[pdftex] writet1, part n for large n

[Tom Kacvinsky]

Hi,

On Wed, 27 Jun 2001, Y&Y Support wrote:

> At 13:36 2001-06-26 -0400, Tom Kacvinsky wrote:
>
> >Hans' test files points out another bug in Acrobat 5.0.  The problem is
> >when one uses pdftex to include another PDF file.  The font resources
> >are scanned over, and sometimes a Type 1 font is written into the PDF
> >file with an encoding array that looks like this:
> >
> >/Enoding 256 array
> >0 1 255 {1 index /.notdef exch put} for
> >readonly def
> >
> >Acrobat 5.0 does not like this.  I will try to come up with a fix this
> >weekend.
>
> Err, what is the problem with that --- other than dyslexia :-):
>

Doh!

> /Encoding 256 array
> 0 1 255 {1 index exch /.notdef put} for
> ...
> dup 49 /one put
> ...
> readonly def
>

Nope, don't have any encoded glyphs (dup 49 /one put).  We just have

  /Encoding 256 array
  0 1 255 {1 index exch /.notdef put} for
  readonly def

In the Type 1 font embedded in the PDF file.

As I have stated many a time, Acrobat 5.0 is disassembling embedded Type 1
fonts and reassembling them to "look nice".  I guess it doesn't like the
fact the the embedded font has not encoded glyphs (if one were to go off of
the array stored in the PFA in the FontFile stream).

Now, I suppose that Adobe will cite section 10.3, the paragraph on tokens
following /Encoding as the reason why this font is not correct.  Even
though it is. :(

Regards,

Tom