48424 - if a message is selected, Double Clicking on column header opens up a stand alone window

Status: VERIFIED FIXED

(SeaMonkey :: MailNews: Message Display, defect, P2)

Description

[Pratik Dhupia]

This happens only with Imap accounts. 
Using Build Id: 2000-08-09-08 M18  

Steps to Reproduce:
1) Open Mail
2) Double Click on a column header

Actual: Opens up a stand alone window
Expected: Should sort the column twice

Comment 1

[lchiang]

Can you do one click, pause, followed by another click to sort?

Comment 2

[Pratik Dhupia]

I hit this bug even when I clock pause for less than a second and click back
again. I found this bug when I was trying to sort by sender to find a message. 

Comment 3

[scottputterman]

mail triage marking nsbeta3+

Comment 4

[scottputterman]

fix checked in.

Comment 5

[Pratik Dhupia]

Using Build Id: 2000-08-21-08 M18 on Mac, Linux and NT4.
Verified: Fixed

Comment 6

[scottputterman]

My fix for this caused worse regressions so I had to back it out.  This bug will 
now exist again.

Comment 7

[lchiang]

second pass: - per mail triage

Comment 8

[Pratik Dhupia]

*** Bug 58306 has been marked as a duplicate of this bug. ***

Comment 9

[Jason Wong]

i take it that the lack of rtm means that this won't get ifxed for N6.  I know
that this isn't data loss, but it sure is annoying.

any chance?  whadddayasay?

Comment 10

[scottputterman]

sorry, this won't make it for NS6.

Comment 11

[scottputterman]

reassigning to sspitzer

Comment 12

[scottputterman]

*** Bug 59536 has been marked as a duplicate of this bug. ***

Comment 13

[scottputterman]

marking nsbeta1+ and moving to mozilla0.8

Comment 14

[lchiang]

note: this is a frequently seen problem according to DecisionOne (support)

Comment 15

[(not reading, please use [e-mail] instead)]

accepting and investigating...

Comment 16

[(not reading, please use [e-mail] instead)]

ok, here's what's going on:

in our threadPane.xul, we have an onclick handler for the treecells in the

<tree onclick="foo()">

<treehead>
	<treerow>
			<treecell onclick="bar()"/>
	</treerow>
</treehead>

</tree>

when the user clicks on the column, both foo() and bar() get called.
first bar(), then foo().

I would expect just bar() to be called.  perhaps hyatt can shed some light?

I've got a simple fix to tree's onclick handler to bail out if we double click
on a treehead's treecell.

I'll attach my fix.

Comment 17

[(not reading, please use [e-mail] instead)]

updating summary.  if a message wasn't selected, this would not open a a
stand-alone message window.  (we'd get a js error that prevented it.)

here comes my fix, seeking feedback from hyatt about the onclick handlers.

(and a sr= from bienvenu?)

Comment 18

[(not reading, please use [e-mail] instead)]

Attachment: fix

Comment 19

[David :Bienvenu]

sr=bienvenu, but feedback from Hyatt would be reassuring :-)

Comment 20

[(not reading, please use [e-mail] instead)]

fix checked in, but I'll talk to hyatt before marking this fixed.

Comment 21

[John Morrison]

The click event will bubble up the tree, so both 'bar()' and 'foo()' should be
called unless you were to cancel the bubbling while in function 'bar()'.

Comment 22

[(not reading, please use [e-mail] instead)]

do I cancel the bubbling by returning false?  I'll try that, and if it works
I'll propose a new fix.

Comment 23

[John Morrison]

Um. Made me go think (and I still don't know if this is optimal, but ...)

You can call 'event.stopPropagation();' to kill the bubbling. (This is what
is done to deal with the case where, on linux, you want a middlemouse to load 
the URL that is on the clipboard, unless that middlemouse is over a scrollbar 
[which should instead just move the scroll position].)

(stopPropagation was known as (apparently) preventBubble in an earlier draft 
of the DOM L2 Events spec, but either will work. However, consult your nearest
hyatt or joki for a blessing that I'm not completely on crack).

So this works for me (prints barMouseDown()\n fooMouseDown()\n barClick()\n)

  <script>
    function barClick(event) { 
      dump("barClick()\n");
      event.stopPropagation();           
    }
    function barMouseDown(event) { dump("barMouseDown()\n"); }
    function fooClick(event)     { dump("fooClick()\n"); }
    function fooMouseDown(event) { dump("fooMouseDown()\n"); }
  </script>
  <tree onclick="fooClick(event);" 
        onmousedown="fooMouseDown(event);">
    <treehead>
      <treerow>
        <treecell onclick="barClick(event);" 
                  onmousedown="barMouseDown(event);" 
                  value="Click me. Look in console."/>
      </treerow>
    </treehead>
  </tree>

Comment 24

[(not reading, please use [e-mail] instead)]

marking fixed to get it off the radar.  I'll open a new bug to make sure we fix
the fix to use "event.stopPropagation();"

Comment 25

[Ninoschka Baca]

Build 2001-01-2204: NT4, Mac 9.04
Build 2001-01-1908: Linux 6.2
Verified Fixed. A stand alone window no longer appears.