Skip Menu |

This queue is for tickets about the Array-Compare CPAN distribution.

Report information
The Basics
Id: 48735
Status: resolved
Priority: 0/
Queue: Array-Compare

People
Owner: DAVECROSS [...] cpan.org
Requestors: ddascalescu+perl [...] gmail.com
Cc:
AdminCc:

Bug Information
Severity: Important
Broken in: 1.17
Fixed in: 2.00



Subject: Bogus version number '33' in 1.17
I was playing with 'r' at the cpan prompt and saw this: (22:55:41) dandv: "Array::Compare 1.15 33 DAVECROSS/Array-Compare-1.17.tar.gz" - but http://search.cpan.org/dist/Array-Compare shows 2.0 as the latest. How reliable is this 'r' thing? (22:56:40) rafl: it's pretty good, unless the author or the indexer screws something up (22:56:53) rafl: the latter is very rare. the former happens every now and then (22:57:51) dandv: I checked that module's source. there is no '33' anywhere (22:58:35) hobbs: dandv: there is in the 1.17 revision of the module (22:58:45) hobbs: ($VERSION) = '$Id: Compare.pm 33 2008-12-18 11:34:24Z dave $' =~ /(\d+)/; (22:58:48) hobbs: (oops) (22:59:13) hobbs: so, 1.18 and 2.00 are newer and exist, but the indexer says "no! I saw a 33!" (22:59:39) rafl: heh. i was just about to paste that as well (22:59:46) rafl: yeah, that's where the author screwed up (23:00:12) rafl: the indexer finds version 33 in 1.17. 1.20 (the dist) has the same module with a lower version (23:00:22) hobbs: I think if he were to delete the 1.17 tarball, and then use the "reset version" button in PAUSE, it would fix it. I think that's what it's there for. :) (23:00:27) rafl: and 'r' asks for the latest version of a module (23:00:31) rafl: and that's what you'll get
On Sat Aug 15 02:05:45 2009, dandv wrote: Show quoted text
> I was playing with 'r' at the cpan prompt and saw this: > > (22:55:41) dandv: "Array::Compare 1.15 33 > DAVECROSS/Array-Compare-1.17.tar.gz" - but > http://search.cpan.org/dist/Array-Compare shows 2.0 as the latest. How > reliable is this 'r' thing? > (22:56:40) rafl: it's pretty good, unless the author or the indexer > screws something up > (22:56:53) rafl: the latter is very rare. the former happens every now > and then > (22:57:51) dandv: I checked that module's source. there is no '33'
anywhere Show quoted text
> (22:58:35) hobbs: dandv: there is in the 1.17 revision of the module > (22:58:45) hobbs: ($VERSION) = '$Id: Compare.pm 33 2008-12-18 11:34:24Z > dave $' =~ /(\d+)/; > (22:58:48) hobbs: (oops) > (22:59:13) hobbs: so, 1.18 and 2.00 are newer and exist, but the indexer > says "no! I saw a 33!" > (22:59:39) rafl: heh. i was just about to paste that as well > (22:59:46) rafl: yeah, that's where the author screwed up > (23:00:12) rafl: the indexer finds version 33 in 1.17. 1.20 (the dist) > has the same module with a lower version > (23:00:22) hobbs: I think if he were to delete the 1.17 tarball, and > then use the "reset version" button in PAUSE, it would fix it. I think > that's what it's there for. :) > (23:00:27) rafl: and 'r' asks for the latest version of a module > (23:00:31) rafl: and that's what you'll get
I'm not convinced that CPAN is doing the right thing here. To my mind it should only ever be looking at the distribution version, not the version numbers in the individual module files. I've recently switched from using svn to git for my modules and (at the same time) from using automatic (i.e. generated from svn) version numbers to manual ones (as git commit identifiers make really bad version numbers!) so I'm going to have this problem with all of my modules. But I'll follow your advice and use the "reset version" button to see if that fixes the problem. Cheers, Dave...