Re: print.m: size support for (e)ps terminals

[Sebastian Schubert]

John W. Eaton wrote:
> | > It seems to work for me.  The last suggestion I have is that the texinfo
> | > should reflect the -S3x2 format.  It currently specifies a comma.
> | 
> | I suppose a "D'oh" would be appropriate here... Thanks for pointing that
> | out. Here is a another patch.
> 
> Sorry for jumping in late on this thread, but I'm not sure that this
> is the right way to specify sizes for plots.  Doesn't Matlab use the
> PaperUnits, PaperOrientation, PaperPosition, PaperPositionMode,
> PaperSize, and PaperType figure properties to handle the size of
> printed figures?

Ok, I've never used Matlab but this is the way it "should" work (I hope
that there will be a time when the Matlab guys ask themselves the same
question concerning octave). :)

I had a look into the online documentation of Matlab and it seems that
all these values should be set similar to paperorientation.  But AFAICS
there aren't any native Matlab functions except orient() for
paperorientation, so one is supposed to use either the "Property Editor"
or use get() and set().  The latter way seems the way to go for octave.

Is Papersize the size of the actual plot or the size of the paper the
plot is supposed to be printed on? If the first one is the case, what
does PaperPosition really mean? If the latter, how do you set the size
of the plot?

So the way to implement all this is AFAICS:
1) add PaperUnits, PaperSize and PaperType (or the values to define the
plot size) to the graphics properties.  IIRC gnuplot only has an option
for the real plot size.
2) write functions to set these properties (similar to orient()) because
having to do it manually with get() and set() seems to me rather cumbersome.
3) change print.m. That should be easy then.

I could do that but it would take time.  In about a month I have a
matlab available to test its behaviour.

Sebastian