myb100@csc.anu.oz (11/22/89)
(posted to c.lang.postscript and c.sys.amiga)
G'day
I recently got a copy of VLT4.226 terminal emulator for the Amiga from xanth,
and WOW what a great package !
However, I've come across a small problem: This relates to the PostScript
output generated by capturing the graphics on the TEK screen. What I have is
a plot generated by a local package, and on the same screen some text
written by the VAX on the TEK screen, the word `INPUT', written before the plot
starts.
My PS interpreter produces a `nocurrentpoint' error at a `show' statement.
What follows is part of the PS header which defines the commands:
/m {stroke newpath moveto} bdef
/d {lineto} bdef
/p {stroke newpath moveto domark} bdef
/s {stroke newpath show} bdef
/sd {stroke newpath setdash} bdef
/sc {stroke newpath setrgbcolor} bdef
[...] | and now the plot...
3248 1400 m |
3248 1400 d |
|
[...] |----- These commands produce the plot
|----- without any errors
3220 1148 m |
3220 1148 d |
0 3068 m (I) s |
52 3068 m (N) s |
104 3068 m (P) s |---- This is the text which causes problems.
156 3068 m (U) s |---- (remove it => no errors)
208 3068 m (T) s |---- and is the only place where the `s' is used
stroke
showpage
Digging out my handy red book I find that nocurrentpoint implies the path
has been wiped and I'm trying to execute a `show'. Hmmmm. Looking at the
definition
/s {stroke newpath show} bdef
I find (i) newpath wipes the path, (ii) stroke wipes the path and
(iii) show comes straight after these commands. This doesn't make any
sense to me, however, I'm not a PS guru.
So, my question is this: Is this a real bug ? (a) the definition of `s'
as it stands is strange, and (ii) text on the TEK screen is usually
generated by (vector)graphics commands, not by the VAX writing to
the sys$output device, so is this a case which VLT wasn't designed to handle ?
Any comments greatly appreciated
adTANXvance
Markus Buchhorn
Mt Stromlo Observatory, Canberra, Australia
markus@mso.anu.oz.au or nssdca::psi%mssso::markus
P.S. Anybody know Willy Langeveld's (sp?) E-mail address (not on BIX) I'd
like to get in contact with him.