iraf.net

Hi Henry, Sorry for the slow reply, I've been trying to ignore my mail
lately. I don't see anything immediately wrong with your graphcap,
but there is already an 'epsl' device entry that will do a landscape
EPS so it may not be needed. Does this produce the plot you want? As for adding PS code a la SM, it can't currently be done
without modifying the sgi2ueps translator code to take extra arguments.
Actually, since the DD string is just a unix commandline, I suppose you
could construct it such that there were some sort of sed/awk command
to add the lines you want. Hope this helps.Cheers,
-Mike
> HI Mike,
> I've gotten this to give a nice sized print, but need it to go
> landscape;
>
> g-lpr-uwo-eps|UNIX generic interface to default printer device (lpr):\
> D=lp,tmp$sgk,!{ sgidispatch sgi2ueps $F -l$(XO) -w$(XW) \
> -b$(YO) -h$(YW) -p$(PW) | lp -d tek; rm $F ; }&:tc=sgi_apl_land:
>
> sgi_apl_land|Apple Laserwriter (Postscript - landscape - UWO test)
> 300dpi:\
> :kf=bin$x_sgikern.e:tn=sgikern:cw#.0125:ch#.0294:\
> :ar#0.762s#0.267:ys#0.203r#3180:yr#2380:\
> :RO:YF:RO:YF:MF#1:XO#55:XW#3180:YO#90:YW#2380W#2.3:
>
>
> Could you tell me how I can flip it over to landscape mode?
> I tried the 'RO:YF:' commands, but they only change the axes.
>
> I notice that supermongo people (of which there seem to be example
> graphcap files on the web) throw postscript commands into the graphcap.
> Is it possible for us to do this and still make use of the SGI kernel
> driver above?
>
> Here is part of an example I came across;
>
> postscript|POSTSCRIPT|Postscript for square plots (8 x 8in):\
> r#2390:yr#2390:ch#.0229:cw#.013:\
> :CW=stroke\ngrestore\nshowpage\nend\n:\
> E=L\n(7#1+!778<#11;#0!7)CS M\n:\
> :OW='%!\ninitgraphics\n:\
> :OX=(#22!9#300!8#0!7) 20 dict begin\n72 300 div dup scale\n\
> 1 setlinejoin 0 setlinecap\n\
> /Helvetica findfont 55 scalefont setfont \n\
> ...
>
> Can we use the OW and OX commands to add postscript control, such as
> rotation?
>
> Thanks,
>
> Henry
>
>
> > Hi Henry,
> > Try using the 'psdump' device to produce a /tmp/irafdmpXXXX
> > (where XXXX is some pid) file. This uses the same sgi2uapl translator
> > to produce the postscript. If that works try
> >
> > prompt% cat /tmp/irafdmpXXXXX | lp
> >
> > to see if you get a print (i.e. simulate the DD string by hand). If
> > that fails, the PS generation works but the printer doesn't like the
> > resulting file. This is possible since that translator uses a simple
> > "%!PS" prefix and some printers want "%!PS-Adobe-something". You might
> > instead try the sgi2ueps translator in place of sgi2uapl to send EPS.
> >
> > Cheers,
> > -Mike
> >
> >
> > On Tue, 29 Nov 2005, Henry Leparskas wrote:
> >
> >> Hi Mike,
> >> I'm at home now, but should add that the attempt I made that is
> >> refered to below caused a core dump each time it was tried.
> >> That's probably significant.
> >>
> >> I always do a gflush, but because I am superstitious I exit 'cl' and
> >> reenter every time I try a new graphcap 'stdplot' setting.
> >>
> >> Yes, both of our printers work just fine for the 'lp' command at the unix
> >> level. I have the 'tek' priter set as default for both unix lpr (via
> >> the PRINTER variable) and for unix 'lp' via the LPDEST variable (both in
> >> the '.cshrc' file, and not '.login'. So by
> >> leaving out the destination printer in the pipe command, I was hoping it
> >> would go to 'tek'. Hmm, I recall that I did make one try at setting it
> >> explicitly to 'lp -d tek' with no success.
> >>
> >> Could it have something to do with the sgi device I am trying to use. Is
> >> it possible to try another one other than 'sgi2uapl'? Is there a way I
> >> would know that the executable for these drivers is okay?
> >>
> >> Thanks,
> >>
> >> Henry
> >
> >
>