150485 - Substituted glyph drawn with too much horizontal whitespace

Status: VERIFIED DUPLICATE of bug 111728

(Core :: Internationalization, defect)

Description

[Ken Harris]

From Bugzilla Helper:
User-Agent: Mozilla/5.0 (Macintosh; U; PPC Mac OS X; en-US; rv:1.0.0+)
Gecko/20020603
BuildID:    2002060303

The ² character looks fine, but the character seems to have
a fair amount of empty white space it thinks is part of the character.
"a²+b", for example, has a gap between the "2" and "+" maybe
3 times as wide as the "2".

Reproducible: Always
Steps to Reproduce:
1. Go to http://www.w3schools.com/html/html_entitiesref.asp
2. Look at the "superscript 2" line of the table
3. Select the little "2" in the first column

Actual Results:  As you can see from the selection, there's a gap after the "2".
(It's even more noticeable in a real equation.)

Expected Results:  There should be no gap, or at least a smaller gap.  (See the
"degree"
entity in the same table for what I think is a more appropriate size.)

I tried in some other apps, too: IE5/Mac looks similar to Mozilla
(gap), but TextEdit and Omniweb look better (no gap).  Is this an
OS bug?

(Will gladly attach screenshots if this is unclear.)

Comment 1

[Stefan Borggraefe]

Attachment: Test document with ² character

This bug WFM on WinXP, Trunk-Build 2002060808. I tested it with the attached
test-document.

Comment 2

[Ken Harris]

Attachment: Sample file (prev. attachment), as rendered by Mozilla and TextEdit on Mac OS X

Thanks for the test case, Stefan.  Just so you can see I'm not crazy, here it
is
(1) in Mozilla at 200%, (2) in Mozilla, with just the ² selected, and (3)
in
TextEdit, Times New Roman 24 (copy-paste directly from Mozilla).  (I combined
the
screenshots and converted them to PNG in Illustrator, so it might not be pixel-

for-pixel, but you can see the spacing.)

Comment 3

[Greg K.]

Confirmed using FizzillaCFM/2002052918 (rv1.0.0). This is a problem with glyph
substitution; -->Internationalization, and editing Summary.

Comment 4

[Roy Yokoyama]

->nhotta ccing myself

Comment 5

[Greg K.]

I think this is also related to the problem of unnecessary font fallback taking
place. I'll attach a second testcase.

I think this is more serious than 

Comment 6

[Greg K.]

Attachment: Testcase for Unicode lowercase l and l+slash glyphs in Lucida Grande

Comment 7

[Greg K.]

It seems the ATSUI patch fixes the problem entirely.

Comment 8

[Greg K.]

See also bug 148361.

Comment 9

[Henri Sivonen (:hsivonen)]

TEC likes to use CJK fonts whose European chars are designed to be as wide as
the Han/Kanji/Hangul blocks.

*** This bug has been marked as a duplicate of 111728 ***

Comment 10

[Yuying Long]

Mark as verified dup.  Please reopen if disagree.