RE: tests that are invalid because of dependence on subpixel metrics of Ahem font

On Wednesday, November 17, 2010 10:36 AM L. David Baron wrote:
> 
> As I mentioned at the face-to-face meeting (see
> http://lists.w3.org/Archives/Public/www-style/2010Nov/0240.html ), I
> believe a significant number of tests are invalid because they use the Ahem
> font at sizes that are not a multiple of 5 pixels.  This means that important
> metrics such as ascent, descent, and x-height are not integral numbers of
> pixels.  Different platforms tend to perform rounding differently.
> 
> See the following bug reports for extended discussion of issues on
> Linux:
> https://bugzilla.mozilla.org/show_bug.cgi?id=604836
> https://bugzilla.mozilla.org/show_bug.cgi?id=611780
> 
> This affects the following tests:
> 
> html4/c43-rpl-bbx-002.htm
> xhtml1/c43-rpl-bbx-002.xht
> html4/c43-rpl-ibx-000.htm
> xhtml1/c43-rpl-ibx-000.xht
> html4/c548-leadin-000.htm
> xhtml1/c548-leadin-000.xht
> html4/c548-ln-ht-002.htm
> xhtml1/c548-ln-ht-002.xht
> html4/c526-font-sz-002.htm
> xhtml1/c526-font-sz-002.xht
> html4/first-line-pseudo-013.htm
> xhtml1/first-line-pseudo-013.xht
> html4/first-line-pseudo-014.htm
> xhtml1/first-line-pseudo-014.xht
> html4/first-line-pseudo-015.htm
> xhtml1/first-line-pseudo-015.xht
> html4/first-line-pseudo-016.htm
> xhtml1/first-line-pseudo-016.xht
> html4/inline-formatting-context-022.htm
> xhtml1/inline-formatting-context-022.xht
> html4/inline-non-replaced-height-002.htm
> xhtml1/inline-non-replaced-height-002.xht
> html4/inline-non-replaced-height-003.htm
> xhtml1/inline-non-replaced-height-003.xht
> html4/letter-spacing-043.htm
> xhtml1/letter-spacing-043.xht
> html4/letter-spacing-044.htm
> xhtml1/letter-spacing-044.xht
> html4/margin-collapse-154.htm
> xhtml1/margin-collapse-154.xht
> html4/max-height-108.htm
> xhtml1/max-height-108.xht
> html4/outline-019.htm
> xhtml1/outline-019.xht
> html4/outline-020.htm
> xhtml1/outline-020.xht
> html4/units-001.htm
> xhtml1/units-001.xht
> html4/units-004.htm
> xhtml1/units-004.xht
> html4/vertical-align-121.htm
> xhtml1/vertical-align-121.xht
> html4/vertical-align-applies-to-008.htm
> xhtml1/vertical-align-applies-to-008.xht
> html4/vertical-align-applies-to-009.htm
> xhtml1/vertical-align-applies-to-009.xht
> html4/vertical-align-applies-to-010.htm
> xhtml1/vertical-align-applies-to-010.xht
> html4/vertical-align-applies-to-011.htm
> xhtml1/vertical-align-applies-to-011.xht
> html4/vertical-align-applies-to-012.htm
> xhtml1/vertical-align-applies-to-012.xht
> html4/vertical-align-applies-to-013.htm
> xhtml1/vertical-align-applies-to-013.xht
> html4/vertical-align-applies-to-014.htm
> xhtml1/vertical-align-applies-to-014.xht
> html4/vertical-align-applies-to-015.htm
> xhtml1/vertical-align-applies-to-015.xht
> html4/white-space-collapsing-001.htm
> xhtml1/white-space-collapsing-001.xht
> html4/white-space-collapsing-002.htm
> xhtml1/white-space-collapsing-002.xht
> html4/white-space-processing-043.htm
> xhtml1/white-space-processing-043.xht
> html4/white-space-processing-044.htm
> xhtml1/white-space-processing-044.xht
> html4/white-space-processing-045.htm
> xhtml1/white-space-processing-045.xht
> 

I fixed all the Ahem font an sub pixel metrics issues I could find for the Microsoft cases. If you find any other cases let me know and I will correct those as well. There were more than 350 cases changed due to this issue that I found so far.

--
Thanks,
Arron Eicholz

Received on Tuesday, 11 January 2011 03:59:51 UTC