FS#2538 - Roll back GTK

Attached to Project: Arch Linux
Opened by Eugenia Loli-Queru (Eugenia) - Sunday, 10 April 2005, 19:40 GMT
Task Type Bug Report
Category Packages: Current
Status Closed
Assigned To No-one
Architecture not specified
Severity Critical
Priority Normal
Reported Version 0.7 Wombat
Due in Version Undecided
Due Date Undecided
Percent Complete 0%
Votes 0
Private No

Details

Come on guys, how could you release GTK+ without first placing it under -Testing and testing it? People are already buzzing about its bugginess: http://www.gnomefiles.org/comment.php?soft_id=370#2500

Please, either patch it and re-distribute it, or roll back the old version. This is not the first time the gtk team has released really buggy (and easy to reproduce bugs) versions.
This task depends upon

Closed by  dorphell (dorphell)
Monday, 11 April 2005, 05:19 GMT
Reason for closing:  Fixed
Comment by Yi Qiang (yi) - Sunday, 10 April 2005, 20:52 GMT
As usual from Eugenia, lots of complaining with nothing constructive to add.

here is the bug on gnome's bugzilla + proposed (and accepted patch)
http://bugzilla.gnome.org/show_bug.cgi?id=169870
patch-->http://bugzilla.gnome.org/attachment.cgi?id=39871&action=view

I would propose that we include this patch, unless gtk rolls out a new tar ball very soon.
Comment by Eugenia Loli-Queru (Eugenia) - Sunday, 10 April 2005, 21:12 GMT
> As usual from Eugenia, lots of complaining with nothing constructive to add.

It is not my job to be "constructive" and hunt down bugs on gnome's bugzilla. My job is to report the bug I found. I can do debugging or extra research if the bug is not reproducible or if I am specifically asked by the maintainer, but this is not the case here. That bug should have been caught not only by Red Hat (Owen has the biggest share of responsibility here for releasing yet another version that doesn't fully work) but by the Arch Linux maintainer as well.

You feel angry because of my tone, but my tone is fully justifiable: gtk got into -current WITHOUT TESTING. This *IS* one thing that deservse "lots of complaining" and be assured you will get some of that from me when deserved.
Comment by Vardyr (Vardyr) - Sunday, 10 April 2005, 22:26 GMT
Many packages don't go through testing when by all means they should. I can never get an explanation beyond "we didn't think it would be a problem."

This happens far too often, so I'm debating creating a script that won't install any updates younger than two days, or something of that sort.
Comment by dorphell (dorphell) - Monday, 11 April 2005, 05:19 GMT
This has been fixed no more than 1 hour after this was opened, no need for the open flame.

Loading...