372404 - font-style: italic; regression on trunk

Status: RESOLVED DUPLICATE of bug 333126

(Core :: Graphics, defect, P2)

Description

[Robert Sayre]

On the branch, the page in the URL field displays with italics on OS X and Linux. It is not italicized on trunk. The computed style seems to be correct.

Comment 1

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

Works for me on Linux nightly 2007-03-02-04-trunk.  (The word "intertwingly" is in italics.  If the expected results are different, it would be nice if you described "Actual results" and "Expected results" more clearly in the bug report.)

Comment 2

[Robert Sayre]

> it would be nice if you described "Actual results" and "Expected results" more clearly in the bug report.

Actual:   Not italicized.
Expected: Italicized.

Comment 3

[Robert Sayre]

Attachment: Screenshot: Firefox 2

Comment 4

[Robert Sayre]

Attachment: Screenshot: Minefield (up-to-date trunk)

Comment 5

[Robert Sayre]

Attachment: Screenshot: Firefox 2 Mac

Comment 6

[Robert Sayre]

Attachment: Screenshot: Minefield Mac Nightly

Comment 7

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

Yeah, I see it on Mac.  We might want separate bugs per platform.

Comment 8

[Marco Perez [:mpk]]

AFAIK there are several fonts (among them "Arial Black" and
"Bitstream Vera Serif") which lack italic. Depending upon
how the fonts are mapped on your systems (or by the mozilla
build you're using) this may trigger the behaviour
described in this bug.

Comment 9

[Robert Sayre]

(In reply to comment #8)
> AFAIK there are several fonts (among them "Arial Black" and
> "Bitstream Vera Serif") which lack italic. 

Yes, could be bug 364713.

Comment 10

[Greg K.]

Yes, also seeing this with trunk Camino 2007032200 (1.2+). Lucida Grande text won’t italicize because there is no actual italic variant, so synthesis is failing I suppose.

Comment 11

[Daniel Holbert [:dholbert]]

Attachment: simple italic testcase

I'm seeing this bug with the default font (serif) on Linux.

See attached testcase...
 expected: italic text
 observed: normal text

Comment 12

[Daniel Holbert [:dholbert]]

(In reply to comment #11)
> I'm seeing this bug with the default font (serif) on Linux.

FWIW, FF2/Linux renders attachment 291758 [details] correctly (italicized) using the same font that's broken in FF3, serif.

Comment 13

[Daniel Holbert [:dholbert]]

(In reply to comment #11)
> I'm seeing this bug with the default font (serif) on Linux.

Interesting... Two other linux boxes I've tested work fine.  I only see the problem in FF3 builds (not FF2) specifically on my main dev machine.

All three computers run Ubuntu 7.10 Gutsy.  Not sure what's causing the difference.

Comment 14

[Daniel Holbert [:dholbert]]

Attachment: simple italic testcase 2 (using "Bitstream Vera Serif")

In this testcase, I specify a specific Linux font-family: Bitstream Vera Serif.

This fails (straight text) in FF3 on all three Linux boxes that I'd previously tested, and works in FF2.

Comment 15

[Daniel Holbert [:dholbert]]

... and from looking at this site:

  http://www.bitstream.com/font_rendering/products/dev_fonts/vera.html

...it looks like Bitstream Vera Serif is the only Bitstream Vera font without an Oblique/Italic version.

So, as mentioned in Comment 10, this is probably an issue with synthesizing italics for fonts with no italic variant.

Comment 16

[Robert O'Callahan (:roc) (email my personal email if necessary)]

Probably a gfx bug.

Comment 17

[Stuart Parmenter]

i think this is 364713 on mac and 333126 on linux

Comment 18

[Daniel Holbert [:dholbert]]

(In reply to comment #17)
> i think this is 364713 on mac and 333126 on linux

Those bugs are ranked P3 and P5, respectively. Either those bugs should be promoted to higher-priority (particularly 333126, the P5), or we should reduce the priority of this bug.

(Or, we could just resolve this as a dupe)

Comment 19

[Daniel Holbert [:dholbert]]

(In reply to comment #18)
> (In reply to comment #17)
>> i think this is 364713 on mac and 333126 on linux

Just a follow up to my previous comment:

Those bugs are now marked P3 and P2, respectively -- cool.  They're also both assigned.

I don't think anything more is needed for this bug. --> resolving as a dupe.