Two problems about CJK fonts with the beta (2010.05.22)

Two problems about CJK fonts with the beta (2010.05.22)

[Li Yanrui (李延瑞)]

[-- Attachment #1: Type: text/plain, Size: 428 bytes --]

Hi,

1. The characters overlaps each other and can not be set size.

The is a mini example.

\definefont[song][name:simsun at 12pt]
\starttext

\song 测试

\stoptext

The attachment is its output.

2. the cid fonts cause MkIV gives a waring as following:

  LuaTeX warning: lua-loaded font [44]
(/usr/share/fonts/adobe/AdobeSongStd-Light) has no characters!


-- 
Best regards,

Li Yanrui (李延瑞)

[-- Attachment #2: a.pdf --]
[-- Type: application/pdf, Size: 8832 bytes --]

[-- Attachment #3: Type: text/plain, Size: 486 bytes --]

___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://tex.aanhet.net
archive  : http://foundry.supelec.fr/projects/contextrev/
wiki     : http://contextgarden.net
___________________________________________________________________________________

Re: Two problems about CJK fonts with the beta (2010.05.22)

[Taco Hoekwater]

Hi,

Li Yanrui (李延瑞) wrote:
> Hi,
> 
> 1. The characters overlaps each other and can not be set size.
> 
> The is a mini example.
> 
> \definefont[song][name:simsun at 12pt]
> \starttext
> 
> \song 测试
> 
> \stoptext
> 
> The attachment is its output.

Verified.

This actually looked fine here with the current 2010.05.21 11:10,
but the beta 2010.05.22 12:06 MKIV is broken, producing the same
output you have.

> 2. the cid fonts cause MkIV gives a waring as following:
> 
>   LuaTeX warning: lua-loaded font [44]
> (/usr/share/fonts/adobe/AdobeSongStd-Light) has no characters!

Verified, same here, even with the official current.

\definefont[song][name:adobesongstd-light at 12pt]
\starttext

\song 测试

\stoptext


Best wishes,
Taco

___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://tex.aanhet.net
archive  : http://foundry.supelec.fr/projects/contextrev/
wiki     : http://contextgarden.net
___________________________________________________________________________________

Re: Two problems about CJK fonts with the beta (2010.05.22)

[Taco Hoekwater]

Taco Hoekwater wrote:
> 
> Hi,
> 
> Li Yanrui (李延瑞) wrote:
>> Hi,
>>
>> 1. The characters overlaps each other and can not be set size.
>>
>> The is a mini example.
>>
>> \definefont[song][name:simsun at 12pt]
>> \starttext
>>
>> \song 测试
>>
>> \stoptext
>>
>> The attachment is its output.
> 
> Verified.
> 
> This actually looked fine here with the current 2010.05.21 11:10,
> but the beta 2010.05.22 12:06 MKIV is broken, producing the same
> output you have.

Some debugging applied. This line in font-otf.lua:

   tfm.format = (metadata.order2 == 1 and 'truetype') or 'opentype'

produces 'opentype' for simsun.ttf, and so all glyphs are printed
at 2.048 times the normal size in the PDF output.

This is because the key metadata.order2 doesn't exist _anywhere_ in
the context source (except in this test). But neither did it exist in
the official current. The context current did tfm.format later.

Patch to revert back to the current's behaviour:

--- font-otf.lua~ 2010-05-23 10:20:10.979586135 +0200
+++ font-otf.lua  2010-05-23 10:31:14.153587544 +0200
@@ -1653,6 +1652,13 @@
          local otfdata = tfmtable.shared.otfdata
          tfmtable.name = specification.name
          tfmtable.sub = specification.sub
+        if otfdata.metadata.order2 == 0 then
+            tfmtable.format = 'opentype'
+        elseif otfdata.metadata.order2 == 1 then
+            tfmtable.format = 'truetype'
+        else
+            tfmtable.format = specification.format
+        end
          local s = specification.size
          local m = otfdata.metadata.math
          if m then


>> 2. the cid fonts cause MkIV gives a waring as following:
>>
>>   LuaTeX warning: lua-loaded font [44]
>> (/usr/share/fonts/adobe/AdobeSongStd-Light) has no characters!
> 
> Verified, same here, even with the official current.

Something is broken in the handling of the 'subfonts' key, but I haven't
figured out what (yet?).

Best wishes,
Taco

___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://tex.aanhet.net
archive  : http://foundry.supelec.fr/projects/contextrev/
wiki     : http://contextgarden.net
___________________________________________________________________________________

Re: Two problems about CJK fonts with the beta (2010.05.22)

[Taco Hoekwater]

Taco Hoekwater wrote:
>>> 2. the cid fonts cause MkIV gives a waring as following:
>>>
>>>   LuaTeX warning: lua-loaded font [44]
>>> (/usr/share/fonts/adobe/AdobeSongStd-Light) has no characters!
>>
>> Verified, same here, even with the official current.
> 
> Something is broken in the handling of the 'subfonts' key, but I haven't
> figured out what (yet?).

The 'merge cid fonts' enhancer is broken. It starts by testing
data.glyphs, but this is the field it is supposed to fill (!).

The patch is a one-liner:

--- font-otf.lua~ 2010-05-23 10:20:10.979586135 +0200
+++ font-otf.lua  2010-05-23 10:39:51.954338034 +0200
@@ -576,7 +576,7 @@
  otf.enhancers["merge cid fonts"] = function(data,filename)
      -- we can also move the names to data.luatex.names which might
      -- save us some more memory (at the cost of harder tracing)
-    if data.subfonts and data.glyphs and next(data.glyphs) then
+    if data.subfonts then
          local cidinfo = data.cidinfo
          local verbose = fonts.verbose
          if cidinfo.registry then
___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://tex.aanhet.net
archive  : http://foundry.supelec.fr/projects/contextrev/
wiki     : http://contextgarden.net
___________________________________________________________________________________

Re: Two problems about CJK fonts with the beta (2010.05.22)

[Khaled Hosny]

On Sun, May 23, 2010 at 10:34:30AM +0200, Taco Hoekwater wrote:
> Taco Hoekwater wrote:
> >
> >Hi,
> >
> >Li Yanrui (李延瑞) wrote:
> >>Hi,
> >>
> >>1. The characters overlaps each other and can not be set size.
> >>
> >>The is a mini example.
> >>
> >>\definefont[song][name:simsun at 12pt]
> >>\starttext
> >>
> >>\song 测试
> >>
> >>\stoptext
> >>
> >>The attachment is its output.
> >
> >Verified.
> >
> >This actually looked fine here with the current 2010.05.21 11:10,
> >but the beta 2010.05.22 12:06 MKIV is broken, producing the same
> >output you have.
> 
> Some debugging applied. This line in font-otf.lua:
> 
>   tfm.format = (metadata.order2 == 1 and 'truetype') or 'opentype'
> 
> produces 'opentype' for simsun.ttf, and so all glyphs are printed
> at 2.048 times the normal size in the PDF output.
> 
> This is because the key metadata.order2 doesn't exist _anywhere_ in
> the context source (except in this test). But neither did it exist in
> the official current. The context current did tfm.format later.
> 
> Patch to revert back to the current's behaviour:
> 
> --- font-otf.lua~ 2010-05-23 10:20:10.979586135 +0200
> +++ font-otf.lua  2010-05-23 10:31:14.153587544 +0200
> @@ -1653,6 +1652,13 @@
>          local otfdata = tfmtable.shared.otfdata
>          tfmtable.name = specification.name
>          tfmtable.sub = specification.sub
> +        if otfdata.metadata.order2 == 0 then
> +            tfmtable.format = 'opentype'
> +        elseif otfdata.metadata.order2 == 1 then
> +            tfmtable.format = 'truetype'
> +        else
> +            tfmtable.format = specification.format
> +        end
>          local s = specification.size
>          local m = otfdata.metadata.math
>          if m then

I'm wondering if the 'if' statment is even necessary, metadata.order2
doesn't exist neither in context nor in luatex generated tables, so the
above statment always resolve to 'tfmtable.format = specification.format'.
-- 
 Khaled Hosny
 Arabic localiser and member of Arabeyes.org team
 Free font developer
___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://tex.aanhet.net
archive  : http://foundry.supelec.fr/projects/contextrev/
wiki     : http://contextgarden.net
___________________________________________________________________________________