some plotting bugs

[Ben Abbott]

On Thursday, July 09, 2009, at 12:03PM, "Petr Mikulik" <mikulik at physics.muni.cz> wrote:
>
>> Regarding the "-tight" option, the problem is likely do my attempt to 
>> simplify the implementation. I'd assumed the new BBOX string would be 
>> shorted than the original. When it isn't the file is left unchanged.
>
>What about calling ghostscript to calculate the bounding box precisely if 
>the "-tight" option is used?

I'm using gs for that now. The process is ...

(1) have GS calcualate the BBOX.
(2) replace the BBOX line in the eps-file with a new line of the same length.

The current implementation avoids reading the entire file into Octave, by simply replacing the BBOX line in the original file. In the event the original BBOX line is shorted than the tight version the file is left as it was.

If anyone has an efficient solution that would work for *nix systems and windows, please share :-)

Ben