Re: [HACKERS] Win32 WEXITSTATUS too - Mailing list pgsql-patches

From Bruce Momjian
Subject Re: [HACKERS] Win32 WEXITSTATUS too
Date
Msg-id 200701230137.l0N1bsF23952@momjian.us
Whole thread Raw
In response to Re: [HACKERS] Win32 WEXITSTATUS too  ("Takayuki Tsunakawa" <tsunakawa.takay@jp.fujitsu.com>)
List pgsql-patches
Takayuki Tsunakawa wrote:
> From: "Bruce Momjian" <bruce@momjian.us>
> > Tom Lane wrote:
> >> Basically this whole idea is misconceived.  Just print the number
> and
> >> have done.
> >
> > And how do people interpret that number?
>
> I understand that "people" Bruce-san is saying means PostgreSQL
> developers, not ordinary users.
> When ordinary users encounter an Win32 exception, what they can do is
> to report the message and the description of phenomenon to PostgreSQL
> developers.  What can they do when they see the descriptive text of an
> exception code, such as "Access violation" for 0xC0000005?  An
> exception means a bug of PostgreSQL.  Win32 exceptions are different
> from errno on UNIX (counterpart of which is Win32 error code.)  Can
> they avoid the exception by changing PostgreSQL settings?  If luckly
> so in one case, can they think of the treatment from the message?
>
> If "people" means PostgreSQL developers, the descriptive text is not
> necessary either.  I think the developers who try to solve the bug
> know where to refer to interpret the exception code.
>
> My opinion is the same as Tom-san's.

Well, I am thinking of cases where we the error can help the user
diagnose the problem.  I have found a way to print descriptions with
FormatMessage(), and the codes without descriptions will just print as
hex.

--
  Bruce Momjian   bruce@momjian.us
  EnterpriseDB    http://www.enterprisedb.com

  + If your life is a hard drive, Christ can be your backup. +

pgsql-patches by date:

Previous
From: "Takayuki Tsunakawa"
Date:
Subject: Re: [HACKERS] Win32 WEXITSTATUS too
Next
From: Bruce Momjian
Date:
Subject: Re: [HACKERS] Win32 WEXITSTATUS too