A table, when sent to a height of 100%, and perhaps with little or no content, will not stretch to fit the entire height of the mother cell as commanded. Please note this bug also seems to exist in Windows IE.

*** Bug 167064 has been marked as a duplicate of this bug. ***

What browsers *does* this work in?

Testcase. A table contains a 200px by 200px cell, that contains a table of 100% height.

WorksForMe using FizzillaCFM/2002090508 on Mac OS X 10.1.5 to access my testcase. konrad, does my testcase work for you?

Karl comments in email: works in IE 5 on both Windows and Mac. may collapse like it does in Mozilla 1.1 in IE 4 haven't tried it in others n.b. please enter comments here through the web interface in Bugzilla; this automatically emails the people cc'd to the bug and will preserve the comment here.

Greg, Your test case did work. However I ammended the code of your test to mimic my problem and got the same bug. My changes to your HTML: I removed the height setting from the parent table (no height setting so now its automatic). I added another column. I threw some text into the left column to give the table some height... http://www.uvm.edu/~kwitte/test/167063.html

It appears your testcase does show the problem, though you added several HTML syntax errors. I fixed them, and it still shows the problem. This could be a dup of bug 10212. Other bugs on the topic are bug 26617, bug 88035, bug 97138, bug 115245, and bug 119409.

The HTML in my testcase (http://www.uvm.edu/~kwitte/test/167063.html) may not be as exotic as 98294 (http://bugzilla.mozilla.org/attachment.cgi?id=98294&action=view) but there aren't any errors in it... You'd probably want to make sure it works in both.

konrad, run yours through validator.w3.org to see the errors.

only happens in standard compliance mode. table in strict dtd doesn't have height attribute.

Fixed in bug 227123, but see bug 241161 and bug 240148 for reversing the change in quirks mode because it breaks sites.