charles wrote: > Responding to my own message with an update... if anybody cares. =} > > charles wrote: >> I've recently run into a new (to me) bug in IE6 and IE7 regarding >> fieldsets and legends. I've been bashing my head against my desk for a >> couple days now and would appreciate it if anybody could let me know >> what the heck is going wrong. >> >> I'm using a position:absolute fix to place the Legend in the right spot, >> targeting IE with a conditional comment. At first, I was having >> problems with the Legend not rendering in IE on a single fieldset, while >> other fieldsets on the same page (with almost identical contents) >> displayed the Legend as expected. >> >> I finally discovered that the Legend could be rendered by segregating >> the content by wrapping everything inside the fieldset except the Legend >> in a div (incidentally, placing an empty div between the Legend and >> following Label tag has the same effect). However, adding that div >> caused IE to add a sizable hunk of whitespace further down the form. >> >> I've created a somewhat minimal, very ugly test case here: >> >> http://www.toxicwombat.com/fieldset_weirdness.html > > The issue appears to be with the input elements being set to 100% width. > If the containing div is dropped, and the input set to width:95%, the > legend mysteriously reappears (but the alignment between the input and > label tag goes to heck). However, if the div inside the fieldset is > retained, and the input tags are set to width:98%, the legend is > rendered *and* the mysterious whitespace vanishes. > > I played around with removing input {float:left;}, but it didn't matter > what width was assigned to the input element. Any width caused the > absolutely positioned Legend to disappear. > > And the fixed version: http://www.toxicwombat.com/fieldset_weirdness.1.html > > As a side discovery, inputs in IE apparently inherit the horizontal > margins from the fieldset. That one was news to me. Wrapping the > content in a div apparently cancels the inheritance (I could also target > IE with a negative horizontal margin, but what a pain in the butt). > > Thanks for looking, > -charles
Charles, I missed your first mail. Check Bruno's bug analysis on disappearing absolutely positioned elements. Some of the results are similar to yours. http://www.brunildo.org/test/IE_raf3.html I played with the whitespace problem you reported, and I think an additional element (<br />) which follows the last <input> could 'eat' the space. Regards, Ingo -- http://www.satzansatz.de/css.html ______________________________________________________________________ css-discuss [EMAIL PROTECTED] http://www.css-discuss.org/mailman/listinfo/css-d IE7 information -- http://css-discuss.incutio.com/?page=IE7 List wiki/FAQ -- http://css-discuss.incutio.com/ Supported by evolt.org -- http://www.evolt.org/help_support_evolt/
