article printing with `A P' is generating bad postscript

article printing with `A P' is generating bad postscript

[Karl Kleinpaste]

I was trying to print a couple articles, and finally had to resort to
a manual execution of (gnus-summary-print-article "/tmp/out.ps") in
order to get a postscript file I could look at.  Executing "gv
/tmp/out.ps" rewards me with this error.
________________

Error: /typecheck in --transform--
Operand stack:
   h0   --nostringval--   --nostringval--
Execution stack:
   %interp_exit   .runexec2   --nostringval--   --nostringval--   --nostringval--   2   %stopped_push   --nostringval--   --nostringval--   --nostringval--   false   1   %stopped_push   1   3   %oparray_pop   1   3   %oparray_pop   .runexec2   --nostringval--   --nostringval--   --nostringval--   2   %stopped_push   --nostringval--   --nostringval--   --nostringval--   --nostringval--
Dictionary stack:
   --dict:1038/1476(ro)(G)--   --dict:0/20(G)--   --dict:138/200(L)--   --dict:13/19(L)--
Current allocation mode is local
GNU Ghostscript 6.52: Unrecoverable error, exit code 1
________________

Am I looking at...
- a Gnus error?
- a ghostscript error?
- an XEmacs ps-print error?

I'm running a very recent Gnus, XEmacs 21.4.8, and the very latest
ghostscript RPM from RedHat.

Re: article printing with `A P' is generating bad postscript

[Jesper Harder]

karl@charcoal.com (Karl Kleinpaste) writes:

> I was trying to print a couple articles, and finally had to resort to
> a manual execution of (gnus-summary-print-article "/tmp/out.ps") in
> order to get a postscript file I could look at.  Executing "gv
> /tmp/out.ps" rewards me with this error.
>
> Am I looking at...
> - a Gnus error?
> - a ghostscript error?
> - an XEmacs ps-print error?
>
> I'm running a very recent Gnus, XEmacs 21.4.8, and the very latest
> ghostscript RPM from RedHat.

I'm using Ghostscript 7.04 and I don't get any errors, FWIW.

Re: article printing with `A P' is generating bad postscript

[Reiner Steib]

On Tue, Jun 11 2002, Karl Kleinpaste wrote:

> I was trying to print a couple articles, and finally had to resort to
> a manual execution of (gnus-summary-print-article "/tmp/out.ps") in
> order to get a postscript file I could look at.  Executing "gv
> /tmp/out.ps" rewards me with this error.
> ________________
>
> Error: /typecheck in --transform--
> Operand stack:
>    h0   --nostringval--   --nostringval--
> Execution stack:
[...]
> Current allocation mode is local
> GNU Ghostscript 6.52: Unrecoverable error, exit code 1

I had problems some time ago with printing articles, when the subject
contained unbalanced brackets, e.g. "Re: foo :-)".

Look at the lines containing "h0" in out.ps. The problem should be
there. Maybe post the output of "grep -2 'h0' out.ps". "h0" is used to
print the subject and page numbers.

If this fails, remove the error handler from out.ps:

%%BeginProcSet: ErrorHandler
[...]
%%EndProcSet

And run "gs out.ps". You should see something like:

| Error: ...
| ...
| Current allocation mode is local
| Current file position is 27395
                           ^^^^^

Open the file in Emacs. `M-x goto-char 27395 RET' brings you to the
position of the error.

Bye, Reiner.
-- 
       ,,,
      (o o)
---ooO-(_)-Ooo--- PGP key available via WWW   http://rsteib.home.pages.de/

Re: article printing with `A P' is generating bad postscript

[Karl Kleinpaste]

Reiner Steib <4uce.02.r.steib@gmx.net> writes:
> Look at the lines containing "h0" in out.ps. The problem should be
> there. Maybe post the output of "grep -2 'h0' out.ps"

I tried this on your own article, right here.  "grep -2 h0 out.ps" gives:
________________________________________________________________

%%BeginSetup
/h0 14 /Helvetica-Bold DefFont
/h1 12 /Helvetica DefFont

% ---- These lines must be kept together because...

/h0 F
/HeaderTitleLineHeight FontHeight def

--
/PageCount 1 def
/HeaderLinesLeft [
  [ /h0 (Re: article printing with `A P' is generating bad postscript) ]
  [ /h1 (Reiner Steib <4uce.02.r.steib@gmx.net>) ]
] def
/HeaderLinesRight [
  [ /h0 /pagenumberstring load ]
  [ /h1 (Tue, 11 Jun 2002 18:17:54 +0200) ]
] def
--
(> Operand stack:) S
HL
(>    h0   --nostringval--   --nostringval--) S
HL
(> Execution stack:) S
--
HL
HL
(Look at the lines containing "h0" in out.ps. The problem should be) S
HL
(there. Maybe post the output of "grep -2 'h0' out.ps". "h0" is used to) S
HL
(print the subject and page numbers.) S
________________________________________________________________

> If this fails, remove the error handler from out.ps:
> %%BeginProcSet: ErrorHandler

There is no such error handler.  There is no other occurrence of the
word "error" in out.ps.  Other thoughts?

Weirdness: out.ps displays fine where I work, under RH7.2 with
ghostscript 6.51.  The same out.ps, scp'd home, does not display fine
under 6.52 under RH7.3, and when I downgrade 7.3's ghostscript to
7.2's 6.51, it still doesn't work.  XEmacs and the sumo tarballs
supporting it are the same at work and home.

confused,
--karl

Re: article printing with `A P' is generating bad postscript

[Reiner Steib]

On Wed, Jun 12 2002, Karl Kleinpaste wrote:

> Reiner Steib <4uce.02.r.steib@gmx.net> writes:
>> Look at the lines containing "h0" in out.ps. The problem should be
>> there. Maybe post the output of "grep -2 'h0' out.ps"
>
> I tried this on your own article, right here.  "grep -2 h0 out.ps" gives:
[...]

I can't see anything strange there. Did you get an error related to
"h0" too ("h0   --nostringval--   --nostringval--")?

> There is no such error handler.  There is no other occurrence of the
> word "error" in out.ps.  

Which version of ps-print do you use (see the variable
ps-print-version)?

> Other thoughts?

You could provide an example file and ask in comp.lang.postscript. If
you have different behavior with different gs versions, it's either an
error in ps-print or in one of the gs versions, IMHO.

> Weirdness: out.ps displays fine where I work, under RH7.2 with
> ghostscript 6.51.  The same out.ps, scp'd home, does not display fine
> under 6.52 under RH7.3, and when I downgrade 7.3's ghostscript to
> 7.2's 6.51, it still doesn't work.  XEmacs and the sumo tarballs
> supporting it are the same at work and home.

Then it's probably no Gnus error, I can reproduce the error with GNU
Ghostscript 6.52 and 5.50 and ps-print v6.5.5 (distributed with Emacs
21.1).

> confused,
> --karl

me too.

Bye, Reiner.
-- 
       ,,,
      (o o)
---ooO-(_)-Ooo--- PGP key available via WWW   http://rsteib.home.pages.de/

Re: article printing with `A P' is generating bad postscript

[Karl Kleinpaste]

Reiner Steib <4uce.02.r.steib@gmx.net> writes:
> I can't see anything strange there. Did you get an error related to
> "h0" too ("h0   --nostringval--   --nostringval--")?

Not that I noticed.  I think that was just a quotation in the text
from your message.

> Which version of ps-print do you use (see the variable
> ps-print-version)?

| `ps-print-version' is a variable declared in Lisp.
|   -- loaded from "ps-print"

| Value: "3.05x1"

| Documentation:
| ps-print.el, v 3.05 <97/01/17 duthen>

That seems really old.  But it's what comes with XEmacs' sumo tarballs
these days.  (I certainly haven't any other ps-print lying around.)

> Then it's probably no Gnus error, I can reproduce the error with GNU
> Ghostscript 6.52 and 5.50 and ps-print v6.5.5 (distributed with Emacs
> 21.1).

OK, so I'm not entirely alone if you can reproduce it.

I will see if there's a more appropriate ps-print out there, but I
wonder why this used to work for me.

Hm, comments in ps-print.el say:

;; 3.05x1 -- Forked for XEmacs/No-Mule.  The official version of
;;  ps-print is Mule-only.

I will hunt for that.

Re: article printing with `A P' is generating bad postscript

[Reiner Steib]

On Wed, Jun 12 2002, Karl Kleinpaste wrote:

> Reiner Steib <4uce.02.r.steib@gmx.net> writes:
>> I can't see anything strange there. Did you get an error related to
>> "h0" too ("h0   --nostringval--   --nostringval--")?
>
> Not that I noticed.  I think that was just a quotation in the text
> from your message.

"grep -2 h0 out.ps" is only interesting if the error is after/near
'h0'. ;-)

> | Documentation:
> | ps-print.el, v 3.05 <97/01/17 duthen>
>
> That seems really old.  But it's what comes with XEmacs' sumo tarballs
> these days.  (I certainly haven't any other ps-print lying around.)

The version that comes with Emacs seems to be the most recent one:
;; Version: 6.5.5
;; X-URL: http://www.cpqd.com.br/~vinicius/emacs/

>> Then it's probably no Gnus error, I can reproduce the error with GNU
>> Ghostscript 6.52 and 5.50 and ps-print v6.5.5 (distributed with Emacs
>> 21.1).
>
> OK, so I'm not entirely alone if you can reproduce it.

Oops, this should read "I can't reproduce ...". Sorry. But as I said
in my first mail, I had problems with older versions, too. So an
upgrade of ps-print.el may solve your problem.

> ;; 3.05x1 -- Forked for XEmacs/No-Mule.  The official version of
> ;;  ps-print is Mule-only.

The web page mentions that is depends on mule.

Bye, Reiner.
-- 
       ,,,
      (o o)
---ooO-(_)-Ooo--- PGP key available via WWW   http://rsteib.home.pages.de/