Almost-certain CSS margin-bottom Bug {3}, & Post-conclusion

To continue, again:

Almost Certain CSS margin-bottom Bug {3} (I would upgrade this to
"Certain" now)
========================================
 [Further proof of "Almost-certain CSS margin-bottom Bug {1, 2}",
  previously reported.]

* To reformulate, Amaya does not honor "{margin-bottom:0}" (or
"{margin-bottom:0px}", the bottom-relevant aspect of "{margin:0}",
etc.),
any way, any how, in a manner that will (as it should when used in
"body" or at bare minimum when used in "html") place no (i.e.,
zero-width/height) borders around documents on all 4 edges (nor will it
do so using "@page", but this is already an admitted issue, so I don't
address it here).

Amaya creates a page-bottom margin *no matter what*.


New, Further Tests:
-------------------

http://www.well.com/~mech/.AmayaTest/NewTEST3.html
http://www.well.com/~mech/.AmayaTest/NewTEST4.html
(also attached here, with .css files, so you don't have to manually download
those to look at them.)

This test differs from the previous one in that
a) it attempts to outright *force* Amaya into honoring {margin-bottom:
0} as applied to the entire document "canvas" (or "viewport" or "box",
whatever term you prefer), by applying it to "html" as well as to
"body".
b) Tests values of "0" and "0px" separately, in NewTEST3.html (and
NewTEST3.css) in the first case, and in NewTEST4.html (and
NewTEST4.css) in the latter case.


Description of the issue:
-------------------------

As with the previous tests, there should be NO bottom margin at all,
yet there is a very large one. Again, Amaya IS properly handling the
left, right and top margins, just not the bottom one.

The new tests, attached, show very clearly that Amaya is ignoring any
"margin-bottom" that applies to the entire document, whatever syntax is
used to arrive at it.  It is as if the stylesheet had "{margin-bottom:
60px}" in "html" or "body", when it does not.

After reading a whole bunch of archived mailing list discussions full
of "deep geeking" about CSS 2, I learned that technically "html", not
"body" is supposed to be the controlling the ultimate dimensions of the
document (that I didn't know this isn't surprising - very few browsers
actually implement it this way, because the CSS2L1 spec. isn't really
clear enough on the topic).

Amaya surely, being the (nascent) reference standard implementation of
CSS and HTML (and XHTML) would conform, right?  Nope.  This is utterly
conclusive proof that this margin-bottom issue is a genuine bug.


POST-CONCLUSION:
================

Didn't surprise me at all. I've had a strong feeling this was an
outright bug from the moment I saw it, because there is no rational
reason for the stray page-bottom whitespace to be there.  It's just
something weird that Amaya does by default, and it doesn't undo this
particular default even in the presence of CSS telling it to do
precisely that.  As I mentioned in previous reports, I believe the
ultimate culprit is Amaya's failure to stop scrolling downward at the
end of document (when document is taller than window height upon
loading the document, and the user manually vertically scrolls to the
bottom), such that it tricks itself into artificially creating a fake
void that "must" be filled with bogus whitespace to make up the
difference.  If Amaya would stop scrolling where it should, this
condition wouldn't ever arise.

DEFINITELY a bug.  No question about it.  Both in the sense of
"application doing something undesirable, in general" (in
implementation) and in the more specific sense of "violation of the
published specs to which the application is attempting to conform (in
effect).

Received on Thursday, 30 September 2004 11:51:21 UTC