[DejaVu-bugs] [Bug 14084] Some erroneous glyphs in DejaVu Sans Mono 2.21

[bugzilla-daemon at freedesktop.org]

http://bugs.freedesktop.org/show_bug.cgi?id=14084





--- Comment #3 from Adrian Bocaniciu <a.bocaniciu at computer.org>  2008-01-16 03:08:52 PST ---
(In reply to comment #0)
> After I have installed DejaVu version 2.21 I have discovered some characters
> that have erroneous glyphs in "DejaVu Sans Mono".  They are the following:
> 
> 1. "U+00AC NOT SIGN", correct in DejaVu Serif and DejaVu Sans
> 2. "U+21D2 RIGHTWARDS DOUBLE ARROW", correct in DejaVu Serif and DejaVu Sans
> 3. "U+00B6 PILCROW SIGN", correct in DejaVu Serif and DejaVu Sans
> 
> 
> There might be still other errors, but I have noticed these three immediately,
> because I use very often these characters.
> 

I am using the DejaVu fonts in OpenOffice 2.3.1 and I use only Unicode
encodings in all my documents.  I have never seen any problems in other fonts,
so I am pretty sure that a bug in OpenOffice  or other system components is
highly unlikely. The computer has been rebooted several times after the
installation of the DejaVu 2.21, so a desktop restart problem is excluded.  Now
I have noticed a fourth character with a wrong glyph:

4. "U+00A7 SECTION SIGN", this has the correct glyph only in DejaVu Sans.  In
both Serif and Mono, it shows a German "scharfes es" instead of the section
sign.  The double arrow mentioned above is double only in Serif and Sans and it
is simple in Mono

However, this bug might be much more subtle than a simple code point
misassignment, because I have discovered that if I use the "insert special
character" dialog box and I cycle through various fonts and character blocks,
when I return to one of the DejaVu fonts, the result is not consistent, i.e.
sometimes I obtain a different map from code points to glyphs than the previous
time when I have selected the same DejaVu font.  This does not happen with any
other font, neither with the standard OpenOffice fonts nor with Microsoft
fonts, regardless how many times I change the font or the character block, no
desynchronization between code points and glyphs is displayed.

Using KWrite instead of OpenOffice, I have also obtained inconsistent results,
i.e. even if some characters are also displayed with wrong glyphs, they are not
exactly the same that are wrong in OpenOffice, so I am totally confused now.

Unfortunately, I do not have at this moment enough time to do exhaustive
experiments to understand what is really happening.


-- 
Configure bugmail: http://bugs.freedesktop.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.