[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: 'send-pr --database' is broken in 4.0 beta

From: Lars Henriksen
Subject: Re: 'send-pr --database' is broken in 4.0 beta
Date: Thu, 16 May 2002 16:57:39 +0200
User-agent: Mutt/1.3.25i

On Tue, May 14, 2002 at 03:24:14PM -0700, Mel Hatzis wrote:

>  1. invoking send-pr --database=xx doesn't work (a space is
>     required between the '--database' and the 'xx'. Given
>     the usage line, this is very misleading.

There is some inconsistency in the way the GNATS user tools
send-pr, query-pr and edit-pr accept their arguments. Or rather
in their usage messages: they all expect a space, but only
send-pr says it wants an equal sign. So presumably the send-pr
usage message should be corrected.

The GNATS manual ("Keeping Track") section on send-pr (2.2.3
Invoking send-pr from the shell) has not yet been updated and
still documents version 3.113.

>  2. the '--database' option in send-pr simply sets the GNATSDB
>     environment variable to whatever value is specified. This
>     is inconsistent with the use of the --database option in
>     all the other gnats programs and furthermore, won't work
>     at ends up pointing to the 'default' database.
> Regarding #2, I've previously mailed a few messages to the
> group with the subject 'use of GNATSDB'....I'd really like
> to see this working and will gladly submit a patch if
> people are interested in my proposed solution - to use
> a GNATSID environment variable in addition to GNATSDB.

I've read your messages, but have not been able to reproduce the
behaviour you describe. I'll return to that in another mail.

> GNATSID would be set to database names that are defined in
> the databases file (e.g. GNATSID=otherdb) and GNATSDB will
> continue to work as documented
> (e.g GNATSDB=host:port:name:user:passwd)

I don't agree. GNATSDB is used in many places and in some contexts
only the local database name is meaningful, e.g. gnatsd and gen-index.
Sure, GNATSDB has two syntactically and semantically different
interpretations, but so have the lines in the databases file.
Of course, bugs should be corrected, but you haven't made me
understand why two environment variables are needed for that

Lars Henriksen

reply via email to

[Prev in Thread] Current Thread [Next in Thread]