GNOME Bugzilla – Bug 88585
The "up arrow" and "down arrow" displayed in the column headers are inconsistent.
Last modified: 2006-12-26 18:29:46 UTC
Tested in Solaris 2.9 64 bit, Sparc, Code taken on 15th July Description: Select view as list and click on the column headers to sort the items in ascending/descending order. Observation =========== File Name Up arrow:- when the items are displayed descending order from top to bottom Down arrow:- when the items are displayed ascending order from top to bottom Size Up arrow:- when the items are displayed ascending order from top to bottom Down arrow:- when the items are displayed descending order from top to bottom
Is this a general gtk treeview bug or specific to nautilus???
I guess it's a bug in the compare function nautilus uses to sort the size column.
Pretty minor. Definitely irritating, though. Tagging as easy fix as I'm betting it's just a matter of one <> confusion. Someone just needs to find which one it is ;)
*** Bug 86371 has been marked as a duplicate of this bug. ***
I think this problem is a bit better described in bug 86371; please see the comments there.
Created attachment 11997 [details] [review] Fixes the sorting of the size column
can you please email your patch to nautilus-list@gnome.org
Apologies for the spam- I'm removing 'bugsquad' from some keywords via the web interface. This is a one-time only thing before I re-add bugsquad via the SQL interface, which will generate no mail. Apologies again.
In regards to bug 86371, and the backwards sorting arrow behavior: This is a big problem, as the arrow is actually behaving in a counter-intuitive manner. I know there was a previous response to this, with an equally valid explanation (and defense) of the current behavior, but this arrow is used in other places in the world, and not only in the Gnome environment, or even computers. The arrow is not really an arrow, but a glyph of a triangle. The point of the triangle is the smallest value, while the base is the largest. Therefore, if the point faces up, the sort should always be in Ascending order (12345), and if it points down, the sort should always be in Descending (54321) order. Triangles are used to depict many things. - Least to greatest - Specific to general - Old to new (smallest numerical date to largest) All of the above orders are ALWAYS depicted as a triangle with its point facing upward. And their reverse is ALWAYS depicted as a triangle with its point facing downward. Therefore, the arrows are backwards, and should be changed to be consistant with the rest of the world.
Excellent point, John.
Comment on attachment 11997 [details] [review] Fixes the sorting of the size column I think we should fix the directory count to sort consistently with the other columns, but I'm reluctant to change the interpretation in a minor version. We'll commit this after the branch.
This patch does not apply to CVS anymore. Tomas, can you modify the patch so it applies ? Thanks !
Comment on attachment 11997 [details] [review] Fixes the sorting of the size column Since the patch doesn't apply anymore I'm changing the status from accepted to needs-work. Smack me if you think that's the wrong thing to do.
Created attachment 38723 [details] [review] Updated patch This is an updated patch for CVS.
I commited this. I also fixed the date sort order.
So, just to clarify, is the behaviour now the same as or opposite to that prescribed in the HIG? http://developer.gnome.org/projects/gup/hig/2.0/controls-lists.html#controls-lists-sortable While I appreciate that this the arrow direction issue seems to be a controversial one, I think the HIG team would prefer not to have active anti-HIG commits, and have bugs filed against the HIG instead for possible rectification.
I don't know the answer to your question; but the HIG is indeed in need of fixing on this point: bug 305277.
calum: The behaviour is now per the hig.
*** Bug 310182 has been marked as a duplicate of this bug. ***
Mass reassigning bugs with 2.2.0 milestone to 2.2.x milestone Grep for "Mass reassigning" to filter out this bug spam.