Closed Bug 537444 Opened 15 years ago Closed 14 years ago

Missing &uArr; glyph in default <h2> typeface

Categories

(Core :: Layout: Text and Fonts, defect)

Product:
Core
Component:
Layout: Text and Fonts
Platform:
x86
macOS
Type:
defect
Priority:
Not set
Severity:
normal

Tracking

()

Status:
RESOLVED FIXED

People

(Reporter: jdawiseman, Unassigned)

References

(
URL
)

Details

Attachments

(11 files)

Screenshot of bug 537444 example.
1.00 KB, image/gif
Details
HTML that generates GIF
583 bytes, text/html
Details
GIF showing problem with headers and italics
8.11 KB, image/gif
Details
251 glyps in <h2>, <p>, <p><b>, and <p><i>.
32.01 KB, text/html
Details
Screen shot of test case 419880.
11.50 KB, image/gif
Details
Screen shot of test case 419880.
11.72 KB, image/gif
Details
character log
39.59 KB, text/plain
Details
Multi-character test case for bug 537444
32.61 KB, text/html
Details
Test case 420619: screenshot 1
13.81 KB, image/gif
Details
Test case 420619: screenshot 2
12.95 KB, image/gif
Details
Test case 420619: screenshot 3
12.39 KB, image/gif
Details 
User-Agent:       Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.6; en-US; rv:1.9.1.6) Gecko/20091201 Firefox/3.5.6
Build Identifier: Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.6; en-US; rv:1.9.1.6) Gecko/20091201 Firefox/3.5.6

<html><body>
<h2>&uArr;</h2>
<p>&uArr;</p>
</body></html>

The uArr glyph is invalid in h2; but works in p. In 2005, and subsequently, both worked. I don't know at which version the break occurred. 

Reproducible: Always
WFM. 
Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.1.6) Gecko/20091215 Ubuntu/9.10 (karmic) Firefox/3.5.6

Your font is not complete?
Using a regular Mac OSX 10.6.2, with no known oddities, and certainly none deliberate. How would I test the hypothesis?

    
    

    
  
And works on my Mac under Chrome 4.0.249.43, Opera 10.10, and Safari 4.0.4. Which suggests, though doesn’t prove, that it might be a bug in Mac Firefox 3.5.6.
Component: General → Layout: Text
Product: Firefox → Core
QA Contact: general → layout.fonts-and-text

    
    

    
  
WFM in Firefox 3.5.7 on 10.6.2.

Note that the character U+21D1 is not present in the default fonts used, so display depends on font fallback behavior. It's not immediately clear why this would be failing for you in one style (h2), whereas it works in another (p). This does sound like a bug in our font-matching behavior, but it'd be easier to track down if I could reproduce it locally.

It'd be interesting to know which of your installed fonts have the character U+21D1 available. You can use the Character Viewer palette (from the Input menu) to determine this.

    
    

    
  
Same broken dysfunctionality on my wife’s much newer Mac, also running 10.6.2. 

Her uArr fonts are: Apple Symbols; Arial Unicode MS; Lucida Sans Unicode; Menio (Regular, Bold, Italic, both); MS Gothic; MS Mincho; MS PGothic; MS PMincho; Symbol.

The <p> uArr looks like the one that appears in the Character Info box, neither of which looks like any of the above typefaces.

    
    

    
  
Also: you report using Firefox 3.5.7; my household is still on 3.5.6, which is also the latest reported on Mozilla.com. Presumably knowledgeable developer people are a step ahead of us mortals. But would it be worth your testing the problem on 3.5.6?

    
    

    
  
Back at my computer, uArr typefaces are the same. I think the p uArr looks like Symbol Regular. Will check later whether it does on wife's machine.

If there are any other diagnostics that I could try for you, please ask.

    
    

    
  


  

    
    

    
  


  
<html><body>
<h1>h1: &uArr;</h1>
<h2>h2: &uArr;</h2>
<h3>h3: &uArr;</h3>
<h4>h4: &uArr;</h4>
<h5>h5: &uArr;</h5>
<h6>h6: &uArr;</h6>
<p>p: &uArr;</p>
<p><big>p big: &uArr;</big></p>
<p><big><big>p big big: &uArr;</big></big></p>
<p><big><big><big>p big big big: &uArr;</big></big></big></p>
<p><small>p small: &uArr;</small></p>
<p><b>p b: &uArr;</b></p>
<p><i>p i: &uArr;</i></p>
<p><b><i>p b i: &uArr;</i></b></p>
<p><tt>p tt: &uArr;</tt></p>
<p><tt><b>p tt b: &uArr;</b></tt></p>
<p><tt><i>p tt i: &uArr;</i></tt></p>
<p><tt><b><i>p tt b i: &uArr;</i></b></tt></p>
</body></html>

    
    

    
  
Broken for <i> and <h1>, <h2>, …, <h6>.

    
    

    
  
Strange. I was running a beta of 3.5.7, but downgraded to 3.5.6 to check... no difference, still works here.

I also disabled additional fonts I had installed, so that I have the same collection of fonts with the U+21D1 character, and it still works OK. Looks like it's choosing the Symbol glyph in some cases (e.g., <p>) and the Menlo in others (<h> styles and italic styles).

So I'm still stumped for how to reproduce this. :(

    
    

    
  
My computer has MS Office and several thousand typefaces added to the machine; with duplicates removed. My wife's computer has the vanilla install, with nothing added except Office and some regular applications like Firefox, Thunderbird, and Skype. The page looks the same on the two machines: identically broken.

Scratching my head as to what I can do to show this better. 

Does your machine have MS Office?

    
    

    
  
I have MS Office 2004 (but not "several thousand typefaces" - this shouldn't matter, though, if they don't include the character in question). I'll try a couple of other machines after the weekend.

Your testcase + screenshot make it pretty clear it's a problem with font fallback, apparently related to trying to find style-matched (i.e., not Regular but bold and/or italic) faces, but I'm currently puzzled why I'm not seeing the same problem here.

    
    

    
  
As expected the same result with dArr, lArr, rArr. Are there any particular glyphs that I could try that might help with differential diagnosis?

    
    

    
  
Our MS Office is 2008. (Would it be nice to blame Redmond for a bug in Mac Firefox? A bit.)

    
    

    
  
I've just realized that my font collection supporting this character doesn't quite match yours, as I thought it did.... to double-check, could you please try disabling Lucida Sans Unicode, and see if that makes any difference to the result you see?

    
    

    
  
• Firefox quit. 
• Lucida Sans Unicode disabled in Font Book.
• Firefox re-opened, and behaving just the same as it was before.
• Lucida Sans Unicode re-enabled in Font Book.

So that’s not the culprit.

    
    

    
  
(In reply to comment #18)
> • Firefox quit. 
> • Lucida Sans Unicode disabled in Font Book.
> • Firefox re-opened, and behaving just the same as it was before.
> • Lucida Sans Unicode re-enabled in Font Book.
> 
> So that’s not the culprit.

Well, it didn't seem very plausible; just trying to eliminate as many factors as possible.

I've now tried this with Firefox 3.5.6 on three different Macs running 10.6.2, two of them with a variety of extra fonts installed and one with a "clean" Snow Leopard installation with no extra fonts - not even the Apple CJK fonts. In all cases, the up arrow displays properly; I still can't reproduce the problem shown in your screenshot.

One other thing to try: could you please clear the font caches on your system, e.g., using a tool such as FontNuke (http://www.jamapi.com/pr/fn/), and see if that makes any difference.

    
    

    
  
Top-left button in FontNuke pressed, machine restarted: SNAFU. No change. Still broken. Sorry.

My wife’s computer has nothing clever done to it. No fonts have ever been added or taken away, except as changed by MS Office 2008. Did any of your three extra test machines have Office ’08?

    
    

    
  
Sigh. No, I don't have a copy of Office 08 available; will look for someone who can try that.

    
    

    
  
Testing other glyphs for you, mostly using the test case in the next post.

• Broken: #8657, the numerically-expressed uArr.

• Broken: crarr, lArr, uArr, rArr, dArr, hArr, forall, exist, empty, nabla, isin, notin, ni, lowast, prop, ang, and, or, cup, sim, cong, sub, sup, nsub, sube, supe, oplus, otimes, sdot, lceil, rceil, lfloor, rfloor. 

• Working well: uarr, part, plusmn, alefsym, weierp, infin, harr, cap, asymp, perp, and all other named glyphs in 
www.jdawiseman.com/papers/trivia/character-entities.html 

The cup/cap difference might help with diagnosis.

    
    

    
  


  
<html><body><table border="1">
<tr><td></td><th>h2</th><th>p</th><th>p b</th><th>p i</th></tr>
<tr><td>iexcl</td><td><h2>&iexcl;</h2></td><td><p>&iexcl;</p></td><td><p><b>&iexcl;</b></p></td><td><p><i>&iexcl;</i></p></td></tr>
<tr><td>cent</td><td><h2>&cent;</h2></td><td><p>&cent;</p></td><td><p><b>&cent;</b></p></td><td><p><i>&cent;</i></p></td></tr>
<tr><td>pound</td><td><h2>&pound;</h2></td><td><p>&pound;</p></td><td><p><b>&pound;</b></p></td><td><p><i>&pound;</i></p></td></tr>
…

    
    

    
  


  

    
    

    
  


  

    
    

    
  
Whilst making the above attachments a problem with bounding boxes was noticed, which might or might not be related, and has been reported in bug 537715.

    
    

    
  
Could you please see if you can reproduce the problem with the current Firefox 3.6 beta (see http://www.mozilla.com/en-US/firefox/all-beta.html) and/or nightly build of Firefox 3.7pre-alpha (see http://ftp.mozilla.org/pub/mozilla.org/firefox/nightly/latest-mozilla-central/).

(You should be able to run either of these directly from the downloaded disk image, if you don't want to install pre-release software for longer-term use.)

I think it's likely you'll see the same issue with those versions, but would appreciate confirmation of that.

    
    

    
  
ftp.mozilla.org/pub/mozilla.org/firefox/nightly/latest-mozilla-central/
Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.6; en-US; rv:1.9.3a1pre) Gecko/20100104 Minefield/3.7a1pre
• Slightly different:
◊ isin prop ang and or cup oplus lceil rceil lfloor rfloor: h2 working; p i missing.
◊ ensp emsp thins: newly broken in both h2 and p i.
◊ Everything else as before.

    
    

    
  
Sorry: spelling corrector changed thinsp to thins.

    
    

    
  
www.mozilla.com/en-US/firefox/all-beta.html
Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.6; en-US; rv:1.9.2b5) Gecko/20091204 Firefox/3.6b5
• Identical problems to those of 3.5.6 reported in comment 22.

    
    

    
  
Sorry: except that ensp emsp thinsp also newly broken.

    
    

    
  
Is it possible for a message to be sent to developers, asking whether they have access to a Mac with Office 2008 on which they would be willing to test attachment 419880 [details]? It would surely be helpful to pin down the problem to MS's fonts.

    
    

    
  
To try and confirm which fonts it's choosing on your system, could you please get the custom build of 3.6b6-prerelease that's at https://build.mozilla.org/tryserver-builds/jkew@mozilla.com-try-6f6b85b1df99/try-6f6b85b1df99-macosx.dmg, and try loading that first testcase (attachment 419785 [details]).

Then look at the "Console Messages" in Console.app; you should see a series of messages, one for each time when the browser had to search for a fallback font. Please let me know what it reports there - that may help indicate which direction to look next.

    
    

    
  
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65227)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65228)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65230)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65231)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65232)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65234)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65235)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65236)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65238)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65239)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65240)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65242)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65243)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65244)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65246)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65247)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65248)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65250)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65251)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65252)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65254)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65255)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65256)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65258)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65259)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65260)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65262)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65264)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65266)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65267)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65268)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65269)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65270)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65271)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65272)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65273)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65274)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65275)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65276)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found TimesNewRomanPSMT
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=65279)
06/01/2010 01:17:54	[0x0-0xd20d2].org.mozilla.firefox	 -> found Thonburi
06/01/2010 01:18:00	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=8657)
06/01/2010 01:18:00	[0x0-0xd20d2].org.mozilla.firefox	 -> found Menlo-BoldItalic
06/01/2010 01:18:00	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=8657)
06/01/2010 01:18:00	[0x0-0xd20d2].org.mozilla.firefox	 -> found Menlo-BoldItalic
06/01/2010 01:18:00	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=8657)
06/01/2010 01:18:00	[0x0-0xd20d2].org.mozilla.firefox	 -> found Menlo-BoldItalic
06/01/2010 01:18:00	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=8657)
06/01/2010 01:18:00	[0x0-0xd20d2].org.mozilla.firefox	 -> found Menlo-BoldItalic
06/01/2010 01:18:00	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=8657)
06/01/2010 01:18:00	[0x0-0xd20d2].org.mozilla.firefox	 -> found Menlo-BoldItalic
06/01/2010 01:18:00	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=8657)
06/01/2010 01:18:00	[0x0-0xd20d2].org.mozilla.firefox	 -> found Menlo-BoldItalic
06/01/2010 01:18:00	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=8657)
06/01/2010 01:18:00	[0x0-0xd20d2].org.mozilla.firefox	 -> found Symbol
06/01/2010 01:18:00	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=8657)
06/01/2010 01:18:00	[0x0-0xd20d2].org.mozilla.firefox	 -> found Symbol
06/01/2010 01:18:00	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=8657)
06/01/2010 01:18:00	[0x0-0xd20d2].org.mozilla.firefox	 -> found Symbol
06/01/2010 01:18:00	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=8657)
06/01/2010 01:18:00	[0x0-0xd20d2].org.mozilla.firefox	 -> found Symbol
06/01/2010 01:18:00	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=8657)
06/01/2010 01:18:00	[0x0-0xd20d2].org.mozilla.firefox	 -> found Symbol
06/01/2010 01:18:00	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=8657)
06/01/2010 01:18:00	[0x0-0xd20d2].org.mozilla.firefox	 -> found Symbol
06/01/2010 01:18:00	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=8657)
06/01/2010 01:18:00	[0x0-0xd20d2].org.mozilla.firefox	 -> found Menlo-Italic
06/01/2010 01:18:00	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=8657)
06/01/2010 01:18:00	[0x0-0xd20d2].org.mozilla.firefox	 -> found Menlo-Italic
06/01/2010 01:18:00	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=8657)
06/01/2010 01:18:00	[0x0-0xd20d2].org.mozilla.firefox	 -> found Symbol
06/01/2010 01:18:00	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=8657)
06/01/2010 01:18:00	[0x0-0xd20d2].org.mozilla.firefox	 -> found Symbol
06/01/2010 01:18:00	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=8657)
06/01/2010 01:18:00	[0x0-0xd20d2].org.mozilla.firefox	 -> found Menlo-Italic
06/01/2010 01:18:00	[0x0-0xd20d2].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=8657)
06/01/2010 01:18:00	[0x0-0xd20d2].org.mozilla.firefox	 -> found Menlo-Italic

    
    

    
  

      
      
      
      

        Attached file
          
        character log
        — 
      

    


  
Note: Using the testcase, I don't see any problems on my set up - 10.6.2

Out of curiosity, I checked what got logged on my side per comment 33 above. Output attached. This is Firefox running with a default profile, default prefs.

One thing I noticed as unexpected while checking the list, the 3rd entry for ch=8242 (that is one example, there are other occurrences in the log) uses a different font, compared to the first two. I think it corresponds to <p><b>.
The fourth entry is expected, as mplus-2p doesn't have italic faces.

1/6/10 9:28:45 AM	[0x0-0x5b05b].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=8242)
1/6/10 9:28:45 AM	[0x0-0x5b05b].org.mozilla.firefox	 -> found mplus-2p-bold

1/6/10 9:28:45 AM	[0x0-0x5b05b].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=8242)
1/6/10 9:28:45 AM	[0x0-0x5b05b].org.mozilla.firefox	 -> found mplus-2p-regular

1/6/10 9:28:45 AM	[0x0-0x5b05b].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=8242)
1/6/10 9:28:45 AM	[0x0-0x5b05b].org.mozilla.firefox	 -> found LiHeiPro

1/6/10 9:28:45 AM	[0x0-0x5b05b].org.mozilla.firefox	gfxQuartzFontCache::FindFontForChar(ch=8242)
1/6/10 9:28:45 AM	[0x0-0x5b05b].org.mozilla.firefox	 -> found TrebuchetMS-Italic

    
    

    
  
(In reply to comment #34)

The interesting part of this relates to ch=8657, the &uArr; character:

> 06/01/2010 01:18:00    [0x0-0xd20d2].org.mozilla.firefox   
> gfxQuartzFontCache::FindFontForChar(ch=8657)
> 06/01/2010 01:18:00    [0x0-0xd20d2].org.mozilla.firefox     -> found
> Menlo-BoldItalic
...etc.

(I guess the earlier messages must relate to some other page, perhaps your default start page, where font fallback also occurred.)

The results -- several occurrences of Menlo-BoldItalic, then also some Symbol and some Menlo-Italic -- exactly match the behavior I see here, where the arrow displays properly.

Could you please confirm the exact version of the Menlo font(s) that you have, by checking the Unique Name as shown in the Font Info display in Font Book? In my case, all four faces show "6.1d8e1; 2009-09-17". As this is an OS X system font, I'd expect yours to be identical, but it'd be good to check.

    
    

    
  
(In reply to comment #35)
> Created an attachment (id=420282) [details]
> character log
> 
> Note: Using the testcase, I don't see any problems on my set up - 10.6.2

OK; but note that you're getting quite different fallback behavior because of the presence of additional fonts on your system - e.g., the mplus family.

> One thing I noticed as unexpected while checking the list, the 3rd entry for
> ch=8242 (that is one example, there are other occurrences in the log) uses a
> different font, compared to the first two. I think it corresponds to <p><b>.

I'm not sure whether this is an issue; there are a number of factors that can affect fallback choices. If you think it's wrong, please try to cut it down to a minimal testcase, and file a separate bug - that would be an issue with which font is chosen, whereas the original problem here is that (AFAICT) a suitable font is being chosen, yet the character gets displayed as a hexbox (as if it were missing).

    
    

    
  
I opened the special Firefox (my first custom Firefox!), cleared the console, and the File-Open'ed the locally-stored page. So everything on that log happened as a result of opening that page. Recall that page has several failures: H1 to H6, and all the italics. 

Menlo Regular; 6.1d8e1; 2009-09-17
Menlo Italic; 6.1d8e1; 2009-09-17
Menlo Bold; 6.1d8e1; 2009-09-17
Menlo Bold Italic; 6.1d8e1; 2009-09-17
All of which seem to be the same as yours.

    
    

    
  
Also, Firefox 3.5.7, to which I have just updated, has exactly the same bug as 3.5.6.

    
    

    
  


  
Better multi-character test case for bug 537444, as rows/characters sorted by broken-ness, and as HTML passed by validator.

    
    

    
  
There follow screenshots for attachment 420619 [details] in Firefox 3.6.

Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.6; en-US; rv:1.9.2) Gecko/20100115 Firefox/3.6

    
    

    
  


  

    
    

    
  


  

    
    

    
  


  

    
    

    
  
Some changes.

Each of the following has been fixed in <h2> (hurray!) but not in <p><i> (boo!): isin prop ang and or cup oplus lceil rceil lfloor rfloor.

    
    

    
  
I've just been looking into bug 554820, and wondered if this might be a symptom of the same underlying issue. I'm marking it as "see also" rather than duplicate or dependent for now, until we see what the relationship really is.
See Also:  → 554820

    
    

    
  
Please consider adding the see-also in both directions.

    
    

    
  
Fixed in Firefox 3.6.3, thank you. Marking as Resolved.

Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.6; en-US; rv:1.9.2.3) Gecko/20100401 Firefox/3.6.3
Status: UNCONFIRMED → RESOLVED
Closed: 14 years ago
Resolution: --- → FIXED

          You need to log in
          before you can comment on or make changes to this bug.
        






  

    
  







  

    

      
Attachment

      

      
      
      
      
    

    

      

        

          

            
General

            

              Creator: 



            

            
Created: 

            
Updated: 

            
Size: