PDA

View Full Version : Mac IE 5.2.3



serge4
10-05-2005, 05:54 PM
Anyone here have any tips on troubleshooting a JS in Mac IE 5.2.3? I'm running into a problem where my JS/CSS is really buggy in that browser and I can't figure out the problem.

jscheuer1
10-06-2005, 05:51 AM
Quirksmode.org's IE5 Mac Page (http://www.quirksmode.org/browsers/explorer5mac.html) Is a great resource. In fact, so is their entire site, which is best accessed via their root domain:

http://www.quirksmode.org/

serge4
10-06-2005, 08:07 PM
Here is the issue i'm having. I need help troubleshooting some problems i'm having with the A List Apart version of vertical centering (http://www.alistapart.com/d/footers/footer_variation1.html) in Mac IE 5.2.3 and Safari?

Here is the problem and all the associated files. The ALA Javascript vertical centering solution worked great as is. But after adding a couple extra DIVs and a few styles I noticed some problems in Mac IE 5.2.3 which unfortunately is an important browser to consider for this. I added some extra JS to the original ALA JS which you will see pointed out in the comments. This seemed to take care of the problem for Mac IE 5.2.3 but now Safari and Firefox are having some issues. If you remove the Mac IE 5 code everything works fine. I need some help ironing out these remaining few kinks in this vertical centering adventure. I'd prefer to have 1 script and not a bunch of browser sniffing junk if possible. Perhaps there is something that could be modified with the JS to make this happen?

mwinter
10-09-2005, 01:12 AM
Any thoughts?At last count, none of the regulars in this forum have access to Macs, so advice won't be particularly forthcoming. Sorry. :o

Although IE/Mac was one of the first browsers (certainly before it's Windows counterpart) to make an effort to support CSS properly, the effort was arguably wasted and resulted in more bugs than successes.

It has been suggested by some to drop IE/Mac and treat it much like other obsolete browsers (Opera 6, IE4, NN4, etc.). That is, style it as much as practical, and hide the rest of the style sheet (http://w3development.de/css/hide_css_from_browsers/summary/) from it.

Good luck,
Mike

serge4
10-10-2005, 02:26 PM
Well, I think this is more of a Javascript problem than CSS. It has a bit to do with CSS but mainly Javascript. I've discovered that in IE 6 you can see a similar problem...

mwinter
10-11-2005, 02:39 PM
Well, I think this is more of a Javascript problem than CSS.If the original solution didn't work, then that is likely due to CSS support issues in IE/Mac than the script itself. However, I can't be sure of that (for obvious reasons).


I've discovered that in IE 6 you can see a similar problem...This is to do with the script. The behaviour that you're seeing is caused by the repeated call of the setContent function: the load event immediately followed by the resize event. If you resize Firefox, you'll see the same problem.

When the #info and #footer elements are absolutely positioned, they no longer contribute to the size of the #content element (absolutely positioned elements are taken out of the flow). This means that when setContent is called a second time, the #content element is only as tall as the #header element. The #header element will now be in the central position, with the #info and #footer elements positioned below it.

You could solve this by adding the heights of each inner element (#header, #info, and #footer) and then calculate the position based on this. However, I really don't think it's worth the effort. Perhaps that's just my bias against fixed width designs.

By the way, that eval function call is totally unnecessary (uses of eval usually are). The parentheses are a good idea, though.

Mike

serge4
10-11-2005, 08:01 PM
What part is the eval function?

mwinter
10-11-2005, 11:05 PM
What part is the eval function?In the code you posted, you have:


document.getElementById('footer').style.top = eval(infoH + headH) + 'px';which you use to set the position of the #footer element. In my last comment, I'm suggesting that you remove 'eval', to just leave the parentheses.

Why? Well, the call does nothing except slow execution. Both infoH and headH are numbers, as the offsetHeight property should yield only integers (and eval wouldn't help even if they weren't). So, before the eval function is called, the two values have already been added together to create your new value.

The eval function works on a string that matches the Program grammar production, so this new number must first be converted to a string. It is then evaluated - converted back into a number - then returned before being converted back into a string and suffixed with 'px'.

A whole lot of work with absolutely no gain.

Mike

rino5000
12-12-2005, 10:03 PM
I wanted to bring this up to the front of the forum as I am having this issue as well and I did not see any replies. The Javascript I am using is read in all browsers except for a Mac running IE 5. Does anyone have any good scripts for drop down menus tested and functional in all browsers?

The model I am using is located at http://www.dynamicdrive.com/dynamicindex1/dropmenuindex.htm

Any suggestions would be greatly appreciated.
Thanks

cyber1
12-13-2005, 02:45 AM
Well, I think this is more of a Javascript problem than CSS. It has a bit to do with CSS but mainly Javascript. I've discovered that in IE 6 you can see a similar problem...


serge-

Forget about Mac IE. Its all but dead.
MS no longer supports it. Last verssion was 3 yrs ago and that was the last and final version.
Heres the MS FAQ (http://www.microsoft.com/mac/products/internetexplorer/internetexplorer.aspx?pid=faq#features) link
When Apple developed Safari that decision was made.

So why bother, the percentage of users has to be minuscule.

-Bill