* TITLE/SUMMARY "nohref" attribute improperly handled on client-side image maps * STEPS TO REPRODUCE 0) Launch Apprunner 1) Execute the area_nohref.html test case (internally located at http://slip/ projects/marvin/html/area_noref.html) * RESULT - What happened Seamonkey treats a click inside of the red box equivalently to a click in the external white region. - What was expected According to the 4.0 Transitional DTD for the "nohref" attribute of the AREA element, I *believe* [but could be mistaken, of course] that no action at all should be taking place: <!ATTLIST AREA [...snip...] nohref (nohref) #IMPLIED -- this region has no action -- [...snip...] > * REGRESSION - Occurs On Mac OS Apprunner (1999/05.25.16 build) Win32 Apprunner (1999/05.25.17 build) - Doesn't Occur On Communicator 4.6 RTM <Can't check Linux; system is totally hosed beyond my ability to recover.> * CONFIGURATIONS TESTED - [Mac] Power Mac 8500/120 (233 Mhz 604e), 64 MB RAM (VM on; 1 MB of VM used), 1024x768 (Thousands of Colors), Mac OS 8.6 - [Win32] Vectra VL (233 Mhz P2), 96 MB RAM, 800x600 (True Color), NT 4.0 SP3. - [Linux] Vectra VL (266 Mhz P2), 96 MB RAM.

[QA Assigned to self, CC:'d Jan Carpenter. Jan, if you really want to be the QA assigned on this, of course, please just reassign it back to yourself. Thanks!]

elig when you get the chance, would you please http://slip/projects/marvin/html/area_noref.html as an attachment so we could have a look at it out here? Thanks.

Attached. (If you need the other 2 or 3 referenced files, I can post 'em, too, but their contents should be obvious from the context...)

Attaching a couple of new test cases. First test case: * What should happen: Both images in this test case should contain areas where no action should be taken if the user clicks inside that area. * What actually happens: Any other links defined appear to bleed-through. On the first image, this is an overlapping client-side map area. On the second image, it is a server-side image map. Second test case: * What should happen: The second test case shows the same scenario for the alternative content model for MAP elements, as defined in HTML 4. Basically, Anchors with coordinates but no HREF defined should be treated the same as AREAs with the NOHREF attribute defined. * What actually happens: Where no HREF is specified by the Anchor, the browser jumps to the parent directory of the current page. Tested on: Mozilla M10, 1999082708, on Windows NT 4 Workstation. See the last example in section 13.6.1 of the HTML 4.0 spec for another example of this type of client-side map.

[As a courtesy FYI, as of the 1999092408 Mac OS & Win32 builds, this bug is still present, and has not received a "Mystery Fix" from other code changes.]

This is an HTML4 compliance issue. Marking dependency.

Moving M16.

Bulk moving [testcase] code to new testcase keyword. Sorry for the spam!

Sorry it took so long to get to this but thanks run2000@geocities.com for the excellent test cases. The test cases actually seem to work now except for cursor errors which I have a fix for locally. I'll mark this fixed when I check it in. Also, the second test currently seems to have a parser (?) error in it. If you don't declare both maps at the end the second test case doesn't show up. Weird. I'm going to open a different but for that but I'm attaching my modified second test case with the maps at the end for verification of this one after I mark it fixed.

fix checked in.

Imagemap issue; QA assigning to petersen for verification.

Fixed in the May 22 builds.