Re: scs2ps

Cc'ed to linux5250@midrange.com...

On Tue, 12 Mar 2002, Tony A. Lambley wrote:

>            "\t\t/Type /Font\n"
>            "\t\t/Subtype /Type1\n"
>            "\t\t/Name /F1\n"
>            "\t\t/BaseFont /Courier\n"
>            "\t\t/Encoding /MacRomanEncoding\n" "\t>>\n" "endobj\n\n", objnum);
>
>
> That `MacRomanEncoding', hmm, if it's what I think it is, it may cause some
> character mapping issues. I've had problems when dealing with apple mac
> peeps and their mac-fied postscript fonts. But it may be something else and
> I'm just ringing alarm bells because it has `Mac' and `Encoding' in the
> string.
>
> Just looked at the PDF specs and there isn't exactly much choice for this
> encoding value is there! Not worth worring about then.

Right.  I don't know that there are other choices.  MacRomanEncoding works
fine so far so I'm not worried about it.  But in the future it would be
nice to make font selection user-definable.

> Have you printed a 16x16 character grid in your testing yet? I'm not in the

Nope, but that might be a good test.  I probably won't have time for it
soon, though.  Maybe when you return you can try it?

> I don't know enough about pdf to know how it handles non-latin1 fonts. My
> experience of doing pdf reports for the likes of Poland, Greece, Turkey and
> Russia says that you have to embed fonts encoded for these different
> character sets. But I'm starting from postscript so it could be different.
> I'm sure it was an issue with the older versions of ghostscript. Apparently
> gs does things by the book now, so when I use non-latin1, it always embeds my
> custom fonts.

Yikes - non-latin1 characters!  But I think it should work fine.  The most
likely problem is that we map EBCDIC to ASCII wrong for non-latin1.  But I
think that tn5250_char_map_to_local() handles everything for us.  Could
someone who uses special characters give this a try?  I remember reading
about handling special characters in a PDF, but it might already work
fine.

James Rich
james@eaerich.com