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

Re: interactive gnuplot

[<man...@il...>]

Le lundi 28 Juin 2004 16:15, Hans-Bernhard Broeker a =E9crit :

> On Mon, 28 Jun 2004, [iso-8859-15] Fr=E9d=E9ric Mantegazza wrote:
> > But, in this mode, we don't have interactivity (mouse, zoom, etc..),
> > but it works well when lauching gnuplot from a shell.
>
> Note: Zooming is part of mousing.
>
> > Gnuplot probably enters in a special mode when not launched from a
> > shell.
>
> Close, but not quite.  The special mode is activated whenever the command
> input is not a tty.  This applies both during execution of script files
> and if the standard input is redirected from a pipe or file.
>
> > Is it possible to turn out this behaviour, and have interactivity even
> > when launched from a pipe ?
>
> In non-interactive mode, mousing is indeed of by default.  You have to
> turn it on explicitly:
>
> 	set mouse
>
> That's basically all there is to it.

Great !!! Thank you guys :o)

PS: I didn't see that this question has been posted a few days ago. Sorry=20
for the noise...

=2D-=20
   Fr=E9d=E9ric

Re: interactive gnuplot

[Hans-Bernhard B. <br...@ph...>]

On Mon, 28 Jun 2004, [iso-8859-15] Fr=E9d=E9ric Mantegazza wrote:

> But, in this mode, we don't have interactivity (mouse, zoom, etc..)=
, but it=20
> works well when lauching gnuplot from a shell.

Note: Zooming is part of mousing.=20

> Gnuplot probably enters in a special mode when not launched from a =
shell.=20

Close, but not quite.  The special mode is activated whenever the com=
mand
input is not a tty.  This applies both during execution of script fil=
es
and if the standard input is redirected from a pipe or file.

> Is it possible to turn out this behaviour, and have interactivity e=
ven
> when launched from a pipe ?

In non-interactive mode, mousing is indeed of by default.  You have t=
o
turn it on explicitly:

=09set mouse

That's basically all there is to it.

--=20
Hans-Bernhard Broeker (br...@ph...)
Even if all the snow were burnt, ashes would remain.

Re: question

[Hans-Bernhard B. <br...@ph...>]

On Mon, 28 Jun 2004, Hartmut Ehmler wrote:

> I have installed gnuplot 4.0 on a windows computer and run it by
> wgnuplot.exe.
> What I get on the screen as gnuplot command line is hardly to read
> because the resolution is too bad. How can I improve it?

Right-click into the text-window, select 'Choose Font...', and do that.
Once you've configured it to your liking, select  'Update wgnuplot.ini'
from the same menu.

-- 
Hans-Bernhard Broeker (br...@ph...)
Even if all the snow were burnt, ashes would remain.

Re: question

[Lars H. <lhe...@us...>]

Hartmut Ehmler writes:
> I have installed gnuplot 4.0 on a windows computer and run it by
> wgnuplot.exe.
> What I get on the screen as gnuplot command line is hardly to read
> because the resolution is too bad. How can I improve it?
> I have attached a screenshot how it looks like.
 
 If you attach a _screenshot_, kindly use some format other than M$ Word.

Re: interactive gnuplot

[Arnd B. <arn...@we...>]

On Mon, 28 Jun 2004, [iso-8859-15] Fr=E9d=E9ric Mantegazza wrote:

> Hi,
>
> We are using gnuplot 4.0 under linux within a python script. To do that, =
we
> just run gnuplot and send commands through a pipe connected to its stdin.
>
> But, in this mode, we don't have interactivity (mouse, zoom, etc..), but =
it
> works well when lauching gnuplot from a shell.
>
> Gnuplot probably enters in a special mode when not launched from a shell.=
 Is
> it possible to turn out this behaviour, and have interactivity even when
> launched from a pipe ?

Yep, see for example

http://www.physik.tu-dresden.de/~baecker/python/gnuplot.html

I.e., you just have to do send
"set mouse" before you start any plotting.

Further information from `help set mouse`
"""
 X11 specific implementation details: Mouse support is turned on by
default if
 standard input comes from a terminal (tty). Mouse support is turned off
if
 standard input does not come from a tty, e.g. a pipe. If you want to use
mouse
 support while writing to gnuplot from a pipe, the mouse must be turned on
 *before* starting the x11 driver, e.g. immediately after startup with the
 explicit command `set mouse`.
"""

Best,

Arnd

interactive gnuplot

[<man...@il...>]

Hi,

We are using gnuplot 4.0 under linux within a python script. To do that, we=
=20
just run gnuplot and send commands through a pipe connected to its stdin.

But, in this mode, we don't have interactivity (mouse, zoom, etc..), but it=
=20
works well when lauching gnuplot from a shell.

Gnuplot probably enters in a special mode when not launched from a shell. I=
s=20
it possible to turn out this behaviour, and have interactivity even when=20
launched from a pipe ?

Thank's,

=2D-=20
   Fr=E9d=E9ric

question

[Hartmut E. <Har...@ip...>]

I have installed gnuplot 4.0 on a windows computer and run it by
wgnuplot.exe.
What I get on the screen as gnuplot command line is hardly to read
because the resolution is too bad. How can I improve it?
I have attached a screenshot how it looks like.

Regards,
Hartmut
----------------------------------------------------------------------
Dr. Hartmut Ehmler

Max Planck Institut f=FCr Plasmaphysik
W7-X Basismaschine/Magnetsystem
Wendelsteinstra=DFe 1
17491 Greifswald (Germany)

phone: +49 (0)3834/88-2525=20
fax:   +49 (0)3834/88-2509
email: eh...@ip...

Re: Access to Gnuplot cvs repository

[Lars H. <lhe...@us...>]

Harald Harders writes:
> Hello,
> 
> at my office, I have a copy of the gnuplot CVS repository. When I try to
> update it using "cvs update", it does not work in most cases. I get the
> error message
> 
> cvs [update aborted]: connect to cvs.gnuplot.sourceforge.net(66.35.250.209):2401
>  failed: Connection timed out
> 
> Very rarely I can connect to the server and get the updated files. Do you
> have the same problem or could it be a misconfiguration at my computer?
 
 Your cvs Root is out ouf date. Remove your files and checkout a fresh copy
 as per http://sourceforge.net/cvs/?group_id=2055.

Access to Gnuplot cvs repository

[Harald H. <h.h...@tu...>]

Hello,

at my office, I have a copy of the gnuplot CVS repository. When I try to
update it using "cvs update", it does not work in most cases. I get the
error message

cvs [update aborted]: connect to cvs.gnuplot.sourceforge.net(66.35.250.209)=
:2401
 failed: Connection timed out

Very rarely I can connect to the server and get the updated files. Do you
have the same problem or could it be a misconfiguration at my computer?

Yours
Harald

--=20
Harald Harders                           Langer Kamp 8
Technische Universit=E4t Braunschweig      D-38106 Braunschweig
Institut f=FCr Werkstoffe                  Germany
E-Mail: h.h...@tu...               Tel: +49 (5 31) 3 91-3062
WWW   : http://www.ifw.tu-bs.de          Fax: +49 (5 31) 3 91-3058

Re: gnuplot library dependancies

[Mark S. <msp...@ho...>]

Hans-Bernhard Broeker <broeker <at> physik.rwth-aachen.de> writes:

We kind of figured it out .  We just changed LDFLAGS to point to the right 
directory.

Thanks for the help.

Mark

Re: gnuplot library dependancies

[Hans-Bernhard B. <br...@ph...>]

On Mon, 21 Jun 2004, Mark Spanglet wrote:

> How can we get gnuplot to look for it's libraries in a place other than
> the default?  We have libpng.a(png.o) in /usr/local/lib and
> libpng.a(libpng.so.2)  in /opt/freeware/lib.  We tried the
> with-png=/opt/freeware/lib thing. No luck.  

Note: that should be --with-png=... (note the double dash in the front).

Please be more specific: how *exactly* did it fail?  Please report what
config.log had to say about its attempts to find libpng.  What platform
did you do this on?  Does it need special linker flags (like -rpath or -R)
to allow the dynamic linker to find things in non-standard places?

-- 
Hans-Bernhard Broeker (br...@ph...)
Even if all the snow were burnt, ashes would remain.

Re: gnuplot library dependancies

[Lars H. <lhe...@us...>]

Mark Spanglet writes:
> How can we get gnuplot to look for it's libraries in a place other than=
 the=20
> default?  We have libpng.a(png.o) in /usr/local/lib and libpng.a(libpng=
.so.2)=20
> in /opt/freeware/lib.  We tried the with-png=3D/opt/freeware/lib thing.=
 No luck.
> There doesn't seem to be a command line option.  Do we have to change t=
he=20
> Makefile?

 There cannot be a command line option, because the shared libraries are
 loaded before the program itself starts.

 man ld.so
...
       ld.so, ld-linux.so* - dynamic linker/loader

DESCRIPTION
       ld.so loads the shared libraries needed by a program, pre=AD
       pares the program  to  run,  and  then  runs  it.   Unless
...

 If configure fails to find libraries, config.log will tell you why.

Re: Pause Mouse and X11 Window Close

[John C. <j-c...@ra...>]

Ethan Merritt wrote:

> On Monday 21 June 2004 06:24 am, John Cerney wrote:
> 
>>I am waiting for mouse input so that mousing is available while the user
>>is viewing the plot. For example, if a 3d surface plot is displayed, I
>>would do a "pause mouse" so that the users could rotate/scale with plot
>>with their mouse.
> 
> 
> But that's not going to work.  If you say "pause mouse" then it will
> terminate at the first mouse click (or actually at the first time the
> button is released).  So you can't do rotate/scale/zoom inside a
> pause mouse as it is currently implemented.  This is something
> that Petr and I were discussing before, but we didn't come to any
> particular resolution.
> 
> But you don't need a 'pause mouse' command for this anyhow.
> What's wrong with a regular 'pause -1' command?  You can still
> exit from the pause via a ctrl-C, or a hotkey, or an external trigger.
> 
> Another possibility is to put gnuplot into a loop that goes something
> like
> 	load 'loop-until-key'
> 
> where the file 'loop-until-key' contains
> 	pause 1
> 	if (MOUSE_KEY != <something>) reread

I am currently doing something similar, but my 'loop-until-key' file 
contains:

pause mouse
if (defined(MOUSE_BUTTON)) \
     print "Mouse button Clicked:", MOUSE_BUTTON;\
else \
     print "No mouse click?" ;\
     exit ;\
reread


I do a "pause mouse" (rather than pause 1), because I want to detect a 
mouse button-3 key event in my main program so I can pop-up a context menu.

The only problem with the approach so far is (as we have discussed) if 
the windows is closed during the "pause mouse", gnuplot hangs.


> 
> Actually I don't think that quite works currently, but it could be made
> to work if people decide it's a useful option.
> 
> 
>> >  Wouldn't it be
>> > better to program a hot key ("bind <key> '<gnuplot commands>'")?
>> > That way the user can either hit the appropriate key or close
>> > the window and in either case the response is handled asynchronously.
>>
>>I could have a hotkey to close the window, but users are so conditioned
>>to just close windows when they are done with a window, I don't think
>>many would actually use the hotkey.
> 
> 
> The cvs version should now be able to detect normal window-close
> events.
> 

I tried this last night with the latest CVS version and it didn't appear 
to be working for me. My simple test was this file:

set mouse
set term x11
splot cos(x)*sin(y)
pause mouse
if (defined(MOUSE_BUTTON)) \
     print "1 Mouse button Clicked:", MOUSE_BUTTON;\
else \
     print "No mouse click?" ;\
     exit ;
print "doing something else";
pause mouse
if (defined(MOUSE_BUTTON)) \
     print "1 Mouse button Clicked:", MOUSE_BUTTON;\
else \
     print "No mouse click?" ;\
     exit ;


When I run gnuplot with this file, and close the plot window when it 
shows up, gnuplot still hangs. With the changes you made, I was 
expecting the close window to break-out of the pause mouse, and set 
MOUSE_BUTTON to undefined or -1 or something. Does the above file work 
the same way with you?

Thanks,

John

Re: Time to branch off 4.0 from 4.1?

[Hans-Bernhard B. <br...@ph...>]

On Mon, 21 Jun 2004, Ethan Merritt wrote:

> Did you do this manually, or were you able to automate the procedure
> somehow?   

Some automation was possible.  I essentially did an emacs
'tags-query-replace' operation to cover entire source directories, and a
rather tricky egrep search to spot things still left to do (search for a
function definition that has only one word before the first comma or
closing parenthesis of the parameter list).  I expect to have to use more
tools to hunt the K&R definitions missed by egrep; like gcc in ultra-picky
mode, or maybe splint.

> I ask with some trepidation since it seems likely that applying
> this to the cvs tree will break every patchset on SourceForge including a
> couple of large ones that I have been maintaining.  

As long as they leave the function definitions alone, most patchsets
should either be fine, or the conflict resolutions obvious.

> That's not an  argument against doing it, but I wonder if there is a way 
> to apply your ANSIfication to both a patched and unpatched CVS tree
> in parallel so that I can regenerate the patchset to apply against the
> ANSIfied version.

Not really.  The method is O(N) in the number of function definitions,
which means using it on two source trees would take twice as long, and the
majority of that would be wasted duplicate effort.

-- 
Hans-Bernhard Broeker (br...@ph...)
Even if all the snow were burnt, ashes would remain.

Re: Time to branch off 4.0 from 4.1?

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

Ethan Merritt wrote:

>On Monday 21 June 2004 07:07 am, Hans-Bernhard Broeker wrote:
>  
>
>>I've pretty much completed this monstrous editing job over the weekend.
>>Nothing in CVS yet, but it can now be pulled off at a moment's notice.
>>    
>>
>
>Did you do this manually, or were you able to automate the procedure
>somehow?   I ask with some trepidation since it seems likely that applying
>this to the cvs tree will break every patchset on SourceForge including a
>couple of large ones that I have been maintaining.  
>
>That's not an  argument against doing it, but I wonder if there is a way 
>to apply your ANSIfication to both a patched and unpatched CVS tree
>in parallel so that I can regenerate the patchset to apply against the
>ANSIfied version.
>

That would be nice.

Dan

gnuplot library dependancies

[Mark S. <msp...@ho...>]

How can we get gnuplot to look for it's libraries in a place other than the 
default?  We have libpng.a(png.o) in /usr/local/lib and libpng.a(libpng.so.2) 
in /opt/freeware/lib.  We tried the with-png=/opt/freeware/lib thing. No luck.  
There doesn't seem to be a command line option.  Do we have to change the 
Makefile?

Re: Time to branch off 4.0 from 4.1?

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

On Monday 21 June 2004 07:07 am, Hans-Bernhard Broeker wrote:
> > *) If we move towards ANSI-C function definitions, I think that should
> > be done *before* we split off a 4.0 branch, to reduce "patch distance"
> > between the two branches.  I'm in favour of doing that, with a tag
> > 4.0.1 before we start it, and another tag (4.0.2) right afterwards. 
> > 4.0.2 would become the branch-off point for the 4.0-release branch.
>
> I've pretty much completed this monstrous editing job over the weekend.
> Nothing in CVS yet, but it can now be pulled off at a moment's notice.

Did you do this manually, or were you able to automate the procedure
somehow?   I ask with some trepidation since it seems likely that applying
this to the cvs tree will break every patchset on SourceForge including a
couple of large ones that I have been maintaining.  

That's not an  argument against doing it, but I wonder if there is a way 
to apply your ANSIfication to both a patched and unpatched CVS tree
in parallel so that I can regenerate the patchset to apply against the
ANSIfied version.



-- 
Ethan A Merritt       merritt@u.washington.edu
Biomolecular Structure Center
Mailstop 357742
University of Washington, Seattle, WA 98195

Re: y-axis reversed again

[Hans-Bernhard B. <br...@ph...>]

On Mon, 21 Jun 2004, Petr Mikulik wrote:

> Wouldn't it suffice in mouse.c, event_do_zoom (or how it is called), which
> is making the string
> 
> sprintf("set yrange [from:to]", mouse.y from, mouse.y to);
> 
> to change it to
> 
> if (is_view_map)
>    sprintf("set yrange [from:to]", mouse.y to, mouse.y from);
>    or whatever can fix the current bad behaviour
> else
>    the current sprintf as above

It might.  But then, so might re-doing this part of 'set view map' in the
place where really should be done, i.e. the same place where all the rest
of 'set view' is applied.  It's quite crazy that we currently have to flag
the y axis of a 3D plot as RANGE_REVERSE to get a display that is *not*
reverted.  'set view map' has no business fiddling with the yrange
settings in the first place.

-- 
Hans-Bernhard Broeker (br...@ph...)
Even if all the snow were burnt, ashes would remain.

Re: y-axis reversed again

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

> set xr[-6.93319487712:6.80166147456]; set yr[7.03235403496:-8.13313499442]
> zoom region finished.
> gnuplot> show yr
>
>         set yrange [ 7.03235 : -8.13313 ] noreverse nowriteback
>
> This state was reached by a combination of the 'backward' set yrange
> issued by the zoom with the action of splot_map_activate(), which flips
> the RANGE_REVERSE flag off again after 'set yrange' turned it on.
>
> I suspect it may take a thorough re-think and rewrite of how 'set view
> map' works before this is fixed for good.

Wouldn't it suffice in mouse.c, event_do_zoom (or how it is called), which
is making the string

sprintf("set yrange [from:to]", mouse.y from, mouse.y to);

to change it to

if (is_view_map)
   sprintf("set yrange [from:to]", mouse.y to, mouse.y from);
   or whatever can fix the current bad behaviour
else
   the current sprintf as above

---
PM

Re: Pause Mouse and X11 Window Close

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

> > I am waiting for mouse input so that mousing is available while the user
> > is viewing the plot. For example, if a 3d surface plot is displayed, I
> > would do a "pause mouse" so that the users could rotate/scale with plot
> > with their mouse.
>
> But you don't need a 'pause mouse' command for this anyhow.
> What's wrong with a regular 'pause -1' command?  You can still
> exit from the pause via a ctrl-C, or a hotkey, or an external trigger.

gnuplot> pause -1

gnuplot_x11: play with mouse as you like, then press <Space> to go to
gnuplot and there <Return>. Thus, just two keystrokes.

---
Petr

Re: y-axis reversed again

[Hans-Bernhard B. <br...@ph...>]

On Mon, 14 Jun 2004, Petr Mikulik wrote:

> There seems to be again the bug of "y-axis gets reversed" in "set pm3d map"
> splots. Now, every 2nd splot zoomed by mouse has reversed y-axis. Can you
> please have a look to this?

Here's the start of an analysis of this.  Brace yourself, for this is
really all a rather despicable mess.  

The core of the problem is that we still have two different possible
results of the 'set yrange' command regarding reverted axes:

	set yrange [10:5] 
	set yrange [5:10] reverse

are (nowadays) supposed to do mean the same thing, but in the reality
of our code, they don't:

gnuplot> set yrange [1:0]
[...]
(gdb) p axis_array[FIRST_Y_AXIS ]
$1 = {autoscale = AUTOSCALE_BOTH, set_autoscale = AUTOSCALE_NONE,
  range_flags = 2, range_is_reverted = 0 '\0', min = -10, max = 10,
  set_min = 1, set_max = 0, writeback_min = -10, writeback_max = 10,
[...]

gnuplot> set yrange [0:1] reverse
[...]
(gdb) p axis_array[FIRST_Y_AXIS ]
$3 = {autoscale = AUTOSCALE_BOTH, set_autoscale = AUTOSCALE_NONE,
  range_flags = 2, range_is_reverted = 0 '\0', min = -10, max = 10,
  set_min = 0, set_max = 1, writeback_min = -10, writeback_max = 10,

The difference is that variant 1 has set_min=1, set_max = 0, whereas
variant 2 has it the opposite way around.  Both of them have the
RANGE_REVERSE flag set (range_flags == 2).  It's this RANGE_REVERSE flag
alone that 'set view map' will modify.  It doesn't touch any of the axis
range endpoints themselves (and it shouldn't).  

RANGE_REVERSE is then consulted by the CHECK_REVERSE() macro to decide
whether to exchange axis->min and axis->max, and set the
axis->range_is_reverted flag, which in turn is read by
axis_revert_and_unlog_range() to decide whether to exchange axis->min and
axis->max *again*.  Sounds insane, doesn't it?

The error at hand seems to happen because the mouse zoom generates a
command as in variant 1, as can be seen in verbose mode.  E.g. in a fresh
gnuplot session:

gnuplot> set view map
gnuplot> sp x*y
gnuplot> communication commands will be echoed.
starting zoom region.
set xr[-6.93319487712:6.80166147456]; set yr[7.03235403496:-8.13313499442]
zoom region finished.
gnuplot> show yr

        set yrange [ 7.03235 : -8.13313 ] noreverse nowriteback

This state was reached by a combination of the 'backward' set yrange
issued by the zoom with the action of splot_map_activate(), which flips
the RANGE_REVERSE flag off again after 'set yrange' turned it on.

I suspect it may take a thorough re-think and rewrite of how 'set view
map' works before this is fixed for good.

-- 
Hans-Bernhard Broeker (br...@ph...)
Even if all the snow were burnt, ashes would remain.

Re: Pause Mouse and X11 Window Close

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

On Monday 21 June 2004 06:24 am, John Cerney wrote:
> I am waiting for mouse input so that mousing is available while the user
> is viewing the plot. For example, if a 3d surface plot is displayed, I
> would do a "pause mouse" so that the users could rotate/scale with plot
> with their mouse.

But that's not going to work.  If you say "pause mouse" then it will
terminate at the first mouse click (or actually at the first time the
button is released).  So you can't do rotate/scale/zoom inside a
pause mouse as it is currently implemented.  This is something
that Petr and I were discussing before, but we didn't come to any
particular resolution.

But you don't need a 'pause mouse' command for this anyhow.
What's wrong with a regular 'pause -1' command?  You can still
exit from the pause via a ctrl-C, or a hotkey, or an external trigger.

Another possibility is to put gnuplot into a loop that goes something
like
	load 'loop-until-key'

where the file 'loop-until-key' contains
	pause 1
	if (MOUSE_KEY != <something>) reread

Actually I don't think that quite works currently, but it could be made
to work if people decide it's a useful option.

>  >  Wouldn't it be
>  > better to program a hot key ("bind <key> '<gnuplot commands>'")?
>  > That way the user can either hit the appropriate key or close
>  > the window and in either case the response is handled asynchronously.
>
> I could have a hotkey to close the window, but users are so conditioned
> to just close windows when they are done with a window, I don't think
> many would actually use the hotkey.

The cvs version should now be able to detect normal window-close
events.


-- 
Ethan A Merritt       merritt@u.washington.edu
Biomolecular Structure Center
Mailstop 357742
University of Washington, Seattle, WA 98195

Re: Time to branch off 4.0 from 4.1?

[Hans-Bernhard B. <br...@ph...>]

On Wed, 16 Jun 2004, Hans-Bernhard Broeker wrote:

> *) If we move towards ANSI-C function definitions, I think that should be
> done *before* we split off a 4.0 branch, to reduce "patch distance"  
> between the two branches.  I'm in favour of doing that, with a tag 4.0.1
> before we start it, and another tag (4.0.2) right afterwards.  4.0.2 would
> become the branch-off point for the 4.0-release branch.

I've pretty much completed this monstrous editing job over the weekend.
Nothing in CVS yet, but it can now be pulled off at a moment's notice.

> *) K&R incompatibilities.  Mainly the use of 'char' in some function's
> parameter lists, most prominently in the terminal API.  

I went ahead and did that in the above-mentioned copy of the sources for
the term->enhanced_writec() entry.  Other instances of this will only be
found by applying a genuine K&R compiler or lint.

The yrange flip is next. The new bug with clipping of both pm3d and
contours in a 'set dgrid3d ; set view map' situation is about equally bad,
IMHO.

-- 
Hans-Bernhard Broeker (br...@ph...)
Even if all the snow were burnt, ashes would remain.

Re: Windows EMF

[Hans-Bernhard B. <br...@ph...>]

On Mon, 21 Jun 2004, Stephane Barbaray wrote:

> Hans-Bernhard Broeker wrote:
> 
> >On Mon, 21 Jun 2004, Stephane Barbaray wrote:
> >
> >  
> >
> >>Made again some new modification on emf.trm, the file is available now
> >>on sourceforge (Patches section), here is a part of the history relating
> >>the change :
> >>    
> >>
> >
> >Stephane, do I take this new patch submission to mean that you consider
> >your earlier patch to be out of date by now, and only the newer one should
> >be used?
> >  
> >
> 
> Yes indeed, this new version replace the earlier one

Ok. I'll close the other one as 'outdated', then, and unless I hear quick
objections, the new version goes into CVS tomorrow.

-- 
Hans-Bernhard Broeker (br...@ph...)
Even if all the snow were burnt, ashes would remain.

Re: Windows EMF

[Stephane B. <gn...@co...>]

Hans-Bernhard Broeker wrote:

>On Mon, 21 Jun 2004, Stephane Barbaray wrote:
>
> =20
>
>>Made again some new modification on emf.trm, the file is available now
>>on sourceforge (Patches section), here is a part of the history relatin=
g
>>the change :
>>   =20
>>
>
>Stephane, do I take this new patch submission to mean that you consider
>your earlier patch to be out of date by now, and only the newer one shou=
ld
>be used?
> =20
>

Yes indeed, this new version replace the earlier one

>[Hint: you could have submitted a new patch file into your old patch=20
>tracker entry instead...]
> =20
>
Except that I did submit as anonymous the two files before deciding the=20
need to register at sourceforce... ;-)

Bye,

--=20
St=E9phane BARBARAY.