Overview description: The 'codebase' attribute of the HTML 4.0 OBJECT element is not supported. See http://www.w3.org/TR/REC-html40/struct/objects.html#adef-codebase-OBJECT for the definition. Steps to reproduce: 1) View the URL. Actual results: A regular, non-inverted photograph of a handsome Labrador retriever embedded on the page. Expected results: An inverted image of the same dog. Build date & platform: 2000-05-22-08, Linux.

Oops, wrong URL. Also adding 'html4' keyword and a dependency on this to bug 7954 (HTML 4 meta bug).

Andrei, if you don't feel confortable with this, try to figure out who should help.

I just realized that codebase _is_ actually supported - it just doesn't affect the 'data' URI, only 'classid'. I haven't tested yet whether it affects 'archive' or not. Adjusting the summary.

massive update for QA contact.

I'll try to come up with a release note for this one ASAP.

Proposed release note (for developer release notes): "The 'codebase' attribute of the OBJECT element doesn't affect relative URIs in the 'data' attribute value. Workaround: use absolute URIs in the OBJECT's 'data' attribute value when 'codebase' attribute is present." Feel free to try and clarify it if you think that sounds clumsy.

Nom. nsbeta1 on correctness-of-standards-compliance grounds. In particular, until we fix this, we're essentially preventing developers from taking advantage of codebase at all in their content and forcing them to use absolute URIs (which are more labor-intensive to maintain).

Reassigning QA Contact for all open and unverified bugs previously under Lorca's care to Gerardo as per phone conversation this morning.

qa contact updated.

moving to m0.9.1

not critical to 0.9.1 moving to 0.9.2

<CLUELESS> Does this bug have anything to do with the fact that the table at http://www.sls.nl/hospiteren.htm fails to be populated in Mozilla yet works in IE? If not, does anyone know if there is a bug report filed yet for this phenomenon? This bug is the closest I could find... </CLUELESS> BTW... shouldn't mozilla0.9.2 keyword be added?

> Does this bug have anything to do with the fact that the table at > http://www.sls.nl/hospiteren.htm fails to be populated in Mozilla yet works in > IE? No. Nothing to do with 'codebase', but everything to do with ActiveX controls. See http://www.iol.ie/~locka/mozilla/mozilla.htm for (rather technical) details about ActiveX support in Mozilla.

Moving to m0.9.3

-->0.9.5

SPAM. HTML Element component is deprecated, changing to Layout component. See bug 88132 for details.

Come on Bugzilla, you can do it...

moving 1.1

this particular issue works for plug-ins, but does not work for images and documents. Reassigning to Peter, since he has done this for plug-ins

that should fix it

oh, it should be noted that this patch depends on the one in bug 213701

Actually, this one is probably better. It cleans up a few other places to use the getters so that it gets absolute urls automatically, and in some cases doesn't resolve the urls twice.

Changing summary so it is clear this is not about data URIs.

sigh, it took me some time to realize that comment 23 should've said "data attributes" instead of "data URIs" anyway, fixed by bug 1156.

What exactly is unclear about comment 23? I believe it describes what I meant...

oh. turns out I just misread it. sorry.

This isn't working anymore. See, for instance, the example in bug 401249.

This was fixed on the trunk (Firefox 3.0), not the 1.8 branch.

Yup, sorry. I had tried it in Gran Paradiso Alpha 7, where it failed; but it does appear to be working in the latest nightly.