Re: Slightly inconsistent behaviour in regproc? - Mailing list pgsql-hackers

From Christopher Kings-Lynne
Subject Re: Slightly inconsistent behaviour in regproc?
Date
Msg-id 3F9E0BE8.20603@familyhealth.com.au
Whole thread Raw
In response to Re: Slightly inconsistent behaviour in regproc?  (Tom Lane <tgl@sss.pgh.pa.us>)
Responses Re: Slightly inconsistent behaviour in regproc?
List pgsql-hackers
>>Only regproc adds the unnecessary pg_catalog. qualification, why is that?
> 
> 
> Er, I couldn't see the part of your example where that happened?

Basically, my question is why ::regproc alone always addes the catalogue 
qualification in this case?

Rows below correspond to:

::regtype
::regtype
::regprocedure
::regproc

-[ RECORD 1 ]----------------------------------------
castsource | "char"
casttarget | text
castfunc   | text("char")
castfunc2  | pg_catalog.text

eg. Why is it not:

-[ RECORD 1 ]----------------------------------------
castsource | "char"
casttarget | text
castfunc   | text("char")
castfunc2  | text

Or even:

-[ RECORD 1 ]----------------------------------------
castsource | pg_catalog."char"
casttarget | pg_catalog.text
castfunc   | pg_catalog.text("char")
castfunc2  | pg_catalog.text

Chris




pgsql-hackers by date:

Previous
From: Bruce Momjian
Date:
Subject: Re: Horology failures
Next
From: "Andrew Dunstan"
Date:
Subject: limiting parallelism in "make check"