User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:2.0b4pre) Gecko/20100817 Minefield/4.0b4pre
1) Third line is same as first two.
2) No slant.
Isn't this invalid ?
Yeah, this test is invalid. I sent a mail to some of the MS QA guys.
The test defines a font family with a single italic face, then tries to compare between the behavior when different font-style attributes are used. There are a few problems with this.
1. If a family includes a single italic face, independent of the setting of font-style, that face will be used.
2. Style matching with @font-face runs is done based on the values of the descriptors, *not* based on the style data in the font. So in this case the normal face of Arial will be treated as italic. No synthetic obliqueing should happen in this case.
3. For font selection, italic and oblique are effectively synonymous. With @font-face you technically could define separate italic and oblique faces that *should* match on or the other but I doubt there are implementations that do this correctly (I know Gecko has lots of "if italic or oblique..." logic).
4. The test also assumes Arial is the default font.
This shouldn't be closed until the MS test is fixed, per several other similar bugs.
I can't see why Mozilla should track MS bugs.
Because we need to make sure they fix this, because we get bad press for it.
This test seems to have changed, and it appears we pass now.