gnuplot development

Tatsuro Matsuoka also noticed that console gnuplot has problems 
executing plotstyles.gnu (creating files). I don't see any of those 
problems here on my system - but then again my build is without libgd, 
pango/cairo etc. and my computers are running XP and Vista.

I cannot test his executables here since my virus scanner claims to find 
a trojan - but 41 scanners at virustotal.com disagree and don't find 
anything.

Have you, by any chance, tested his previous version (18th March)?

   Bastian

Am 23.03.2011 12:48, schrieb Peter Juhasz:
> Dear list,
>
> I've downloaded the latest gnuplot 4.5 CVS build from Tatsuro
> Matsuoka's site ( http://www.tatsuromatsuoka.com/gnuplot/Eng/winbin/
> ).
>
> The console executable gnuplot.exe displays a strange behavior:
> instead on exiting, it hangs.
> If I start it from the console then type "exit", it doesn't return to
> the command line.
> Even worse, it also hangs when I invoke it as "gnuplot.exe -V".
> (That's how I noticed this: one of my applications that uses gnuplot
> checks for its presence and up-to-date-ness with gnuplot -V, and
> because of this, it appeared to hang on startup. And I spent a few
> confused minutes trying to debug my application before realizing that
> gnuplot's at fault.)
>
> Péter Juhász

I didn't test with his previous build (I don't know where to get it).

However, I did test the current (03/23) build on a Vista system and it
didn't exhibit this problem (it exited normally). It starts slowly,
though: there is a 5-10 second delay at startup.

Péter Juhász

On Wed, 2011-03-23 at 14:39 +0100, Bastian Märkisch wrote:
> Tatsuro Matsuoka also noticed that console gnuplot has problems 
> executing plotstyles.gnu (creating files). I don't see any of those 
> problems here on my system - but then again my build is without libgd, 
> pango/cairo etc. and my computers are running XP and Vista.
> 
> I cannot test his executables here since my virus scanner claims to find 
> a trojan - but 41 scanners at virustotal.com disagree and don't find 
> anything.
> 
> Have you, by any chance, tested his previous version (18th March)?
> 
>    Bastian
> 
> Am 23.03.2011 12:48, schrieb Peter Juhasz:
> > Dear list,
> >
> > I've downloaded the latest gnuplot 4.5 CVS build from Tatsuro
> > Matsuoka's site ( http://www.tatsuromatsuoka.com/gnuplot/Eng/winbin/
> > ).
> >
> > The console executable gnuplot.exe displays a strange behavior:
> > instead on exiting, it hangs.
> > If I start it from the console then type "exit", it doesn't return to
> > the command line.
> > Even worse, it also hangs when I invoke it as "gnuplot.exe -V".
> > (That's how I noticed this: one of my applications that uses gnuplot
> > checks for its presence and up-to-date-ness with gnuplot -V, and
> > because of this, it appeared to hang on startup. And I spent a few
> > confused minutes trying to debug my application before realizing that
> > gnuplot's at fault.)
> >
> > Péter Juhász

On Wednesday, March 23, 2011 11:37:09 am Juhász Péter wrote:
> I didn't test with his previous build (I don't know where to get it).
> 
> However, I did test the current (03/23) build on a Vista system and it
> didn't exhibit this problem (it exited normally). It starts slowly,
> though: there is a 5-10 second delay at startup.
> 
> Péter Juhász

For what it's worth, I can no longer run the current Windows build
under linux+wine.   

gnuplot.exe hangs on entry, after printing 
  gnuplot changed the codepage of this console to 437 to match
  the graph window. Some characters might only display correctly
  if you change the font to a non-raster type.
I am dubious that setting the codepage to 437 is the right thing
to do under linux+wine (earlier builds seemed happy to accept
input from my UTF-8 environment), but I can't really test this
because it is hung.

wgnuplot.exe gives me a text window to type in, but if I try to plot
anything it exits with an error message   
   "This version of GLib requires NT-based Windows"
This message comes from libglib-2.0-0.dll

	Ethan

> On Wed, 2011-03-23 at 14:39 +0100, Bastian Märkisch wrote:
> > Tatsuro Matsuoka also noticed that console gnuplot has problems 
> > executing plotstyles.gnu (creating files). I don't see any of those 
> > problems here on my system - but then again my build is without libgd, 
> > pango/cairo etc. and my computers are running XP and Vista.
> > 
> > I cannot test his executables here since my virus scanner claims to find 
> > a trojan - but 41 scanners at virustotal.com disagree and don't find 
> > anything.
> > 
> > Have you, by any chance, tested his previous version (18th March)?
> > 
> >    Bastian
> > 
> > Am 23.03.2011 12:48, schrieb Peter Juhasz:
> > > Dear list,
> > >
> > > I've downloaded the latest gnuplot 4.5 CVS build from Tatsuro
> > > Matsuoka's site ( http://www.tatsuromatsuoka.com/gnuplot/Eng/winbin/
> > > ).
> > >
> > > The console executable gnuplot.exe displays a strange behavior:
> > > instead on exiting, it hangs.
> > > If I start it from the console then type "exit", it doesn't return to
> > > the command line.
> > > Even worse, it also hangs when I invoke it as "gnuplot.exe -V".
> > > (That's how I noticed this: one of my applications that uses gnuplot
> > > checks for its presence and up-to-date-ness with gnuplot -V, and
> > > because of this, it appeared to hang on startup. And I spent a few
> > > confused minutes trying to debug my application before realizing that
> > > gnuplot's at fault.)
> > >
> > > Péter Juhász
> 
> 
> 
> ------------------------------------------------------------------------------
> Enable your software for Intel(R) Active Management Technology to meet the
> growing manageability and security demands of your customers. Businesses
> are taking advantage of Intel(R) vPro (TM) technology - will your software 
> be a part of the solution? Download the Intel(R) Manageability Checker 
> today! http://p.sf.net/sfu/intel-dev2devmar
> _______________________________________________
> gnuplot-beta mailing list
> gnuplot-beta@...
> https://lists.sourceforge.net/lists/listinfo/gnuplot-beta
> 

-- 
Ethan A Merritt
Biomolecular Structure Center,  K-428 Health Sciences Bldg
University of Washington, Seattle 98195-7742

On Wed, 2011-03-23 at 12:22 -0700, Ethan Merritt wrote:
> On Wednesday, March 23, 2011 11:37:09 am Juhász Péter wrote:
> > I didn't test with his previous build (I don't know where to get it).
> > 
> > However, I did test the current (03/23) build on a Vista system and it
> > didn't exhibit this problem (it exited normally). It starts slowly,
> > though: there is a 5-10 second delay at startup.
> > 
> > Péter Juhász
> 
> For what it's worth, I can no longer run the current Windows build
> under linux+wine.   
> 
> gnuplot.exe hangs on entry, after printing 
>   gnuplot changed the codepage of this console to 437 to match
>   the graph window. Some characters might only display correctly
>   if you change the font to a non-raster type.
> I am dubious that setting the codepage to 437 is the right thing
> to do under linux+wine (earlier builds seemed happy to accept
> input from my UTF-8 environment), but I can't really test this
> because it is hung.
> 
> wgnuplot.exe gives me a text window to type in, but if I try to plot
> anything it exits with an error message   
>    "This version of GLib requires NT-based Windows"
> This message comes from libglib-2.0-0.dll
> 
> 	Ethan

To add a further data point:

gnuplot.exe with linux+wine prints the gnuplot> prompt for me, and
allows to type in one command, but hangs after that. I don't know if
this is the same as you get.

wgnuplot.exe works (plotting is possible with wxt and windows terminals,
very little delay on startup and clean exit).

Péter Juhász

Hello

Confirmed.  Thank you for your report.

I have re-made windows binaries using the source at 2011-03-17 as a temporal treatment.
I will see what is wrong for gnuplot.exe with the changes on 2011-03-18.

Regards

Tatsuro
 

--- Peter Juhasz  wrote:

> Dear list,
> 
> I've downloaded the latest gnuplot 4.5 CVS build from Tatsuro
> Matsuoka's site ( http://www.tatsuromatsuoka.com/gnuplot/Eng/winbin/
> ).
> 
> The console executable gnuplot.exe displays a strange behavior:
> instead on exiting, it hangs.
> If I start it from the console then type "exit", it doesn't return to
> the command line.
> Even worse, it also hangs when I invoke it as "gnuplot.exe -V".
> (That's how I noticed this: one of my applications that uses gnuplot
> checks for its presence and up-to-date-ness with gnuplot -V, and
> because of this, it appeared to hang on startup. And I spent a few
> confused minutes trying to debug my application before realizing that
> gnuplot's at fault.)
> 
> P将アter Juh将。sz
> 
> ------------------------------------------------------------------------------
> Enable your software for Intel(R) Active Management Technology to meet the
> growing manageability and security demands of your customers. Businesses
> are taking advantage of Intel(R) vPro (TM) technology - will your software 
> be a part of the solution? Download the Intel(R) Manageability Checker 
> today! http://p.sf.net/sfu/intel-dev2devmar
> _______________________________________________
> gnuplot-beta mailing list
> gnuplot-beta@...
> https://lists.sourceforge.net/lists/listinfo/gnuplot-beta
>

On Wednesday, March 23, 2011 01:14:05 pm Juhász Péter wrote:
> On Wed, 2011-03-23 at 12:22 -0700, Ethan Merritt wrote:
> > On Wednesday, March 23, 2011 11:37:09 am Juhász Péter wrote:
> > > I didn't test with his previous build (I don't know where to get it).
> > > 
> > > However, I did test the current (03/23) build on a Vista system and it
> > > didn't exhibit this problem (it exited normally). It starts slowly,
> > > though: there is a 5-10 second delay at startup.
> > > 
> > > Péter Juhász
> > 
> > For what it's worth, I can no longer run the current Windows build
> > under linux+wine.   
> > 
> > gnuplot.exe hangs on entry, after printing 
> >   gnuplot changed the codepage of this console to 437 to match
> >   the graph window. Some characters might only display correctly
> >   if you change the font to a non-raster type.
> > I am dubious that setting the codepage to 437 is the right thing
> > to do under linux+wine (earlier builds seemed happy to accept
> > input from my UTF-8 environment), but I can't really test this
> > because it is hung.
> > 
> > wgnuplot.exe gives me a text window to type in, but if I try to plot
> > anything it exits with an error message   
> >    "This version of GLib requires NT-based Windows"
> > This message comes from libglib-2.0-0.dll
> > 
> > 	Ethan
> 
> To add a further data point:
> 
> gnuplot.exe with linux+wine prints the gnuplot> prompt for me, and
> allows to type in one command, but hangs after that. I don't know if
> this is the same as you get.

Yes. Same thing.
It does accept and executes commands in non-interactive mode, e.g.
    wine gnuplot.exe -e "plot sin(x)" -persist


> wgnuplot.exe works (plotting is possible with wxt and windows terminals,
> very little delay on startup and clean exit).

Now works for me also, using a different installation of wine.

Now that Tatsuro has provided a built of the CVS state as of March 17th,
could somebody please confirm that the problems go away with
that version? As I can't reproduce anything here, it's like fishing in 
murky waters...

As the codepage switching is the only thing that specifically changed 
the behaviour of the console, I am going to remove that from CVS 
again.

On Wednesday, March 23, 2011 01:14:05 pm Juhász Péter wrote:
>> On Wed, 2011-03-23 at 12:22 -0700, Ethan Merritt wrote:
>> > On Wednesday, March 23, 2011 11:37:09 am Juhász Péter wrote:
>> > > I didn't test with his previous build (I don't know where to get it).
>> > > 
>> > > However, I did test the current (03/23) build on a Vista system and it
>> > > didn't exhibit this problem (it exited normally). It starts slowly,
>> > > though: there is a 5-10 second delay at startup.

This slow start up may be related to the windows help system or fontconfig.
Does it also start slowly if you try to start it a second time?

  Bastian

"Bastian Märkisch" <bmaerkisch@...> wrote>

> Now that Tatsuro has provided a built of the CVS state as of March 17th,
> could somebody please confirm that the problems go away with
> that version? As I can't reproduce anything here, it's like fishing in 
> murky waters...

The executable in gp45-winbin.zip downloaded from 
   http://www.tatsuromatsuoka.com/gnuplot/Eng/winbin/
on 24 March  still contains the codepage-switching and still
hangs after the first carriage return on input.

So I don't think the executable that I found is the one from 17 March
that Tatsuro refers to.

It would be helpful if the windows build script would do something
equivalent to 
   head -1 ChangeLog | sed 's/ .*$//' > PATCHLEVEL

So that the date of the most recent entry in ChangeLog appears
in "show version".   I have that in my build script for linux,
but I don't know if that command works on windows.

	Ethan

On Thu, 2011-03-24 at 20:22 +0100, =?UTF-8?Q? Bastian_M=C3=A4rkisch ?=
wrote:
> Now that Tatsuro has provided a built of the CVS state as of March 17th,
> could somebody please confirm that the problems go away with
> that version? As I can't reproduce anything here, it's like fishing in 
> murky waters...
> 
> As the codepage switching is the only thing that specifically changed 
> the behaviour of the console, I am going to remove that from CVS 
> again.
> 

- Windows Vista:
gnuplot.exe starts with a delay (always, not just the first time), but
then it executes commands and exits normally.

- Linux + wine:
gnuplot.exe starts with a slight delay (less than one second, but
noticeable), it accepts one command then hangs.
Non-interactive mode works (e.g. wine gnuplot.exe 'plot x' or wine
gnuplot.exe -V)

- Windows 7:
This is where it gets interesting! 
If I start gnuplot.exe interactively, plot something, then exit, then it
exits normally. However, if I don't plot anything, just do "sh var" or
"exit", it hangs!
Non-interactively, 'gnuplot.exe -e "plot x"' works and exits but
'gnuplot -V' doesn't.

This is all with Tatsuro's latest binary package which does have
2011/03/17 as its last Changelog entry.


> On Wednesday, March 23, 2011 01:14:05 pm Juhász Péter wrote:
> >> On Wed, 2011-03-23 at 12:22 -0700, Ethan Merritt wrote:
> >> > On Wednesday, March 23, 2011 11:37:09 am Juhász Péter wrote:
> >> > > I didn't test with his previous build (I don't know where to get it).
> >> > > 
> >> > > However, I did test the current (03/23) build on a Vista system and it
> >> > > didn't exhibit this problem (it exited normally). It starts slowly,
> >> > > though: there is a 5-10 second delay at startup.
> 
> This slow start up may be related to the windows help system or fontconfig.
> Does it also start slowly if you try to start it a second time?
> 

Yes, there is a delay at every startup.

>   Bastian
> 

Péter Juhász

On Thu, 2011-03-24 at 12:50 -0700, Ethan A Merritt wrote:
> "Bastian Märkisch" <bmaerkisch@...> wrote>
> 
> > Now that Tatsuro has provided a built of the CVS state as of March 17th,
> > could somebody please confirm that the problems go away with
> > that version? As I can't reproduce anything here, it's like fishing in 
> > murky waters...
> 
> The executable in gp45-winbin.zip downloaded from 
>    http://www.tatsuromatsuoka.com/gnuplot/Eng/winbin/
> on 24 March  still contains the codepage-switching and still
> hangs after the first carriage return on input.
> 
> So I don't think the executable that I found is the one from 17 March
> that Tatsuro refers to.
> 
> It would be helpful if the windows build script would do something
> equivalent to 
>    head -1 ChangeLog | sed 's/ .*$//' > PATCHLEVEL
> 
> So that the date of the most recent entry in ChangeLog appears
> in "show version".   I have that in my build script for linux,
> but I don't know if that command works on windows.
> 
> 	Ethan

I think the build script should write the date of the last Changelog
entry right into gnuplot_date[] in version.c, regardless of the
platform. (It still says "September 2010", by the way...)

Péter Juhász

Hello

I have tested cvs binaries for windows provided by Prof. Kakuto

http://www.ring.gr.jp/pub/text/TeX/ptex-win32/w32/
gnuplot-45pl0w32.zip

The binaries are built by the MSVC complier.  The latest ChangeLog date is 2011-03-22.
Using this binary, exit problem seems not to occur for gnuplot.exe.

At the moment, I do not find the reason why the difference occur between mine and Kakuto's binaries.

Regards

Tatsuro

On Thursday, March 24, 2011, Juhász Péter wrote:
> I think the build script should write the date of the last Changelog
> entry right into gnuplot_date[] in version.c, regardless of the
> platform. (It still says "September 2010", by the way...)

One way to force this is "touch PATCHLEVEL".
But that puts in todays date, which is not necessarily what we want.
(This has bugged me for a while).

OK.  I've modifed src/Makefile.maint to take the "last modified"
date from the top of ChangeLog.  It works here, but I can't vouch
for whether it will work under Windows.

	Ethan

> - Linux + wine:
> gnuplot.exe starts with a slight delay (less than one second, but
> noticeable), it accepts one command then hangs.
> Non-interactive mode works (e.g. wine gnuplot.exe 'plot x' or wine
> gnuplot.exe -V)

You have to run gnuplot.exe through "wineconsole" instead of "wine", see
http://old.nabble.com/gnuplot.exe-in-wine-td29878883.html

The "wineconsole cmd" is much more DOS box-like environment than "wine cmd" 
in an xterm.

(Oh yes, it has taken me a long time to figure this out.)

Petr

Hello,

To exclude build problems with MinGW, I have done and uploaded a build 
of the current CVS version to
http://www.gnuplot.info/development/binaries/gp45-Mar25_2011-winbin-test.zip 


It is built with a fresh install of current MinGW and current stable 
releases of wxWidgets/cairo/pango. It's not a full release build, 
though, as most of the support files and the lua and gd terminals are 
missing. Also, plotstyles.gnu will fail with this version since libgd is 
missing. gnuplot-switch.exe is compiled with -DCONSOLE_SWITCH_CP,
gnuplot.exe without.

Again, on my XP system, I see none of the behaviour described earlier: 
immediate startup and exit and console gnuplot (both versions) works as 
expected (started with msys, cmd, "-V", w/ and without plots ...). I 
will try later on with Vista.

On thing, though: closing gnuplot*.exe via the "close"-button of the 
console seems to cause it to hang or crash. That is not the case for 
Tatsuro Matsuoka's build.

   Bastian


Am 25.03.2011 00:19, schrieb Tatsuro MATSUOKA:
> Hello
>
> I have tested cvs binaries for windows provided by Prof. Kakuto
>
> http://www.ring.gr.jp/pub/text/TeX/ptex-win32/w32/
> gnuplot-45pl0w32.zip
>
> The binaries are built by the MSVC complier.  The latest ChangeLog date is 2011-03-22.
> Using this binary, exit problem seems not to occur for gnuplot.exe.
>
> At the moment, I do not find the reason why the difference occur between mine and Kakuto's binaries.
>
> Regards
>
> Tatsuro
>
> ------------------------------------------------------------------------------
> Enable your software for Intel(R) Active Management Technology to meet the
> growing manageability and security demands of your customers. Businesses
> are taking advantage of Intel(R) vPro (TM) technology - will your software
> be a part of the solution? Download the Intel(R) Manageability Checker
> today! http://p.sf.net/sfu/intel-dev2devmar
> _______________________________________________
> gnuplot-beta mailing list
> gnuplot-beta@...
> https://lists.sourceforge.net/lists/listinfo/gnuplot-beta

Hello


I have built the binaries with the cvs source on 2011-03-26 on my win 7 PC.
(Windows 7 64 bit Home Premium sp1.)

../src/gnuplot.exe -e "winhelp=1" ../docs/plotstyles.gnu

set output out . 'figure_lines' . ext
                                     ^
cannot open file; output not changed
"../docs/plotstyles.gnu", line 21: util.c: No such file or directory

make hangs at this stage. 

As pointed out by Juh将。sz P将アter,
If I start gnuplot.exe interactively, plot something, then exit, then it
exits normally. However, if I don't plot anything, just do "sh var" or
"exit", it hangs!

I have tested
http://www.gnuplot.info/development/binaries/gp45-Mar25_2011-winbin-test.zip 

The situation is the same as that I have built today.
The gnuplot.exe and gnuplot-switch.exe caused the same behaviors.


Hmmm? Thus codepage modification is  not origin of exit fail on win 7 64 bit.

Regards

Tatsuro
--- Bastian M将」rkisch  wrote:

> Hello,
> 
> To exclude build problems with MinGW, I have done and uploaded a build 
> of the current CVS version to
> http://www.gnuplot.info/development/binaries/gp45-Mar25_2011-winbin-test.zip 
> 
> 
> It is built with a fresh install of current MinGW and current stable 
> releases of wxWidgets/cairo/pango. It's not a full release build, 
> though, as most of the support files and the lua and gd terminals are 
> missing. Also, plotstyles.gnu will fail with this version since libgd is 
> missing. gnuplot-switch.exe is compiled with -DCONSOLE_SWITCH_CP,
> gnuplot.exe without.
> 
> Again, on my XP system, I see none of the behaviour described earlier: 
> immediate startup and exit and console gnuplot (both versions) works as 
> expected (started with msys, cmd, "-V", w/ and without plots ...). I 
> will try later on with Vista.
> 
> On thing, though: closing gnuplot*.exe via the "close"-button of the 
> console seems to cause it to hang or crash. That is not the case for 
> Tatsuro Matsuoka's build.
> 
>    Bastian
> 
> 
> Am 25.03.2011 00:19, schrieb Tatsuro MATSUOKA:
> > Hello
> >
> > I have tested cvs binaries for windows provided by Prof. Kakuto
> >
> > http://www.ring.gr.jp/pub/text/TeX/ptex-win32/w32/
> > gnuplot-45pl0w32.zip
> >
> > The binaries are built by the MSVC complier.  The latest ChangeLog date is 2011-03-22.
> > Using this binary, exit problem seems not to occur for gnuplot.exe.
> >
> > At the moment, I do not find the reason why the difference occur between mine and Kakuto's
> binaries.
> >
> > Regards
> >
> > Tatsuro
> >
> > ------------------------------------------------------------------------------
> > Enable your software for Intel(R) Active Management Technology to meet the
> > growing manageability and security demands of your customers. Businesses
> > are taking advantage of Intel(R) vPro (TM) technology - will your software
> > be a part of the solution? Download the Intel(R) Manageability Checker
> > today! http://p.sf.net/sfu/intel-dev2devmar
> > _______________________________________________
> > gnuplot-beta mailing list
> > gnuplot-beta@...
> > https://lists.sourceforge.net/lists/listinfo/gnuplot-beta
> 
> ------------------------------------------------------------------------------
> Enable your software for Intel(R) Active Management Technology to meet the
> growing manageability and security demands of your customers. Businesses
> are taking advantage of Intel(R) vPro (TM) technology - will your software 
> be a part of the solution? Download the Intel(R) Manageability Checker 
> today! http://p.sf.net/sfu/intel-dev2devmar
> _______________________________________________
> gnuplot-beta mailing list
> gnuplot-beta@...
> https://lists.sourceforge.net/lists/listinfo/gnuplot-beta
>

Hello

The exit fail also occurred at windows 7 starter 32 bit. (netbook)
Thus this problem seem to be specific windows 7.

I do not have any idea on this fail.

Regards

Tatsuro


--- Tatsuro MATSUOKA wrote:

> Hello
> 
> 
> I have built the binaries with the cvs source on 2011-03-26 on my win 7 PC.
> (Windows 7 64 bit Home Premium sp1.)
> 
> ../src/gnuplot.exe -e "winhelp=1" ../docs/plotstyles.gnu
> 
> set output out . 'figure_lines' . ext
>                                      ^
> cannot open file; output not changed
> "../docs/plotstyles.gnu", line 21: util.c: No such file or directory
> 
> make hangs at this stage. 
> 
> As pointed out by Juh将。sz P将アter,
> If I start gnuplot.exe interactively, plot something, then exit, then it
> exits normally. However, if I don't plot anything, just do "sh var" or
> "exit", it hangs!
> 
> I have tested
> http://www.gnuplot.info/development/binaries/gp45-Mar25_2011-winbin-test.zip 
> 
> The situation is the same as that I have built today.
> The gnuplot.exe and gnuplot-switch.exe caused the same behaviors.
> 
> 
> Hmmm? Thus codepage modification is  not origin of exit fail on win 7 64 bit.
> 
> Regards
> 
> Tatsuro
> --- Bastian M将」rkisch  wrote:
> 
> > Hello,
> > 
> > To exclude build problems with MinGW, I have done and uploaded a build 
> > of the current CVS version to
> > http://www.gnuplot.info/development/binaries/gp45-Mar25_2011-winbin-test.zip 
> > 
> > 
> > It is built with a fresh install of current MinGW and current stable 
> > releases of wxWidgets/cairo/pango. It's not a full release build, 
> > though, as most of the support files and the lua and gd terminals are 
> > missing. Also, plotstyles.gnu will fail with this version since libgd is 
> > missing. gnuplot-switch.exe is compiled with -DCONSOLE_SWITCH_CP,
> > gnuplot.exe without.
> > 
> > Again, on my XP system, I see none of the behaviour described earlier: 
> > immediate startup and exit and console gnuplot (both versions) works as 
> > expected (started with msys, cmd, "-V", w/ and without plots ...). I 
> > will try later on with Vista.
> > 
> > On thing, though: closing gnuplot*.exe via the "close"-button of the 
> > console seems to cause it to hang or crash. That is not the case for 
> > Tatsuro Matsuoka's build.
> > 
> >    Bastian
> > 
> > 
> > Am 25.03.2011 00:19, schrieb Tatsuro MATSUOKA:
> > > Hello
> > >
> > > I have tested cvs binaries for windows provided by Prof. Kakuto
> > >
> > > http://www.ring.gr.jp/pub/text/TeX/ptex-win32/w32/
> > > gnuplot-45pl0w32.zip
> > >
> > > The binaries are built by the MSVC complier.  The latest ChangeLog date is 2011-03-22.
> > > Using this binary, exit problem seems not to occur for gnuplot.exe.
> > >
> > > At the moment, I do not find the reason why the difference occur between mine and Kakuto's
> > binaries.
> > >
> > > Regards
> > >
> > > Tatsuro
> > >
> > > ------------------------------------------------------------------------------
> > > Enable your software for Intel(R) Active Management Technology to meet the
> > > growing manageability and security demands of your customers. Businesses
> > > are taking advantage of Intel(R) vPro (TM) technology - will your software
> > > be a part of the solution? Download the Intel(R) Manageability Checker
> > > today! http://p.sf.net/sfu/intel-dev2devmar
> > > _______________________________________________
> > > gnuplot-beta mailing list
> > > gnuplot-beta@...
> > > https://lists.sourceforge.net/lists/listinfo/gnuplot-beta
> > 
> > ------------------------------------------------------------------------------
> > Enable your software for Intel(R) Active Management Technology to meet the
> > growing manageability and security demands of your customers. Businesses
> > are taking advantage of Intel(R) vPro (TM) technology - will your software 
> > be a part of the solution? Download the Intel(R) Manageability Checker 
> > today! http://p.sf.net/sfu/intel-dev2devmar
> > _______________________________________________
> > gnuplot-beta mailing list
> > gnuplot-beta@...
> > https://lists.sourceforge.net/lists/listinfo/gnuplot-beta
> > 
> 
> 
> ------------------------------------------------------------------------------
> Enable your software for Intel(R) Active Management Technology to meet the
> growing manageability and security demands of your customers. Businesses
> are taking advantage of Intel(R) vPro (TM) technology - will your software 
> be a part of the solution? Download the Intel(R) Manageability Checker 
> today! http://p.sf.net/sfu/intel-dev2devmar
> _______________________________________________
> gnuplot-beta mailing list
> gnuplot-beta@...
> https://lists.sourceforge.net/lists/listinfo/gnuplot-beta
>

Hello

When I have checked out by  
cvs update -D '2011-03-21'

2011-03-20  Bastian Maerkisch  <bmaerkisch@...>

	* src/win/wmenu.c (LoadMacros), src/win/wgnuplot.mnu ([Button]): Add
	icons to the text-window toolbar.

	* docs/windows/doc2html.c (process_line): Don't include '=' items in
	index. This	avoids duplicate entries which caused wgnuplot to stall 
	when an e.g. 'set term png; help "png"' was issued.

the above were included and no error happened on windows 7.

When I have checked out by  
cvs update -D '2011-03-22'

2011-03-20  Ethan A Merritt  <merritt@...>

	* src/util.c (equals): Test here for token index in range, rather than
	trusting all callers to get it right.

	* src/command.c src/parse.c: Token index range test is now redundant.

	* docs/gnuplot.doc: Document that ^D from an empty line acts as EOF.

the above were included and caused error on windows 7.

I will see in more details.

Regards

Tatsuro

--- Tatsuro MATSUOKA  wrote:

> Hello
> 
> The exit fail also occurred at windows 7 starter 32 bit. (netbook)
> Thus this problem seem to be specific windows 7.
> 
> I do not have any idea on this fail.
> 
> Regards
> 
> Tatsuro
> 
> 
> --- Tatsuro MATSUOKA wrote:
> 
> > Hello
> > 
> > 
> > I have built the binaries with the cvs source on 2011-03-26 on my win 7 PC.
> > (Windows 7 64 bit Home Premium sp1.)
> > 
> > ../src/gnuplot.exe -e "winhelp=1" ../docs/plotstyles.gnu
> > 
> > set output out . 'figure_lines' . ext
> >                                      ^
> > cannot open file; output not changed
> > "../docs/plotstyles.gnu", line 21: util.c: No such file or directory
> > 
> > make hangs at this stage. 
> > 
> > As pointed out by Juh将。sz P将アter,
> > If I start gnuplot.exe interactively, plot something, then exit, then it
> > exits normally. However, if I don't plot anything, just do "sh var" or
> > "exit", it hangs!
> > 
> > I have tested
> > http://www.gnuplot.info/development/binaries/gp45-Mar25_2011-winbin-test.zip 
> > 
> > The situation is the same as that I have built today.
> > The gnuplot.exe and gnuplot-switch.exe caused the same behaviors.
> > 
> > 
> > Hmmm? Thus codepage modification is  not origin of exit fail on win 7 64 bit.
> > 
> > Regards
> > 
> > Tatsuro
> > --- Bastian M将」rkisch  wrote:
> > 
> > > Hello,
> > > 
> > > To exclude build problems with MinGW, I have done and uploaded a build 
> > > of the current CVS version to
> > > http://www.gnuplot.info/development/binaries/gp45-Mar25_2011-winbin-test.zip 
> > > 
> > > 
> > > It is built with a fresh install of current MinGW and current stable 
> > > releases of wxWidgets/cairo/pango. It's not a full release build, 
> > > though, as most of the support files and the lua and gd terminals are 
> > > missing. Also, plotstyles.gnu will fail with this version since libgd is 
> > > missing. gnuplot-switch.exe is compiled with -DCONSOLE_SWITCH_CP,
> > > gnuplot.exe without.
> > > 
> > > Again, on my XP system, I see none of the behaviour described earlier: 
> > > immediate startup and exit and console gnuplot (both versions) works as 
> > > expected (started with msys, cmd, "-V", w/ and without plots ...). I 
> > > will try later on with Vista.
> > > 
> > > On thing, though: closing gnuplot*.exe via the "close"-button of the 
> > > console seems to cause it to hang or crash. That is not the case for 
> > > Tatsuro Matsuoka's build.
> > > 
> > >    Bastian
> > > 
> > > 
> > > Am 25.03.2011 00:19, schrieb Tatsuro MATSUOKA:
> > > > Hello
> > > >
> > > > I have tested cvs binaries for windows provided by Prof. Kakuto
> > > >
> > > > http://www.ring.gr.jp/pub/text/TeX/ptex-win32/w32/
> > > > gnuplot-45pl0w32.zip
> > > >
> > > > The binaries are built by the MSVC complier.  The latest ChangeLog date is 2011-03-22.
> > > > Using this binary, exit problem seems not to occur for gnuplot.exe.
> > > >
> > > > At the moment, I do not find the reason why the difference occur between mine and Kakuto's
> > > binaries.
> > > >
> > > > Regards
> > > >
> > > > Tatsuro
> > > >
> > > > ------------------------------------------------------------------------------
> > > > Enable your software for Intel(R) Active Management Technology to meet the
> > > > growing manageability and security demands of your customers. Businesses
> > > > are taking advantage of Intel(R) vPro (TM) technology - will your software
> > > > be a part of the solution? Download the Intel(R) Manageability Checker
> > > > today! http://p.sf.net/sfu/intel-dev2devmar
> > > > _______________________________________________
> > > > gnuplot-beta mailing list
> > > > gnuplot-beta@...
> > > > https://lists.sourceforge.net/lists/listinfo/gnuplot-beta
> > > 
> > > ------------------------------------------------------------------------------
> > > Enable your software for Intel(R) Active Management Technology to meet the
> > > growing manageability and security demands of your customers. Businesses
> > > are taking advantage of Intel(R) vPro (TM) technology - will your software 
> > > be a part of the solution? Download the Intel(R) Manageability Checker 
> > > today! http://p.sf.net/sfu/intel-dev2devmar
> > > _______________________________________________
> > > gnuplot-beta mailing list
> > > gnuplot-beta@...
> > > https://lists.sourceforge.net/lists/listinfo/gnuplot-beta
> > > 
> > 
> > 
> > ------------------------------------------------------------------------------
> > Enable your software for Intel(R) Active Management Technology to meet the
> > growing manageability and security demands of your customers. Businesses
> > are taking advantage of Intel(R) vPro (TM) technology - will your software 
> > be a part of the solution? Download the Intel(R) Manageability Checker 
> > today! http://p.sf.net/sfu/intel-dev2devmar
> > _______________________________________________
> > gnuplot-beta mailing list
> > gnuplot-beta@...
> > https://lists.sourceforge.net/lists/listinfo/gnuplot-beta
> > 
> 
>

Hello

Hmmmm

I cannot figure out what is happening
I have checked out the source by 
cvs update -D '2011-03-22'

and I have copied the  src/util.c, src/command.c and src/parse.c before change from the cvs
repository.

I have tried to build.

The build error occur and exit command make failure on gnuplot.exe on windows 7.
???????????????????????????

I cannot find the reason at the moment :-( .
However, at least codepage change is not origin of the error.

Regards

Tatsuro

 
--- Tatsuro MATSUOKA  wrote:

> Hello
> 
> When I have checked out by  
> cvs update -D '2011-03-21'
> 
> 2011-03-20  Bastian Maerkisch  <bmaerkisch@...>
> 
> 	* src/win/wmenu.c (LoadMacros), src/win/wgnuplot.mnu ([Button]): Add
> 	icons to the text-window toolbar.
> 
> 	* docs/windows/doc2html.c (process_line): Don't include '=' items in
> 	index. This	avoids duplicate entries which caused wgnuplot to stall 
> 	when an e.g. 'set term png; help "png"' was issued.
> 
> the above were included and no error happened on windows 7.
> 
> When I have checked out by  
> cvs update -D '2011-03-22'
> 
> 2011-03-20  Ethan A Merritt  <merritt@...>
> 
> 	* src/util.c (equals): Test here for token index in range, rather than
> 	trusting all callers to get it right.
> 
> 	* src/command.c src/parse.c: Token index range test is now redundant.
> 
> 	* docs/gnuplot.doc: Document that ^D from an empty line acts as EOF.
> 
> the above were included and caused error on windows 7.
> 
> I will see in more details.
> 
> Regards
> 
> Tatsuro
> 
> --- Tatsuro MATSUOKA  wrote:
> 
> > Hello
> > 
> > The exit fail also occurred at windows 7 starter 32 bit. (netbook)
> > Thus this problem seem to be specific windows 7.
> > 
> > I do not have any idea on this fail.
> > 
> > Regards
> > 
> > Tatsuro
> > 
> > 
> > --- Tatsuro MATSUOKA wrote:
> > 
> > > Hello
> > > 
> > > 
> > > I have built the binaries with the cvs source on 2011-03-26 on my win 7 PC.
> > > (Windows 7 64 bit Home Premium sp1.)
> > > 
> > > ../src/gnuplot.exe -e "winhelp=1" ../docs/plotstyles.gnu
> > > 
> > > set output out . 'figure_lines' . ext
> > >                                      ^
> > > cannot open file; output not changed
> > > "../docs/plotstyles.gnu", line 21: util.c: No such file or directory
> > > 
> > > make hangs at this stage. 
> > > 
> > > As pointed out by Juh将。sz P将アter,
> > > If I start gnuplot.exe interactively, plot something, then exit, then it
> > > exits normally. However, if I don't plot anything, just do "sh var" or
> > > "exit", it hangs!
> > > 
> > > I have tested
> > > http://www.gnuplot.info/development/binaries/gp45-Mar25_2011-winbin-test.zip 
> > > 
> > > The situation is the same as that I have built today.
> > > The gnuplot.exe and gnuplot-switch.exe caused the same behaviors.
> > > 
> > > 
> > > Hmmm? Thus codepage modification is  not origin of exit fail on win 7 64 bit.
> > > 
> > > Regards
> > > 
> > > Tatsuro
> > > --- Bastian M将」rkisch  wrote:
> > > 
> > > > Hello,
> > > > 
> > > > To exclude build problems with MinGW, I have done and uploaded a build 
> > > > of the current CVS version to
> > > > http://www.gnuplot.info/development/binaries/gp45-Mar25_2011-winbin-test.zip 
> > > > 
> > > > 
> > > > It is built with a fresh install of current MinGW and current stable 
> > > > releases of wxWidgets/cairo/pango. It's not a full release build, 
> > > > though, as most of the support files and the lua and gd terminals are 
> > > > missing. Also, plotstyles.gnu will fail with this version since libgd is 
> > > > missing. gnuplot-switch.exe is compiled with -DCONSOLE_SWITCH_CP,
> > > > gnuplot.exe without.
> > > > 
> > > > Again, on my XP system, I see none of the behaviour described earlier: 
> > > > immediate startup and exit and console gnuplot (both versions) works as 
> > > > expected (started with msys, cmd, "-V", w/ and without plots ...). I 
> > > > will try later on with Vista.
> > > > 
> > > > On thing, though: closing gnuplot*.exe via the "close"-button of the 
> > > > console seems to cause it to hang or crash. That is not the case for 
> > > > Tatsuro Matsuoka's build.
> > > > 
> > > >    Bastian
> > > > 
> > > > 
> > > > Am 25.03.2011 00:19, schrieb Tatsuro MATSUOKA:
> > > > > Hello
> > > > >
> > > > > I have tested cvs binaries for windows provided by Prof. Kakuto
> > > > >
> > > > > http://www.ring.gr.jp/pub/text/TeX/ptex-win32/w32/
> > > > > gnuplot-45pl0w32.zip
> > > > >
> > > > > The binaries are built by the MSVC complier.  The latest ChangeLog date is 2011-03-22.
> > > > > Using this binary, exit problem seems not to occur for gnuplot.exe.
> > > > >
> > > > > At the moment, I do not find the reason why the difference occur between mine and
> Kakuto's
> > > > binaries.
> > > > >
> > > > > Regards
> > > > >
> > > > > Tatsuro
> > > > >
> > > > > ------------------------------------------------------------------------------
> > > > > Enable your software for Intel(R) Active Management Technology to meet the
> > > > > growing manageability and security demands of your customers. Businesses
> > > > > are taking advantage of Intel(R) vPro (TM) technology - will your software
> > > > > be a part of the solution? Download the Intel(R) Manageability Checker
> > > > > today! http://p.sf.net/sfu/intel-dev2devmar
> > > > > _______________________________________________
> > > > > gnuplot-beta mailing list
> > > > > gnuplot-beta@...
> > > > > https://lists.sourceforge.net/lists/listinfo/gnuplot-beta
> > > > 
> > > > ------------------------------------------------------------------------------
> > > > Enable your software for Intel(R) Active Management Technology to meet the
> > > > growing manageability and security demands of your customers. Businesses
> > > > are taking advantage of Intel(R) vPro (TM) technology - will your software 
> > > > be a part of the solution? Download the Intel(R) Manageability Checker 
> > > > today! http://p.sf.net/sfu/intel-dev2devmar
> > > > _______________________________________________
> > > > gnuplot-beta mailing list
> > > > gnuplot-beta@...
> > > > https://lists.sourceforge.net/lists/listinfo/gnuplot-beta
> > > > 
> > > 
> > > 
> > > ------------------------------------------------------------------------------
> > > Enable your software for Intel(R) Active Management Technology to meet the
> > > growing manageability and security demands of your customers. Businesses
> > > are taking advantage of Intel(R) vPro (TM) technology - will your software 
> > > be a part of the solution? Download the Intel(R) Manageability Checker 
> > > today! http://p.sf.net/sfu/intel-dev2devmar
> > > _______________________________________________
> > > gnuplot-beta mailing list
> > > gnuplot-beta@...
> > > https://lists.sourceforge.net/lists/listinfo/gnuplot-beta
> > > 
> > 
> > 
> 
>

On 25.03.2011 20:21, Bastian Märkisch wrote:
> When trying to build with MinGW and cairo/pango, I noticed a weird
> problem: make seems to have problems with variables defined like this:
>
> 	CAIRO_CFLAGS = $(shell pkg-config --cflags cairo)

Note that the make documentation pretty strongly recommends not to do 
that.  They recommend the other type of variable definition instead:

	CAIRO_CFLAGS := $(shell pkg-config --cflags cairo)

One of the issues is that with the '=' style definition, pkg-config 
would be called every single time that you use any command (indirectly) 
including $(CAIRO_CFLAGS).  Not something I fell particularly 
comfortable doing on MS Windows.

> -DGNUPLOT_PS_DIR="share/PostScript" -DGNUPLOT_JS_DIR="share/js"
> <snip>
> libs/include/freetype2 -Id:/Software/mingw-libs/include/libpng14
                          --^^--
> <snip>

Hmmm... I'm not sure drive letters are a good idea in makefiles (or 
anywhere in a MinGW environment, for that matter).  They tend to royally 
screw up make's usual interpretation of ':' for starters.

Is that raw MinGW, as opposed to MSYS+MinGW you're using?  AFAICS, 
that's not been the typical setup for quite some time now.

>On 25.03.2011 20:21, Bastian Märkisch wrote:
>> When trying to build with MinGW and cairo/pango, I noticed a weird
>> problem: make seems to have problems with variables defined like this:
>>
>> 	CAIRO_CFLAGS = $(shell pkg-config --cflags cairo)
>
>Note that the make documentation pretty strongly recommends not to do 
>that.  They recommend the other type of variable definition instead:
>
>	CAIRO_CFLAGS := $(shell pkg-config --cflags cairo)
>
>One of the issues is that with the '=' style definition, pkg-config 
>would be called every single time that you use any command (indirectly) 
>including $(CAIRO_CFLAGS).  Not something I fell particularly 
>comfortable doing on MS Windows.

Thanks for the clarification. I have updated config/mingw/Makefile accordingly.

>
>> -DGNUPLOT_PS_DIR="share/PostScript" -DGNUPLOT_JS_DIR="share/js"
>> 
>> libs/include/freetype2 -Id:/Software/mingw-libs/include/libpng14
>                          --^^--
>> 
>
>Hmmm... I'm not sure drive letters are a good idea in makefiles (or 
>anywhere in a MinGW environment, for that matter).  They tend to royally 
>screw up make's usual interpretation of ':' for starters.
>
>Is that raw MinGW, as opposed to MSYS+MinGW you're using?  AFAICS, 
>that's not been the typical setup for quite some time now.
>

That is MinGW+MSYS. It is really the pkg-config tool which returns these drive 
letters. Tatsuro Matsuoka's solution was to edit all the .pc files. The updated 
Makefile filters them out with the help of sed instead.

  Bastian

On Friday, March 25, 2011 10:15:28 am Douglas Mason wrote:
> Dear Developers,
> 
> There's a little discussion going on at the bug request tracker about making
> an explicit color column for pm3d. The idea is that rgbimage and "w p lc
> var" allow you to color each point with an explicit color, not interpolating
> on a palette. Since pm3d offers options like "corner2colors", which tells
> pm3d to use a color that is NOT interpolated among nearby points, it seems
> reasonable that a pm3d option could be made for explicit colors.
> 
> I was wondering what to make of this related paragraph in the documentation:
> 
> "Another drawing algorithm, which would draw quadrangles around a given node
> by taking corners from averaged (x,y)-coordinates of its surrounding 4 nodes
> while using node�s color, could be implemented in the future. This is
> already done for drawing images (2D grids) via image and rgbimage styles."
> 
> I'd love to hear people's thoughts on the possibility of implementing such a
> feature. If not, can anyone suggest another program that does this?

I think that particular hypothetical feature is not what you want.
It would guarantee that every color value is the result of interpolation,
which generally works badly for RGB colors.

Instead, as summarized on the tracker, I think you want an additional option
to pm3d that specifies a mapping function Gray_value -> RGB_Color, where
Gray_value is not a true continuous value running from [0:1] but instead
runs from [0 : 2^24] so that  int(Gray_value) is a 24-bit RGB color spec.

Although I am not so familiar with the code for pm3d, this sounds possible.
Am I failing to see some fundamental difficulty?

On Friday, March 25, 2011 10:32:05 am Ethan A Merritt wrote:
> On Friday, March 25, 2011 10:15:28 am Douglas Mason wrote:
> > Dear Developers,
> > 
> > There's a little discussion going on at the bug request tracker about making
> > an explicit color column for pm3d. The idea is that rgbimage and "w p lc
> > var" allow you to color each point with an explicit color, not interpolating
> > on a palette. Since pm3d offers options like "corner2colors", which tells
> > pm3d to use a color that is NOT interpolated among nearby points, it seems
> > reasonable that a pm3d option could be made for explicit colors. 
 
> Although I am not so familiar with the code for pm3d, this sounds possible.
> Am I failing to see some fundamental difficulty?

Trial patch and sample PNG output now on the tracker.
https://sourceforge.net/tracker/?func=detail&atid=352055&aid=3237990&group_id=2055

I can't think of any drawback to allowing "with pm3d lc rgb variable",
but go ahead and try to find something that breaks :-)

	Ethan