367931 - mac.com - (.mac/MobileMe) iWeb font size script conflicts with the minimum font-size setting in Gecko

Status: RESOLVED WORKSFORME

(Tech Evangelism Graveyard :: English US, defect)

Description

[Eric Manchester]

User-Agent:       Mozilla/5.0 (Macintosh; U; Intel Mac OS X; en; rv:1.8.1.2pre) Gecko/20070123 Camino/1.1a2+
Build Identifier: Mozilla/5.0 (Macintosh; U; Intel Mac OS X; en; rv:1.8.1.2pre) Gecko/20070123 Camino/1.1a2+

Blog page generated by iWeb renders correctly until the search box appears.  Once search box is rendered, various elements on right hand side of the page disappear and about text on left side becomes far too spaced out.

Reproducible: Always

Steps to Reproduce:
1.Go to page
2.Look at page before search box loads
3.Repeat and let the search box load
Actual Results:  
Same as every time.  About me info spreads out vertically and pictures and sites on right side of page disappear.

Expected Results:  
Photos and listed web sites should not disappear and info on left side should not be spaced far apart.

The version of Firefox I have:
Mozilla/5.0 (Macintosh; U; Intel Mac OS X; en-US; rv:1.8.1.1) Gecko/20061204 Firefox/2.0.0.1
renders the page properly.

Comment 1

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

Hmm, I see the same bug in Fx 2.0.0.1.

I'd suspect this is a bug in whatever iWeb is doing with its HTML, but I'm not sure.  philippe, can you have look at this?

Comment 2

[philippe (part-time)]

The search form is only a red herring here.
Eric and Smokey, do you have a minimum font-size set in the browser ?
(In Camino: Preferences>Appearance>Fonts>Advanced)

In both Camino (trunk and branch) and Firefox, the page works fine only if and when I have NO minimum font-size set. But with my settings (minimum font-size set to 12), then the layout breaks when the script kicks in.

There is a script that checks/sets a minimum font-size, and bases part of the layout on this. There are no problems with this in Safari, but that browser has a completely different way of handling the user-set minimum font-size.
Very shortly:
In Gecko browsers: set the minimum font-size on the root element and scale from there.
In WebKit (safari): check on every single element and scale per element.
(note to self - ask Georg where he has put his article with our complete analysis of this).

When I disable that script on a local copy of the page, all works correctly.

I'd say: this is a TE bug.

For practical purposes: Eric, I don't know how much you can customise those iWeb templates - maybe there is an option to disable that script, or maybe one can do that manually (aks if you need help).

Comment 3

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

Indeed, I have one set in both; thanks for figuring out the issue, philippe!  (Note that in Camino, you have to select another prefPane or close prefs when changing any font prefs due to bug 346209; otherwise, it will seem like you made the change and it did no good).

I'm not sure how to re-summarize this to kick it to TE: iWeb minimum font size script makes bad assumptions about Gecko?

It's also probably worth someone who understands this filing an issue with Bug Reporter or Radar....

Comment 4

[philippe (part-time)]

Attachment: zipped archive of offending page

For reference, in case the code changes: the page and the relevant linked files (.css, .js). I omited the main images

Comment 5

[philippe (part-time)]

What about:
iWeb  font size script conflicts with the minimum font-size setting in Gecko ?

To be clear, the main issue is actually the generated line-height

(sorry for the delay, loads of work)

Comment 6

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

> (sorry for the delay, loads of work)

No problem, and thanks for working on this!

Comment 7

[Chris Lawson (gone)]

Smokey, now that we know what the problem is, who should file the rdar: on this?

Comment 8

[Felix Miata]

(In reply to comment #2)
> (note to self - ask Georg where he has put his article with our complete
> analysis of this).
 
Probably http://www.gunlaug.no/contents/wd_1_03_04.html

Comment 9

[Hallvord R. M. Steen [:hallvors]]

404
Mozilla/5.0 (X11; Ubuntu; Linux i686; rv:22.0) Gecko/20100101 Firefox/22.0