Re: Printing problem with 3.2.4

[Ben Abbott]

On May 22, 2010, at 8:46 AM, murlox wrote:

> I performed a simple fix by changing the following code on print.m:
> 
> OLD
> 
> if (doprint)
>      if (isunix ())
>        prn_opt = "l";
> 
> NEW
> 
> if (doprint)
>      if (isunix ())
>        prn_opt = "";
> 
> This allowed me to print again (I'm very glad), but the outcome of the print
> was no longer stretched to the sides when adding the option "-landscape". I
> kinda missed the capability from old 3.0.3. When I saw the old print.m from
> 3.0.3, it looked much simpler, yet it worked very well. I guess, my problem
> is now partially solved, albeit a little bit unsatisfactory.
> 
> Many thanks to Ben Abbott who allowed me to isolate the problem.
> 
> Regards,
> 
> Al

Thanks for following up.

To adjust the side of the result, change the "paperposition" property of the 
figure. This change was implemented to be consistent with how Matlab works.

Regarding the original problem, I think I have a clear understanding of the 
problem now.

(1) For postscript printers

        lpr -l myfile.ps

(2) For non-postscript printers using ghostscript, epson for example

        gs -sDEVICE=epson sOutputFile=foo.bar myfile.ps
        lpr -l foo.bar

(3) To print using the CUPS drivers

        lpr myfile.ps

Octave is able to print directly to postscript printers, or to use ghostscript 
to filter postscript into a format understood by many printers. So octave 
supports (1) and (2), but not (3).

I'll start a discussion on this with the other developers to see if we can 
provide a solution for all 3 cases.

Ben