385997 - Zoom button does not resize webpages with frames

Status: RESOLVED INCOMPLETE

(Camino Graveyard :: General, defect)

Description

[Mike]

User-Agent:       Mozilla/5.0 (Macintosh; U; Intel Mac OS X; en; rv:1.8.1.4) Gecko/20070509 Camino/1.5
Build Identifier: Mozilla/5.0 (Macintosh; U; Intel Mac OS X; en; rv:1.8.1.4) Gecko/20070509 Camino/1.5

Webpages which include frames do not resize when the zoom button is pressed.

Reproducible: Always

Steps to Reproduce:
1.Navigate to a site with frames such as http://mail.google.com/ or http://victorian.fortunecity.com/brambles/4/frames/
2.Make the window as small as possible
3.Click the zoom button
Actual Results:  
Nothing happened.

Comment 1

[Smokey Ardisson (offline for a while; not following bugs - do not email)]

For reference, this is mentioned several times in bug 360880 (comment 20, and maybe comment 8), but it's a different bug from 360880.

Comment 2

[Chris Lawson (gone)]

OK, here's what I see on

http://victorian.fortunecity.com/brambles/4/frames/

with trunk if I open it by itself in a new window (didn't see any point in
messing up my good window with my current tabs :-p):

If you resize the window manually to some arbitrarily small size (where "small"
means "smaller than the maximum size of the screen"), then click the zoom
widget, you get resize behaviour in increments.

If the window is less than half the height of the screen, the first click gets
the window to about 3/4 full-screen height.

If the window is more than half the height of the screen, the first click
resizes the window to full height.

Once the window is full-height on its associated screen, each subsequent click
on the resize widget expands the window to the right by the width of a
scrollbar until the right edge of the window contacts the right edge of the
screen. The next click after this will resize the window to full-screen size.
The next click after this will start the whole cycle over again with the window
at its original (manually resized) size.

There's definitely a bug here somewhere, probably in how we're calculating the
size that a massively complex frameset "should" be.

Comment 4

[philippe (part-time)]

I made some very basic frame/iframe tests stored here (and may add some more):
http://dev.l-c-n.com/camino/iframe/

Fwiw, Safari doesn't handle frames perfectly either. With the URL in comment 2, I can more or less replicate the behaviour described in comment 2, but to a lesser extend (testing with Safari 5.0.2).

Comment 5

[philippe (part-time)]

Attachment: iframe testcase

Comment 6

[philippe (part-time)]

Attachment: frames testcase