Hi,

I noticed you started to file bugs for non-working debcheckouts. Was
this discussed anywhere as suggested by the developer's reference[1]?

The ones I saw just asked to drop "+ssh" from Vcs-* fields which I
imagine a new lintian check could also achieve.

Regards,
Ansgar

[1] <http://www.debian.org/doc/manuals/developers-reference/beyond-pkging.html#submit-many-bugs>


--
To UNSUBSCRIBE, email to debian-devel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Archive: 871undw15g.fsf@deep-thought.43-1.org">http://lists.debian.org/871undw15g.fsf@deep-thought.43-1.org

Ansgar Burchardt <ansgar@debian.org> writes:

> The ones I saw just asked to drop "+ssh" from Vcs-* fields which I
> imagine a new lintian check could also achieve.

Merely dropping the “+ssh� is unlikely to result in a working URL, at
least for many such repositories. Often, the publicly-accessible URL is
quite different from the SSH-accessible URL.

This is because the public-accessible URL is made available relative to
the root of a (usually) web service; whereas the SSH-accessible URL is
relative to the filesystem root directory, or the home directory of
whoever logs in. Switching from one to the other needs to consider this
difference.

So I would suggest that the reports you file should not talk
particularly about “+ssh�; they should instead point out that the URLs
are not currently accessible by the public, and request that the
public-accessible repository URL be specified.

--
“The Vatican is not a state.… a state must have people. There |
` are no Vaticanians.… No-one gets born in the Vatican except by |
_o__) an unfortunate accident.� —Geoffrey Robertson, 2010-09-18 |
Ben Finney


--
To UNSUBSCRIBE, email to debian-devel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Archive: 87vckobrlm.fsf@benfinney.id.au">http://lists.debian.org/87vckobrlm.fsf@benfinney.id.au

On 2012-04-24 18:22, Ansgar Burchardt wrote:
|
| I noticed you started to file bugs for non-working debcheckouts. Was
| this discussed anywhere as suggested by the developer's reference[1]?

Hi Ansgar,

There are only handful of packages that mistakenly have their Vcs-*
headers set up incorrectly so a minor bug reports and corrective
instructions were sent. In this regard I dind't consider it worthy of
"mass bug filing" process although it may have been in place.

| imagine a new lintian check could also achieve.

Correct. The problem is that Lintian's returned severity level does
not adequately correspond to the problem level this error is
causing. Many package maintainers don't even see the Lintian warnings
if they don't crank up the reporting levels. I've asked Lintian
maintainer to reconsider the severity of the Vcs header check.

This particular case directly affects usability because every
debcheckout command fails for all non-members that don't have
development access to Vcs repositories.

A minor bug report serves as a reminder to make maintainers aware of
the problem and to consider including a fix in future uploads.

Jari

| [1] <http://www.debian.org/doc/manuals/developers-reference/beyond-pkging.html#submit-many-bugs>


--
To UNSUBSCRIBE, email to debian-devel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Archive: 20120511055436.GI13646@taiko.cante.net">http://lists.debian.org/20120511055436.GI13646@taiko.cante.net