Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.postscript > #3832

Re: encoding vectors: different ones returned from GS vs Adobe Distiller

Show all headers | View raw

In article <cdc04d05-9855-40e1-8dde-9c26dafe89c7n@googlegroups.com>, 
douglas.mcallaster@gmail.com says...
> 
> Folks,
> 
> The encoding vectors for fonts are different in Adobe and Ghostscript.
> I ran this postscript on Adobe (Distiller) and then Ghostscript:
> /LucidaBright 12 selectfont
> currentfont 
> /FontName get =
> currentfont 
> /Encoding get
> dup
> type =
> {=} forall
> 
> Below are a couple of the many differences in the returned vector:
> (1) bullet is not in the GS vector 
> (2) germandbls is in both vectors but in position 224 in GS and 252 in Abobe 

Which tells you that Adobe has used StandardEncoding and Ghostscript has 
used ISOLatin1Encoding. Assuming the actual positions are decimal 223 
and 251 that is, the first element  in the Encoding array is 0.

According to the Latin character set, bullet is present in 
StandardEncoding but not in ISOLatin1Encoding.

 
> I thougth the above postscript code would read the vector from the font itself.
> Thus, I expected the same vector from either distiller and GS.

There is no requirement that fonts should have a specific Encoding, if 
you need a specific Encoding for a font then you are expected to apply 
it yourself.

Beyond that, LucidaBright is not one of the base 35 fonts, so both 
interpreters will be either:

1) Using a substitute font
or
2) Using a system font.

Exactly which will depend on the OS you are running and the fonts 
installed, as well as how you have Ghostscript configured.

If they are using substitute fonts then they could easily be using 
different substitutions (eg Helvetica instead of Courier).

One not unlikely scenario is that you are on Windows, Ghostscript has 
been installed so that it can use the Windows/Fonts directory and 
LucidaBright is available there as a TrueType font.

In that case both Ghostscript and Acrobat Distiller willl load the 
TrueType font and return that. 

But.....

PostScript doesn't permit the use of naked TrueType fonts, they should 
be Type 42 fonts. In order to load a TT font both interpreters must 
create the missing information which includes the Encoding vector. Since 
there is no specification for loading a TT font, there is no defined way 
this should be done.

So its no real surprise that they should differ.

 
> Is there a way to get the same vector?

Yes, encode the font the way you want it.


> Otherwise, using the (octal) position code to show a glyph will return different glyph if the postscript is run under GS verses Adobe Distiller.

You can validate the Encoding before you start, to see if it's what you 
want. You can also check the CharStrings dictionary of the font to see 
if it includes the glyphs you want and encode the font the way you want 
it.

But you shouldn't really be using random fonts off disk. If you are 
using a font other than the base 35 standard fonts then you should embed 
the font in the PostScript program.

This is why PostScript programs generally contain the fonts they need.



					Ken

Back to comp.lang.postscript | Previous | Next — Previous in thread | Find similar

Thread

encoding vectors: different ones returned from GS vs Adobe Distiller douglas mcallaster <douglas.mcallaster@gmail.com> - 2023-02-21 13:26 -0800
  Re: encoding vectors: different ones returned from GS vs Adobe Distiller ken <ken@spamcop.net> - 2023-02-22 08:25 +0000

csiph-web