RE: About "warning: implicit conversion from matrix to string"

[Richardson, Anthony]

> -----Original Message-----
> From: Dr.cryo [mailto:address@hidden
> Sent: Thursday, May 20, 2010 9:33 AM
> To: Richardson, Anthony; address@hidden
> Cc: address@hidden
> Subject: RE: About "warning: implicit conversion from matrix to
string"
> > Subject: RE: About "warning: implicit conversion from matrix to
> string"
> > Date: Thu, 20 May 2010 08:33:59 -0500
> > From: address@hidden
> > To: address@hidden; address@hidden
> > CC: address@hidden
> >
> >
> >
> > > From: Dr.cryo [mailto:address@hidden
> > > Sent: Thursday, May 20, 2010 7:11 AM
> > > To: address@hidden
> > > Cc: address@hidden
> > > Subject: RE: About "warning: implicit conversion from matrix to
> > string"
> > > > Subject: Re: About "warning: implicit conversion from matrix to
> > > string"
> > > > From: address@hidden
> > > > Date: Thu, 20 May 2010 07:48:15 -0400
> > > > CC: address@hidden
> > > > To: address@hidden
> > > >
> > > > On May 20, 2010, at 7:30 AM, Dr.cryo wrote:
> > > > >
> > > > > > Subject: Re: About "war! ning: implicit conversion from
> matrix
> > > > > > to
> > > string"
> > > > > > From: address@hidden
> > > > > > Date: Thu, 20 May 2010 07:23:23 -0400
> > > > > > CC: address@hidden
> > > > > > To: address@hidden
> > > > > >
> > > > > >
> > > > > > On May 20, 2010, at 1:48 AM, cookyzhao wrote:
> > > > > >
> > > > > > >
> > > > > > > Each time, when i try to export an emf formatted picture,
> > there
> > > > > > > is a warning message "warning: implicit conversion from
> > > > > > > matrix to string" for octave 3.2.4.
> > > > > > >
> > > > > > &g! t; I really wonder why i get this message, and what
> should
> > > > > > i do to correctly
> > > > > > > export such emf files?
> > > >! ; > > >
> > > > > > >
> > > > &g t; > > for example:
> > > > > > >
> > > > > > > print -demf "-S640,480" test.emf
> > > > > >
> > > > > > When I print (using your example) I don't get the error. I
> > > suspect the error is from gnuplot.
> > > > > >
> > > > > > What is the version of gnuplot you hav installed?
> > > > > >
> > > > > > Ben
> > > > > >
> > > > > >
> > > > >
> > > > > in the fold of octave, i find my gnuplot, version 4.4, and i
> > > > > have
> > > also installed gnuplot 4.5 seperately.
> > > > > i think octave will call version 4.4
> > > >
> > > > You can determine which one by typing ...
> > > >
> > > > __gnuplot_version__
> > > >
> > > > Another possibility is that the error is related so a specific
> > > > plot
> > > obje! ct. Do you get the error for ...
> > > &! gt;
> > > > clf
> > > > plolt (0:10)
> > > > print -demf "-S640,480" test.emf
> > > >
> > > > If not, can you provide a simple example which does?
> > > >
> > > > I don't presently have version 4.4 or 4.5 installed, but have
> been
> > > meaning to do so. I'll take a look to see if that is straight
> forward.
> > > >
> > > > Ben
> > > >
> > > > p.s. Please "reply-all" so that others can benefit, and reply at
> > > > the
> > > bottom so that those arriving late can follow along. Thanks.
> > >
> > >
> > > i copy gnuplot 4.5.0 into the octave fold, and i find that as is
> > > also the fact.
> >
> > I get the same warning with Octave version 3.2.4 for Windows. It
> comes
> > from the following line in print.m:
> >
> > have_ghostscript = (exist (ghostscript_bi! nary, "file") == 2);
> >
> > when ghostscript_binary is an em pty matrix. ghostscript is used to
> > produce an output file only for certain output formats. gnuplot is
> > used for most, including the emf output format that you want, so the
> > warning is harmless.
> > You can eliminate it though by setting the GSC environment variable
> to
> > a dummy value:
> >
> > setenv("GSC","GSC");
> 
> 
> it does work, great!
> 
> >
> > in an octave startup file (or at the octave command prompt before
> > using print).
> > Better yet would be to set GSC to the actual ghostscript executable
> > filename.
> >
> > It appears that if ghostscript is installed and is in a directory in
> > your PATH, that this warning should not appear even if GSC is not
> set.
> > I expect that Ben has ghostscript installed and that is why he is
not
> > seeing the message.
> >
> > Alternatively, replace the above line in print.m with:
> >
> > if(isempty(ghostscript_binary))
> > have_ghostscript = 0;
> > else
> > have_ghostscript = (exist (ghostscript_binary, "file") == 2); endif
> >
> > to make the warning go away.
> >
> > Tony
> 
> 
> However, only by modifying print.m file, it has no effect for me.

You will need to enter "clear print" after making the change to the
file (or close and restart Octave).  m-files in standard directories
are not automatically reread after modification.

Tony