294376 - [Mac] Wrong default font sometimes used when non-default language specified

Status: RESOLVED FIXED

(Core Graveyard :: GFX: Mac, defect)

Description

[Uri Bernstein]

When a document specifies a language using the "lang" attribute, but does not
specify a font (other than perhaps "serif"/"sans-serif"), any run of text
containing non-ASCII characters is displayed using the default font of the
default language (rather than the default font of the specified language).

Moreover, the spacing between such a run of text and any following text is
incorrect.

Testcase and screenshot coming up.

Comment 1

[Uri Bernstein]

Attachment: testcase

Note that the empty <span>s are just for separating the non-ASCII text from the
ASCII text. Any other markup would do.

The &rsquo;, &ldquo; etc. are just samples of non-ASCII characters. The actual
UTF characters would have worked as well.

The third paragraph is just for reference - to show that the metrics used for
the first part of the text in line 2 are *not* those of "Lucida Grande" (as you
might have expected).

Screenshot will follow.

Comment 2

[Uri Bernstein]

Attachment: Screenshot

Configuration used in this screenshot:

Mozilla/5.0 (Macintosh; U; PPC Mac OS X Mach-O; en-US; rv:1.8b2) Gecko/20050514
Firefox/1.0+

Locale: en-US
Default fonts for Western: Times / Helvetica
Default fonts for Hebrew: Lucida Grande / Lucida Grande

Comment 3

[Uri Bernstein]

*** Bug 273003 has been marked as a duplicate of this bug. ***

Comment 4

[Uri Bernstein]

Attachment: testcase #2

In this testcase, an actual Hebrew character is added somewhere in the
document. This causes the following differences from the previous testcase:
1. ASCII-only text now also appears using the wrong font.
2. The spacing problem for ASCII text (but not for non-ASCII text) now does
look as if the metrics of the correct (Hebrew default) font (Lucida Grande) are
being used.

The third line starting with "Please" is again just for reference. Screenshot
will follow.

Comment 5

[Uri Bernstein]

Attachment: screenshot #2

Comment 6

[Uri Bernstein]

Changing summary to reflect testcase #2 as well.

Comment 7

[Uri Bernstein]

This appears to be fixed in Cairo/Cocoa builds.

Comment 8

[Uri Bernstein]

Fixed by Cairo (bug 323934).