Re: postscript printing from emacs

[stromme@mi.uib.no (Stein Arild Strømme)]

Peter Dyballa <Peter_Dyballa@Web.DE> writes:

> Am 23.03.2007 um 16:41 schrieb Stein Arild Strømme:
>
>> I also find it suspicious that Preview.app in OS X cannot display
>> the file correctly (it does with other postscript files).
>
> I reported this error before. Ps-print is pretty useless, in any case
> in Mac OS X. Besides this it only supports one or two encodings: US 7- 
> bit ASCII and ISO 8859-1.

I should have stated more about my situation: I'm running Linux in a
centralized university environment.  Printing is via a cups server
which in turn talks to a Windows print server common for the whole
university.  My mention of Mac OS X is peripheral to the issue, it was
just a test naively trying to narrow down the search for the culprit.
Involving a Mac is no solution in any case.

> From the failure with Preview you can deduce that the Mac OS X and
> CUPS related mechanisms to prepare the PS output for printing are the
> reason that your printer reports a failure. It's nothing but non- 
> sense that gets into the printer queue. (You can halt the queue and
> inspect what's in it!)

Unfortunately, I have absolutely no control of the queue at all.

>> So what is it about the ps-print-generated postscript files fools the
>> printer and Preview.app, that's the question.
>
> I presume it's like MSIE "optimised" web sites: only Ghostscript can
> convert and only Ghostscript can display. So the function family
> should better, before release of GNU Emacs 22.1, be renamed gs-print.

OK, but both a2ps and enscript *do* work, in so far as the buffers get
printed, and even correctly, if encoded in 8859-1.  ps-print gives
nothing, even on the simplest of buffers.

>> | Use htmlize to convert the buffer or region content to UTF-8 encoded
>> | HTML, view it in a capable browser, print from the browser.
>>
>> That is tongue-in-cheek, surely?
>
> No. It's a proven way. And it's the only way a few members on this
> list could find. 

OK, I have no doubts that it works, but it is too cumbersome.  

I suppose it is a luxury problem for me, I don't *really* need utf-8
(I can get by with 8859-1), I simply want to try to migrate to utf-8.
What you are saying below about postscript and utf-8 gives me second
thoughts indeed.

> The problem with PostScript is that it usually only supports 8 bit
> encodings. So you can't print Unicode. You can try to find a CID
> keyed PostScript font and with some effort you can print CJK. No such
> font is available for free or public domain. TrueType and OpenType
> fonts have better Unicode support. Modern PostScript can use such
> TrueType fonts, which are quite often free and acceptable quality.
> But I have no idea how to create a mapping from an UTF-8 encoded text
> to a Unicode encoded font in PostScript.

It seems that I have two distinct problems: 1) ps-print does not work
at all, even for straight ascii, and 2) nothing simpler than your
solution works for utf-8 encoded files.

I'd still like to understand 1) better.

>
> --
> Greetings
>
>   Pete

Thanks for informative posts!

SA