcfg schema, and run the same query, this is
the result (appcfg qualification disappears, other schema qualification is
fine).
SELECT cfg_msg.id_msg,
cfg_msg.id_src,
cfg_src.src_name,
cfg_msg.msg_type,
cfg_msg.msg_version,
cfg_msg.id_class,
cfg_cls.class_name,
cfg_msg.descript
FROM ((cfg_msg cfg_msg
JOIN cfg_data_src cfg_src ON (((cfg_msg.id_src)::integer =
(cfg_src.id_src)::integer)))
JOIN cmdb.cfg_class cfg_cls ON (((cfg_msg.id_class)::integer =
(cfg_cls.id_class)::integer)));
^ permalink raw reply [nested|flat] 3+ messages in thread
* Re: BUG #17634: Inconsistent view_definition in information_schema.views
@ 2022-10-11 14:58 Tom Lane <tgl@sss.pgh.pa.us>
parent: PG Bug reporting form <noreply@postgresql.org>
0 siblings, 1 reply; 3+ messages in thread
From: Tom Lane @ 2022-10-11 14:58 UTC (permalink / raw)
To: kimber7@illinois.edu; +Cc: pgsql-bugs@lists.postgresql.org
PG Bug reporting form <noreply@postgresql.org> writes:
> If I run my query against information_schema.views while my session is
> active in a schema that is referenced in my query, information_schema.views
> strips out that schema name. All other objects remain schema qualified.
This is not a bug, it's normal behavior of the rule decompiler:
object names are qualified only if necessary (where "necessary"
means "would not be found in a search_path search").
If you don't like it, set search_path to empty.
regards, tom lane
^ permalink raw reply [nested|flat] 3+ messages in thread
* RE: BUG #17634: Inconsistent view_definition in information_schema.views
@ 2022-10-11 15:57 Blum, Kimber <kimber7@illinois.edu>
parent: Tom Lane <tgl@sss.pgh.pa.us>
0 siblings, 0 replies; 3+ messages in thread
From: Blum, Kimber @ 2022-10-11 15:57 UTC (permalink / raw)
To: Tom Lane <tgl@sss.pgh.pa.us>; +Cc: pgsql-bugs@lists.postgresql.org <pgsql-bugs@lists.postgresql.org>
Then perhaps the rule decompiler should be used when populating a column that implies it contains the view definition - because it is not the correct definition. No snark intended, just my thoughts. On a related note, having dependencies for all object (including functions and procedures) available in postgres would be awesome. You all provide a great product for us opensource users, and it's appreciated.
-----Original Message-----
From: Tom Lane <tgl@sss.pgh.pa.us>
Sent: Tuesday, October 11, 2022 9:58 AM
To: Blum, Kimber <kimber7@illinois.edu>
Cc: pgsql-bugs@lists.postgresql.org
Subject: Re: BUG #17634: Inconsistent view_definition in information_schema.views
PG Bug reporting form <noreply@postgresql.org> writes:
> If I run my query against information_schema.views while my session is
> active in a schema that is referenced in my query,
> information_schema.views strips out that schema name. All other objects remain schema qualified.
This is not a bug, it's normal behavior of the rule decompiler:
object names are qualified only if necessary (where "necessary"
means "would not be found in a search_path search").
If you don't like it, set search_path to empty.
regards, tom lane
^ permalink raw reply [nested|flat] 3+ messages in thread
end of thread, other threads:[~2022-10-11 15:57 UTC | newest]
Thread overview: 3+ messages (download: mbox mbox.gz follow: Atom feed)
-- links below jump to the message on this page --
2022-10-11 13:32 BUG #17634: Inconsistent view_definition in information_schema.views PG Bug reporting form <noreply@postgresql.org>
2022-10-11 14:58 ` Tom Lane <tgl@sss.pgh.pa.us>
2022-10-11 15:57 ` Blum, Kimber <kimber7@illinois.edu>
This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox