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

Re: install TkiZ terminal gnuplot-4.4.0-rc1

[Tatsuro M. <tma...@ya...>]

Hello

Although the below patch is mainly for cygwin and windows build for lua terminal, please try it if you
use the source of the 4.4-rc1.
 
http://sourceforge.net/tracker/?func=detail&aid=2906430&group_id=2055&atid=302055


BTW, I recommend you to see config.log to get detailed information during the configure execution.

Regards

Tatsuro
--- gerion wrote:

> 
> Thanks for the quick response. Can't find it in the INSTALL though.
> now the terminal gets installed, but doesn't seem to work properly. The
> "./configure --with-lua yes" gives
> 
> checking for LUA... yes
> checking for library containing luaL_openlibs... none required
> checking lua.h usability... no
> checking lua.h presence... no
> checking for lua.h... no
> configure: WARNING: please add path to lua.h to CPPFLAGS in Makefile
> 
> however, the Makefile contains "CPPFLAGS =   -I/usr/include/lua5.1" which
> has the lua.h
> 
> Cheers, Christian
> -- 
> View this message in context:
> http://old.nabble.com/install-TkiZ-terminal-gnuplot-4.4.0-rc1-tp27479233p27523924.html
> Sent from the Gnuplot - Dev mailing list archive at Nabble.com.
> 
> 
> ------------------------------------------------------------------------------
> SOLARIS 10 is the OS for Data Centers - provides features such as DTrace,
> Predictive Self Healing and Award Winning ZFS. Get Solaris 10 NOW
> http://p.sf.net/sfu/solaris-dev2dev
> _______________________________________________
> gnuplot-beta mailing list
> gnu...@li...
> https://lists.sourceforge.net/lists/listinfo/gnuplot-beta
> 


--------------------------------------
VANCOUVER 2010 Olympic News [Yahoo! Sports/sportsnavi]
http://pr.mail.yahoo.co.jp/olympic/

Re: install TkiZ terminal gnuplot-4.4.0-rc1

[adad <ad...@gm...>]

it actually works. The problem was that in the terminal list and 
examples on the web, it's said to be "set term lua", what should be "set 
term tikz". Thanks, Christian

Ethan Merritt wrote:
> On Tuesday 09 February 2010 15:26:41 gerion wrote:
>> Thanks for the quick response. Can't find it in the INSTALL though.
> 
> Right. The note has been added for inclusion in the 4.4.0 release based on
> feedback from the release candidate.
> 
>> now the terminal gets installed, but doesn't seem to work properly. The
>> "./configure --with-lua yes" gives
>>
>> checking for LUA... yes
>> checking for library containing luaL_openlibs... none required
>> checking lua.h usability... no
>> checking lua.h presence... no
>> checking for lua.h... no
>> configure: WARNING: please add path to lua.h to CPPFLAGS in Makefile
>>
>> however, the Makefile contains "CPPFLAGS =   -I/usr/include/lua5.1" which
>> has the lua.h
> 
> That warning is from the configure script, not from make.
> I would have expected make to succeed if CPPFLAGS is correct.
> If "make" fails, could you attach the error messages?
> 
> 	Ethan
> 
>> Cheers, Christian
>> -- 
>> View this message in context: http://old.nabble.com/install-TkiZ-terminal-gnuplot-4.4.0-rc1-tp27479233p27523924.html
>> Sent from the Gnuplot - Dev mailing list archive at Nabble.com.
>>
>>
>> ------------------------------------------------------------------------------
>> SOLARIS 10 is the OS for Data Centers - provides features such as DTrace,
>> Predictive Self Healing and Award Winning ZFS. Get Solaris 10 NOW
>> http://p.sf.net/sfu/solaris-dev2dev
>> _______________________________________________
>> gnuplot-beta mailing list
>> gnu...@li...
>> https://lists.sourceforge.net/lists/listinfo/gnuplot-beta
>>
> 
> 
> 

Re: install TkiZ terminal gnuplot-4.4.0-rc1

[Allin C. <cot...@wf...>]

On Tue, 9 Feb 2010, gerion wrote:

> Thanks for the quick response. Can't find it in the INSTALL though.
> now the terminal gets installed, but doesn't seem to work properly. The
> "./configure --with-lua yes" gives
>
> checking for LUA... yes
> checking for library containing luaL_openlibs... none required
> checking lua.h usability... no
> checking lua.h presence... no
> checking for lua.h... no
> configure: WARNING: please add path to lua.h to CPPFLAGS in Makefile
>
> however, the Makefile contains "CPPFLAGS =   -I/usr/include/lua5.1" which
> has the lua.h

I'd have thought that CFLAGS would be more relevant than CPPFLAGS.

Allin Cottrell

Re: install TkiZ terminal gnuplot-4.4.0-rc1

[gerion <ad...@gm...>]

Thanks for the quick response. Can't find it in the INSTALL though.
now the terminal gets installed, but doesn't seem to work properly. The
"./configure --with-lua yes" gives

checking for LUA... yes
checking for library containing luaL_openlibs... none required
checking lua.h usability... no
checking lua.h presence... no
checking for lua.h... no
configure: WARNING: please add path to lua.h to CPPFLAGS in Makefile

however, the Makefile contains "CPPFLAGS =   -I/usr/include/lua5.1" which
has the lua.h

Cheers, Christian
-- 
View this message in context: http://old.nabble.com/install-TkiZ-terminal-gnuplot-4.4.0-rc1-tp27479233p27523924.html
Sent from the Gnuplot - Dev mailing list archive at Nabble.com.

Re: install TkiZ terminal gnuplot-4.4.0-rc1

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

On Tuesday 09 February 2010 13:17:58 gerion wrote:
> 
> Hello,
> 
> I tried to install the TkiZ terminal in v4.4.0-rc1 on Ubuntu 9.10, lua 5.1.

Known problem.  
We have added instructions in the INSTALL file for 4.4.0 to cover this case.

The problem is that Ubuntu does not install a lua package configuration file
with a generic name.

  204 Ubuntu:
  205 	./configure fails to find lua support because Ubuntu packages it as
  206 	"lua5.1" rather than "lua".  You can fix this by adding a symlink
  207 	prior to running ./configure
  208 	    ln -s /usr/lib/pkgconfig/lua5.1.pc  /usr/lib/pkgconfig/lua.pc
  209 
  210

		Ethan


> The configure command reads like this:
> 
> LUA_CFLAGS="-I /usr/include/lua5.1/lua.h" LUA_LIBS="-L /usr/lib/liblua5.1.a"
> ./configure --with-lua=yes
> 
> The command gives the following output:
> 
> ...
> checking for LUA... yes
> checking for library containing luaL_openlibs... no
> ...
> === configuring in lisp (/home/christian/Desktop/gnuplot-4.4.0-rc1/lisp)
> configure: running /bin/bash ./configure --disable-option-checking
> '--prefix=/usr/local' '--with-lua=yes' 'LUA_CFLAGS=-I
> /usr/include/lua5.1/lua.h' 'LUA_LIBS=-L /usr/lib/liblua5.1.a'
> --cache-file=/dev/null --srcdir=.
> ..
> ** Configuration summary for gnuplot 4.4.0-rc1:
> 
> gnuplot will be compiled with the following terminals:
> ...
> lua/TikZ terminal: no
> ...
> 
> Any ideas how to make this work? cheers
> -- 
> View this message in context: http://old.nabble.com/install-TkiZ-terminal-gnuplot-4.4.0-rc1-tp27479233p27479233.html
> Sent from the Gnuplot - Dev mailing list archive at Nabble.com.
> 
> 
> ------------------------------------------------------------------------------
> SOLARIS 10 is the OS for Data Centers - provides features such as DTrace,
> Predictive Self Healing and Award Winning ZFS. Get Solaris 10 NOW
> http://p.sf.net/sfu/solaris-dev2dev
> _______________________________________________
> gnuplot-beta mailing list
> gnu...@li...
> https://lists.sourceforge.net/lists/listinfo/gnuplot-beta
> 



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

install TkiZ terminal gnuplot-4.4.0-rc1

[gerion <ad...@gm...>]

Hello,

I tried to install the TkiZ terminal in v4.4.0-rc1 on Ubuntu 9.10, lua 5.1.
The configure command reads like this:

LUA_CFLAGS="-I /usr/include/lua5.1/lua.h" LUA_LIBS="-L /usr/lib/liblua5.1.a"
./configure --with-lua=yes

The command gives the following output:

...
checking for LUA... yes
checking for library containing luaL_openlibs... no
...
=== configuring in lisp (/home/christian/Desktop/gnuplot-4.4.0-rc1/lisp)
configure: running /bin/bash ./configure --disable-option-checking
'--prefix=/usr/local' '--with-lua=yes' 'LUA_CFLAGS=-I
/usr/include/lua5.1/lua.h' 'LUA_LIBS=-L /usr/lib/liblua5.1.a'
--cache-file=/dev/null --srcdir=.
..
** Configuration summary for gnuplot 4.4.0-rc1:

gnuplot will be compiled with the following terminals:
...
lua/TikZ terminal: no
...

Any ideas how to make this work? cheers
-- 
View this message in context: http://old.nabble.com/install-TkiZ-terminal-gnuplot-4.4.0-rc1-tp27479233p27479233.html
Sent from the Gnuplot - Dev mailing list archive at Nabble.com.

Re: Parsing bug?

[Thomas S. <t.s...@fz...>]

this has been repaired in the development version of gnuplot:


        G N U P L O T
        Version 4.2 patchlevel 6 
gnuplot> print 6000**4
-2086600704


        G N U P L O T
        Version 4.4 patchlevel 0-rc1
gnuplot> print 6000**4
-2086600704


        G N U P L O T
        Version 4.5 patchlevel 0
gnuplot> print 6000**4
1.296e+15


Daniel Farrell wrote:
> 
> Hi folks,
> 
> I just tried plotting this equation,
> 
> T0 = 300
> T1 = 6000
> set xrange [T0:T1]
> f(x) = (1 - x**4/T1**4) * (1 - T0/x)
> 
> Which should have two zeros at (x=300 and x=6000). It seems that because
> the numbers have been defined as integers it give the wrong result.
> Redefining,
> 
> T0 = 300.
> T1 = 6000.
> 
> plots the correct function.
> 
> Is there a case for converting ints to floats as standard in the gnuplot
> interpreter? When using gnuplot as a user, I would expect the maths to
> just work without having to think about programming details like ints of
> floats. In a programming environment of course this is what I would
> expected.
> 
> Cheers,
> 
> Dan
> 
> 
> 
> ------------------------------------------------------------------------------
> The Planet: dedicated and managed hosting, cloud storage, colocation
> Stay online with enterprise data centers and the best network in the
> business
> Choose flexible plans and management services without long-term contracts
> Personal 24x7 support from experience hosting pros just a phone call away.
> http://p.sf.net/sfu/theplanet-com
> _______________________________________________
> gnuplot-beta mailing list
> gnu...@li...
> https://lists.sourceforge.net/lists/listinfo/gnuplot-beta
> 
> 

-- 
View this message in context: http://old.nabble.com/Parsing-bug--tp27514663p27521028.html
Sent from the Gnuplot - Dev mailing list archive at Nabble.com.

Re: save and multiplot

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

On Monday 01 February 2010 16:11:43 adhemar wrote:
> 
> Hello everybody,
> 
> I am unable to save all the "plot" commands that I've given inside the
> multiplot mode, but it appears to only save the last plot command that I've
> issued.

Correct.
Only the most recent "plot" command is available.

> Is this working as intended?

Yes.

> Is there a way to save all the plot commands I 
> have given in multiplot mode?

You could issue a save after each plot command.

	Ethan

Parsing bug?

[Daniel <boy...@gm...>]

Hi folks,

I just tried plotting this equation,

T0 = 300
T1 = 6000
set xrange [T0:T1]
f(x) = (1 - x**4/T1**4) * (1 - T0/x)

Which should have two zeros at (x=300 and x=6000). It seems that because the numbers have been defined as integers it give the wrong result. Redefining,

T0 = 300.
T1 = 6000.

plots the correct function.

Is there a case for converting ints to floats as standard in the gnuplot interpreter? When using gnuplot as a user, I would expect the maths to just work without having to think about programming details like ints of floats. In a programming environment of course this is what I would expected.

Cheers,

Dan

Re: Windows terminal ignores fontname+size of labels

[Benjamin L. <lin...@gm...>]

> 
> Note: the windows terminal lacks yet another options:
> 	set term win close
> It should close the graph window, see set term {x11|wxt} close.
> Benjamin, can you add this functionality?

Yes, I'll look into it and hava a try.

benjamin
-- 
NEU: Mit GMX DSL über 1000,- ¿ sparen!
http://portal.gmx.net/de/go/dsl02

Re: Add 'size' and 'position' to windows terminal options

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

On Monday 08 February 2010, Tatsuro MATSUOKA wrote:
> Hello
> 
> --- Ethan Merritt  wrote:
> 
> > The only patch I see in the ChangeLogs that might explain this is the one from
> > Shige Takeno applied to 4.5 CVS 30-Dec-2009.   I have now applied that same patch 
> > to the 4.4 CVS source as well.   Please see if that fixes the problem.
> > 
> > 	Ethan
> 
> I have attached the patch by Shige Takeno applied to 4.5 CVS 30-Dec-2009 is applied.
> http://sourceforge.net/tracker/?func=detail&aid=2918370&group_id=2055&atid=302055
> 
> After that I attached the Benjamin's  patches. 
> The 'size' and 'position' to windows terminal options worked correctly.
> 
> 
> BTW,  how can I access the 4.4 CVS source?

Same root repository, but a special branch label

  CVSROOT=:pserver:ano...@gn...:/cvsroot/gnuplot
  cvs login
  cvs co -r branch-4-4-stable gnuplot


	Ethan

> I do not know the address. So I have tried
> 
> $ export CVSROOT=:pserver:ano...@gn...:/cvsroot/gnuplot-4.4
> $ cvs login
> Logging in to :pserver:ano...@gn...:2401/cvsroot/gnuplot-4.4
> CVS password:
> cvs login: authorization failed: server gnuplot.cvs.sourceforge.net rejected access to
> /cvsroot/gnuplot-4.4 for user anonymous
> 
> Any suggestions?
> 
> Regards
> 
> Tatsuro
> 
> --------------------------------------
> VANCOUVER 2010 Olympic News [Yahoo! Sports/sportsnavi]
> http://pr.mail.yahoo.co.jp/olympic/
> 

Re: Add 'size' and 'position' to windows terminal options

[Tatsuro M. <tma...@ya...>]

Hello

--- Ethan Merritt  wrote:

> The only patch I see in the ChangeLogs that might explain this is the one from
> Shige Takeno applied to 4.5 CVS 30-Dec-2009.   I have now applied that same patch 
> to the 4.4 CVS source as well.   Please see if that fixes the problem.
> 
> 	Ethan

I have attached the patch by Shige Takeno applied to 4.5 CVS 30-Dec-2009 is applied.
http://sourceforge.net/tracker/?func=detail&aid=2918370&group_id=2055&atid=302055

After that I attached the Benjamin's  patches. 
The 'size' and 'position' to windows terminal options worked correctly.


BTW,  how can I access the 4.4 CVS source?
I do not know the address. So I have tried

$ export CVSROOT=:pserver:ano...@gn...:/cvsroot/gnuplot-4.4
$ cvs login
Logging in to :pserver:ano...@gn...:2401/cvsroot/gnuplot-4.4
CVS password:
cvs login: authorization failed: server gnuplot.cvs.sourceforge.net rejected access to
/cvsroot/gnuplot-4.4 for user anonymous

Any suggestions?

Regards

Tatsuro

--------------------------------------
VANCOUVER 2010 Olympic News [Yahoo! Sports/sportsnavi]
http://pr.mail.yahoo.co.jp/olympic/

Re: Windows terminal ignores fontname+size of labels

[Tatsuro M. <tma...@ya...>]

Hello

I have tested also on 4.4rc-1 
--- Petr Mikulik  wrote:

> > The first two labels should look alike and the third one should
> > be the same in arial.
> > 
> > With the 4.4.0-rc1 binaries, the first and third labels look alike
> > and the second one only is in courier.
> > 
> > What's the effect of the patch when using linux/wine?
> 
> The looks like  labeltest-patched.png  i.e. the correct one (in both cvs and 
> 44rc1).

I have confirmed the same results as Petr.

Regards

Tatsuro

--------------------------------------
VANCOUVER 2010 Olympic News [Yahoo! Sports/sportsnavi]
http://pr.mail.yahoo.co.jp/olympic/

Re: Add 'size' and 'position' to windows terminal options

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

On Monday 08 February 2010 15:59:41 Tatsuro MATSUOKA wrote:
> Hello
> 
> --- Petr Mikulik wrote:
> > I tried again, and "set term win pos ... size ..." works for patched 
> > gnuplot-cvs but not for gnuplot 44rc1 even though the patch applies 
> > correctly. Can you try it?
> 
> 
> I have built gnuplot.exe for windows from 4.4rc-1 source with the patch of Add 'size' and 'position'
> to windows terminal options.
> 
> For 4.4rc-1, I have confirmed that the same behavior occurs as Petr reported.

The only patch I see in the ChangeLogs that might explain this is the one from
Shige Takeno applied to 4.5 CVS 30-Dec-2009.   I have now applied that same patch 
to the 4.4 CVS source as well.   Please see if that fixes the problem.

	Ethan




> 
> The position and size of graph window follow indicated in wignuplot.ini even though I set size and
> position option.
> 
> Regards
> 
> Tatsuro
>  
> 
> > > I have tried a new patch.
> > > 
> > > In my case (Msys+mingw on win xp), size and position seem to work correct with the new patch.
> > 
> > 
> > Petr
> > 
> > > > > One thing I am not sure about is, what is to be called the 'size'
> > > > > of the graph window.
> > > > > Is it the total extension of the window on screen? 
> > > > > Or is it the extension of the actual plotting canvas?
> > > > >
> > > > > I tend to view the 'size' as the extension of the canvas, but 
> > > > > I'm not convinced everybody else does...
> > > > 
> > > > Yes, it should be the canvas size.
> > > > 
> > > > I have tried the patch  gp_add-size-to-win-term.patch  on Linux/Wine but 
> > > > it does not work for me. It takes position from wgnuplot.ini or the default 
> > > > one.
> > > > 
> > > > ---
> > > > PM
> > 
> 
> 
> --------------------------------------
> VANCOUVER 2010 Olympic News [Yahoo! Sports/sportsnavi]
> http://pr.mail.yahoo.co.jp/olympic/
> 
> ------------------------------------------------------------------------------
> The Planet: dedicated and managed hosting, cloud storage, colocation
> Stay online with enterprise data centers and the best network in the business
> Choose flexible plans and management services without long-term contracts
> Personal 24x7 support from experience hosting pros just a phone call away.
> http://p.sf.net/sfu/theplanet-com
> _______________________________________________
> gnuplot-beta mailing list
> gnu...@li...
> https://lists.sourceforge.net/lists/listinfo/gnuplot-beta
> 



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

Re: Add 'size' and 'position' to windows terminal options

[Tatsuro M. <tma...@ya...>]

Hello

--- Petr Mikulik wrote:
> I tried again, and "set term win pos ... size ..." works for patched 
> gnuplot-cvs but not for gnuplot 44rc1 even though the patch applies 
> correctly. Can you try it?


I have built gnuplot.exe for windows from 4.4rc-1 source with the patch of Add 'size' and 'position'
to windows terminal options.

For 4.4rc-1, I have confirmed that the same behavior occurs as Petr reported.

The position and size of graph window follow indicated in wignuplot.ini even though I set size and
position option.

Regards

Tatsuro
 

> > I have tried a new patch.
> > 
> > In my case (Msys+mingw on win xp), size and position seem to work correct with the new patch.
> 
> 
> Petr
> 
> > > > One thing I am not sure about is, what is to be called the 'size'
> > > > of the graph window.
> > > > Is it the total extension of the window on screen? 
> > > > Or is it the extension of the actual plotting canvas?
> > > >
> > > > I tend to view the 'size' as the extension of the canvas, but 
> > > > I'm not convinced everybody else does...
> > > 
> > > Yes, it should be the canvas size.
> > > 
> > > I have tried the patch  gp_add-size-to-win-term.patch  on Linux/Wine but 
> > > it does not work for me. It takes position from wgnuplot.ini or the default 
> > > one.
> > > 
> > > ---
> > > PM
> 


--------------------------------------
VANCOUVER 2010 Olympic News [Yahoo! Sports/sportsnavi]
http://pr.mail.yahoo.co.jp/olympic/

Re: towards 4.4

[Hans-Bernhard B. <HBB...@t-...>]

Petr Mikulik wrote:

> key ... and Win key was on available on old keyboards? The -Insert shortcuts 
> are still working even they seem not to be documented!

wgnuplot does document them, of sorts: as hints in the context menu of 
the text window (right-click into the text menu, or select "Options->" 
from the text window's program menu (top-left corner icon).

Re: towards 4.4

[Hans-Bernhard B. <HBB...@t-...>]

Benjamin Lindner wrote:

> Well, actually, for a windows GUI application Ctrl+C means "copy to 
> clipboard"

Nowadays --- but wgnuplot is older than that tradition.  Back then 
Ctrl-Insert meant copy.  And it still does, in just about any 
application, including wgnuplot.

And wgnuplot is a GUI application only by necessity, not by choice. 
It's a console application forced into a GUI corset because that was the 
only way to create a Windows application at all, back in the day (we're 
talking about 16-bit Windows here).

> Coming from the windows world I find it more confusing if an application
> does not follow these "standards".

Well, something's gotta give.  A massively portable program like gnuplot 
can't follow each and every platform's conventions completely.

In the long time wgnuplot.exe has been around, the "standard" on MS 
Windows has changed several times over --- roughly every two years, when 
either a new version of Windows or Office hits the streets.  In the same 
time, wgnuplot barely changed at all.  That's both a strength and a 
weakness.  It means it still looks and feels the same to people 
accustomed to it from way back when, but it also means that it'll 
alienate newcomers to some extent.

Re: Add 'size' and 'position' to windows terminal options

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

> I have tried a new patch.
> 
> In my case (Msys+mingw on win xp), size and position seem to work correct with the new patch.

I tried again, and "set term win pos ... size ..." works for patched 
gnuplot-cvs but not for gnuplot 44rc1 even though the patch applies 
correctly. Can you try it?

Petr

> > > One thing I am not sure about is, what is to be called the 'size'
> > > of the graph window.
> > > Is it the total extension of the window on screen? 
> > > Or is it the extension of the actual plotting canvas?
> > >
> > > I tend to view the 'size' as the extension of the canvas, but 
> > > I'm not convinced everybody else does...
> > 
> > Yes, it should be the canvas size.
> > 
> > I have tried the patch  gp_add-size-to-win-term.patch  on Linux/Wine but 
> > it does not work for me. It takes position from wgnuplot.ini or the default 
> > one.
> > 
> > ---
> > PM

Re: Windows terminal ignores fontname+size of labels

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

> The first two labels should look alike and the third one should
> be the same in arial.
> 
> With the 4.4.0-rc1 binaries, the first and third labels look alike
> and the second one only is in courier.
> 
> What's the effect of the patch when using linux/wine?

The looks like  labeltest-patched.png  i.e. the correct one (in both cvs and 
44rc1).


Note: the windows terminal lacks yet another options:
	set term win close
It should close the graph window, see set term {x11|wxt} close.
Benjamin, can you add this functionality?

---
PM

Re: Add 'size' and 'position' to windows terminal options

[Tatsuro M. <tma...@ya...>]

Hello

I have tried a new patch.

In my case (Msys+mingw on win xp), size and position seem to work correct with the new patch.

Very nice!!!

Regards

Tatsuro
 
--- Petr Mikulik  wrote:

> > One thing I am not sure about is, what is to be called the 'size'
> > of the graph window.
> > Is it the total extension of the window on screen? 
> > Or is it the extension of the actual plotting canvas?
> >
> > I tend to view the 'size' as the extension of the canvas, but 
> > I'm not convinced everybody else does...
> 
> Yes, it should be the canvas size.
> 
> I have tried the patch  gp_add-size-to-win-term.patch  on Linux/Wine but 
> it does not work for me. It takes position from wgnuplot.ini or the default 
> one.
> 
> ---
> PM
> 
> 


--------------------------------------
VANCOUVER 2010 Olympic News [Yahoo! Sports/sportsnavi]
http://pr.mail.yahoo.co.jp/olympic/

Re: Windows terminal ignores fontname+size of labels

[Tatsuro M. <tma...@ya...>]

Hello

> Tatsuro, can you confirm this?

At the moment I have tested only on the cvs source

> 
> The first two labels should look alike and the third one should
> be the same in arial.

I have confirmed! Fine!
 
> With the 4.4.0-rc1 binaries, the first and third labels look alike
> and the second one only is in courier.

Please wait until tomorrow.

Regards

Tatsuro

--------------------------------------
VANCOUVER 2010 Olympic News [Yahoo! Sports/sportsnavi]
http://pr.mail.yahoo.co.jp/olympic/

Re: Windows terminal ignores fontname+size of labels

[Benjamin L. <lin...@gm...>]

> Hmm, I guess it's not that simple to solve this.

Okay, here's another try in the attached patch which looks now 
better (at least for me here).

Passing WIN_font to enhanced_recusion() in the initial call does 
the trick, however one cannot pass WIN_font directly, rather a 
local copy, since WIN_font as global variable gets modified if 
you alter the font inside the label string and then it 
is not switched back again for the remaining string.

set term windows enhanced font "Arial,15"
set label 1 "Courier 15: e^{i{/Symbol p}}=-1" at graph 0.1,0.9 font "Courier,15"
set label 2 "{/Courier=15 Courier 15: e^{i{/Symbol p}}=-1}" at graph 0.1,0.8 
set label 3 "Arial 15: e^{i{/Symbol p}}=-1" at graph 0.1,0.7
plot x with linespoints

The first two labels should look alike and the third one should
be the same in arial.

With the 4.4.0-rc1 binaries, the first and third labels look alike
and the second one only is in courier.

Tatsuro, can you confirm this?
What's the effect of the patch when using linux/wine?

(see the png's)

benjamin
-- 
Jetzt kostenlos herunterladen: Internet Explorer 8 und Mozilla Firefox 3.5 -
sicherer, schneller und einfacher! http://portal.gmx.net/de/go/atbrowser

Re: Add 'size' and 'position' to windows terminal options

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

> One thing I am not sure about is, what is to be called the 'size'
> of the graph window.
> Is it the total extension of the window on screen? 
> Or is it the extension of the actual plotting canvas?
>
> I tend to view the 'size' as the extension of the canvas, but 
> I'm not convinced everybody else does...

Yes, it should be the canvas size.

I have tried the patch  gp_add-size-to-win-term.patch  on Linux/Wine but 
it does not work for me. It takes position from wgnuplot.ini or the default 
one.

---
PM

Re: Add 'size' and 'position' to windows terminal options

[Benjamin L. <lin...@gm...>]

> gnuplot> set term win size 1024, 768 position 200, 200
> Terminal type set to 'windows'
> Options are 'color noenhanced font "Arial, 10"'
> gnuplot> plot sin(x)
> 
> However size are small compared to that expected (t) and position option
> was not changed.

Oops, my mistake. and a very stupid one.
Second try, again patch is attached.

One thing I am not sure about is, what is to be called the 'size'
of the graph window.
Is it the total extension of the window on screen? 
Or is it the extension of the actual plotting canvas?

The first currently also contains the status bar and it might 
possibly contain a menu bar or icon toolbars (like the wxt term).

If I copy the graph window to the clipboard, then I don't get the 
full window as displayed on screen but only the plotting canvas.
So the size of the copied plot differs from the size of the window
shown on screen.

I tend to view the 'size' as the extension of the canvas, but 
I'm not convinced everybody else does...

comments?
benjamin

-- 
Jetzt kostenlos herunterladen: Internet Explorer 8 und Mozilla Firefox 3.5 -
sicherer, schneller und einfacher! http://portal.gmx.net/de/go/chbrowser

Re: Windows terminal ignores fontname+size of labels

[Benjamin L. <lin...@gm...>]

> > > 
> > > Huh.  Does the "enhancedtext.dem" demo not work under [real] Windows?
> > > It works from the Windows terminal when run under linux+wine.
> > 
> > Yes it works, but none of the demos contain a "set label" with explicit 
> > fontname/size. *this* does not work.
> 
> Ah. I misunderstood.
>   
> I thought you were talking about the  enhanced text constructs, as in 
>   set label 11 "{/Symbol=18 \362@_{/=9.6 0}^{/=12 ^\245}} {e^{-{/Symbol
> m}^2/2} d}"
> where both the font and the size are correctly selected.

Right. Sorry, I wasn't precise in my description.

set term windows enhanced font "arial,9"
set label 1 "e^{i{/symbol p}}=-1" at graph 0.5,0.5 font "Times,15"
set label 2 "{/Courier=15 e^{i{/symbol p}}=-1}" at graph 0.5,0.6
plot sin(x) with linespoints

Here the second label works, but the first doesn't.

>From debugging I saw that the flow was:
     term.c: place_labels(), calling
 gadgetgs.c: write_label(), calling
     term.c: write_multiline()
now in write_multiline() there is the call to set_font in line 965

    /* EAM 9-Feb-2003 - Set font before calculating sizes */
    if (font && *font && t->set_font)
	(*t->set_font) (font);

which applies the explicit font for label 1. Ok.
Then write_multiline() calls WIN_enhanced_put_text() in win.trm where
I read in line 895

	/* This will restore the default font 
	   and update WIN_font and WIN_fontsize */
	WIN_set_font(NULL);

which then revokes the previously set fontname/size.

However, I see that enhanced_recursion() also has a parameter
'fontname' which is set to be NULL from within WIN_enhanced_put_text()
which, in line 2755 of enhanced_recursion()

	default:
	    /*{{{  print it */
	    (term->enhanced_open)(fontname, fontsize, base, widthflag, showflag, overprint);
	    (term->enhanced_writec)(*p);

leads to NULL passed in the variable fontname to WIN_enhanced_open()
consequently leading to the default fontname being substituted *again*.

And then I got a bit confused as to the many calls to set the 
fontname/size.

Hmm, I guess it's not that simple to solve this.

benjamin
-- 
GRATIS für alle GMX-Mitglieder: Die maxdome Movie-FLAT!
Jetzt freischalten unter http://portal.gmx.net/de/go/maxdome01