Skip Menu |

This queue is for tickets about the Glib CPAN distribution.

Report information
The Basics
Id: 76422
Status: resolved
Priority: 0/
Queue: Glib

People
Owner: XAOC [...] cpan.org
Requestors: 'spro^^*%*^6ut# [...] &$%*c
Cc:
AdminCc:

Bug Information
Severity: (no value)
Broken in: (no value)
Fixed in: (no value)



Subject: Indexing problem
This may be related to #75948. There appears to be an indexing problem. modules/02packages.details.txt.gz on CPAN lists 1.242 as the current version of Glib, but I see there is already a XAOC/Glib-1.250.tar.gz on CPAN. Do you need to re-index XAOC/Glib-1.250.tar.gz in the PAUSE interface?
Subject: Re: [rt.cpan.org #76422] Indexing problem
Date: Mon, 09 Apr 2012 02:37:48 +0200
To: bug-Glib [...] rt.cpan.org, Brian Manning <elspicyjack [...] gmail.com>
From: Torsten Schoenfeld <kaffeetisch [...] gmx.de>
On 09.04.2012 02:25, Father Chrysostomos via RT wrote: Show quoted text
> There appears to be an indexing problem. modules/02packages.details.txt.gz on CPAN lists > 1.242 as the current version of Glib, but I see there is already a XAOC/Glib-1.250.tar.gz on > CPAN.
1.250 is what we call an "unstable" release and was not actually meant to appear on CPAN. We don't want it to be installed automatically if someone tells their CPAN client to "install Glib". We use the underscore mechanism now to handle this, see <https://metacpan.org/module/XAOC/Glib-1.252_001/lib/Glib.pm> for the latest unstable release. Brian, maybe it would make sense to delete 1.250 and 1.251.TRIAL from PAUSE to avoid confusion. What do you think?
On Sun Apr 08 20:37:57 2012, TSCH wrote: Show quoted text
> On 09.04.2012 02:25, Father Chrysostomos via RT wrote:
> > There appears to be an indexing problem.
> modules/02packages.details.txt.gz on CPAN lists
> > 1.242 as the current version of Glib, but I see there is already a
> XAOC/Glib-1.250.tar.gz on
> > CPAN.
> > 1.250 is what we call an "unstable" release and was not actually meant > to appear on CPAN. We don't want it to be installed automatically if > someone tells their CPAN client to "install Glib". We use the > underscore mechanism now to handle this, see > <https://metacpan.org/module/XAOC/Glib-1.252_001/lib/Glib.pm> for the > latest unstable release. > > Brian, maybe it would make sense to delete 1.250 and 1.251.TRIAL from > PAUSE to avoid confusion. What do you think?
Delete and then reupload with the 001 tails? In any case, let me make an announcement to the list, and then I'll do it tomorrow afternoon my time (0000UTC 10Apr2012).
Subject: Re: [rt.cpan.org #76422] Indexing problem
Date: Mon, 09 Apr 2012 11:49:27 +0200
To: bug-Glib [...] rt.cpan.org
From: Torsten Schoenfeld <kaffeetisch [...] gmx.de>
On 09.04.2012 06:38, Brian Manning via RT wrote: Show quoted text
>> Brian, maybe it would make sense to delete 1.250 and 1.251.TRIAL from >> PAUSE to avoid confusion. What do you think?
> > Delete and then reupload with the 001 tails? In any case, let me make an announcement to the > list, and then I'll do it tomorrow afternoon my time (0000UTC 10Apr2012).
I don't think re-uploading them is necessary, as 1.252_001 is found just fine: <https://metacpan.org/module/XAOC/Glib-1.252_001/lib/Glib.pm>. Similarly, I don't think triggering a reindex of 1.242 is necessary as it is found just fine too: <https://metacpan.org/module/Glib>. (This also holds when using CPAN.pm and CPANPLUS.pm, as far as I have tested.)
On Mon Apr 09 05:49:40 2012, TSCH wrote: Show quoted text
> I don't think re-uploading them is necessary, as 1.252_001 is found > just > fine: <https://metacpan.org/module/XAOC/Glib-1.252_001/lib/Glib.pm>. > Similarly, I don't think triggering a reindex of 1.242 is necessary as > it is found just fine too: <https://metacpan.org/module/Glib>. (This > also holds when using CPAN.pm and CPANPLUS.pm, as far as I have > tested.)
I posted to the mailing list how to obtain the two files that will be deleted via git tags, so I will not be reuploading them with 001 tails. I will be deleting the two files April 10th at 00:00UTC (about 8.5 hours from now) unless someone says something on the mailing list.
This issue should now be resolved.