🚀 go-pugleaf

I was using a standard example of composite font definition as follows...

      /LetterFont
      <<
      /FontType 0
      /FontMatrix [1 0 0 1 0 0]
      /FDepVector
      [
      >/Times-Roman findfont
      >/Times-Bold findfont
      >/Times-Italic findfont
      >/Times-BoldItalic findfont
      ]
      /Encoding [0 1 2 3]
      /FMapType 3
      >> definefont pop

      As it turns out, a couple hours of researching the font encoding and definitions,
      I finally landed upon the ever-so-subtle discovery that after an escape/offset
      sequence like "/337/001", postscript printers don't like spaces. If there
      is not a visible character after this escape/offset, postscript will attempt
      to interpret them as characters, which led it to the ISO vector, which
      output '�' instead of escaping to the composite font.





So programmers beware, keep your composite font escapes followed tightly by actual text. White space is bad.

Example:
/LetterFont 11 selectfont
(Your mom\377\001 likes\377\000 spam.) show

Error:rangecheck
Offending command: stringwidth
Stack:
mom�
wordwidth
....(and so on)
=====================================
THE CORRECT WAY

/LetterFont 11 selectfont
(Your mom \377\001likes \377\000spam.) show


Your mom likes spam.