nanogui: Thread: displaying latin1, latin2 or european Unicode characters.

I would like to ask whether UNICODE support in microwindows (nanoX) covers
European latin1, latin2 and russian characters? We have used the atached
patch to get latin1 support with T1-1.0/Fonts/enc/IsoLatin1.enc but how to
get also latin2 and other characters in the same program/window ?

I can accept the need to write all input text in unicode, but how to match
unicode characters to the fonts?

diff -bur microwin-0.88pre11/src/engine/devfont.c ../microwindows-0.88pre11/src/engine/devfont.c
--- microwin-0.88pre11/src/engine/devfont.c	Tue May 23 20:16:11 2000
+++ ../microwindows-0.88pre11/src/engine/devfont.c	Thu Dec 21 16:09:02 2000
@@ -583,6 +583,7 @@
 t1lib_init(PSD psd)
 {
 	static int inited = 0;
+	char **encoding;/* m.r.*/
 
 	if (inited)
 		return 1;
@@ -590,6 +591,13 @@
 	T1_SetBitmapPad(8);
 	if (!T1_InitLib(0))
 		return 0;
+
+	/* m.r. */
+	encoding = T1_LoadEncoding ("IsoLatin1.enc");
+	T1_SetDefaultEncoding (encoding);
+	/* end m.r. */
+
+
 #ifdef T1LIB_USE_AA_HIGH	 
 	T1_AASetLevel(T1_AA_HIGH);
 #else

I have tried to add latin2 characters to ftdemo.txt and run it in
microwindows-0.89pre6b but unfortunately it did not work - only latin1
characters were displaied. I have edited the file with "yudit" using
"european" font selection. 

Best regards,
--
Tomasz Motylewski
BFAD GmbH & Co.KG http://www.bfad.de/

gEût été ôté de là ...aććaąĆŹŚśżŻęĘł�

: I would like to ask whether UNICODE support in microwindows (nanoX) covers
: European latin1, latin2 and russian characters? We have used the atached
: patch to get latin1 support with T1-1.0/Fonts/enc/IsoLatin1.enc but how to
: get also latin2 and other characters in the same program/window ?

I assume that you're running using GrText(..., MWTF_UC16);
It seems we may need to load various truetype/T1 fonts w/separate
decoding .enc files, something that we might need an additional API
for.  Try loading a different .enc file for each of the fonts that
you are trying to display.


:
: I can accept the need to write all input text in unicode, but how to match
: unicode characters to the fonts?

Isn't that what the .enc file does?  It's my understanding that you
ask T1lib for a glyph using a unicode index.  Certainly Microwindows
carries the unicode index down to that level.

Regards,

Greg



:
: diff -bur microwin-0.88pre11/src/engine/devfont.c
../microwindows-0.88pre11/src/engine/devfont.c
: --- microwin-0.88pre11/src/engine/devfont.c Tue May 23 20:16:11 2000
: +++ ../microwindows-0.88pre11/src/engine/devfont.c Thu Dec 21 16:09:02 2000
: @@ -583,6 +583,7 @@
:  t1lib_init(PSD psd)
:  {
:   static int inited = 0;
: + char **encoding;/* m.r.*/
:
:   if (inited)
:   return 1;
: @@ -590,6 +591,13 @@
:   T1_SetBitmapPad(8);
:   if (!T1_InitLib(0))
:   return 0;
: +
: + /* m.r. */
: + encoding = T1_LoadEncoding ("IsoLatin1.enc");
: + T1_SetDefaultEncoding (encoding);
: + /* end m.r. */
: +
: +
:  #ifdef T1LIB_USE_AA_HIGH
:   T1_AASetLevel(T1_AA_HIGH);
:  #else
:
: I have tried to add latin2 characters to ftdemo.txt and run it in
: microwindows-0.89pre6b but unfortunately it did not work - only latin1
: characters were displaied. I have edited the file with "yudit" using
: "european" font selection.
:
: Best regards,
: --
: Tomasz Motylewski
: BFAD GmbH & Co.KG http://www.bfad.de/
:
:


--------------------------------------------------------------------------------


: gEût été ôté de là ...aÄ?Ä?aÄ.Ä?ŹÅsÅ>żŻÄTÄ~Å,Å�
:


--------------------------------------------------------------------------------


: ---------------------------------------------------------------------
: To unsubscribe, e-mail: ####@####.####
: For additional commands, e-mail: ####@####.####

On Tue, 16 Jan 2001, Greg Haerr wrote:

> I assume that you're running using GrText(..., MWTF_UC16);
> It seems we may need to load various truetype/T1 fonts w/separate
> decoding .enc files, something that we might need an additional API
> for.  Try loading a different .enc file for each of the fonts that
> you are trying to display.

OK, this is second step. First step is that if I want to use GrText(...,
MWTF_ASCII) Latin1 encoding is not loaded. May I do direct T1lib calls
from the client?

> : I can accept the need to write all input text in unicode, but how to match
> : unicode characters to the fonts?
> 
> Isn't that what the .enc file does?  It's my understanding that you
> ask T1lib for a glyph using a unicode index.  Certainly Microwindows
> carries the unicode index down to that level.

The problem is T1lib does not accept unicode indices. It accepts 0-255
index and uses currently loaded encoding to display it. So to do it
correctly Microwindows would need to have a map of "unicode-index  : font
and encoding to use in T1" and reconfigure T1 dynamically. Or just load
fonts with different encoding as different fonts.

Best regards,
--
Tomasz Motylewski
BFAD GmbH & Co.KG http://www.bfad.de/

P.S. I have seen GdConvertEncoding
I would not call 

*ostr8++ = (unsigned char)ch; 
a "UNICODE suport".

: > I assume that you're running using GrText(..., MWTF_UC16);
: > It seems we may need to load various truetype/T1 fonts w/separate
: > decoding .enc files, something that we might need an additional API
: > for.  Try loading a different .enc file for each of the fonts that
: > you are trying to display.
:
: OK, this is second step. First step is that if I want to use GrText(...,
: MWTF_ASCII) Latin1 encoding is not loaded. May I do direct T1lib calls
: from the client?

Not easily, unless you use LINK_APP_WITH_SERVER=Y


:
: > : I can accept the need to write all input text in unicode, but how to
match
: > : unicode characters to the fonts?
: >
: > Isn't that what the .enc file does?  It's my understanding that you
: > ask T1lib for a glyph using a unicode index.  Certainly Microwindows
: > carries the unicode index down to that level.
:
: The problem is T1lib does not accept unicode indices. It accepts 0-255
: index and uses currently loaded encoding to display it. So to do it
: correctly Microwindows would need to have a map of "unicode-index  : font
: and encoding to use in T1" and reconfigure T1 dynamically. Or just load
: fonts with different encoding as different fonts.

I think the fix here is to always load a .enc file, like you patch
suggested, and then basically T1lib look like it supports unicode
indices.  I would suggest that we patch Microwindows to always
attempt to load a matching .enc file, and then if MWTF_UC16 is specified,
to use unicode indices.

Regards,

Greg