539120 - Dragging a Tab onto the Home Button does nothing

Status: RESOLVED FIXED

(Firefox :: General, defect)

Description

[Drew Ruderman]

User Agent: Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US; rv:1.9.2) Gecko/20100105 Firefox/3.6 (.NET CLR 3.5.30729)

Dragging a tab to the home button does not prompt the user if he/she wants to set it as the homepage.  Dragging bookmarks or website icons, however, does prompt the user regarding the homepage.  For consistency, dragging a tab should also prompt the user regarding the homepage.

Comment 1

[Ria Klaassen (not reading all bugmail)]

Confirmed with Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US; rv:1.9.3a1pre) Gecko/20100111 Minefield/3.7a1pre

Regression range is:
http://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=c4e00b34bbdd&tochange=a8daae5ef5ce

Comment 2

[Mike Beltzner [:beltzner, not reading bugmail]]

I'm guessing that this is to do with the drag'n'drop pieces.

Enn?

Comment 3

[Jesse Ruderman]

Mano fixed this the last two times it happened (bug 491530, bug 465910), so he might be interested in helping to fix it again.

One of those bugs had been blocking-firefox3.5+, fwiw.

Comment 4

[Mike Beltzner [:beltzner, not reading bugmail]]

One wonders how the hell we don't have a test for this, if it keeps breaking.

Mano's kind of out of pocket. Copying his past reviewers for similar patches in bug 487263 and bug 471499. Can one of you guys take a look at this? I think we might have to suck up the functional regression for release day, but would want a patch ready in the wings for an RC ridealong if that happens.

Comment 5

[Marco Bonardo [:mak]]

(In reply to comment #4)
> One wonders how the hell we don't have a test for this, if it keeps breaking.

d&d tests are hard, complete d&d tests are practically impossible afaik. we can just simlulate dragstart and drop, not the full interaction.

the only thing that jumps to my eyes is that browserDragAndDrop.dragOver is missing "text/x-moz-text-internal" flavor, but i don't know if that's all offhand.

Comment 6

[Marco Bonardo [:mak]]

Attachment: Patch v1.0

yeah, was just not marked as accepting the drop

now, i could try to make a test with synthesizeDrop, but i'm not sure it will catch these kind of things, indeed i think it won't be really useful.

Comment 7

[Marco Bonardo [:mak]]

http://hg.mozilla.org/mozilla-central/rev/973dab910358

Comment 8

[Mike Shaver (:shaver -- probably not reading bugmail closely)]

(In reply to comment #5)
> (In reply to comment #4)
> > One wonders how the hell we don't have a test for this, if it keeps breaking.
> 
> d&d tests are hard, complete d&d tests are practically impossible afaik. we can
> just simlulate dragstart and drop, not the full interaction.

There are ways to synthesize drag-and-drop native events on every platform we support, I'm pretty sure.  If we need to extend our test or event tools to support simulation of more parts of D&D in our own toolkit, though, then why don't we do that?  Given that we've now fixed this bug for the 3rd time, and only narrowly avoided regressing it in 3.6, it would seem a wise investment for the people who are doing work with app-level D&D!

Comment 9

[Mike Beltzner [:beltzner, not reading bugmail]]

Comment on attachment 421629 [details] [diff] [review]
Patch v1.0

a192=beltzner, please land on mozilla-1.9.2 and GECKO192_20100105_RELBRANCH

Comment 10

[:Gavin Sharp [email: [e-mail]]]

Landed on 1.9.2 and the 1.9.2 GECKO192_20100105_RELBRANCH for RC2:
https://hg.mozilla.org/releases/mozilla-1.9.2/rev/b8bd375179cd
https://hg.mozilla.org/releases/mozilla-1.9.2/rev/7f62239d8417

Comment 11

[juan becerra [:juanb]]

Verified on xp, mac, and linux using the RC2.