[Q] [Q] StrikeFontSet display problem for separators in 3.8

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi,

I've installed version 4 of AccuFont package and changed default font as
accujen. The problem is that if I replace the StrikeFont with
StrikeFontSet made with original font like this:

accujenArray _ (TextConstants at: #Accujen) fontArray.
a12 _ accujenArray at: 3.
af12 _ StrikeFontSet newFontArray: {a12}.
accujenArray at: 3 put: af12.

then, the separator characters like tab, cr and others displayed as
something like '?' glyph. If I revert it as StrikeFont then, it just
displayed as normal invisible glyph.

Can anyone know this problem? Can anyone help me? I'm using 3.8

Thank in advance.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.2 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFFGIbPQqspS1+XJHgRArcUAJ0TIqBstYjAoWu2jKfr9DaRj7B1+ACfYomN
BQRzFXxhIXHGUKbw3KRSqDE=
=wCjc
-----END PGP SIGNATURE-----

  Sungjin,

  I tried the same thing in 3.8.  I copied and pasted the code above
into a workspace evaluated it, and then changed the code in the
workspace to "Accujen 12", but it looks ok.

  Is there something not shown in the above code?  Are you using a
character with different language tab other than 0?  If so, you may
set up the fallbackFont instance variable.

  Sorry for slow response.  Hope this helps,

-- Yoshiki

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

I've found the source of the problem. When glyph of a character is
requested and the character's ascii value is less than font's minimum
ascii value, then ? is returned. This is why ? is displayed for tab or
other special characters.

Yoshiki Ohshima wrote:
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.2 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFFIN5pQqspS1+XJHgRAll2AJ9xkEJM0pDCN5puDvC3N5fhYE+BmgCghLHE
2/+zJZDSYyaGw4LjCHiwuqY=
=4MZI
-----END PGP SIGNATURE-----

  Sungjin,

> I've found the source of the problem. When glyph of a character is
> requested and the character's ascii value is less than font's minimum
> ascii value, then ? is returned. This is why ? is displayed for tab or
> other special characters.

  Hmm.  while it is true, the Accujen fonts in 3.8-6665 are tweaked
so that they have 0 as their minAscii.

((StrikeFont allInstances) select: [:e | e familyName beginsWith: 'Accujen']) collect: [:f | f minAscii]
  => #(0 0 0 0 0 0)

  Now I can't point exactly, but the installation script of m17n
system has expression to change the minAscii value of fonts.

(and also...)

((StrikeFont allInstances) select: [:e | e familyName beginsWith: 'Accujen']) collect: [:f | f maxAscii]
  => #(255 255 255 255 255 255)

  Can you tell me a bit more to reproduce the problem?

-- Yoshiki

 First, I've already fixed my problem with resetting minascii value of original fonts. Second,
for my image, asa my previous mail said, that's all I did for producing problem. I think original
accufont package in SqueakMap does have this problem, which I'm not sure on this. I've downloaded
Squeak 3.8 and updated to 3.8.1(?) and installed accufont package from SqueakMap.

Hope this can help you.

----- Original Message -----
   From: Yoshiki Ohshima <[hidden email]>
   To: The general-purpose Squeak developers list <[hidden email]>
   Sent: 06-10-02 18:46:15
   Subject: Re: Re: [Q] [Q] StrikeFontSet display problem for separators in 3.8

  > I've found the source of the problem. When glyph of a character is
> requested and the character's ascii value is less than font's minimum
> ascii value, then ? is returned. This is why ? is displayed for tab or
> other special characters.

  Hmm.  while it is true, the Accujen fonts in 3.8-6665 are tweaked
so that they have 0 as their minAscii.

((StrikeFont allInstances) select: [:e | e familyName beginsWith: 'Accujen']) collect: [:f | f minAscii]
  => #(0 0 0 0 0 0)

  Now I can't point exactly, but the installation script of m17n
system has expression to change the minAscii value of fonts.

(and also...)

((StrikeFont allInstances) select: [:e | e familyName beginsWith: 'Accujen']) collect: [:f | f maxAscii]
  => #(255 255 255 255 255 255)

  Can you tell me a bit more to reproduce the problem?

-- Yoshiki

 First, I've already fixed my problem with resetting minascii value of original fonts. Second,
for my image, asa my previous mail said, that's all I did for producing problem. I think original
accufont package in SqueakMap does have this problem, which I'm not sure on this. I've downloaded
Squeak 3.8 and updated to 3.8.1(?) and installed accufont package from SqueakMap.

Hope this can help you.

----- Original Message -----
   From: Yoshiki Ohshima <[hidden email]>
   To: The general-purpose Squeak developers list <[hidden email]>
   Sent: 06-10-02 18:46:15
   Subject: Re: Re: [Q] [Q] StrikeFontSet display problem for separators in 3.8

  > I've found the source of the problem. When glyph of a character is
> requested and the character's ascii value is less than font's minimum
> ascii value, then ? is returned. This is why ? is displayed for tab or
> other special characters.

  Hmm.  while it is true, the Accujen fonts in 3.8-6665 are tweaked
so that they have 0 as their minAscii.

((StrikeFont allInstances) select: [:e | e familyName beginsWith: 'Accujen']) collect: [:f | f minAscii]
  => #(0 0 0 0 0 0)

  Now I can't point exactly, but the installation script of m17n
system has expression to change the minAscii value of fonts.

(and also...)

((StrikeFont allInstances) select: [:e | e familyName beginsWith: 'Accujen']) collect: [:f | f maxAscii]
  => #(255 255 255 255 255 255)

  Can you tell me a bit more to reproduce the problem?

-- Yoshiki