From maintainers-request at octave dot org Thu Jan 27 11:15:43 2005
Subject: Plotting Documentation (was Re: "no data point found" trying to plot  2D)
From: "John W. Eaton" <jwe at bevo dot che dot wisc dot edu>
To: Quentin Spencer <qspencer at ieee dot org>
Cc: maintainers at octave dot org
Date: Thu, 27 Jan 2005 12:19:09 -0500

On 27-Jan-2005, Quentin Spencer <qspencer at ieee dot org> wrote:

| I see a lot of people on the mailing lists using the low-level "gplot" 
| functions to do printing. I find using "plot" much easier, and I seem to 
| recall from John's past postings that he feels this is the preferred 
| way. I came to Octave from Matlab, so I've never read this documentation 
| in much detail but looking at it now, it seems to give very little 
| information about the higher level plot functions. Perhaps this explains 
| why so many people try to do it the "hard" way (I realize that some 
| users have good reasons for using gplot, but that should be the 
| exception). I think the plotting documentation could use a rewrite,

Yes.  I think the reason that the gplot functions are emphasized is
that they were implemented and documented first and unfortunately the
documentation has never caught up to the idea that the gplot functions
should be avoided unless absolutely necessary.

| I'm volunteering to submit some patches when I get some time to work on 
| it.

That would be helpful.  Thanks.

| I want to know if there's anything else in doc/interpreter/plot.txi 
| that is inadequately documented.

I'm sure there are quite a few things, but I don't have a list.

| One other FAQ on the help list is how 
| to export plots. I would like to propose including the "print" function 
| from Octave-Forge in octave. Is there a reason to not do this? Are there 
| modifications necessary to consider including it?

The problem is that after printing, you want to go back to the
previous gnuplot terminal setting, but when I last looked, there was
no really good way to get that information from gnuplot.  I see that
the print function from octave-forge uses gget, also from octave-forge,
which is a bit of a hack.  But looking at the problem again, it also
appears that since version 3.8f, gnuplot supports "set term push" and
"set term pop", so we could fix this properly now.

The print function from octave-forge also tries to restore the "set
output" value, so it would be nice if we could also do "set output
push|pop".  Maybe we can also get that change into gnuplot?

jwe