plplot-general — PLplot user discussion group

[Plplot-general] real-time plplot

[Owens, T. <Tho...@gt...>]

I have a real-time plotting application with multiple plots in one window that I would like to speed up some.  The code is

plspage(0,0,2000,430,0,0);
plsdev("xwin");
plsetopt("drvopt","nobuffered=1");
plinit();

plssub(2,1);

while(1){
  //get updated data
  getdataPlot1(data_1);
  getdataPlot2(data_2);

//update subplot 1
pladv(1);
plenv0(0.0, 1000, 0, 1000, 0, 1);
pllab("time","amp","data plot 1");
plshades(data_1, 1000,1000,NULL,0,1000,0,1000,......);

//update subplot 2
pladv(2);
plenv0(0.0, 1000, 0, 1000, 0, 1);
pllab("time","amp","data plot 2");
plshades(data_2, 1000,1000,NULL,0,1000,0,1000,......);
}//end while(1)

Is there a way to get to subplot 1 or subplot 2 without calling pladv and having the entire plot, axis, title, etc. erased?  Can I avoid having to recall plenv0 and pllab?  Is there a more efficient way to do real-time plotting?

Thank you for your assistance.

Tom

Re: [Plplot-general] Any way to get a list of available devices to a program?

[Alan W. I. <ir...@be...>]

Hi James:

I am switching further discussion of this topic to the plplot-devel
mailing list.

Alan
__________________________
Alan W. Irwin

Astronomical research affiliation with Department of Physics and Astronomy,
University of Victoria (astrowww.phys.uvic.ca).

Programming affiliations with the FreeEOS equation-of-state
implementation for stellar interiors (freeeos.sf.net); the Time
Ephemerides project (timeephem.sf.net); PLplot scientific plotting
software package (plplot.sf.net); the libLASi project
(unifont.org/lasi); the Loads of Linux Links project (loll.sf.net);
and the Linux Brochure Project (lbproject.sf.net).
__________________________

Linux-powered Science
__________________________

[Plplot-general] Any way to get a list of available devices to a program?

[James T. <jt...@gm...>]

When a program using plplot is run without a device specified (or with an
unknown device), the program produces a list of devices and prompts for a
device.

Is there any way to get that list to a program (e.g. to put them into a
selector widget, or to find available print-like devices)? [Preferably to
Fortran].

James

Re: [Plplot-general] Build documentation

[Alexis P. <ale...@ma...>]

Le 2013-11-17 02:11, Alan W. Irwin a écrit :
> On 2013-11-15 16:32-0800 Alan W. Irwin wrote:
> 
>> On 2013-11-15 14:34-0800 Alan W. Irwin wrote:
>> 
>>> If you trigger the segfault, but we cannot spot anything wrong in
>>> those files, I already have some ideas about continuing this segfault
>>> chase by narrowing down the test even more leading eventually to a
>>> simplest possible example of the problem that we can then analyze 
>>> with
>>> valgrind on your platform and also on ours.
>>> 
>>> Thanks very much for reporting this segfault issue in the first 
>>> place.
>>> They are the very bugger to track down (since the memory management
>>> issue that is typically the cause, may be symptomless on other
>>> platforms except for valgrind warnings).  I hope you will stick with
>>> us (probably for several more iterations) until we have it nailed.
>> 
>> Hi Alexis:
>> 
>> Never mind.  The whole point of those detailed instructions was to
>> figure out why you were getting the segfault, and I was not.  But
>> something changed from yesterday until today for me (I am not sure
>> what), and now I see that segfault as well.  So I can take it
>> from here and will let you know when I have that solved.
> 
> This turns out to be a tough issue which I have had to hand off to
> those with a lot more Tcl/Tk expertise than me (see discussion on the
> plplot-devel list).  The upshot is that the
> test_wish_standard_examples target (the one that segfaults at the very
> end of all the examples in the process of exiting) has been
> (temporarily until the segfault is fixed) taken off the list of
> targets that are run by the test_interactive target.  I have also
> fixed some Tcl-related build-system issues in the installed examples
> tree.  The result is now (revision 12710) I can run "make
> test_interactive" for either the core build system or the installed
> examples build system without encountering any errors, and I hope that
> is the case for you as well on your Fedora platform.
> 
> Alan
> __________________________
> Alan W. Irwin
> 
> Astronomical research affiliation with Department of Physics and 
> Astronomy,
> University of Victoria (astrowww.phys.uvic.ca).
> 
> Programming affiliations with the FreeEOS equation-of-state
> implementation for stellar interiors (freeeos.sf.net); the Time
> Ephemerides project (timeephem.sf.net); PLplot scientific plotting
> software package (plplot.sf.net); the libLASi project
> (unifont.org/lasi); the Loads of Linux Links project (loll.sf.net);
> and the Linux Brochure Project (lbproject.sf.net).
> __________________________
> 
> Linux-powered Science
> __________________________

Hi Alan,

I can confirm the test suite works on my machine, both in the build tree 
and
  in the installed directory, for the latest (rev. 12710) version.


Cheers,

-- 
Alexis Praga
GPG key : 66F9 4BD4 70EC E5D5 2AA4  9097 0673 7CE2 4671 306F

Re: [Plplot-general] Build documentation

[Alan W. I. <ir...@be...>]

On 2013-11-15 16:32-0800 Alan W. Irwin wrote:

> On 2013-11-15 14:34-0800 Alan W. Irwin wrote:
>
>> If you trigger the segfault, but we cannot spot anything wrong in
>> those files, I already have some ideas about continuing this segfault
>> chase by narrowing down the test even more leading eventually to a
>> simplest possible example of the problem that we can then analyze with
>> valgrind on your platform and also on ours.
>>
>> Thanks very much for reporting this segfault issue in the first place.
>> They are the very bugger to track down (since the memory management
>> issue that is typically the cause, may be symptomless on other
>> platforms except for valgrind warnings).  I hope you will stick with
>> us (probably for several more iterations) until we have it nailed.
>
> Hi Alexis:
>
> Never mind.  The whole point of those detailed instructions was to
> figure out why you were getting the segfault, and I was not.  But
> something changed from yesterday until today for me (I am not sure
> what), and now I see that segfault as well.  So I can take it
> from here and will let you know when I have that solved.

This turns out to be a tough issue which I have had to hand off to
those with a lot more Tcl/Tk expertise than me (see discussion on the
plplot-devel list).  The upshot is that the
test_wish_standard_examples target (the one that segfaults at the very
end of all the examples in the process of exiting) has been
(temporarily until the segfault is fixed) taken off the list of
targets that are run by the test_interactive target.  I have also
fixed some Tcl-related build-system issues in the installed examples
tree.  The result is now (revision 12710) I can run "make
test_interactive" for either the core build system or the installed
examples build system without encountering any errors, and I hope that
is the case for you as well on your Fedora platform.

Alan
__________________________
Alan W. Irwin

Astronomical research affiliation with Department of Physics and Astronomy,
University of Victoria (astrowww.phys.uvic.ca).

Programming affiliations with the FreeEOS equation-of-state
implementation for stellar interiors (freeeos.sf.net); the Time
Ephemerides project (timeephem.sf.net); PLplot scientific plotting
software package (plplot.sf.net); the libLASi project
(unifont.org/lasi); the Loads of Linux Links project (loll.sf.net);
and the Linux Brochure Project (lbproject.sf.net).
__________________________

Linux-powered Science
__________________________

Re: [Plplot-general] Build documentation

[Alan W. I. <ir...@be...>]

On 2013-11-15 14:34-0800 Alan W. Irwin wrote:

> If you trigger the segfault, but we cannot spot anything wrong in
> those files, I already have some ideas about continuing this segfault
> chase by narrowing down the test even more leading eventually to a
> simplest possible example of the problem that we can then analyze with
> valgrind on your platform and also on ours.
>
> Thanks very much for reporting this segfault issue in the first place.
> They are the very bugger to track down (since the memory management
> issue that is typically the cause, may be symptomless on other
> platforms except for valgrind warnings).  I hope you will stick with
> us (probably for several more iterations) until we have it nailed.

Hi Alexis:

Never mind.  The whole point of those detailed instructions was to
figure out why you were getting the segfault, and I was not.  But
something changed from yesterday until today for me (I am not sure
what), and now I see that segfault as well.  So I can take it
from here and will let you know when I have that solved.

Alan
__________________________
Alan W. Irwin

Astronomical research affiliation with Department of Physics and Astronomy,
University of Victoria (astrowww.phys.uvic.ca).

Programming affiliations with the FreeEOS equation-of-state
implementation for stellar interiors (freeeos.sf.net); the Time
Ephemerides project (timeephem.sf.net); PLplot scientific plotting
software package (plplot.sf.net); the libLASi project
(unifont.org/lasi); the Loads of Linux Links project (loll.sf.net);
and the Linux Brochure Project (lbproject.sf.net).
__________________________

Linux-powered Science
__________________________

Re: [Plplot-general] Build documentation

[Alan W. I. <ir...@be...>]

Hi Alexis:

On 2013-11-15 14:53+0100 Alexis Praga wrote:

> Unfortunately, there is a segmentation fault for the build-tree interactive 
> tests.
> Here are the steps I followed (from the build directory) :
>
> cmake -DCMAKE_INSTALL_PREFIX=$SOFTWARES/plplot -DBUILD_TEST=ON 
> ../../plplot-5.9.10
> make clean && make
> make VERBOSE=1 -j4 test_noninteractive >& test_noninteractive_build.out
> make VERBOSE=1 -j4 test_interactive >& test_interactive_build.out
>
> Corresponding files are attached. I hope this helps.

Yes it does.

Please run "svn update" from the top-level of your checked
out source tree to make sure you are consistent with the
latest code that the developers are testing.

I am also a bit concerned by that "make clean" which is fine as something
quick and dirty to do to get rebuilds, but still leaves you
potentially subject to stale cached build-system results from previous
bad builds.  The recommended way to make sure you are not subject to
such stale results is to completely remove the contents of your build
tree, i.e.,

rm -rf /home/pae/praga/Downloads/plplot/builddir/*

That, of course, is an potentially dangerous command so please
check that I have identified the top directory of your build tree
correctly.

Afterward,

cd /home/pae/praga/Downloads/plplot/builddir
and proceed as you did above, but this time (and everytime thereafter)
please capture the cmake output in cmake.out, e.g.,

cmake -DCMAKE_INSTALL_PREFIX=$SOFTWARES/plplot -DBUILD_TEST=ON \
../../plplot-5.9.10 >& cmake.out

Also, this time (and from now on as long as you are starting from an
empty build tree) you can drop the unnecessary make clean step.

If the segfault still persists (which I am pretty sure it will), there
is no need to send those results.  Instead, please
narrow it down further as follows.

Again start with an empty build tree and run cmake then
without doing any other make steps, do

make VERBOSE=1 test_wish_standard_examples >& \
test_wish_standard_examples.out

I am pretty sure from the test_interactive results you sent me before,
that this will trigger the segfault.  Note also, the lack of parallel
build option above (normally an excellent idea) that I deliberately
requested that makes the output easier to understand if there are
errors.

Assuming this narrowed down try triggers the segfault, I would like
you to send to the list (so Arjen and Orion can see the results as
well) cmake.out, test_wish_standard_examples.out, and
examples/tk/wish_standard_examples (a compressed tarball would be a
convenient format for those) to see if we can spot anything obvious
that the build system has done wrong in your case.

If you trigger the segfault, but we cannot spot anything wrong in
those files, I already have some ideas about continuing this segfault
chase by narrowing down the test even more leading eventually to a
simplest possible example of the problem that we can then analyze with
valgrind on your platform and also on ours.

Thanks very much for reporting this segfault issue in the first place.
They are the very bugger to track down (since the memory management
issue that is typically the cause, may be symptomless on other
platforms except for valgrind warnings).  I hope you will stick with
us (probably for several more iterations) until we have it nailed.

Alan
__________________________
Alan W. Irwin

Astronomical research affiliation with Department of Physics and Astronomy,
University of Victoria (astrowww.phys.uvic.ca).

Programming affiliations with the FreeEOS equation-of-state
implementation for stellar interiors (freeeos.sf.net); the Time
Ephemerides project (timeephem.sf.net); PLplot scientific plotting
software package (plplot.sf.net); the libLASi project
(unifont.org/lasi); the Loads of Linux Links project (loll.sf.net);
and the Linux Brochure Project (lbproject.sf.net).
__________________________

Linux-powered Science
__________________________

Re: [Plplot-general] Build documentation

[Alexis P. <ale...@ma...>]

Le 2013-11-15 14:14, Alan W. Irwin a écrit :
> On 2013-11-15 12:59+0100 Alexis Praga wrote:
> 
>> Alan,
>> 
>> At the moment, I'm only testing PLplot on a Fedora machine that is not 
>> my personnal computer.
>> Therefore, I prefer not to install too many dependencies, as long as 
>> it works.
>> 
>> As for the tests, I could not figure how to run both interactive and 
>> non-interactive in the build tree
>> prior to the install.
> 
> Hi Alexis:
> 
> Here is a quick overview without many details.
> 
> cmake <various options including the install prefix> \
> <path to top directory in source tree>
> 
> Those options should include DBUILD_TEST=ON which is the
> bit I forgot to mention before.  (Sorry about that.)  After that,
> 
> make test_noninteractive
> make test_interactive
> 
> should just work in the build tree.  Although they appear redundant
> with the tests you
> ran for the the installed examples tree, the conditions are quite
> different so you may get different results (see comments about that
> below).
> 
> For the others here, to run the installed examples tests that Alexis
> just ran, do the following:
> 
> make install
> mkdir <separate build directory for installed examples>
> cd <separate build directory for installed examples>
> cmake <no options> $prefix/share/plplot$version/examples
> make test_noninteractive
> make test_interactive
> 
>> [...] I also ran the tests in the separated build directory with the 
>> installed examples. However, after running the interactive
>> tests (and pressed "Enter" quite a few times), I was stuck with two 
>> plots which I had to close, thus producing
>> an error in the output files (attached to this mail).
>> 
>> Does that help you ?
> 
> Yes it does!
> 
> Clearly, everything is fine with your noninteractive test. (The
> standard example 22 difference is because we have not yet propagated a
> C improvement to that example to the other languages yet because we
> are still debating how else to improve that example).  In constrast to
> your noninteractive tests, your interactive test shows a clear
> build-system issue for the Tcl component for the installed examples.
> So thanks very much for providing those interesting results.
> 
> @Arjen: The explanation of that result is my horrible hack we have been
> discussing which distinguishes whether tkdemos.tcl has been sourced
> from a plserver environment or a wish environment depending on the
> location of the important pkgIndex.tcl file.  I now realize that hack
> cannot possibly work in the installed examples tree because for that
> case that file is joint between Pltcl, Pltk, and Plplotter so it can
> only appear in one installed location.
> 
> @Alexis: Fortunately, Arjen, who has much better Tcl skills than I do,
> is on the case, and should hopefully come up soon with a good
> alternative to my hack that will work in both build tree and installed
> examples tree.  Try the build-tree version of the tests for now.  They
> are currently working perfectly for me because of the hack which does
> work in that case.
> 
> Alan
> __________________________
> Alan W. Irwin
> 
> Astronomical research affiliation with Department of Physics and 
> Astronomy,
> University of Victoria (astrowww.phys.uvic.ca).
> 
> Programming affiliations with the FreeEOS equation-of-state
> implementation for stellar interiors (freeeos.sf.net); the Time
> Ephemerides project (timeephem.sf.net); PLplot scientific plotting
> software package (plplot.sf.net); the libLASi project
> (unifont.org/lasi); the Loads of Linux Links project (loll.sf.net);
> and the Linux Brochure Project (lbproject.sf.net).
> __________________________
> 
> Linux-powered Science
> __________________________

Unfortunately, there is a segmentation fault for the build-tree 
interactive tests.
Here are the steps I followed (from the build directory) :

cmake -DCMAKE_INSTALL_PREFIX=$SOFTWARES/plplot -DBUILD_TEST=ON 
../../plplot-5.9.10
make clean && make
make VERBOSE=1 -j4 test_noninteractive >& test_noninteractive_build.out
make VERBOSE=1 -j4 test_interactive >& test_interactive_build.out

Corresponding files are attached. I hope this helps.


-- 
Alexis Praga
GPG key : 66F9 4BD4 70EC E5D5 2AA4  9097 0673 7CE2 4671 306F

Re: [Plplot-general] Build documentation

[Alan W. I. <ir...@be...>]

On 2013-11-15 12:59+0100 Alexis Praga wrote:

> Alan,
>
> At the moment, I'm only testing PLplot on a Fedora machine that is not my 
> personnal computer.
> Therefore, I prefer not to install too many dependencies, as long as it 
> works.
>
> As for the tests, I could not figure how to run both interactive and 
> non-interactive in the build tree
> prior to the install.

Hi Alexis:

Here is a quick overview without many details.

cmake <various options including the install prefix> \
<path to top directory in source tree>

Those options should include DBUILD_TEST=ON which is the
bit I forgot to mention before.  (Sorry about that.)  After that,

make test_noninteractive
make test_interactive

should just work in the build tree.  Although they appear redundant with the tests you
ran for the the installed examples tree, the conditions are quite
different so you may get different results (see comments about that
below).

For the others here, to run the installed examples tests that Alexis just ran, do the following:

make install
mkdir <separate build directory for installed examples>
cd <separate build directory for installed examples>
cmake <no options> $prefix/share/plplot$version/examples
make test_noninteractive
make test_interactive

> [...] I also ran the tests in the separated build directory with the installed 
> examples. However, after running the interactive
> tests (and pressed "Enter" quite a few times), I was stuck with two plots 
> which I had to close, thus producing
> an error in the output files (attached to this mail).
>
> Does that help you ?

Yes it does!

Clearly, everything is fine with your noninteractive test. (The
standard example 22 difference is because we have not yet propagated a
C improvement to that example to the other languages yet because we
are still debating how else to improve that example).  In constrast to
your noninteractive tests, your interactive test shows a clear
build-system issue for the Tcl component for the installed examples.
So thanks very much for providing those interesting results.

@Arjen: The explanation of that result is my horrible hack we have been
discussing which distinguishes whether tkdemos.tcl has been sourced
from a plserver environment or a wish environment depending on the
location of the important pkgIndex.tcl file.  I now realize that hack
cannot possibly work in the installed examples tree because for that
case that file is joint between Pltcl, Pltk, and Plplotter so it can
only appear in one installed location.

@Alexis: Fortunately, Arjen, who has much better Tcl skills than I do,
is on the case, and should hopefully come up soon with a good
alternative to my hack that will work in both build tree and installed
examples tree.  Try the build-tree version of the tests for now.  They
are currently working perfectly for me because of the hack which does
work in that case.

Alan
__________________________
Alan W. Irwin

Astronomical research affiliation with Department of Physics and Astronomy,
University of Victoria (astrowww.phys.uvic.ca).

Programming affiliations with the FreeEOS equation-of-state
implementation for stellar interiors (freeeos.sf.net); the Time
Ephemerides project (timeephem.sf.net); PLplot scientific plotting
software package (plplot.sf.net); the libLASi project
(unifont.org/lasi); the Loads of Linux Links project (loll.sf.net);
and the Linux Brochure Project (lbproject.sf.net).
__________________________

Linux-powered Science
__________________________

Re: [Plplot-general] Build documentation

[Alexis P. <ale...@ma...>]

Le 2013-11-15 11:21, Alan W. Irwin a écrit :
> On 2013-11-15 10:13+0100 Alexis Praga wrote:
> 
>> Hi Alan,
>> 
>> I successfully built and installed the latest SVN (revision 12702). As 
>> a reference,
>> I have include the output of plplot-test.sh in share/examples, which 
>> looks OK
>> to me (only warnings).
>> Also, I consider it verbose enough for a normal install, even though 
>> the verbosity
>> would be more interesting had the build/install failed.
>> 
>> Finally, this version was validated by an install of PerlDL (and all 
>> of its testing
>> suite) and some plotting I did with it.
> 
> Hi Alexis:
> 
> I agree that the plplot-test.sh.out results are OK (and will be
> considerably cleaned up if you install development packages for
> shapelib and qhull. I assume official rpms of those are available on
> Fedora.) If you want to do even more extensive testing of PLplot, run
> 
> make VERBOSE=1 -j4 test_noninteractive >& test_noninteractive.out
> 
> and
> 
> make VERBOSE=1 -j4 test_interactive >& test_interactive.out
> 
> in the build tree and also after you run "make install", run those
> same targets for the separate CMake-based build system that we install
> along with the installed examples.  You configure that separate build
> system by running cmake pointing to the installed examples top-level
> source-tree directory at $prefix/share/plplot$version/examples. Be
> sure to run that from a separate build tree so it does not interfere
> with another separate build system for the installed examples that is
> based on Makefiles and pkg-config.  But that traditional build system
> is not as powerful as the CMake-based one.
> 
> The above non-interactive tests will generate something roughly like
> 3GB of plot files, and the above interactive tests will produce many
> plots on your screen (mostly in non-pause mode so you don't have to
> hit the enter key that much to keep them moving along).  Of course, it
> is essential to capture the VERBOSE=1 results in the above *.out files
> in case there are any errors.
> 
> Your good results for the current svn trunk version are not only good
> news from your perspective but also from our perspective since we plan
> to do a release of PLplot fairly soon (near the end of the year) based
> on the svn trunk version.  No more two-year waits between PLplot
> releases! I can say that because that unprecedented delay between
> 5.9.9 and 5.9.10 was caused by our extensive search for a release
> manager volunteer who finally turned out (much to my surprise) to be
> me.  :-) I am really looking forward to releasing 5.9.11 soon since
> svn trunk is looking pretty good.  Also, that release should be much
> easier for me than the release of 5.9.10 due to my gained experience
> and the much shorter release cycle.
> 
> Alan
> __________________________
> Alan W. Irwin
> 
> Astronomical research affiliation with Department of Physics and 
> Astronomy,
> University of Victoria (astrowww.phys.uvic.ca).
> 
> Programming affiliations with the FreeEOS equation-of-state
> implementation for stellar interiors (freeeos.sf.net); the Time
> Ephemerides project (timeephem.sf.net); PLplot scientific plotting
> software package (plplot.sf.net); the libLASi project
> (unifont.org/lasi); the Loads of Linux Links project (loll.sf.net);
> and the Linux Brochure Project (lbproject.sf.net).
> __________________________
> 
> Linux-powered Science
> __________________________

Alan,

At the moment, I'm only testing PLplot on a Fedora machine that is not 
my personnal computer.
Therefore, I prefer not to install too many dependencies, as long as it 
works.

As for the tests, I could not figure how to run both interactive and 
non-interactive in the build tree
prior to the install. The Makefile clearly mentionned the targets were 
to be run in the install folder
and not the build tree. Which is why I ran the examples in the install 
directory
(in $prefix/share/plplot$version/examples).
Was that what you wanted ? Otherwise, I'm sorry but could explain more 
the different steps ?

I also ran the tests in the separated build directory with the installed 
examples. However, after running the interactive
tests (and pressed "Enter" quite a few times), I was stuck with two 
plots which I had to close, thus producing
an error in the output files (attached to this mail).

Does that help you ?

-- 
Alexis Praga
GPG key : 66F9 4BD4 70EC E5D5 2AA4  9097 0673 7CE2 4671 306F

Re: [Plplot-general] Build documentation

[Alan W. I. <ir...@be...>]

On 2013-11-15 10:13+0100 Alexis Praga wrote:

> Hi Alan,
>
> I successfully built and installed the latest SVN (revision 12702). As a 
> reference,
> I have include the output of plplot-test.sh in share/examples, which looks OK
> to me (only warnings).
> Also, I consider it verbose enough for a normal install, even though the 
> verbosity
> would be more interesting had the build/install failed.
>
> Finally, this version was validated by an install of PerlDL (and all of its 
> testing
> suite) and some plotting I did with it.

Hi Alexis:

I agree that the plplot-test.sh.out results are OK (and will be
considerably cleaned up if you install development packages for
shapelib and qhull. I assume official rpms of those are available on
Fedora.) If you want to do even more extensive testing of PLplot, run

make VERBOSE=1 -j4 test_noninteractive >& test_noninteractive.out

and

make VERBOSE=1 -j4 test_interactive >& test_interactive.out

in the build tree and also after you run "make install", run those
same targets for the separate CMake-based build system that we install
along with the installed examples.  You configure that separate build
system by running cmake pointing to the installed examples top-level
source-tree directory at $prefix/share/plplot$version/examples. Be
sure to run that from a separate build tree so it does not interfere
with another separate build system for the installed examples that is
based on Makefiles and pkg-config.  But that traditional build system
is not as powerful as the CMake-based one.

The above non-interactive tests will generate something roughly like
3GB of plot files, and the above interactive tests will produce many
plots on your screen (mostly in non-pause mode so you don't have to
hit the enter key that much to keep them moving along).  Of course, it
is essential to capture the VERBOSE=1 results in the above *.out files
in case there are any errors.

Your good results for the current svn trunk version are not only good
news from your perspective but also from our perspective since we plan
to do a release of PLplot fairly soon (near the end of the year) based
on the svn trunk version.  No more two-year waits between PLplot
releases! I can say that because that unprecedented delay between
5.9.9 and 5.9.10 was caused by our extensive search for a release
manager volunteer who finally turned out (much to my surprise) to be
me.  :-) I am really looking forward to releasing 5.9.11 soon since
svn trunk is looking pretty good.  Also, that release should be much
easier for me than the release of 5.9.10 due to my gained experience
and the much shorter release cycle.

Alan
__________________________
Alan W. Irwin

Astronomical research affiliation with Department of Physics and Astronomy,
University of Victoria (astrowww.phys.uvic.ca).

Programming affiliations with the FreeEOS equation-of-state
implementation for stellar interiors (freeeos.sf.net); the Time
Ephemerides project (timeephem.sf.net); PLplot scientific plotting
software package (plplot.sf.net); the libLASi project
(unifont.org/lasi); the Loads of Linux Links project (loll.sf.net);
and the Linux Brochure Project (lbproject.sf.net).
__________________________

Linux-powered Science
__________________________

Re: [Plplot-general] Build documentation

[Alexis P. <ale...@ma...>]

Le 2013-11-14 20:19, Alan W. Irwin a écrit :
> On 2013-11-13 15:42+0100 Alexis Praga wrote:
> 
>>> Could you just check the Tcl version number in that file tcl.h?
>> 
>> It's the 8.5 version:
>> 
>> #define TCL_VERSION	    "8.5"
>> #define TCL_PATCH_LEVEL	    "8.5.9"
>> 
>> If you know how to activate a more verbose for CMake, I would be happy
>> to do it.
> 
> Hi Alexis:
> 
> I am joining this conversation late because I couldn't participate
> earlier (incoming mail-chain troubles that are now fixed).  I am also
> CCing Orion since he is very experienced at dealing with PLplot issues
> on Fedora. (He maintains the Fedora rpms for PLplot.)
> 
> Coincidentally, I have just finished (see my latest post to the
> plplot-devel list) a big effort on improving the parts of the build
> system having to do with Tcl, Tk, Itcl, Itk, and Iwidgets.  The result
> is more demanding of consistency for all components and also more
> verbose about what it finds.  So I would appreciate you giving the
> plplot svn trunk version a try to see if (a) it works on your
> Fedora system, and (b) it is verbose enough for you.
> 
> That version also has Arjen's fix concerning TclFormatInt.  I am
> virtually positive the fundamental issue there is the PLplot build
> historically relied on "private" Tcl headers. In 2000 or so when I was
> still building PLplot on a RedHat platform, I ran into a lot of such
> issues which Maurice (our Tcl/Tk guru who has mostly retired from
> working on PLplot now) fixed at the time.  But after I switched to
> Debian (whose Tcl devel package includes all private headers) a few of
> these issues may have crept back (because Tcl developers likely made
> different decisions about what was private and what not). So Arjen's
> fix is a good step in the right direction, and if you now try the svn
> trunk version of PLplot on Fedora you may find all is well now with
> the Tcl component of the build.  Or you might run into a few more
> private header issues which we would be happy to fix if you were
> willing to do the necessary build experiments.
> 
> I also have access to a Tcl8.6 build I have done myself which I am
> currently using to test PLplot's Tcl8.6 capabilities.  I notice that
> build has a switch that can be used to turn off the install of private
> headers.  So now you have reminded me of the private header issue with
> Tcl, I plan to use that switch to see if there is anything more to
> deal with.
> 
> Alan
> __________________________
> Alan W. Irwin
> 
> Astronomical research affiliation with Department of Physics and 
> Astronomy,
> University of Victoria (astrowww.phys.uvic.ca).
> 
> Programming affiliations with the FreeEOS equation-of-state
> implementation for stellar interiors (freeeos.sf.net); the Time
> Ephemerides project (timeephem.sf.net); PLplot scientific plotting
> software package (plplot.sf.net); the libLASi project
> (unifont.org/lasi); the Loads of Linux Links project (loll.sf.net);
> and the Linux Brochure Project (lbproject.sf.net).
> __________________________
> 
> Linux-powered Science
> __________________________

Hi Alan,

I successfully built and installed the latest SVN (revision 12702). As a 
reference,
I have include the output of plplot-test.sh in share/examples, which 
looks OK
to me (only warnings).
Also, I consider it verbose enough for a normal install, even though the 
verbosity
would be more interesting had the build/install failed.

Finally, this version was validated by an install of PerlDL (and all of 
its testing
suite) and some plotting I did with it.

Cheers,

-- 
Alexis Praga
GPG key : 66F9 4BD4 70EC E5D5 2AA4  9097 0673 7CE2 4671 306F

Re: [Plplot-general] Build documentation

[Alan W. I. <ir...@be...>]

On 2013-11-13 15:42+0100 Alexis Praga wrote:

>> Could you just check the Tcl version number in that file tcl.h?
>
> It's the 8.5 version:
>
> #define TCL_VERSION	    "8.5"
> #define TCL_PATCH_LEVEL	    "8.5.9"
>
> If you know how to activate a more verbose for CMake, I would be happy
> to do it.

Hi Alexis:

I am joining this conversation late because I couldn't participate
earlier (incoming mail-chain troubles that are now fixed).  I am also
CCing Orion since he is very experienced at dealing with PLplot issues
on Fedora. (He maintains the Fedora rpms for PLplot.)

Coincidentally, I have just finished (see my latest post to the
plplot-devel list) a big effort on improving the parts of the build
system having to do with Tcl, Tk, Itcl, Itk, and Iwidgets.  The result
is more demanding of consistency for all components and also more
verbose about what it finds.  So I would appreciate you giving the
plplot svn trunk version a try to see if (a) it works on your
Fedora system, and (b) it is verbose enough for you.

That version also has Arjen's fix concerning TclFormatInt.  I am
virtually positive the fundamental issue there is the PLplot build
historically relied on "private" Tcl headers. In 2000 or so when I was
still building PLplot on a RedHat platform, I ran into a lot of such
issues which Maurice (our Tcl/Tk guru who has mostly retired from
working on PLplot now) fixed at the time.  But after I switched to
Debian (whose Tcl devel package includes all private headers) a few of
these issues may have crept back (because Tcl developers likely made
different decisions about what was private and what not). So Arjen's
fix is a good step in the right direction, and if you now try the svn
trunk version of PLplot on Fedora you may find all is well now with
the Tcl component of the build.  Or you might run into a few more
private header issues which we would be happy to fix if you were
willing to do the necessary build experiments.

I also have access to a Tcl8.6 build I have done myself which I am
currently using to test PLplot's Tcl8.6 capabilities.  I notice that
build has a switch that can be used to turn off the install of private
headers.  So now you have reminded me of the private header issue with
Tcl, I plan to use that switch to see if there is anything more to
deal with.

Alan
__________________________
Alan W. Irwin

Astronomical research affiliation with Department of Physics and Astronomy,
University of Victoria (astrowww.phys.uvic.ca).

Programming affiliations with the FreeEOS equation-of-state
implementation for stellar interiors (freeeos.sf.net); the Time
Ephemerides project (timeephem.sf.net); PLplot scientific plotting
software package (plplot.sf.net); the libLASi project
(unifont.org/lasi); the Loads of Linux Links project (loll.sf.net);
and the Linux Brochure Project (lbproject.sf.net).
__________________________

Linux-powered Science
__________________________

Re: [Plplot-general] Build documentation

[Arjen M. <Arj...@de...>]

Hi Alexis,

hm, then I am truly puzzled. Anyway, CMake with --debug-output should produce more information on what it is doing.

But I will simply change the code so that this will no longer be a problem.

Regards,

Arjen

> -----Original Message-----
> From: Alexis Praga [mailto:ale...@ma...]
> Sent: Wednesday, November 13, 2013 3:42 PM
> To: Arjen Markus
> Cc: plp...@li...
> Subject: RE: [Plplot-general] Build documentation
>
> > Could you just check the Tcl version number in that file tcl.h?
>
> It's the 8.5 version:
>
> #define TCL_VERSION       "8.5"
> #define TCL_PATCH_LEVEL           "8.5.9"
>
> If you know how to activate a more verbose for CMake, I would be happy to do it.
>
> Cheers,
>
> ---
> Alexis Praga
> GPG key : 66F9 4BD4 70EC E5D5 2AA4  9097 0673 7CE2 4671 306F
>
> Le 2013-11-13 14:59, Arjen Markus a écrit :
> > Hi Alexis,
> >
> >
> >
> >> -----Original Message-----
> >> From: Alexis Praga [mailto:ale...@ma...]
> >> Sent: Wednesday, November 13, 2013 2:52 PM
> >> To: Arjen Markus
> >> Cc: plp...@li...
> >> Subject: RE: [Plplot-general] Build documentation
> >>
> >> > Hm, could it be that the version of tclsh is 8.5 but the include
> >> > files are 8.4? That would trigger this kind of error.
> >>
> >> It's possible, there are some remnants of a 8.4 installation. The
> >> current include file is at version 1.254.2.16 2010/08/04.
> >>
> >
> > The tcl.h include file contains the Tcl version number somewhere near
> > the beginning, like:
> >
> > #define TCL_VERSION         "8.6"
> > #define TCL_PATCH_LEVEL     "8.6.0"
> >
> > I am mostly curious as to why CMake reports a version 8.5 and the
> > build still fails (I use both
> > 8.5 and 8.6 and have not seen this problem myself).
> >
> > The remedy is of course to get rid of TclFormatInt altogether :).
> >
> >> > Well, have a look at what I committed in response to your question.
> >> > Hopefully that makes things clearer.
> >>
> >> Totally, I was just pointing out it needed a careful reading.
> >>
> >> Anyway, the output of cmake and make are attached.
> >>
> >
> > Thanks for that. Unfortunately the files do not contain anything you
> > did not already
> > report.
> >
> > Could you just check the Tcl version number in that file tcl.h?
> >
> > Regards,
> >
> > Arjen
> > DISCLAIMER: This message is intended exclusively for the addressee(s)
> > and may contain confidential and privileged information. If you are
> > not the intended recipient please notify the sender immediately and
> > destroy this message. Unauthorized use, disclosure or copying of this
> > message is strictly prohibited. The foundation 'Stichting Deltares',
> > which has its seat at Delft, The Netherlands, Commercial Registration
> > Number 41146461, is not liable in any way whatsoever for consequences
> > and/or damages resulting from the improper, incomplete and untimely
> > dispatch, receipt and/or content of this e-mail.

DISCLAIMER: This message is intended exclusively for the addressee(s) and may contain confidential and privileged information. If you are not the intended recipient please notify the sender immediately and destroy this message. Unauthorized use, disclosure or copying of this message is strictly prohibited. The foundation 'Stichting Deltares', which has its seat at Delft, The Netherlands, Commercial Registration Number 41146461, is not liable in any way whatsoever for consequences and/or damages resulting from the improper, incomplete and untimely dispatch, receipt and/or content of this e-mail.

Re: [Plplot-general] Build documentation

[Alexis P. <ale...@ma...>]

> Could you just check the Tcl version number in that file tcl.h?

It's the 8.5 version:

#define TCL_VERSION	    "8.5"
#define TCL_PATCH_LEVEL	    "8.5.9"

If you know how to activate a more verbose for CMake, I would be happy 
to do it.

Cheers,

---
Alexis Praga
GPG key : 66F9 4BD4 70EC E5D5 2AA4  9097 0673 7CE2 4671 306F

Le 2013-11-13 14:59, Arjen Markus a écrit :
> Hi Alexis,
> 
> 
> 
>> -----Original Message-----
>> From: Alexis Praga [mailto:ale...@ma...]
>> Sent: Wednesday, November 13, 2013 2:52 PM
>> To: Arjen Markus
>> Cc: plp...@li...
>> Subject: RE: [Plplot-general] Build documentation
>> 
>> > Hm, could it be that the version of tclsh is 8.5 but the include files
>> > are 8.4? That would trigger this kind of error.
>> 
>> It's possible, there are some remnants of a 8.4 installation. The 
>> current include file is
>> at version 1.254.2.16 2010/08/04.
>> 
> 
> The tcl.h include file contains the Tcl version number somewhere near
> the beginning, like:
> 
> #define TCL_VERSION         "8.6"
> #define TCL_PATCH_LEVEL     "8.6.0"
> 
> I am mostly curious as to why CMake reports a version 8.5 and the
> build still fails (I use both
> 8.5 and 8.6 and have not seen this problem myself).
> 
> The remedy is of course to get rid of TclFormatInt altogether :).
> 
>> > Well, have a look at what I committed in response to your question.
>> > Hopefully that makes things clearer.
>> 
>> Totally, I was just pointing out it needed a careful reading.
>> 
>> Anyway, the output of cmake and make are attached.
>> 
> 
> Thanks for that. Unfortunately the files do not contain anything you
> did not already
> report.
> 
> Could you just check the Tcl version number in that file tcl.h?
> 
> Regards,
> 
> Arjen
> DISCLAIMER: This message is intended exclusively for the addressee(s)
> and may contain confidential and privileged information. If you are
> not the intended recipient please notify the sender immediately and
> destroy this message. Unauthorized use, disclosure or copying of this
> message is strictly prohibited. The foundation 'Stichting Deltares',
> which has its seat at Delft, The Netherlands, Commercial Registration
> Number 41146461, is not liable in any way whatsoever for consequences
> and/or damages resulting from the improper, incomplete and untimely
> dispatch, receipt and/or content of this e-mail.

Re: [Plplot-general] Build documentation

[James T. <jt...@gm...>]

Sorry; that's the itcl version, rather than tcl.

James


On 13 November 2013 12:58, Arjen Markus <Arj...@de...> wrote:

>  Hi James,
>
>
>
> I would be surprised if someone is actively using Tcl 4.0 – that was
> released at least 15 years ago. Maybe you mean Tcl 8.4? I saw something in
> the source code referring to Tcl 8.4 or earlier.
>
>
>
> Anyway, if that is the case, we can fairly easily correct that.
>
>
>
> Regards,
>
>
>
> Arjen
>
>
>
> *From:* James Tappin [mailto:jt...@gm...]
> *Sent:* Wednesday, November 13, 2013 1:56 PM
> *To:* Arjen Markus
> *Subject:* Re: [Plplot-general] Build documentation
>
>
>
> I'm not sure if this is relevant to your problem, but the current versions
> of the cmake scripts don't work with Tcl 4.0. The attached patch is what I
> used to produce the package for Arch (& derivatives).
>
>  James
>
>
>
> On 13 November 2013 12:16, Arjen Markus <Arj...@de...> wrote:
>
> Hi Alexis,
>
> I added a short explanation of the example to the INSTALL file, but I saw
> that a more elaborate explanation appears lower down. You seem to have
> missed that, but I hope that with this change it is clearer.
>
> Regards,
>
> Arjen
>
> DISCLAIMER: This message is intended exclusively for the addressee(s) and
> may contain confidential and privileged information. If you are not the
> intended recipient please notify the sender immediately and destroy this
> message. Unauthorized use, disclosure or copying of this message is
> strictly prohibited. The foundation 'Stichting Deltares', which has its
> seat at Delft, The Netherlands, Commercial Registration Number 41146461, is
> not liable in any way whatsoever for consequences and/or damages resulting
> from the improper, incomplete and untimely dispatch, receipt and/or content
> of this e-mail.
>
>
> ------------------------------------------------------------------------------
> DreamFactory - Open Source REST & JSON Services for HTML5 & Native Apps
> OAuth, Users, Roles, SQL, NoSQL, BLOB Storage and External API Access
> Free app hosting. Or install the open source package on any LAMP server.
> Sign up and see examples for AngularJS, jQuery, Sencha Touch and Native!
> http://pubads.g.doubleclick.net/gampad/clk?id=63469471&iu=/4140/ostg.clktrk
> _______________________________________________
> Plplot-general mailing list
> Plp...@li...
> https://lists.sourceforge.net/lists/listinfo/plplot-general
>
>
>   DISCLAIMER: This message is intended exclusively for the addressee(s)
> and may contain confidential and privileged information. If you are not the
> intended recipient please notify the sender immediately and destroy this
> message. Unauthorized use, disclosure or copying of this message is
> strictly prohibited. The foundation 'Stichting Deltares', which has its
> seat at Delft, The Netherlands, Commercial Registration Number 41146461, is
> not liable in any way whatsoever for consequences and/or damages resulting
> from the improper, incomplete and untimely dispatch, receipt and/or content
> of this e-mail.
>

Re: [Plplot-general] Build documentation

[Arjen M. <Arj...@de...>]

Hi Alexis,



> -----Original Message-----
> From: Alexis Praga [mailto:ale...@ma...]
> Sent: Wednesday, November 13, 2013 2:52 PM
> To: Arjen Markus
> Cc: plp...@li...
> Subject: RE: [Plplot-general] Build documentation
>
> > Hm, could it be that the version of tclsh is 8.5 but the include files
> > are 8.4? That would trigger this kind of error.
>
> It's possible, there are some remnants of a 8.4 installation. The current include file is
> at version 1.254.2.16 2010/08/04.
>

The tcl.h include file contains the Tcl version number somewhere near the beginning, like:

#define TCL_VERSION         "8.6"
#define TCL_PATCH_LEVEL     "8.6.0"

I am mostly curious as to why CMake reports a version 8.5 and the build still fails (I use both
8.5 and 8.6 and have not seen this problem myself).

The remedy is of course to get rid of TclFormatInt altogether :).

> > Well, have a look at what I committed in response to your question.
> > Hopefully that makes things clearer.
>
> Totally, I was just pointing out it needed a careful reading.
>
> Anyway, the output of cmake and make are attached.
>

Thanks for that. Unfortunately the files do not contain anything you did not already
report.

Could you just check the Tcl version number in that file tcl.h?

Regards,

Arjen
DISCLAIMER: This message is intended exclusively for the addressee(s) and may contain confidential and privileged information. If you are not the intended recipient please notify the sender immediately and destroy this message. Unauthorized use, disclosure or copying of this message is strictly prohibited. The foundation 'Stichting Deltares', which has its seat at Delft, The Netherlands, Commercial Registration Number 41146461, is not liable in any way whatsoever for consequences and/or damages resulting from the improper, incomplete and untimely dispatch, receipt and/or content of this e-mail.

Re: [Plplot-general] Build documentation

[Alexis P. <ale...@ma...>]

> Hm, could it be that the version of tclsh is 8.5 but the include files 
> are 8.4? That would trigger
> this kind of error.

It's possible, there are some remnants of a 8.4 installation. The 
current include file is at version 1.254.2.16 2010/08/04.

> Well, have a look at what I committed in response to your question.
> Hopefully that makes things clearer.

Totally, I was just pointing out it needed a careful reading.

Anyway, the output of cmake and make are attached.


---
Alexis Praga
GPG key : 66F9 4BD4 70EC E5D5 2AA4  9097 0673 7CE2 4671 306F

Le 2013-11-13 14:32, Arjen Markus a écrit :
> Hi Alexis,
> 
>> -----Original Message-----
>> From: Alexis Praga [mailto:ale...@ma...]
>> Sent: Wednesday, November 13, 2013 2:18 PM
>> To: Arjen Markus
>> Cc: plp...@li...
>> Subject: RE: [Plplot-general] Build documentation
>> 
>> Thanks for the update.
>> 
> 
> You're welcome.
> 
>> I am using the latest PLplot (5.9.10) on Fedora 14, 64 bit. Compilers 
>> are : gcc 4.5.1,
>> g++ and gfortran.
>> 
> 
> Okay, that should definitely do fine :).
> 
>> Concerning TCL, the version 8.5 is installed but not on a standard 
>> location
>> (/usr/lib64). Is-it really a requirement for plplot ?
> 
> NO, PLplot can use a number of optional external libraries or be built
> for a variety
> of programming languages. Tcl is merely one of these.
> 
> You can turn off individual components explicitly: -DENABLE_tcl=OFF
> 
> However, if you leave it to CMake it will include any optional
> component it finds.
> In this case something odd is happening. Could you send me the output 
> from
> both CMake and make?
> 
> cmake ... 1>cmake.out 2>&1
> make 1>make.out 2>&1
> 
> (So that both standard output and standard error are captured in the 
> file)
> 
>> Otherwise, I don't see it mentionned in the INSTALL file.
>> Anyway, it is found by cmake (see below for an excerpt of the 
>> configuration), but
>> build fails with :
>> 
>> ../../src/libplplotd.so.12.0.0: undefined reference to `TclFormatInt'
>> collect2: ld returned 1 exit status
>> make[2]: *** [bindings/tk/plserver] Error 1
>> make[1]: *** [bindings/tk/CMakeFiles/plserver.dir/all] Error 2
>> make: *** [all] Error 2
>> 
> 
> The function TclFormatInt is actually a "private" function, not meant
> to be used outside of
> the Tcl library itself. If you are using Tcl 8.5, then it is actually
> locally defined as a macro in
> the file tclMain.c.
> 
> CMake reports 8.5, but clearly the compile error you get indicates
> something else.
> 
>> Concerning the source directory, it's my fault not to have read the 
>> INSTALL file
>> thoroughly. However, I like to be able to install softwares with just 
>> a quick glance at
>> the example.
>> Indeed, an explanation concerning plplot_cmake would help a lot.
>> 
> 
> Well, have a look at what I committed in response to your question.
> Hopefully that makes
> things clearer.
> 
> Hm, could it be that the version of tclsh is 8.5 but the include files
> are 8.4? That would trigger
> this kind of error.
> 
> The simplest solution is to remove TclFormatInt from the PLplot source
> code. That is not
> hard to do.
> 
>> 
>> 
>> Cmake configuration (excerpt):
>> -- Found Tclsh: /usr/bin/tclsh (found version "8.5")
>> -- Found TCL: /usr/lib64/libtcl.so
>> -- Found TCLTK: /usr/lib64/libtcl.so
>> -- Found TK: /usr/lib64/libtk.so
>> -- Looking for include paths and libraries for Tcl/Tk - found
>> -- Looking for tclsh
>> -- Looking for tclsh - found
>> -- TCL_TCLSH = /usr/bin/tclsh
>> -- TCL_INCLUDE_PATH = /usr/include
>> -- TCL_LIBRARY = /usr/lib64/libtcl.so
>> -- Itcl not available or not compatible with current Tcl shell
>> -- TK_INCLUDE_PATH = /usr/include;/usr/include
>> -- TK_LIBRARY = /usr/lib64/libtk.so
>> -- Looking for itk.h
>> -- Looking for itk.h - not found
>> -- WARNING: Disabling incr Tk interface code
>> 
> 
> Regards,
> 
> Arjen
> DISCLAIMER: This message is intended exclusively for the addressee(s)
> and may contain confidential and privileged information. If you are
> not the intended recipient please notify the sender immediately and
> destroy this message. Unauthorized use, disclosure or copying of this
> message is strictly prohibited. The foundation 'Stichting Deltares',
> which has its seat at Delft, The Netherlands, Commercial Registration
> Number 41146461, is not liable in any way whatsoever for consequences
> and/or damages resulting from the improper, incomplete and untimely
> dispatch, receipt and/or content of this e-mail.

Re: [Plplot-general] Build documentation

[Arjen M. <Arj...@de...>]

Hi Alexis,

> -----Original Message-----
> From: Alexis Praga [mailto:ale...@ma...]
> Sent: Wednesday, November 13, 2013 2:18 PM
> To: Arjen Markus
> Cc: plp...@li...
> Subject: RE: [Plplot-general] Build documentation
>
> Thanks for the update.
>

You're welcome.

> I am using the latest PLplot (5.9.10) on Fedora 14, 64 bit. Compilers are : gcc 4.5.1,
> g++ and gfortran.
>

Okay, that should definitely do fine :).

> Concerning TCL, the version 8.5 is installed but not on a standard location
> (/usr/lib64). Is-it really a requirement for plplot ?

NO, PLplot can use a number of optional external libraries or be built for a variety
of programming languages. Tcl is merely one of these.

You can turn off individual components explicitly: -DENABLE_tcl=OFF

However, if you leave it to CMake it will include any optional component it finds.
In this case something odd is happening. Could you send me the output from
both CMake and make?

cmake ... 1>cmake.out 2>&1
make 1>make.out 2>&1

(So that both standard output and standard error are captured in the file)

> Otherwise, I don't see it mentionned in the INSTALL file.
> Anyway, it is found by cmake (see below for an excerpt of the configuration), but
> build fails with :
>
> ../../src/libplplotd.so.12.0.0: undefined reference to `TclFormatInt'
> collect2: ld returned 1 exit status
> make[2]: *** [bindings/tk/plserver] Error 1
> make[1]: *** [bindings/tk/CMakeFiles/plserver.dir/all] Error 2
> make: *** [all] Error 2
>

The function TclFormatInt is actually a "private" function, not meant to be used outside of
the Tcl library itself. If you are using Tcl 8.5, then it is actually locally defined as a macro in
the file tclMain.c.

CMake reports 8.5, but clearly the compile error you get indicates something else.

> Concerning the source directory, it's my fault not to have read the INSTALL file
> thoroughly. However, I like to be able to install softwares with just a quick glance at
> the example.
> Indeed, an explanation concerning plplot_cmake would help a lot.
>

Well, have a look at what I committed in response to your question. Hopefully that makes
things clearer.

Hm, could it be that the version of tclsh is 8.5 but the include files are 8.4? That would trigger
this kind of error.

The simplest solution is to remove TclFormatInt from the PLplot source code. That is not
hard to do.

>
>
> Cmake configuration (excerpt):
> -- Found Tclsh: /usr/bin/tclsh (found version "8.5")
> -- Found TCL: /usr/lib64/libtcl.so
> -- Found TCLTK: /usr/lib64/libtcl.so
> -- Found TK: /usr/lib64/libtk.so
> -- Looking for include paths and libraries for Tcl/Tk - found
> -- Looking for tclsh
> -- Looking for tclsh - found
> -- TCL_TCLSH = /usr/bin/tclsh
> -- TCL_INCLUDE_PATH = /usr/include
> -- TCL_LIBRARY = /usr/lib64/libtcl.so
> -- Itcl not available or not compatible with current Tcl shell
> -- TK_INCLUDE_PATH = /usr/include;/usr/include
> -- TK_LIBRARY = /usr/lib64/libtk.so
> -- Looking for itk.h
> -- Looking for itk.h - not found
> -- WARNING: Disabling incr Tk interface code
>

Regards,

Arjen
DISCLAIMER: This message is intended exclusively for the addressee(s) and may contain confidential and privileged information. If you are not the intended recipient please notify the sender immediately and destroy this message. Unauthorized use, disclosure or copying of this message is strictly prohibited. The foundation 'Stichting Deltares', which has its seat at Delft, The Netherlands, Commercial Registration Number 41146461, is not liable in any way whatsoever for consequences and/or damages resulting from the improper, incomplete and untimely dispatch, receipt and/or content of this e-mail.

Re: [Plplot-general] Build documentation

[Alexis P. <ale...@ma...>]

Thanks for the update.

I am using the latest PLplot (5.9.10) on Fedora 14, 64 bit. Compilers 
are : gcc 4.5.1, g++ and gfortran.

Concerning TCL, the version 8.5 is installed but not on a standard 
location (/usr/lib64). Is-it really a requirement for plplot ?
Otherwise, I don't see it mentionned in the INSTALL file.
Anyway, it is found by cmake (see below for an excerpt of the 
configuration), but build fails with :

../../src/libplplotd.so.12.0.0: undefined reference to `TclFormatInt'
collect2: ld returned 1 exit status
make[2]: *** [bindings/tk/plserver] Error 1
make[1]: *** [bindings/tk/CMakeFiles/plserver.dir/all] Error 2
make: *** [all] Error 2

Concerning the source directory, it's my fault not to have read the 
INSTALL file thoroughly. However, I like to be able to install softwares 
with just a quick glance at the example.
Indeed, an explanation concerning plplot_cmake would help a lot.



Cmake configuration (excerpt):
-- Found Tclsh: /usr/bin/tclsh (found version "8.5")
-- Found TCL: /usr/lib64/libtcl.so
-- Found TCLTK: /usr/lib64/libtcl.so
-- Found TK: /usr/lib64/libtk.so
-- Looking for include paths and libraries for Tcl/Tk - found
-- Looking for tclsh
-- Looking for tclsh - found
-- TCL_TCLSH = /usr/bin/tclsh
-- TCL_INCLUDE_PATH = /usr/include
-- TCL_LIBRARY = /usr/lib64/libtcl.so
-- Itcl not available or not compatible with current Tcl shell
-- TK_INCLUDE_PATH = /usr/include;/usr/include
-- TK_LIBRARY = /usr/lib64/libtk.so
-- Looking for itk.h
-- Looking for itk.h - not found
-- WARNING: Disabling incr Tk interface code


Cheers,


---
Alexis Praga

Le 2013-11-13 13:06, Arjen Markus a écrit :
> Hello Alexis,
> 
> wrt PLplot in general: are you using the latest version? What platform
> (OS, compiler) are you using?
> 
> wrt your first question: what version of Tcl do you have installed? I
> am not familiar with this error and I am a trifle surprised at this.
> Also the message you refer is from 4 years ago, a lot has changed
> since then.
> 
> wrt the base/source directory: the confusion may be that the build
> documentation refers to the source directory as the directory
> containing ALL the source files. There is a subdirectory there called
> "src" which contains the source files of the PLplot core only. I can
> understand some of the confusion, but it is a rather common
> convention.
> 
> wrt "plplot_cmake": we should indeed make this example a bit clearer -
> "plplot_cmake" is merely a place holder and that should have been
> explained.
> 
> Regards,
> 
> Arjen
> 
>> -----Original Message-----
>> From: Alexis Praga [mailto:ale...@ma...]
>> Sent: Wednesday, November 13, 2013 12:34 PM
>> To: plp...@li...
>> Subject: [Plplot-general] Build documentation
>> 
>> Hi,
>> 
>> I have just installed PLplot. For that, I had to disable TCL as 
>> mentioned here :
>> http://sourceforge.net/p/plplot/discussion/8607/thread/d8bed9d5
>> and use the base directory instead of the source directory, as 
>> mentioned here :
>> http://sourceforge.net/p/plplot/support-requests/31/
>> 
>> Why isn't this in the documentation, especially the second "hack" ?
>> Furthermore, the INSTALL file mentions we need to use a certain 
>> "plplot_cmake" as
>> a source directory, which does not exist.
>> 
>> Thanks
>> 
>> --
>> Alexis Praga
>> 
>> ------------------------------------------------------------------------------
>> DreamFactory - Open Source REST & JSON Services for HTML5 & Native 
>> Apps
>> OAuth, Users, Roles, SQL, NoSQL, BLOB Storage and External API Access 
>> Free
>> app hosting. Or install the open source package on any LAMP server.
>> Sign up and see examples for AngularJS, jQuery, Sencha Touch and 
>> Native!
>> http://pubads.g.doubleclick.net/gampad/clk?id=63469471&iu=/4140/ostg.clktrk
>> _______________________________________________
>> Plplot-general mailing list
>> Plp...@li...
>> https://lists.sourceforge.net/lists/listinfo/plplot-general
> 
> DISCLAIMER: This message is intended exclusively for the addressee(s)
> and may contain confidential and privileged information. If you are
> not the intended recipient please notify the sender immediately and
> destroy this message. Unauthorized use, disclosure or copying of this
> message is strictly prohibited. The foundation 'Stichting Deltares',
> which has its seat at Delft, The Netherlands, Commercial Registration
> Number 41146461, is not liable in any way whatsoever for consequences
> and/or damages resulting from the improper, incomplete and untimely
> dispatch, receipt and/or content of this e-mail.

Re: [Plplot-general] Build documentation

[Arjen M. <Arj...@de...>]

Hi Alexis,

I added a short explanation of the example to the INSTALL file, but I saw that a more elaborate explanation appears lower down. You seem to have missed that, but I hope that with this change it is clearer.

Regards,

Arjen
DISCLAIMER: This message is intended exclusively for the addressee(s) and may contain confidential and privileged information. If you are not the intended recipient please notify the sender immediately and destroy this message. Unauthorized use, disclosure or copying of this message is strictly prohibited. The foundation 'Stichting Deltares', which has its seat at Delft, The Netherlands, Commercial Registration Number 41146461, is not liable in any way whatsoever for consequences and/or damages resulting from the improper, incomplete and untimely dispatch, receipt and/or content of this e-mail.

Re: [Plplot-general] Build documentation

[Arjen M. <Arj...@de...>]

Hello Alexis,

wrt PLplot in general: are you using the latest version? What platform (OS, compiler) are you using?

wrt your first question: what version of Tcl do you have installed? I am not familiar with this error and I am a trifle surprised at this. Also the message you refer is from 4 years ago, a lot has changed since then.

wrt the base/source directory: the confusion may be that the build documentation refers to the source directory as the directory containing ALL the source files. There is a subdirectory there called "src" which contains the source files of the PLplot core only. I can understand some of the confusion, but it is a rather common convention.

wrt "plplot_cmake": we should indeed make this example a bit clearer - "plplot_cmake" is merely a place holder and that should have been explained.

Regards,

Arjen

> -----Original Message-----
> From: Alexis Praga [mailto:ale...@ma...]
> Sent: Wednesday, November 13, 2013 12:34 PM
> To: plp...@li...
> Subject: [Plplot-general] Build documentation
>
> Hi,
>
> I have just installed PLplot. For that, I had to disable TCL as mentioned here :
> http://sourceforge.net/p/plplot/discussion/8607/thread/d8bed9d5
> and use the base directory instead of the source directory, as mentioned here :
> http://sourceforge.net/p/plplot/support-requests/31/
>
> Why isn't this in the documentation, especially the second "hack" ?
> Furthermore, the INSTALL file mentions we need to use a certain "plplot_cmake" as
> a source directory, which does not exist.
>
> Thanks
>
> --
> Alexis Praga
>
> ------------------------------------------------------------------------------
> DreamFactory - Open Source REST & JSON Services for HTML5 & Native Apps
> OAuth, Users, Roles, SQL, NoSQL, BLOB Storage and External API Access Free
> app hosting. Or install the open source package on any LAMP server.
> Sign up and see examples for AngularJS, jQuery, Sencha Touch and Native!
> http://pubads.g.doubleclick.net/gampad/clk?id=63469471&iu=/4140/ostg.clktrk
> _______________________________________________
> Plplot-general mailing list
> Plp...@li...
> https://lists.sourceforge.net/lists/listinfo/plplot-general

DISCLAIMER: This message is intended exclusively for the addressee(s) and may contain confidential and privileged information. If you are not the intended recipient please notify the sender immediately and destroy this message. Unauthorized use, disclosure or copying of this message is strictly prohibited. The foundation 'Stichting Deltares', which has its seat at Delft, The Netherlands, Commercial Registration Number 41146461, is not liable in any way whatsoever for consequences and/or damages resulting from the improper, incomplete and untimely dispatch, receipt and/or content of this e-mail.

[Plplot-general] Build documentation

[Alexis P. <ale...@ma...>]

Hi,

I have just installed PLplot. For that, I had to disable TCL as 
mentioned here :
http://sourceforge.net/p/plplot/discussion/8607/thread/d8bed9d5
and use the base directory instead of the source directory, as mentioned 
here :
http://sourceforge.net/p/plplot/support-requests/31/

Why isn't this in the documentation, especially the second "hack" ?
Furthermore, the INSTALL file mentions we need to use a certain 
"plplot_cmake" as a source directory, which does not exist.

Thanks

-- 
Alexis Praga

Re: [Plplot-general] some error in GUI app

[Arjen M. <Arj...@de...>]

Hi Hubin,

I have not had time yet to look into your question. I will try to do so tomorrow.

Regards,

Arjen

From: hubin [mailto:fly...@ou...]
Sent: Sunday, November 03, 2013 8:40 AM
To: Arjen Markus; plp...@li...
Subject: RE: some error in GUI app

HI Arjen,
 I have learn plplot almost a week,I find some problem ,my APP is  WIN32 GUI APP,plplot seem to unfriendly with GUI app.The first  problem is the func "plio_fgets" in \src\plstdio.c;the func "plio_fgets" use "stdin";"stdin" can work fine in a CUI app.But in a GUI APP,how can i  send my string buffer to a console?because the "plio_fgets" just want to get user's option(1,2,3 etc) and set the "response" parameter ,so i want to modify the code and don't
call "plio_fgets",Then i built plplot again ,i get a new "plplotd.dll"and "plplotd.lib",but i find the "plplotd.dll" still
contain the "plio_fgets",i guess may be the cmake ignore my modification; so i have to use "hook",I spent three
day to make "hook" work fine ,then  i see the math graph again;but when i click the "open"button again,oh my god,it close my app ;so i decide debug it again ,and i find the error,it's the func "plD_init_wingcc"in "wingcc.c";
when CreateWindowEx call twice,second time the CreateWindowEx will return null;so CreateWindowEx fails.i try to solve it ,but after 2 days .i still can't have any idea about it .Can you test x02c.c in you GUI app,you use a button,and click it twice;Thank you



DISCLAIMER: This message is intended exclusively for the addressee(s) and may contain confidential and privileged information. If you are not the intended recipient please notify the sender immediately and destroy this message. Unauthorized use, disclosure or copying of this message is strictly prohibited. The foundation 'Stichting Deltares', which has its seat at Delft, The Netherlands, Commercial Registration Number 41146461, is not liable in any way whatsoever for consequences and/or damages resulting from the improper, incomplete and untimely dispatch, receipt and/or content of this e-mail.

Re: [Plplot-general] some error in GUI app

[hubin <fly...@ou...>]

HI Arjen, I have learn plplot almost a week,I find some problem ,my APP is  WIN32 GUI APP,plplot seem to unfriendly with GUI app.The first  problem is the func "plio_fgets" in \src\plstdio.c;the func "plio_fgets" use "stdin";"stdin" can work fine in a CUI app.But in a GUI APP,how can i  send my string buffer to a console?because the "plio_fgets" just want to get user's option(1,2,3 etc) and set the "response" parameter ,so i want to modify the code and don'tcall "plio_fgets",Then i built plplot again ,i get a new "plplotd.dll"and "plplotd.lib",but i find the "plplotd.dll" stillcontain the "plio_fgets",i guess may be the cmake ignore my modification; so i have to use "hook",I spent threeday to make "hook" work fine ,then  i see the math graph again;but when i click the "open"button again,oh my god,it close my app ;so i decide debug it again ,and i find the error,it's the func "plD_init_wingcc"in "wingcc.c";when CreateWindowEx call twice,second time the CreateWindowEx will return null;so CreateWindowEx fails.i try to solve it ,but after 2 days .i still can't have any idea about it .Can you test x02c.c in you GUI app,you use a button,and click it twice;Thank you