Closed (fixed)
Project:
Drupal core
Version:
7.x-dev
Component:
update.module
Priority:
Normal
Category:
Bug report
Assigned:
Issue tags:
Reporter:
Created:
6 Apr 2011 at 16:54 UTC
Updated:
3 Jan 2014 at 02:58 UTC
Jump to comment: Most recent, Most recent file

Comments
Comment #1
damien tournoud commentedIt's not really pretty, but there is nothing better we can actually do :(
Comment #2
bfroehle commentedI'm marking this as a duplicate of #1024990: "Manual updates required" table has an extra column, however you solve the problem in a different way. I'll like that other issue to this one.
Comment #3
bfroehle commentedThe patch in #1 is likely the simplest fix to this problem. I've tested it out and it works wonders. :) I marked the other issue, referred to in #2, as a duplicate of this one.
Comment #4
dries commentedIs the following summary correct?
We have a function that builds up an array that is either passed into theme('tableselect') or theme('table').
The problem is that theme('tableselect') and theme('table') take a different array as input.
This hack turns a theme('tableselect') array into theme('table') array.
If it is correct, I think we can better document it in the code. Setting back to 'needs review' as I'd like some extra input.
Comment #5
keichee commentedthis is a much simpler way that works..
Comment #6
catchPatch looks good, but this isn't critical, and the patch doesn't look like it needs porting to me, ought to apply straight.
Comment #7
catchTested this manually, fixes the problem fine. I think we could still use a comment here though - the code is much simpler but we should explain why we're unsetting attributes.
Comment #8
keichee commentedthats per entry, so core entry attributes have no attributes set, others will have.
thats a simple issue that no one will reuse and free some space.
Comment #9
bfroehle commentedDries' hypothesis in #4 is correct.
Comment #10
damien tournoud commented#5 really looks like a hack.
Comment #11
catch@Damien, without inline docs, both #1 and #5 look like hacks, except #5 is a lot more concise. So if #1 is the correct fix here and #5 is a nasty workaround, the docs should make that obvious. Marking CNW since neither are adding those.
Comment #12
bfroehle commentedJust needs a comment like
Comment #13
keichee commentedlook, i just followed the existing code to handle core
existing code is
but after i traced, weight is not even existing, i think weight is a typo and it must be an attributes.
try to print_r the $entry variable outside that if condition code that handle manual updates and see what i mean.
my code is legit as per complying to the existing code how to handle manual updates.
Comment #14
bfroehle commentedThe patch in #1 with a comment similar to #12 should suffice.
Comment #15
bfroehle commentedMarked #1200834: Ugly border on manual updates table on available updates report as a duplicate.
Comment #16
bfroehle commentedI've updated the comment patch in #1 to address Dries' concern in #4.
#5 is a hack in the sense that it unnecessarily removes the attributes (i.e., 'class' => 'update-recommended') which may be used by the site's CSS.
Comment #17
bfroehle commentedMarking this as RTBC since it is the same code as in #1 (which was marked RTBC in #3) and hopefully addresses Dries concerns with the comments.
Comment #18
catchThe new comment covers this well I think.
Committed to 8.x, moving back to 7.x for webchick to look at.
Comment #19
webchickw00t! Great to see this fixed!
Committed and pushed to 7.x.