gnuplot-beta — Developers and beta testers of new gnuplot code

Re: www.gnuplot.info

[Ethan M. <merritt@u.washington.edu>]

On Tuesday 20 February 2007 13:59, Petr Mikulik wrote:
> Thanks for the update! http://www.gnuplot.info/ is now up-to-date.
> 
> Maybe the fact that rsync does not work is the same as it is currently not 
> possible to ssh gnuplot.sf.net?

Probably.  
The site status page indicates that the shell server machine
died and has not yet been replaced.

By the way.  I made a quick pass over the FAQ and committed the
revision to CVS.  But I don't actually know the procedure for 
propagating the revised *.tex document to other media. Once the
shell server is back up, it will be possible to upload a new
html file.  But what about pure text and pdf versions - is there
a list of places they need to go?

-- 
Ethan A Merritt                                 Courier Deliveries: 1959 NE Pacific
Dept of Biochemistry    M/S 357742
Health Sciences Building
University of Washington - Seattle

Re: www.gnuplot.info

[Petr M. <mi...@ph...>]

> > It seems to me the current largest show-stopper for gnuplot is its
> > official web page, www.gnuplot.info. It is not mirroring
> > gnuplot.sf.net for several months.
> 
> > Further, in January 2006, there was thread in this mailing list:
> > Subject: Re: ftp://ftp.gnuplot.info/pub/gnuplot/testing


Thanks for the update! http://www.gnuplot.info/ is now up-to-date.

Maybe the fact that rsync does not work is the same as it is currently not 
possible to ssh gnuplot.sf.net?

---
PM

Re: www.gnuplot.info

[Clark G. <ga...@di...>]

On Thu, 15 Feb 2007 22:46:23 +0100 (CET), "Petr Mikulik"
<mi...@ph...> said:
> It seems to me the current largest show-stopper for gnuplot is its
> official web page, www.gnuplot.info. It is not mirroring
> gnuplot.sf.net for several months.

Sorry I hadn't noticed this. It is the first it's been mentioned to me.

Unfortunately, sf has broken their shell access (and hence rsync), so
I can't get the sync running exactly the way it was before. Hopefully
they will have that fixed soon. I can do some wget sync, but they
frown on that.

More important, though is having a more high-availability and higher
bandwidth site. I've made preparations for this to be hosted at the
Virginia Bioinformatics Institute at Virginia Tech, so this should be
coming up in the next couple days. This will also give very good access
to Internet2 and National Lambda Rail (and hence much better access to
most of our European and Asian users).

> Further, in January 2006, there was thread in this mailing list:
> Subject: Re: ftp://ftp.gnuplot.info/pub/gnuplot/testing where I listed
> files to be deleted on that ftp server. Again, nobody has access to
> that machine? What to do?

Done. Sorry, I do not monitor this list very closely at all.

The ftp site currently isn't moving to VBI, so that will remain on my
home machine for the time being.

--ckg

Re: strange behaviour of lc

[Mojca M. <moj...@gm...>]

On 2/17/07, Hans-Bernhard Br=F6ker wrote:
> Mojca Miklavec wrote:
>
> > I'm not sure if this is a bug or my misunderstanding of how "lc" is
> > supposed to work.
>
>   > I get different colors of lines on the first and the second plot. (Th=
e
> > second plot uses colors from a palette rather than from linetype).
>
> This is an interaction between 'lc' and hidden3d's default option state
> "offset 1".  That option makes hidden3d mode use different linetypes for
> the front and back side of a surface.  Which means that the back side
> colour cannot be controlled by 'lc'.
>
> OTOH: if you didn't want a change of colours, why use 'lc'?

There was "lt 3", but the lines were dashed. Yes, I know that I could
ask the terminal to plot all the lines as "solid", but I have tried
"lt 1 lc 3" instead, which works in most other cases, but produced
strange results in this particular case.

I thought that lt 1 lc 3 should create a solid line with the same
color as if I would select lt 3.

But I don't know much about 3D plots in gnuplot anyway, so I might
have been mistaken.

Mojca

Re: strange behaviour of lc

[<HBB...@t-...>]

Mojca Miklavec wrote:

> I'm not sure if this is a bug or my misunderstanding of how "lc" is
> supposed to work.

  > I get different colors of lines on the first and the second plot. (The
> second plot uses colors from a palette rather than from linetype).

This is an interaction between 'lc' and hidden3d's default option state 
"offset 1".  That option makes hidden3d mode use different linetypes for 
the front and back side of a surface.  Which means that the back side 
colour cannot be controlled by 'lc'.

OTOH: if you didn't want a change of colours, why use 'lc'?

Re: Help on building gnuplot!!

[Eligio P.M <e__...@ho...>]

Hi Ethan,
Thank you for your answer! I 'll post output about the missing line when i 
issue the command plot
for several functions. I think it's something with red color.
Ethan wouldl you like help with customization of wxt window. I want to 
change the background white for another color better to my eyes. Where can I 
find the init file to do that and how to customize some other features for 
this new terminal?
The .gnuplot file does not work and don't know what to put in there.
Thanks in advance..

>From: Ethan Merritt <merritt@u.washington.edu>
>To: gnu...@li...
>CC: "Eligio P.M" <e__...@ho...>
>Subject: Re: Help on building gnuplot!!
>Date: Fri, 16 Feb 2007 13:44:17 -0800
>
>On Thursday 15 February 2007 09:24, Eligio P.M wrote:
> > To: gnu...@li...
> > From: Eligio Perez
> > Errors: While building gnuplot version-beta 4.2 patchlevel rc4
> > CVS-23Jan2007, I got these errors after applying command "make".
>
>
> > >color.c: In function 'draw_color_smooth_box':
> > >color.c:523: warning: pointer targets in passing argument 2 of
> > >'map3d_position_r' differ in signedness
> > >color.c:523: warning: pointer targets in passing argument 3 of
> > >'map3d_position_r' differ in signedness
> > >set.c: In function 'set_mouse':
> > >set.c:2309: warning: pointer targets in passing argument 2 of
> > >'map_position' differ in signedness
> > >set.c:2309: warning: pointer targets in passing argument 3 of
> > >'map_position' differ in signedness
>
>These warnings are expected, and harmless.
>
>
> > >In gnuplot-comint-start-function:
> > >gnuplot.el:1838:18:Warning: `make-variable-buffer-local' should be 
>called
> > >at
> > >toplevel
> >
> > >In gnuplot-mode:
> > >gnuplot.el:2516:41:Warning: `make-variable-buffer-local' should be 
>called
> > >at toplevel
>
>That seems to be some Emacs problem. I know nothing about Emacs, and can't 
>help.
>Anyhow, it should not affect gnuplot itself.
>
>
> > Wrote /home/eligio/gnuplot-4.2.rc4/lisp/gnuplot.elc
> > Wrote /home/eligio/gnuplot-4.2.rc4/lisp/gnuplot-gui.elc
> > Expected X11 driver: /usr/local/libexec/gnuplot/4.2/gnuplot_x11
> > Exec failed: No such file or directory
>
>The gnuplot x11 driver is placed in 
>/usr/local/libexec/gnuplot/4.2/gnuplot_x11
>during installation.  Your error message is saying that it is not there.
>Did you do "make install"?
>If you want to test out the program before final installation, you can 
>force
>it to look in the build directory for gnuplot_x11. Assuming you are current 
>in
>the same directory ../src
>
>    setenv GNUPLOT_DRIVER_DIR `pwd`
>
> > Also, I applied this command "
> > gnuplot> plot cos(x),tan(x),tan(x)*cos(x),sin(x)
> > but I don't get line and color on the plot for cos(x).
>
>I have no idea what could cause that.
>Certainly no one else has reported anything similar.
>Terminal type? other commands?  Could you post the output?
>
>
>--
>Ethan A Merritt

_________________________________________________________________
Windows Live Spaces: share your New Year pictures! 
http://spaces.live.com/?mkt=en-ca

strange behaviour of lc

[Mojca M. <moj...@gm...>]

Hello,

I'm not sure if this is a bug or my misunderstanding of how "lc" is
supposed to work. I've taken this example from demos (the needed two
data files reside in demos too):

set angles degrees
set ticslevel 0
set view 70,40,0.8,1.2
set mapping spherical
set parametric
set samples 32
set isosamples 9
set urange [-90:90]
set vrange [0:360]

set title "3D solid version through hiddenlining"
set hidden3d
set arrow from 0,0,-1.2 to 0,0,1.2 lt 5 lw 2
set arrow from -1.2, 0, 0 to 1.2, 0, 0 nohead lt 5 lw 1
set arrow from 0, -1.2, 0 to 0, 1.2, 0 nohead lt 5 lw 1

splot cos(u)*cos(v),-cos(u)*sin(v),sin(u) with lines lt 5,\
'world.dat' u 1:2:(1.001) with lines lt 3, \
'world.cor' using 1:2:(1):(5.*rand(0))  with points lt 1 pt 6 ps var

pause -1

splot cos(u)*cos(v),-cos(u)*sin(v),sin(u) with lines lt 1 lc 5,\
'world.dat' u 1:2:(1.001) with lines lt 1 lc 3, \
'world.cor' using 1:2:(1):(5.*rand(0))  with points lt 1 pt 6 ps var


I get different colors of lines on the first and the second plot. (The
second plot uses colors from a palette rather than from linetype).

Mojca

Re: Help on building gnuplot!!

[Ethan M. <merritt@u.washington.edu>]

On Thursday 15 February 2007 09:24, Eligio P.M wrote:
> To: gnu...@li...
> From: Eligio Perez
> Errors: While building gnuplot version-beta 4.2 patchlevel rc4 
> CVS-23Jan2007, I got these errors after applying command "make". 


> >color.c: In function 'draw_color_smooth_box':
> >color.c:523: warning: pointer targets in passing argument 2 of 
> >'map3d_position_r' differ in signedness
> >color.c:523: warning: pointer targets in passing argument 3 of 
> >'map3d_position_r' differ in signedness
> >set.c: In function 'set_mouse':
> >set.c:2309: warning: pointer targets in passing argument 2 of 
> >'map_position' differ in signedness
> >set.c:2309: warning: pointer targets in passing argument 3 of 
> >'map_position' differ in signedness

These warnings are expected, and harmless.

 
> >In gnuplot-comint-start-function:
> >gnuplot.el:1838:18:Warning: `make-variable-buffer-local' should be called 
> >at
> >toplevel
> 
> >In gnuplot-mode:
> >gnuplot.el:2516:41:Warning: `make-variable-buffer-local' should be called 
> >at toplevel

That seems to be some Emacs problem. I know nothing about Emacs, and can't help.
Anyhow, it should not affect gnuplot itself.


> Wrote /home/eligio/gnuplot-4.2.rc4/lisp/gnuplot.elc
> Wrote /home/eligio/gnuplot-4.2.rc4/lisp/gnuplot-gui.elc
> Expected X11 driver: /usr/local/libexec/gnuplot/4.2/gnuplot_x11
> Exec failed: No such file or directory

The gnuplot x11 driver is placed in /usr/local/libexec/gnuplot/4.2/gnuplot_x11
during installation.  Your error message is saying that it is not there.
Did you do "make install"?
If you want to test out the program before final installation, you can force
it to look in the build directory for gnuplot_x11. Assuming you are current in
the same directory ../src

   setenv GNUPLOT_DRIVER_DIR `pwd`

> Also, I applied this command " 
> gnuplot> plot cos(x),tan(x),tan(x)*cos(x),sin(x)
> but I don't get line and color on the plot for cos(x).

I have no idea what could cause that. 
Certainly no one else has reported anything similar.
Terminal type? other commands?  Could you post the output?


-- 
Ethan A Merritt

Re: splot broken in rc4?

[Ethan M. <merritt@u.washington.edu>]

On Thursday 15 February 2007 13:47, Ethan Merritt wrote:
> > "set key auto column", either just before the splot command, or in the
> > .gnuplot file (that's where I found it).
> 
> Aha. You didn't tell us that part.
> But then the error message is correct.
> You have told it that you will provide a separate line of data containing
> the key titles, but you didn't do so.  So you got an error message.

But you are correct that the 3D code should explicitly handle
the return codes DF_FOUND_KEY_TITLE and DF_KEY_TITLE_MISSING.
I will fix that.  

Also it is more difficult than it should be to turn off the option
"set key autotitle columnheader".  There should be a complementary setting
"set key autotile file", or something like that. 

	thanks for the bug report,

		Ethan

-- 
Ethan A Merritt                                 Courier Deliveries: 1959 NE Pacific
Dept of Biochemistry    M/S 357742
Health Sciences Building
University of Washington - Seattle

Re: www.gnuplot.info

[Ethan M. <merritt@u.washington.edu>]

On Thursday 15 February 2007 13:46, Petr Mikulik wrote:
> It seems to me the current largest show-stopper for gnuplot is its offical 
> web page, www.gnuplot.info. It is not mirroring gnuplot.sf.net for several 
> months. 

There are at least 2 obvious possibilities.

(1) Change all the docs to refer only to the SourceForge site itself, thus
    removing mention of www.gnuplot.info

(2) Change the registry of www.gnuplot.info to point to some machine other
    than e028121.vtacs.vt.edu
    DNS service can be changed easily enough.
    Since that vt.edu machine is responding to both http and ftp, pointing
    directly to a single SourceForge address won't work equivalently.
    But we could either split it into ftp.gnuplot.info and www.gnuplot.info,
    or find a machine elsewhere that can be used for both.  






> Old messages on this mailing list indicate that Clark Gaylord is the  
> only one having access to the site... and whois supports this. Could you 
> please restart mirroring? Or, could you give access to this machine to other 
> gnuplot developers to launch rsync'ing manually?
> 
> Without the web page, releasing 4.2 cannot happen.
> 
> Further, in January 2006, there was thread in this mailing list:
>    Subject: Re: ftp://ftp.gnuplot.info/pub/gnuplot/testing
> where I listed files to be deleted on that ftp server. Again, nobody has 
> access to that machine? What to do?
> 
> ---
> PM
> 
> -------------------------------------------------------------------------
> Take Surveys. Earn Cash. Influence the Future of IT
> Join SourceForge.net's Techsay panel and you'll get the chance to share your
> opinions on IT & business topics through brief surveys-and earn cash
> http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
> _______________________________________________
> gnuplot-beta mailing list
> gnu...@li...
> https://lists.sourceforge.net/lists/listinfo/gnuplot-beta
> 

-- 
Ethan A Merritt                                 Courier Deliveries: 1959 NE Pacific
Dept of Biochemistry    M/S 357742
Health Sciences Building
University of Washington - Seattle

Re: docs issues

[Ethan M. <merritt@u.washington.edu>]

On Thursday 15 February 2007 13:26, Petr Mikulik wrote:
>   There was some discussion on this list about the differences of
> 	splot 'bla' vs splot 'bla' using 1:2:3
>   but I cannot find it in the docs. Could somebody summarize it there,
>   and cross-link it with "help missing"?

Not sure what you mean.
The existing section for "help missing" explains what happens when
not all columns on a datafile line are usable.  The 3D splot case
is they same:  If all lines have at least the first 3 columns usable,
then "splot with <foo>" is the same as "splot using 1:2:3 with <foo>".

However, if some of the lines contain only 2 readable entries,
then "splot using 1:2:3 with <foo>" will work as expected by treating
the defective lines as "missing".
In this case "splot with <foo>" is not usable, because the plot style
is re-evaluated for each line and lines with only 2 entries can
switch the plotting style in mid-plot, yielding garbage.

This should probably be considered a bug in gnuplot, but it is not
at all clear what the proper fix is other than to recommend always
giving an explicit "using 1:2:3".


>   Further, I propose to label "help using" by "help nan" as well.

help gnuplot-defined NaN       (at least, that's where "pi" is explained)

-- 
Ethan A Merritt                                 Courier Deliveries: 1959 NE Pacific
Dept of Biochemistry    M/S 357742
Health Sciences Building
University of Washington - Seattle

Re: splot broken in rc4?

[Ethan M. <merritt@u.washington.edu>]

On Thursday 15 February 2007 10:59, Jacques Le Bourlot wrote:
> > >
> > I cannot reproduce this with your test file, and really I can't imagine
> > how you could get this error message with any input file.
> >
> This is a bug in plot3d.c. In the switch starting at line 632, the case
> "(j == DF_FOUND_KEY_TITLE)" is not taken into account.
> 
> It is triggered by setting:
> 
> "set key auto column", either just before the splot command, or in the
> .gnuplot file (that's where I found it).

Aha. You didn't tell us that part.
But then the error message is correct.
You have told it that you will provide a separate line of data containing
the key titles, but you didn't do so.  So you got an error message.

> I turned the crash into a nasty warning by adding (line 633):

Crash? What crash?  You said nothing about that either.
You just reported an error message.

So now I am testing with
   set key autotitle columnheader 
   splot 'ex_splot.dat' using 1:2:3 with lines

and indeed I get the same [correct] error message you originally
reported.  But no crash.

-- 
Ethan A Merritt                                 Courier Deliveries: 1959 NE Pacific
Dept of Biochemistry    M/S 357742
Health Sciences Building
University of Washington - Seattle

www.gnuplot.info

[Petr M. <mi...@ph...>]

It seems to me the current largest show-stopper for gnuplot is its offical 
web page, www.gnuplot.info. It is not mirroring gnuplot.sf.net for several 
months. Old messages on this mailing list indicate that Clark Gaylord is the 
only one having access to the site... and whois supports this. Could you 
please restart mirroring? Or, could you give access to this machine to other 
gnuplot developers to launch rsync'ing manually?

Without the web page, releasing 4.2 cannot happen.

Further, in January 2006, there was thread in this mailing list:
   Subject: Re: ftp://ftp.gnuplot.info/pub/gnuplot/testing
where I listed files to be deleted on that ftp server. Again, nobody has 
access to that machine? What to do?

---
PM

docs issues

[Petr M. <mi...@ph...>]

According to the past discussion, I propose to change the following issues 
in the docs:


(A)
3 dgrid3d
 The `set dgrid3d` command enables, and can set parameters for, non-grid
 to grid data mapping.
=>
 The `set dgrid3d` command enables, and can set parameters for, non-grid
 to grid data mapping. See `splot grid_data` for more details about the
 grid data structure.


(B)
3 contour
 `set contour` enables contour drawing for surfaces. This option is 
 available for `splot` only.
 ...
 If contours are desired from non-grid data, `set dgrid3d` can be used to
 create an appropriate grid.  See `set dgrid3d` for more information.
=>
 `set contour` enables contour drawing for surfaces. This option is 
 available for `splot` only. It requires grid data, see `grid_data` for more 
 details. If contours are desired from non-grid data, `set dgrid3d` can be 
 used to create an appropriate grid.


(C)
  There was some discussion on this list about the differences of
	splot 'bla' vs splot 'bla' using 1:2:3
  but I cannot find it in the docs. Could somebody summarize it there,
  and cross-link it with "help missing"?
  Further, I propose to label "help using" by "help nan" as well.

---
PM

Re: splot broken in rc4?

[Jacques Le B. <Jac...@ob...>]

> >
> I cannot reproduce this with your test file, and really I can't imagine
> how you could get this error message with any input file.
>
This is a bug in plot3d.c. In the switch starting at line 632, the case
"(j == DF_FOUND_KEY_TITLE)" is not taken into account.

It is triggered by setting:

"set key auto column", either just before the splot command, or in the
.gnuplot file (that's where I found it).

setting "set key noauto" gives back a normal behaviour.

DF_FOUND_KEY_TITLE is set in datafile.c, but I could not deceipher the logic.

I guess that "auto title" is handled properly for 2D plots, but was not
completely checked for 3D. I turned the crash into a nasty warning by
adding (line 633):
if (j == DF_FOUND_KEY_TITLE) break;

(I get complains about limits not set), but that's not a patch.

Cheers.

Jacques

Help on building gnuplot!!

[Eligio P.M <e__...@ho...>]

To: gnu...@li...
From: Eligio Perez
Errors: While building gnuplot version-beta 4.2 patchlevel rc4 
CVS-23Jan2007, I got these errors after applying command "make". Also, I 
applied this command " gnuplot> plot cos(x),tan(x),tan(x)*cos(x),sin(x)" but 
I don't get line and color on the plot for cos(x).
Would you like to help solve these problems? Or is it any kind of bug? 
Anyway, I think it is the reason why I'm getting  errors ang bad plotting!
Thanks a lot for help and answer.

>color.c: In function 'draw_color_smooth_box':
>color.c:523: warning: pointer targets in passing argument 2 of 
>'map3d_position_r' differ in signedness
>color.c:523: warning: pointer targets in passing argument 3 of 
>'map3d_position_r' differ in signedness
>set.c: In function 'set_mouse':
>set.c:2309: warning: pointer targets in passing argument 2 of 
>'map_position' differ in signedness
>set.c:2309: warning: pointer targets in passing argument 3 of 
>'map_position' differ in signedness

>In gnuplot-comint-start-function:
>gnuplot.el:1838:18:Warning: `make-variable-buffer-local' should be called 
>at
>toplevel

>In gnuplot-mode:
>gnuplot.el:2516:41:Warning: `make-variable-buffer-local' should be called 
>at
 >toplevel
Wrote /home/eligio/gnuplot-4.2.rc4/lisp/gnuplot.elc
Wrote /home/eligio/gnuplot-4.2.rc4/lisp/gnuplot-gui.elc
Expected X11 driver: /usr/local/libexec/gnuplot/4.2/gnuplot_x11
Exec failed: No such file or directory
See 'help x11' for more details
Expected X11 driver: /usr/local/libexec/gnuplot/4.2/gnuplot_x11
Exec failed: No such file or directory
See 'help x11' for more details
Expected X11 driver: /usr/local/libexec/gnuplot/4.2/gnuplot_x11
Exec failed: No such file or directory
See 'help x11' for more details
Expected X11 driver: /usr/local/libexec/gnuplot/4.2/gnuplot_x11
Exec failed: No such file or directory
See 'help x11' for more details
Expected X11 driver: /usr/local/libexec/gnuplot/4.2/gnuplot_x11
Exec failed: No such file or directory
See 'help x11' for more details
Expected X11 driver: /usr/local/libexec/gnuplot/4.2/gnuplot_x11
Exec failed: No such file or directory
See 'help x11' for more details
Expected X11 driver: /usr/local/libexec/gnuplot/4.2/gnuplot_x11
Exec failed: No such file or directory
See 'help x11' for more details
Expected X11 driver: /usr/local/libexec/gnuplot/4.2/gnuplot_x11
Exec failed: No such file or directory
See 'help x11' for more details

###############################################################################
$ "<gnu...@li...>
>  G N U P L O T
>         Version 4.2 patchlevel rc4 CVS-23Jan2007
>         last modified Tue Jan 23 21:40:03 PST 2007
>         System: Linux 2.6.12-29mdksmp

_________________________________________________________________
Don’t waste time standing in line—try shopping online. Visit Sympatico / MSN 
Shopping today! http://shopping.sympatico.msn.ca

Re: splot broken in rc4?

[Ethan A M. <merritt@u.washington.edu>]

On Wednesday 14 February 2007 01:38, Jacques Le Bourlot wrote:
> Bonjour,
> 
> I'm experiencing a possible bug with 4.2rc4. This used to work fine with
> rc2 and previous releases. I try to do a elementary 3D (s)plot:
> 
> splot "ex_splot.dat" u 1:2:3 w l
> 
> I get an error message:
> "Wrong number of columns in input data - line 1"

I cannot reproduce this with your test file, and really I can't imagine
how you could get this error message with any input file.  

If the program does not like your splot command, it should say:
  "Need 1 or 3 columns for cartesian data"

If the program has trouble reading the file for some reason, 
it should rather give you an error message about 
  "Skipping data file with no valid points".

In no case that I can think of would it give the error message you
report.  So I am mystified.  It makes no sense to me.

What happens if you bypass the actual file contents by saying?

   splot 'ex_splot.dat' u (rand(0)):(rand(0)):(rand(0)) w l
  

Is it possible that you have some previous commands active, 
read in from ~/.gnuplot when you started the program?

> PS: I compiled on a SuSE 9.3 (with all latest patches and upgrades). wxt
> works fine, x11 too, etc...

Could you please try again after compiling with   CFLAGS  -g -O0 ?

 

-- 
Ethan A Merritt
Biomolecular Structure Center
University of Washington, Seattle 98195-7742

Re: latex/pslatex terminals (was Re: grouping available terminals together)

[Mojca M. <moj...@gm...>]

On 2/13/07, Daniel J Sebald wrote:
> While on the topic if LaTeX related terminals, here's a few comments
>
> 2)  In the documentation are these examples:
>
>  Use gnuplot defaults (mostly sensible, but sometimes not really best):
>         set title '\LaTeX\ -- $ \gamma $'
>
> However, my experience is that to get a backslash into the LaTeX code requires \\, not just \.  So the above should probably be
>
>  Use gnuplot defaults (mostly sensible, but sometimes not really best):
>         set title '\\LaTeX\\ -- $ \\gamma $'

No, either
    set label '{\LaTeX\ -- $ \gamma $}' at 0,0
or
    set label "{\\LaTeX\\ -- $ \\gamma $}" at 0,0

Note the difference between single and double quote.

> Am I overlooking a backslash behavior switch somewhere?  The "most sensible, but sometimes not really best" comment above leaves one wondering "Why?".  Of course, there is always the possibility of defining ones own words outside gnuplot, like "\mytext1", "\eq1", that sort of thing.  But I'm not sure why the original author put that comment there.  (Not arguing against, just wondering what was meant.)

I'm not sure if I understood the question, but in LaTeX there's a big
problem, because alignment usually doesn't work properly. If you write
'\varepsilon^{2}', gnuplot will reserve 15 characters of space, but
you need less than two. The examples above show how to overcome this
problem.

> 3)  Is the
>
> font '<font>,#'
>
> terminal independent?  At least in terms of size.  I mean, it would be nice if there were a way of translating
>
> font ',16'
>
> so that the pslatex terminal places an approximate "\\large" in the text.  For example, say I want
>
> set title "Here is a title" font ",16"
>
> which creates a bigger title in x11, but not in pslatex.  If I do the following:
>
> set title "\\large Here is a title"
>
> then that "\\large " shows up in other terminals that aren't LaTeX.  Any objection if I write such a patch some day?  Or might there be a reason for a more broader use of the word "size".  We've got "pointsize", maybe "titlesize", "linesize".  Thoughts?

What would be the difference between "titlesize" and the size of the
font that you specify in the header?

Which font sizes are you planning to map to which LaTeX commands??? I
don't think that's good idea. In that case it might make more sense to
simply scale the title (if you assume that the font size is 10pt and
user asks for 16pt, you scale it to 160%). I bet there there are some
packages which allow you to specify font size explicitly, but fonts
have a high tendency to change (at least their names, if not the
format) and to become obsolete. Perhaps you don't notice that, but
just for example ... the euro symbol in ConTeXt is broken on TeXLive
this year, because the author has changed the font name and nobody
spotted that until it was too late.

In the ConTeXt terminal I pass the font string almost literally to
ConTeXt (I needed to add "pt" or some other explicit measure next to
font size - actually one of the ugliest part of the code). Saying
\switchtobodyfont[somename,16pt] works there, but I don't know about
LaTeX.

True, labels are not truly compatible between (La)TeX/ConTeXt and
other terminals, but they won't even be for some time and I would
expect the users to be aware of that:
- if you ever tried or will try to run some demos with the LaTeX
terminal, you'll see that appoximately half of the demo files will
fail because there is %, _, ^ or $ present in text labels.
- if someone decides to use LaTeX labels, he's probably also planning
to type {\bf ...}, $\alpha$, $\gamma$ or similar commands. And those
will fail with other terminals as well. I also use utf-8, which
doesn't work with ps.

Mojca

splot broken in rc4?

[Jacques Le B. <Jac...@ob...>]

Bonjour,

I'm experiencing a possible bug with 4.2rc4. This used to work fine with
rc2 and previous releases. I try to do a elementary 3D (s)plot:

splot "ex_splot.dat" u 1:2:3 w l

the file "ex_splot.dat" can be e.g. (only the 11 lines between the ----):

-------------------
1.0 1.0 1.0
1.0 2.0 3.0
1.0 3.0 4.0

2.0 1.0 3.0
2.0 2.0 4.0
2.0 3.0 5.0

3.0 1.0 4.0
3.0 2.0 5.0
3.0 3.0 6.0
------------------------

I get an error message:

"Wrong number of columns in input data - line 1"

with a "^" point at the digit "3" in the command line. I tried lots of
variations (terminals, files, fancy commands, etc...) without any success.
(s)plotting functions works (including pm3d and so). Older versions of
gnuplot accept all my files without problem.

Any idea?

Jacques Le Bourlot

PS: I compiled on a SuSE 9.3 (with all latest patches and upgrades). wxt
works fine, x11 too, etc...

Re: latex/pslatex terminals (was Re: grouping available terminals together)

[Daniel J S. <dan...@ie...>]

Daniel J Sebald wrote:

> The other label -- account for long ticlabels: set ylabel
> '[r]{\LaTeX\ -- $ \gamma $\rule{7mm}{0pt}}'
> 
> However, my experience is that to get a backslash into the LaTeX code
> requires \\, not just \.  So the above should probably be

Oh, that's single quotes, I use double quotes... (double oy).

Dan

Re: Keybox too small for user defined location when using splot

[Daniel J S. <dan...@ie...>]

m sutton wrote:
>> ----- Original Message ----- From: "Daniel J Sebald"
>> <dan...@ie...> To: Subject: Re: Keybox too small for user
>> defined location when using splot Date: Tue, 13 Feb 2007 21:35:54
>> -0600
>> 
>> 
>> Well, not exactly.  I updated the placement of key, but mainly in 
>> "plot" mode, not "splot" mode.  (As discussed previously, unifying
>>  the two versions of key and margin layout would be nice.)
>> 
> 
> 
> Today was a snow day, so I was messing with Gnuplot today.  I was
> looking at combining splot mode and the 2D key stuff.  No real
> success.

Well you're willing to try, that's something.  Keep in mind there is a lot of work here if at some point we agree it is worth doing.  But the idea would be to have plot layout very similar in terms of keys and colorboxes.  (They should be treated similarly and we had in mind to allow multiple keys.)  At the center of the layout is the plotting area which can be either 2D or 3D.  Something like

  Screen (outermost)
  Plot Area (set by "size", can have multiple on screen)
  Gridded Area (where the drawn function is, i.e., the coordinate grid be it 2D or 3D)

So, there are margins inside the plot area that determine the extent of the gridded area.  Key placement, colorbox placement will effect those margins. Etc.

I'll send a PNG file from a while ago to illustrate what graph3D.c currently does (top fig) and the more general idea (bot fig).  Don't think there was ever total agreement.  But, that's the idea.

So, right now in 3D if one zooms in far lines go to the edge of the screen.  But if ever a general ->crop() function comes about, cropping to the "gridded area" might be nice.  I don't know.

What to do if multiple plots but only one colorbox?  Turn one off for one of the plots?  Not sure.

Where to start?  I'd suggest taking the key layout code in graphics.c and creating a routine

  ComputeOrDrawKey(key, ptitl_cnt, &size, ...)

Something similar with colorboxes perhaps?  (That one may already be like that.)

  ComputeOrDrawColorbox(cbox, &size, ...)

and make the current code the same for testing.  That way, when computing margins it is a matter of simply calling the above routine.  Go from there.

This is a summer project, Mike.  Too much for a day or two.

Dan

Re: Keybox too small for user defined location when using splot

[m s. <mw...@us...>]

> ----- Original Message -----
> From: "Daniel J Sebald" <dan...@ie...>
> To:=20
> Subject: Re: Keybox too small for user defined location when using splot
> Date: Tue, 13 Feb 2007 21:35:54 -0600
>=20
>=20
> Well, not exactly.  I updated the placement of key, but mainly in=20
> "plot" mode, not "splot" mode.  (As discussed previously, unifying=20
> the two versions of key and margin layout would be nice.)
>=20

Today was a snow day, so I was messing with Gnuplot today.  I was looking a=
t combining splot mode and the 2D key stuff.  No real success.  I just stum=
bled on this bug.

Mike Sutton


=3D
Jewish Personals JewishAmericanSingles
We make Jewish dating easier by offering free, photo profile, advanced sear=
ch, chat, receive and reply to e-mail free and much more. Tell your friends.
http://a8-asy.a8ww.net/a8-ads/adftrclick?redirectid=3Dc275028a6bda4bf7b17a5=
b98dc57c36a

Re: Keybox too small for user defined location when using splot

[Daniel J S. <dan...@ie...>]

Ethan Merritt wrote:
> On Tuesday 13 February 2007 14:52, Mike Sutton wrote:
> 
>>for 4.2 CVS branch.
>>
>>When I run the the script below the keybox is incorrectly drawn.  The bottom 
>>of the keybox cuts right between the cos and tan entries in the key.  If I 
>>omit the key title the box is drawn correctly.
> 
> 
> Yeah, but more to the point -
> The bug is only present if you use the form "set key at <xx>,<yy>".
> If you use a keyword placement (top right) or let it default, then
> there is no problem.
> 
> cc'ing Dan Sebald because I think that's his code.

Well, not exactly.  I updated the placement of key, but mainly in "plot" mode, not "splot" mode.  (As discussed previously, unifying the two versions of key and margin layout would be nice.)

There is a patch on SourceForge that fixes the immediate issue Mike raised, if it breaks something else I'm not sure.  (One line, can't hurt too much.)

Dan

Another texinfo error

[Daniel J S. <dan...@ie...>]

Introduced within the passed few days, I think:

Loading /usr/share/emacs/site-lisp/site-start.d/rpm-spec-mode-init.el (source)...
Inserting help for terminals ...
Analyzing doc file ...
Converting to texinfo ...
Menus, nodes, xrefs ...
Wrong type argument: stringp, nil
make[1]: *** [gnuplot.texi] Error 255
make[1]: Leaving directory `/usr/local/src/gnuplot-key/gnuplot/docs'
make: *** [install-recursive] Error 1

Re: Keybox too small for user defined location when using splot

[Ethan M. <merritt@u.washington.edu>]

On Tuesday 13 February 2007 14:52, Mike Sutton wrote:
> for 4.2 CVS branch.
> 
> When I run the the script below the keybox is incorrectly drawn.  The bottom 
> of the keybox cuts right between the cos and tan entries in the key.  If I 
> omit the key title the box is drawn correctly.

Yeah, but more to the point -
The bug is only present if you use the form "set key at <xx>,<yy>".
If you use a keyword placement (top right) or let it default, then
there is no problem.

cc'ing Dan Sebald because I think that's his code.

 
> set term x11 persist
> set key box
> set key title "KEY TITLE LONG"
> set key box at screen .9, screen .98
> splot x,x,sin(x),cos(x),tan(x)

-- 
Ethan A Merritt