Error in distiller or acrobat? clip; <large> setlinewidth; stroke

jdaw1

unread,

Nov 8, 2006, 6:42:36 PM11/8/06

to

Take a path entirely outside the clipping region, try a large
setlinewidth, and stroke. Some of the path should appear. But not with
Distiller 6.0 Professional Version 6.0.0 5/15/2003 viewed with Acrobat
Professional 6.0.0 5/19/2003. Please, do any wise people know why? Is
it the distilling or the viewing?

Example code follows:

%!

/TimesNewRomanPS-BoldMT 10 selectfont
/TestString (Testing 123) def

1 setlinecap 1 setlinejoin [] 0 setdash

/DoIt
{
[7 3 1 0]
{
setlinewidth
/thisgray 1 thisgray sub def
thisgray setgray
X 42 moveto
currentlinewidth 0 gt
{
TestString false charpath
WhichVariant {stroke} {strokepath fill} ifelse
}
{
TestString
WhichVariant {show} {true charpath fill} ifelse
} ifelse
} forall
} def

[ false true]
{
/WhichVariant exch def
/X WhichVariant {10} {120} ifelse def

/thisgray 0.25 def
DoIt
0.875 setgray
X 10 100 30
clipsave
4 copy rectfill rectclip
/thisgray 0 def
DoIt
cliprestore
} forall

showpage

(pstack start) ==
pstack
(pstack done) ==

jdaw1

unread,

Nov 8, 2006, 6:51:29 PM11/8/06

to

Indeed, the example is clearer if
> [7 3 1 0]
is replaced with
> [63 31 15 7 3 1 0]

jdaw1

unread,

Nov 8, 2006, 6:52:21 PM11/8/06

to

jdaw1

unread,

Nov 8, 2006, 11:23:45 PM11/8/06

to

When emailed to my home Mac and opened with Preview, the PDF (that was
distilled on the PC) looks good. So the problem is with Acrobat
Professional's rendering of the path segments that are disjoint from
the clipping region.

jdaw1

unread,

Nov 17, 2006, 9:32:06 AM11/17/06

to

I've had a variety of related rendering problems, which have been fixed
by distilling with Apple's Preview (which presumably calls Ghostscript)
rather than the Adobe product. I'm surprised that Adobe has been
out-qualitied, but live and learn.

John Doherty

unread,

Nov 17, 2006, 10:03:07 PM11/17/06

to

In <1163773926....@m7g2000cwm.googlegroups.com>, jdaw1 wrote:

> I've had a variety of related rendering problems, which have been
> fixed by distilling with Apple's Preview (which presumably calls
> Ghostscript)

I do not think that's true, and here's a shell transcript to show why:

pmg5 $ uname -a
Darwin jldg5 8.8.0 Darwin Kernel Version 8.8.0: Fri Sep 8 17:18:57
PDT 2006; root:xnu-792.12.6.obj~1/RELEASE_PPC Power Macintosh
powerpc
pmg5 $ whence pstopdf
/usr/bin/pstopdf
pmg5 $ pstopdf -i
executive
^D
PostScript(r) Version 3011.104
(c) Copyright 1984-1999 Adobe Systems Incorporated.
All Rights Reserved.
PS>^D
%%[ Warning: PostScript error. No PDF file produced. ] %%
pstopdf failed on PS read from stdin with error code -31000
pmg5 $

> rather than the Adobe product. I'm surprised that Adobe
> has been out-qualitied, but live and learn.

--

jdaw1

unread,

Nov 18, 2006, 12:56:22 AM11/18/06

to

WYSIWYG-type user not familar with shell scripts would welcome
explanation. I can send you PostScript and PDF made in various ways, to
demonstrate the problem, if a multi-megabyte email would be welcomed.

John Doherty

unread,

Nov 18, 2006, 9:35:58 PM11/18/06

to

In <1163829382.1...@f16g2000cwb.googlegroups.com>, jdaw1
wrote:

> WYSIWYG-type user not familar with shell scripts would welcome
> explanation.

Well, my point was just that the standard MacOSX pstopdf (not the same
as ghostscript's ps2pdf) is a PS interpreter that is (apparently)
copyright Adobe Systems.

Therefore, I think it's unlikely that Preview uses ghostscript.

--