Re: Improving PL/Tcl's error context reports - Mailing list pgsql-hackers

From Pavel Stehule
Subject Re: Improving PL/Tcl's error context reports
Date
Msg-id CAFj8pRBxwHgfWWyJvMrVTiLOvxP88Knhj1NT=M-4rpSk95bK_Q@mail.gmail.com
Whole thread Raw
In response to Re: Improving PL/Tcl's error context reports  (Tom Lane <tgl@sss.pgh.pa.us>)
Responses Re: Improving PL/Tcl's error context reports
List pgsql-hackers
Hi

čt 4. 7. 2024 v 21:42 odesílatel Tom Lane <tgl@sss.pgh.pa.us> napsal:
I wrote:
> Pavel Stehule <pavel.stehule@gmail.com> writes:
>> PLpgSQL uses more often function signature
>> (2024-07-04 19:49:20) postgres=# select bx(0);
>> ERROR:  division by zero
>> CONTEXT:  PL/pgSQL function fx(integer) line 1 at RETURN
>> PL/pgSQL function bx(integer) line 1 at RETURN

> Oh that's a good idea!  So let's use format_procedure(), same as
> plpgsql does, to generate the final context line that currently
> reads like

> in PL/Tcl function "bogus"

> Then, we could apply the "pull out just alphanumerics" rule to
> the result of format_procedure() to generate the internal Tcl name.
> That should greatly reduce the number of cases where we have duplicate
> internal names we have to unique-ify.

Here's a v2 that does it like that.

I like it.

- patching and compilation without any issue
- check world passed

I'll mark this as ready for commit

Regards

Pavel


                        regards, tom lane

pgsql-hackers by date:

Previous
From: Frank Streitzig
Date:
Subject: XML test error on Arch Linux
Next
From: Alexander Lakhin
Date:
Subject: Re: Use generation memory context for tuplestore.c