Re: [ft] Segfault in FT_Get_Name_Index, Solaris 9, FreeType 2.1.8+

2006-05-24 Thread Jan-Åke Larsson
Jan-Åke Larsson wrote:
 Update: Just tried compiling the library with SUN cc (Workshop 11). I no
 longer get a segfault when linking my program to ft 2.1.8.

Scratch that. It works with gcc too. Today. Perhaps it is the phase of
the Moon or something, I give up.

/JÅ


-- 
The box said Windows 95, Windows NT 4.0, or better, so I installed Linux.


___
Freetype mailing list
Freetype@nongnu.org
http://lists.nongnu.org/mailman/listinfo/freetype


[Fwd: Re: [ft] Segfault in FT_Get_Name_Index, Sola...]

2006-05-22 Thread Jan-Åke Larsson
I'd suggest you remove this individual from the freetype list.
Obviously, he doesn't want mail from it.
/JÅ

 Original Message 
Subject: Re: [ft] Segfault in FT_Get_Name_Index, Sola...
Date: Mon, 22 May 2006 15:38:40 +0200 (MEST)
From: [EMAIL PROTECTED], Thibault Jamme@mai.liu.se
To: [EMAIL PROTECTED]



Hi,

You have just sent me, Thibault Jamme, a message at:
[EMAIL PROTECTED]

I am now using an anti-spam system called NIKITA,
which will help me verify that you are a real person
and not a computer program sending automated, unsolicited,
advertising messages (Spam).

To carry out this verification, just click here -
:  http://www.nikitamail.com/cgi-bin/nikita?x=84172.UQUCMHVDFODD

and follow the simple steps explained on the page.

If you don't go to this web page, I will never see your
e-mail and your message will be deleted automatically
after 10 days.

You only need to do this once for this address.


IMPORTANT:
-

If you have never sent a message to [EMAIL PROTECTED]
and don't know me at all then it is possible
someone is using your identity to send unsolicited
e-mails. In this case, ignore this message and the
reminders that will follow for a few days.


Thank you for helping to fight e-mail abuse.

Thibault Jamme


___
Freetype mailing list
Freetype@nongnu.org
http://lists.nongnu.org/mailman/listinfo/freetype


Re: [ft] In freetype how can one set font size in pt, but keep resolution independent?

2006-05-18 Thread Jan-Åke Larsson
Werner LEMBERG wrote:
 Well, the very problem is that neither TrueType nor PS fonts have the
 concept of a `design size'.  This is something from the TeX world.

Correct. In the TeX world, a font designed for 5pt output is visibly
different from the same font as designed for 10pt. For instance Computer
Modern (TeX's standard font) comes in no less than eight design sizes
(5,6,7,8,9,10,12,17). If the effect you are after is that of choosing
one different from the usual cmr10, note that on a modern system (say
teTeX-3) Bluesky's Type1 fonts come in all these varieties also.

I have written a png driver for the dvi format and I used to worry about
this too, design sizes and suchlike. I stopped worrying once I noted
that in the dvi, the correct design sizes were specified as part of the
font selection. Relevant parts of a test run are:

@114 DRAW CMD:  FNT_DEF1 10 195060286 655360 655360 0 6 'cmmi10'
  FONT 10:  New entry created
@136 DRAW CMD:  FONT_10
  FIND FONT:cmmi10 400
  OPEN METRICS:
'/sw/tex/teTeX/3.0/texmf-dist/fonts/tfm/public/cm/cmmi10.tfm'
  OPEN FT FONT:
'/sw/tex/teTeX/3.0/texmf-dist/fonts/type1/bluesky/cm/cmmi10.pfb'



___
Freetype mailing list
Freetype@nongnu.org
http://lists.nongnu.org/mailman/listinfo/freetype


[ft] Segfault in FT_Get_Name_Index, Solaris 9, FreeType 2.1.8+

2006-05-17 Thread Jan-Åke Larsson
I get a segfault (or bus error, sometimes) from FT_Get_Name_Index on
Solaris 9, with FreeType 2.1.8, 2.1.9 and 2.2.1. FreeType 2.1.7 works
fine (but there, FT_LOAD_TARGET_LIGHT does not work)

Steps to reproduce: Load font, map char name d to char index.
Sometimes the map returns index 0, sometimes you get:

Program received signal SIGSEGV, Segmentation fault.
0xff319318 in FT_Get_Name_Index (face=0x1ed858, glyph_name=0x33548 d)
at /home/jalar/build/freetype-2.1.8/src/base/ftobjs.c:2378
2378  if ( service  service-name_index )


I build 2.1.7 (working) and 2.1.8 (SIGSEGV) exactly the same way:
./configure --prefix=/home/jalar; gmake (uses gcc)

Am I missing some advice for Solaris here?
/Jan-Åke



___
Freetype mailing list
Freetype@nongnu.org
http://lists.nongnu.org/mailman/listinfo/freetype


Re: [ft] Segfault in FT_Get_Name_Index, Solaris 9, FreeType 2.1.8+

2006-05-17 Thread Jan-Åke Larsson
David Turner wrote:
 I don't have a Solaris 9 machine here, and won't be able to due much here.
 Given the line were the segfault is detected, I suppose that the value
 of the 'service' variable is bogus (and not NULL).

Yes, but might be overwritten somewhere else, no?

I don't suppose that the changes from 2.1.7 to 2.1.8 are small enough to
give a hint? No, didn't think so.

 In GDB, could you dump the content of  and send it
 to me, this would be a good start.

I don't really know gdb well but I took a stab and did

  dump value services face-internal-services

The result is attached.


About optimization on/off, you mean in my code?

/JÅ


services
Description: Binary data
___
Freetype mailing list
Freetype@nongnu.org
http://lists.nongnu.org/mailman/listinfo/freetype


Re: [ft] Segfault in FT_Get_Name_Index, Solaris 9, FreeType 2.1.8+

2006-05-17 Thread Jan-Åke Larsson
Ian Brown wrote:
 On the SPARC architecture, you can only access 16 bit values on 16
 bit address boundaries, 32 bit values on 32 bit address boundaries
 etc.

Yep: But dereferencing a bogus pointer would sometimes cause a bus
error, sometimes a segfault, I'd imagine.

/JÅ


-- 
ceci n'est pas une .signature


___
Freetype mailing list
Freetype@nongnu.org
http://lists.nongnu.org/mailman/listinfo/freetype