Neil Marshall Reporter
	Description • 25 years ago

This one is odd.  The left frame displays and works perfectly.
The right frame on the other hand doesn't fully display.  It doesn't respond to
mouse clicks.  It doesn't draw the scrollbar properly.  It doesn't repaint.

This _doesn't_ appear to be a frames problem as this page (On the same site)
does the same thing:
http://www.papercomputer.com/PpC-Folder/Willard-bio.htm

	Mats Palmgren (inactive)
	Comment 1 • 25 years ago

Using build 1999122708 on Windows 98 SE, most of the problems are gone. URL
does fully display, no scrollbar problems, no repaint problems. However, the top
navigation links does not respond to clicks and the cursor does not change when
the mouse is over them. Other links in the page works as expected.

http://www.papercomputer.com/PpC-Folder/Willard-bio.htm have no problems AFAICT.

	Mats Palmgren (inactive)
	Comment 3 • 25 years ago

The links in the testcase above are to far down on the page and doesn't work.
The links works and are displayed closer to the top in IE5 and Nav4.08.

	troy
	Comment 4 • 25 years ago

Sounds like an event problem

	David Baron :dbaron:
	Comment 5 • 25 years ago

The links seem to work fine now.  And the positioning looks correct to me - it
looks like the page relied on bugs in NN 4.x.  I don't think the carried-out
margins of the P should be ignored.

Should this be marked WORKSFORME??

	Mats Palmgren (inactive)
	Comment 6 • 25 years ago

I agree that the positioning is correct now and that the link is working in the
testcase. However, there is one more problem with
http://www.papercomputer.com/PpC-Folder/ppc-Opening.htm
that my first testcase failed to show. The links are below the <HR> in Mozilla,
but above the <HR> in IE5 and Nav4. Adding new testcase below.

	ekrock's old account (dead)
	Comment 8 • 25 years ago

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

	joki (gone) Assignee
	Comment 9 • 25 years ago

Okay, my guess is this is a parser issue.  The content model shows the abs. 
positioned div pulled out of the rel. positioned span.  The HR then isn't flowed 
after the div and the content overlaps.  I think maybe the div should still be 
in the span.  Anyway, I'll drop it to Harish and he can fight for whether he or 
rickg gets to fix it.

	harishd
	Comment 10 • 25 years ago

This seems to be fixed.  Petersen, could you verify this?

Here is the content-model that shows DIV inside SPAN.

docshell=01533190
html@0153C7EC refcount=3<
  head@0153C6BC refcount=2<
    title@01FFAF0C refcount=2<
      Text@01FFAE90 refcount=2<Testcase 2 for bug 22636>
    >
  >
  body@01FFAD7C bgcolor=#ffffff refcount=3<
    Text@0206DAC0 refcount=3<\n\n>
    p@0206DA5C refcount=3<
      Text@0206D930 refcount=3<\n  >
      span@0206D8CC style=position: relative; left: 0px; top: 0px; width: 623px; 
height: 23px;  refcount=3<
        Text@0206D800 refcount=3<\n    >
        p@0206D79C refcount=3<
          Text@0206D4A0 refcount=3<\n    >
        >
        div@0206D36C style=position: absolute; left: 0px; top: 0px; width: 
700px; height: 19px;  refcount=4<
          Text@0206CFB0 refcount=3<\n      >
          p@0206CF4C refcount=3<
            a@0206CE2C href=http://www.mozilla.org refcount=3<
              Text@0206CCF0 refcount=3<\n          The <HR> is below this link 
in IE5 and Nav4.08,\n          but above in Mozilla build 2000011314>
            >
            Text@0206CBE0 refcount=3<\n    >
          >
        >
        Text@0206CA10 refcount=3<\n  >
      >
      Text@0206C950 refcount=3<\n>
    >
    Text@0206C850 refcount=3<\n>
    p@0206C7EC refcount=3<
      Text@0206C770 refcount=3<\n  >
    >
    hr@0206C6DC refcount=3<>
    Text@0206C660 refcount=3<\n>
    p@0206C5CC refcount=3<
    >
    Text@0206C550 refcount=3<\n\n>
  >
>

	harishd
	Comment 12 • 25 years ago

FIXED on the parser side:

Here is the content-model:

docshell=00CE28C0
html@0228F338 refcount=3<
  head@0228F2D8 refcount=2<
    title@01F205A8 refcount=2<
      Text@01F20530 refcount=2<Testcase 2 for bug 22636>
    >
  >
  body@01F20478 bgcolor=#ffffff refcount=3<
    Text@0223F0A0 refcount=3<\n\n>
    p@0223F3D8 refcount=3<
      Text@0223F030 refcount=3<\n  >
      span@0223CF68 style=position: relative; left: 0px; top: 0px; width: 623px; 
height: 23px;  refc
ount=3<
        Text@0223D560 refcount=3<\n    >
        p@0223D4C8 refcount=3<
          Text@0223D2E0 refcount=3<\n    >
        >
        div@0223C778 style=position: absolute; left: 0px; top: 0px; width: 
700px; height: 19px;  ref
count=4<
          Text@0223C700 refcount=3<\n      >
          p@0223C668 refcount=3<
            a@0223DDFC href=http://www.mozilla.org refcount=3<
              Text@0223DBD0 refcount=3<\n          The <HR> is below this link 
in IE5 and Nav4.08,\n
          but above in Mozilla build 2000011314>
            >
            Text@0223DB60 refcount=3<\n    >
          >
        >
        Text@0223C250 refcount=3<\n  >
      >
      Text@0223C190 refcount=3<\n>
    >
    Text@0223BC10 refcount=3<\n>
    p@0223BBB8 refcount=3<
      Text@0223BB40 refcount=3<\n  >
    >
    hr@0223BAE8 refcount=3<>
    Text@0223BA70 refcount=3<\n>
    p@0223BA18 refcount=3<
    >
    Text@0223FAB0 refcount=3<\n\n>
  >
>

sending bug back to joki.

	ekrock's old account (dead)
	Comment 13 • 25 years ago

Is it even valid HTML to have a SPAN contain a DIV?

Marking 4xp. Nom. nsbeta2, recc. nsbeta2+ as this is a "backward compatibility 
with actual HTML 3.2 on the web" issue--key part of b2 reqs.

	David Baron :dbaron:
	Comment 14 • 25 years ago

It's not valid HTML, but I'm attaching a valid testcase for the bug.

	leger
	Comment 16 • 25 years ago

Putting on [nsbeta2-] radar. Not critical to beta2.  Setting on nsbeta3 keyword.

	joki (gone) Assignee
	Comment 17 • 25 years ago

My build is currently crashing in this testcase but the link is actually 
working.  I'm thinking its fixed but I'll update my tree and verify the crash 
location first.

	joki (gone) Assignee
	Comment 18 • 25 years ago

Okay, my build is still crashing but the link is defintely working.  I don't 
like to drag bugs out too long so I'm marking this bug fixed and opening a new 
bug for the crash.

	Chris Petersen
	Comment 19 • 25 years ago

Marking verified in July 6th build.