plplot-general — PLplot user discussion group

Re: [Plplot-general] CMake one-line patch required for fortran to work on Linux

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

On 2006-07-28 13:51-0700 Alan W. Irwin wrote:

> This patch needs to be applied to cmake-2.4.2 in order for the fortran
> shared libraries to build properly on Linux.

Forget that.  I have just made a change to our own CMake build system to set
this variable correctly so you will not have to apply any patches to
cmake-2.4.2.

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); PLplot scientific plotting software
package (plplot.org); the Yorick front-end to PLplot (yplot.sf.net); the
Loads of Linux Links project (loll.sf.net); and the Linux Brochure Project
(lbproject.sf.net).
__________________________

Linux-powered Science
__________________________

Re: [Plplot-general] Question about C 2D arrays for Ada binding

[<hba...@ma...>]

On Jul 28, 2006, at 6:15 PM, Jerry wrote:
>
> How does C map 2D arrays such as PLFLT **z to memory (e.g. plmesh)?
> Is this an array of pointers that then each point to the first
> element of a row (column??) of z, or is this simply a pointer to a
> pointer to z[0, 0] which merely needs to be twice de-referenced?

PLFLT **z is an array of pointers each of which point to the first  
element of a row/column.

> Let me try to be even more confusing. Say z is M x N. Is the address
> that is pointed to by PLFLT **z the first byte of a blob of memory
> that is M*N*sizeof(int), with M and N used in parsing up this blob
> into addressable array elements? Or are there actually M pointers to
> M memory locations, each of which represents N array elements as a
> contiguous blob of memory amounting to N*sizeof(int) bytes? If the
> latter is true, then it seems that z would require M*N*sizeof(int)
> +M*sizeof(pointer) bytes, and somehow this seems weird. If the latter

It does allow you to use M non-contiguous blocks of memory of size N  
* sizeof(int), rather than one large contiguous block of size  
M*N*sizeof(int). This might have been handy back in the day when  
computers had small fragmented memories :).

-Hazen

[Plplot-general] Question about C 2D arrays for Ada binding

[Jerry <lan...@qw...>]

I've had a nearly-complete thin binding of PLplot to Ada for a few  
weeks now, and am working on a thick binding. However, even at the  
thin binding level, I've encountered a problem with passing 2D Ada  
arrays to C. I refer the interested to this thread at comp.lang.ada  
(as hosted by Google):

http://groups.google.com/group/comp.lang.ada/browse_thread/thread/ 
6b3a3c920575b35a/193ba675e86cb69a?

Being a relative Ada newbie and just learning enough C in the process  
to get by, I thought the gurus on the list would have the answer.  
However, it hasn't been forthcoming and even they seem befuzzled =  
befuddled + puzzled.

I don't expect any Ada per se help from this post, and this is really  
a C question, but since there might be a vested interest from the  
PLplot community to have functioning Ada bindings, maybe you will  
humor this question:

How does C map 2D arrays such as PLFLT **z to memory (e.g. plmesh)?  
Is this an array of pointers that then each point to the first  
element of a row (column??) of z, or is this simply a pointer to a  
pointer to z[0, 0] which merely needs to be twice de-referenced?

Let me try to be even more confusing. Say z is M x N. Is the address  
that is pointed to by PLFLT **z the first byte of a blob of memory  
that is M*N*sizeof(int), with M and N used in parsing up this blob  
into addressable array elements? Or are there actually M pointers to  
M memory locations, each of which represents N array elements as a  
contiguous blob of memory amounting to N*sizeof(int) bytes? If the  
latter is true, then it seems that z would require M*N*sizeof(int) 
+M*sizeof(pointer) bytes, and somehow this seems weird. If the latter  
is true, then could one access array element i,j by an offset M + M*i  
+ j or something like that?

Note that the standard C interfaces provided by Ada deal perfectly  
and easily with 1D arrays.

Thanks for any comments.
Jerry

[Plplot-general] CMake one-line patch required for fortran to work on Linux

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

This patch needs to be applied to cmake-2.4.2 in order for the fortran
shared libraries to build properly on Linux.  All other PLplot platforms are
unaffected by this Cmake bug.

--- share/CMake/Modules/Platform/Linux.cmake_original   2006-07-04
18:16:42.000000000 -0700
+++ share/CMake/Modules/Platform/Linux.cmake    2006-07-28
13:13:45.000000000 -0700
@@ -7,4 +7,5 @@
  SET(CMAKE_SHARED_LIBRARY_RUNTIME_C_FLAG_SEP ":")
  SET(CMAKE_SHARED_LIBRARY_SONAME_C_FLAG "-Wl,-soname,")
  SET(CMAKE_SHARED_LIBRARY_SONAME_CXX_FLAG "-Wl,-soname,")
+SET(CMAKE_SHARED_LIBRARY_SONAME_Fortran_FLAG "-Wl,-soname,")
  INCLUDE(Platform/UnixPaths)

I hope to get this patch into the next release of cmake.

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); PLplot scientific plotting software
package (plplot.org); the Yorick front-end to PLplot (yplot.sf.net); the
Loads of Linux Links project (loll.sf.net); and the Linux Brochure Project
(lbproject.sf.net).
__________________________

Linux-powered Science
__________________________

[Plplot-general] Request for testing of the experimental CMake build system

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

To PLplot testers everywhere:

As you may be aware the KDE developers have recently switched from an
autotools-based build system to the CMake build system.  The stated
advantages of CMake are substantially faster build results, easier build
system maintenance, and full support for builds on bare windows platforms as
well as Unix, Mac OS X, and Linux.

Inspired by this story, Andrew Ross and I have been implementing a CMake
build system for PLplot.  Although still experimental, we have good reports
for Linux, Mac OS X, and MinGW on windows.  Buoyed by this success, I would
like to encourage widespread testing of the current PLplot cvs version of
our CMake build system.

The CMake build system for PLplot does have some current limitations. We are
still missing the Tcl/Tk interface and a documentation build system. We are
also missing the wxwidgets device, the Tk-related devices, and specialized
windows and Mac OS X devices as well as some more obscure devices.

Despite these current limitations, the results of the PLplot CMake build
have a lot of capability which is already suitable for many users. For
example, on Linux (Debian stable and Ubuntu dapper) we have the C, C++, f77,
f95, java, and python interfaces and examples working as well as png, jpeg,
gif, gcw, ps, psttf, and xwin device drivers.  For the postscript results
produced by ./plplot-test.sh in the installed examples directory, we are
getting the same results (aside from occasional floating-point rounding
errors) as previous results with the autotools build system.  This is most
encouraging.

Here is how to get involved if you would like to help us out with some test
results, learn more about CMake, and/or if you are simply curious about the
speed and convenience of this experimental build system for PLplot.

(1) Download, build and install cmake-2.4.2 (that version is essential and
available from http://www.cmake.org/HTML/Download.html).

(2) Initialize the PLplot build as follows:

cmake -DCMAKE_INSTALL_PREFIX=/your_install_prefix \
-DCMAKE_VERBOSE_MAKEFILE=ON ../plplot_cmake >& cmake.out

plplot_cmake must be a clean PLplot source tree checked out from CVS HEAD
following the directions at http://sourceforge.net/cvs/?group_id=2915

The above cmake command is initially run from an empty build tree.  After
running it you will probably also want to run

ccmake ../plplot_cmake

which is a command-line-interface that makes it convenient to set PLplot
options.

The above cmake command completely replaces the cf/bootstrap.sh and
./configure commands that you are used to (and is much faster than those
commands as well which is one of our motivations for implementing this new
CMake build system for PLplot).

(3) From the same separate build directory that was populated by Makefiles
with the above cmake command, finish the PLplot build and install following
my usual recommended pattern for initial testing of new versions of PLplot
software, i.e.,

make >& make.out
make install >& make_install.out
cd /tmp
cp -a /your_install_prefix/share/plplot-5.6.1/examples .
cd examples
make >& make_examples.out
./plplot-test.sh >& plplot-test.out

The above testing pattern should work on any Unix (including Cygwin and
MinGW on windows).  Currently, we have no experience with the Plplot CMake
build system on bare windows (i.e., windows without Cygwin or MinGW).  For
that case, you will have to consult the CMake documentation (collected
below) to see how to mimic the above test pattern, but we will be most
interested in the results.

(4) If there are problems in any of the above steps, please collect into a
compressed tarball all the important CakeCache.txt,
CMakeFiles/CMakeError.log, and CMakeFiles/CMakeOutput.log files produced by
the cmake command as well as _all_ of the above *.out files up to and
including the step where the problem occurred.  You can send that compressed
tarball to me or the plplot_devel list as an attachment and I will take
it from there.

(5) Success stories are welcome as well!

(6) If you want to delve more deeply into CMake (especially for the bare
windows case), I have collected some useful references in
cmake/README.cmake_doc in the CVS version of PLplot.

Good luck with your testing of the experimental build system!

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); PLplot scientific plotting software
package (plplot.org); the Yorick front-end to PLplot (yplot.sf.net); the
Loads of Linux Links project (loll.sf.net); and the Linux Brochure Project
(lbproject.sf.net).
__________________________

Linux-powered Science
__________________________

Re: [Plplot-general] PLPlot DLL

[Ferrell, S. <sfe...@ya...>]

The latest revision of PLPLOT generates a DLL when you run NMAKE.  It
also 
creates the static link lib as well.  I've attached the DLL to save you
some 
time.


----- Original Message ----- 
From: "Henning Thielemann" <pl...@he...>
To: "plplot_general" <plp...@li...>
Sent: Wednesday, July 19, 2006 7:32 PM
Subject: [Plplot-general] PLPlot DLL


>
> I have some very stupid FAQ: Is there some DLL of plplot availabe? I
have
> some program developed with Modula-3 using plplot on Linux and now I
want
> to run that on plain Windows XP without going through the nightmare
of
> installing a C compiler on Windows or even a Linux emulation
environment.
> I hope that a precompiled DLL of plplot would save me much trouble.
But I
> only see a source archive on sourceforge.net, packed with tar, which
is
> already a problem for a bare Windows, you know. :-)
>
>
-------------------------------------------------------------------------
> Take Surveys. Earn Cash. Influence the Future of IT
> Join SourceForge.net's Techsay panel and you'll get the chance to
share 
> your
> opinions on IT & business topics through brief surveys -- and earn
cash
>
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
> _______________________________________________
> Plplot-general mailing list
> Plp...@li...
> https://lists.sourceforge.net/lists/listinfo/plplot-general
> 


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 

Re: [Plplot-general] PLPlot DLL

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

On 2006-07-19 17:32+0200 Henning Thielemann wrote:

>
> I have some very stupid FAQ: Is there some DLL of plplot availabe? I have
> some program developed with Modula-3 using plplot on Linux and now I want
> to run that on plain Windows XP without going through the nightmare of
> installing a C compiler on Windows or even a Linux emulation environment.
> I hope that a precompiled DLL of plplot would save me much trouble. But I
> only see a source archive on sourceforge.net, packed with tar, which is
> already a problem for a bare Windows, you know. :-)

Hi Henning:

I am not aware of any publically available PLplot DLL for bare Windows.
However, we are making excellent progress with the new CMake build system
which should make it much easier to build PLplot on Linux, Unix, Mac OS X,
and especially bare Windows.  Of course, a compiler is required for the
build, but you cannot have everything. :-)

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); PLplot scientific plotting software
package (plplot.org); the Yorick front-end to PLplot (yplot.sf.net); the
Loads of Linux Links project (loll.sf.net); and the Linux Brochure Project
(lbproject.sf.net).
__________________________

Linux-powered Science
__________________________

[Plplot-general] PLPlot DLL

[Henning T. <pl...@he...>]

I have some very stupid FAQ: Is there some DLL of plplot availabe? I have
some program developed with Modula-3 using plplot on Linux and now I want
to run that on plain Windows XP without going through the nightmare of
installing a C compiler on Windows or even a Linux emulation environment.
I hope that a precompiled DLL of plplot would save me much trouble. But I
only see a source archive on sourceforge.net, packed with tar, which is
already a problem for a bare Windows, you know. :-)

Re: [Plplot-general] Notes on Ada binding (also documentation and Modula-3)

[Henning T. <pl...@he...>]

On Fri, 7 Jul 2006, Jerry wrote:

> I took the API reference literally if it said that a parameter is an  
> output, and made the corresponding Ada parameter say "out", rather  
> than "in out" which would be allowed by a C pointer or a Pascal or  
> Modula-3 "var".

Since Modula-3 does not know OUT parameters, I turn such parameters into
return values, possibly bundled in RECORDs, when there are multiple ones.  
Because of Modula-3's garbage collector, I can nicely work with
references, without destruction of input parameters. This allows a very
functional style of programming. But, as far as I know, Ada does not have
a garbage collector and thus treating output parameters as OUT is
certainly the best way.

> The following subprograms in the Common API had output scalars which  
> required special attention to make sure that Ada understands that  
> some arguments are output scalars and not pointers to arrays. I've  
> included a few notes that may be of interest to the documentation  
> writers and to the Modula-3 binding writer. Please note that I have  
> used 5.5.3 for plplot.h and 5.6.1 for the API documentation. Also,  
> note that not all routines which return results begin with "plg".

I started on some specialised interface file (was it Java or Python?), and
I have probably missed additions to the main header file. Originally I
hoped that SWIG allows me to keep to the main development path. However
since the main plplotcapi.i has not enough type information, this has
failed so far.  Either we come to an agreement to replace plplotcapi.i by
a file with more specific types (which wouldn't bother C compilers,
because they keep on seeing 'int's) or I'm happy if you add missing
functions to the interface prepared for Modula-3, such that we can share
our work.

[Plplot-general] Notes on Ada binding (also documentation and Modula-3)

[Jerry <lan...@qw...>]

This is a continuation of an earlier thread about identifying which C  
subroutines in plplot.h return scalars, since the C syntax is  
ambiguous with respect to whether the intention is to return e.g. a  
float scalar or to pass an array of floats.

I downloaded the documentation in HTML format and made a list of  
files that begin with "pl" and which contain the word "output." (The  
Spotlight function in OS X rocks for this kind of thing.) Some of the  
hits were spurious in their use of "output" which files I ignored.

I took the API reference literally if it said that a parameter is an  
output, and made the corresponding Ada parameter say "out", rather  
than "in out" which would be allowed by a C pointer or a Pascal or  
Modula-3 "var".

The following subprograms in the Common API had output scalars which  
required special attention to make sure that Ada understands that  
some arguments are output scalars and not pointers to arrays. I've  
included a few notes that may be of interest to the documentation  
writers and to the Modula-3 binding writer. Please note that I have  
used 5.5.3 for plplot.h and 5.6.1 for the API documentation. Also,  
note that not all routines which return results begin with "plg".

plcalc_world
plgchr
plgcol0  API reference says this is 8 bits but plplot.h says integer
plgcolbg API reference says this is 8 bits but plplot.h says integer
plgcompression
plgdev Missing from Modula-3
plgdidev
plgdiori
plgdiplt
plgfam
plgfci Missing from Modula-3
plglevel
plgpage
plgriddata API description of parameter "type" is incomplete. Missing  
from Modula-3
plgspa
plgstrm
plgver Missing from Modula-3
plgvpd
plgvpw
plgxax
plgyax
plgzax
plhlsrgb Missing from Modula-3
plmkstrm
plrgbhls API description reverses "input" and "output". Missing from  
Modula-3
plstripc
plxormod



Other Notes:

plexit is listed in the API reference but does not appear in  
plplot.h. Note that I am using version 5.5.3 and the API appears to  
be for 5.6.1.

plsdev In API description, the parameter is marked as output but is  
input.

plsetopt Ditto both args (Note: not part of common API)

Jerry

Re: [Plplot-general] Bindings for Ada and a dumb C question

[<lan...@qw...>]

On Jul 3, 2006, at 2:55 AM, lan...@qw... wrote:

> Hi list,
>
> Sorry to bother with such a trivial question that should properly be
> asked elsewhere, but maybe you'll humor me.
>
> I'm making some serious progress as a spare-time project in making
> Ada bindings for PLplot, and some of the examples are working well.
> Here's my dumb C question: If there is a C procedure such as
>
> void plfoo(PLFLT *bar);
>
> how does one tell if the parameter is intended to be a pointer to an
> array of PLFLT or whether it is simply a scalar (single number) PLFLT
> that is being passed by reference? Am I missing something here? Will
> I have to consult the API reference for each function to discover
> what is intended if it's not obvious otherwise?
>
> Jerry (aka C idiot)

Thanks to everyone for their suggestions. I'm finding the Modula-3  
bindings very helpful, and I'll take a look at the FORTRAN bindings-- 
I've found a handy FORTRAN-to-Ada converter that worked pretty well  
for converting some of the examples to Ada--maybe I should have  
applied it to the bindings themselves. I also found a c2ada converter  
that looks very polished but unfortunately needs a bit of updating in  
order to work correctly. I looked at the SWIG approach but decided  
that for this rather smallish job, given Ada's relatively  
friendliness with C, and given that learning about SWIG was still  
another layer of stuff I didn't know (added to C and Ada itself),  
that a more direct approach would work OK. I've used a few regex  
lines applied to plplot.h to great advantage. About the only thing  
puzzling me now is what in the heck is this:

typedef struct {
     char *opt;
     int  (*handler)	(char *, char *, void *);  <<<<<<< ?
     void *client_data;                             <<<<<<< ?
     void *var;                                     <<<<<<< ?
     long mode;
     char *syntax;
     char *desc;
} PLOptionTable;

Jerry

Re: [Plplot-general] Bindings for Ada and a dumb C question

[Henning T. <pl...@he...>]

On Mon, 3 Jul 2006, lan...@qw... wrote:

> I'm making some serious progress as a spare-time project in making
> Ada bindings for PLplot, and some of the examples are working well.
> Here's my dumb C question: If there is a C procedure such as
> 
> void plfoo(PLFLT *bar);
> 
> how does one tell if the parameter is intended to be a pointer to an
> array of PLFLT or whether it is simply a scalar (single number) PLFLT
> that is being passed by reference?

I encountered the same problems with Modula-3 bindings. In order to resolve 
them, I had to define dummy types which carry the additional information but 
are equal in the view of a C compiler. You might find my extended PLPlot 
interfaces useful:
  http://www.elegosoft.com/cgi-bin/cvsweb.cgi/cm3/m3-libs/plplot/swig/plplotcapi.i?rev=1.18&content-type=text/x-cvsweb-markup


I want to add, that I assume that the Modula-3 interface contains more 
details than Java and Python interfaces (however not always tested ones), 
because Java is less type safe than Modula-3, not to speak of Python. I 
remember that I started with one of these interfaces (Java or Python) and 
that I had still a lot to add.

Re: [Plplot-general] Dumb Link Question

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

On 2006-07-03 09:48-0700 Ferrell, Stephen wrote:

> I'm integrating a build of Plplot and WxWindows.  I built the WxWindows library successfully...all models and also built the wxplot library (plplot 5.51 with WxWindows driver) by German Carrera.
>
>  Everything compiles fine but when I attempt to   compile and link the demowxplot program, I get the following link error in   Visual Stuido 6:
>
>  wxplot.lib(wxPlot.obj) : error LNK2001:   unresolved external symbol "struct PLStream * plsc" (?plsc@@3PAUPLStream@@A)
>
>  I assume it is because the C++ compiler is   mangling a C structure in plcore.c, but I don't know how to fix   it.
>
>  Any suggestions?  Thanks in   advance.
>
>
>  Steve

PLplot-5.5.1 is out-of-date and no longer supported.  Please, use
PLplot-5.6.1 instead.

Also, you may not be aware of it, but we already have an integrated
wxwindows device driver (donated by Werner Smekal) that works well on at
least my Linux platform for PLplot-5.6.1. What are your results for that
device?

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); PLplot scientific plotting software
package (plplot.org); the Yorick front-end to PLplot (yplot.sf.net); the
Loads of Linux Links project (loll.sf.net); and the Linux Brochure Project
(lbproject.sf.net).
__________________________

Linux-powered Science
__________________________

[Plplot-general] Dumb Link Question

[Ferrell, S. <sfe...@ya...>]

I'm integrating a build of Plplot and WxWindows.  I built the WxWindows library successfully...all models and also built the wxplot library (plplot 5.51 with WxWindows driver) by German Carrera.
   
  Everything compiles fine but when I attempt to   compile and link the demowxplot program, I get the following link error in   Visual Stuido 6:
   
  wxplot.lib(wxPlot.obj) : error LNK2001:   unresolved external symbol "struct PLStream * plsc" (?plsc@@3PAUPLStream@@A)
   
  I assume it is because the C++ compiler is   mangling a C structure in plcore.c, but I don't know how to fix   it.
   
  Any suggestions?  Thanks in   advance.
   
   
  Steve

 		
---------------------------------
Yahoo! Music Unlimited - Access over 1 million songs.Try it free. 

Re: [Plplot-general] Bindings for Ada and a dumb C question

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

On 2006-07-03 12:07+0200 Arjen Markus wrote:

> lan...@qw... wrote:
>
>> Hi list,
>>
>> Sorry to bother with such a trivial question that should properly be
>> asked elsewhere, but maybe you'll humor me.
>>
>> I'm making some serious progress as a spare-time project in making
>> Ada bindings for PLplot, and some of the examples are working well.
>> Here's my dumb C question: If there is a C procedure such as
>>
>> void plfoo(PLFLT *bar);
>>
>> how does one tell if the parameter is intended to be a pointer to an
>> array of PLFLT or whether it is simply a scalar (single number) PLFLT
>> that is being passed by reference? Am I missing something here? Will
>> I have to consult the API reference for each function to discover
>> what is intended if it's not obvious otherwise?
>>
>>
> Unfortunately, C does not make any distinction between the two cases
> you mention. So, indeed, you will have to consult the documentation instead
> (and ultimately even the source code for that particular function).
>
> You may find it useful to look at the Fortran 90/95 interface, because for
> that interface we had to solve the same (or at least a similar) problem.
> We have not used the INTENT(IN/OUT) attribute there, because it
> might interfere with the way C expects the arguments to be passed,
> but at least it will give you a way other than the documentation of
> identifying which are scalar and which are array arguments.
>
> PLplot has a helpful naming convention:
> plg - for functions that return attributes
> pls - for functions that set attributes

To add to what Arjen said, our swig-based interfaces (currently Java and
Python) require exactly the information you requested.  It is collected in
bindings/swig-support/plplotcapi.i in an easy-to-understand form.

BTW, have you considered using Swig for this Ada interface?  Swig provides a
framework for interfacing any language with C libraries such as libplplot.
It's huge advantage is it recognizes all distinct C argument patterns (see
plplotcapi.i above) and treats them in a uniform manner rather than relying
on human recognition of the argument pattern that is required for
hand-crafted interfaces and the associated errors from failing to recognize
the pattern or treat it in a uniform way. Also, since we have plplotcapi.i
already set up, adding a new PLplot language interface using swig is
straightforward.

Case in point, we had worked on a hand-crafted Java interface to PLplot off
and on for years, but I was getting frustrated by how much effort it would
take to complete it.  Therefore, I tried the swig approach instead following
the pioneering effort that had been done with the swig-based python
interface.  That new swig-based Java interface took me, a relative Java
newbie, only a week or so to set up, and the result was automatically
complete.  It's because of this huge saving of programming effort that
I always recommend swig for new language interfaces to PLplot.

The only difficulty I can see with generating a swig-based Ada interface to
PLplot is that language is not yet officially supported by swig (see
http://www.swig.org/compat.html#SupportedLanguages which details the ~20
computer languages officially supported by Swig). However, a superficial
google search for (swig ada) showed there has been at least one attempt to
add an ada module to swig.  At that point, I didn't dig any deeper, but if
there is continuing development of such a module, you might benefit a whole
lot of Ada users (not just PLplot ones) by adding your Ada interfacing
expertise to that effort.  The whole effort might actually take less time
than a hand-crafted interface to PLplot simply because of the argument
pattern recognition factor discussed above which makes the swig-based
approach so much easier to programme.

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); PLplot scientific plotting software
package (plplot.org); the Yorick front-end to PLplot (yplot.sf.net); the
Loads of Linux Links project (loll.sf.net); and the Linux Brochure Project
(lbproject.sf.net).
__________________________

Linux-powered Science
__________________________

Re: [Plplot-general] Bindings for Ada and a dumb C question

[Arjen M. <arj...@wl...>]

lan...@qw... wrote:

>Hi list,
>
>Sorry to bother with such a trivial question that should properly be  
>asked elsewhere, but maybe you'll humor me.
>
>I'm making some serious progress as a spare-time project in making  
>Ada bindings for PLplot, and some of the examples are working well.  
>Here's my dumb C question: If there is a C procedure such as
>
>void plfoo(PLFLT *bar);
>
>how does one tell if the parameter is intended to be a pointer to an  
>array of PLFLT or whether it is simply a scalar (single number) PLFLT  
>that is being passed by reference? Am I missing something here? Will  
>I have to consult the API reference for each function to discover  
>what is intended if it's not obvious otherwise?
>  
>
Unfortunately, C does not make any distinction between the two cases
you mention. So, indeed, you will have to consult the documentation instead
(and ultimately even the source code for that particular function).

You may find it useful to look at the Fortran 90/95 interface, because for
that interface we had to solve the same (or at least a similar) problem.
We have not used the INTENT(IN/OUT) attribute there, because it
might interfere with the way C expects the arguments to be passed,
but at least it will give you a way other than the documentation of
identifying which are scalar and which are array arguments.

PLplot has a helpful naming convention:
plg - for functions that return attributes
pls - for functions that set attributes

Regards,

Arjen

[Plplot-general] Bindings for Ada and a dumb C question

[<lan...@qw...>]

Hi list,

Sorry to bother with such a trivial question that should properly be  
asked elsewhere, but maybe you'll humor me.

I'm making some serious progress as a spare-time project in making  
Ada bindings for PLplot, and some of the examples are working well.  
Here's my dumb C question: If there is a C procedure such as

void plfoo(PLFLT *bar);

how does one tell if the parameter is intended to be a pointer to an  
array of PLFLT or whether it is simply a scalar (single number) PLFLT  
that is being passed by reference? Am I missing something here? Will  
I have to consult the API reference for each function to discover  
what is intended if it's not obvious otherwise?

Jerry (aka C idiot)

Re: [Plplot-general] Want to share a c++ wrapper plplot dll with libgd

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

On 2006-06-30 15:54-0400 Ruwei Liu wrote:

> Hi, All,
>
>    I've worked out a .Net c++ wrapper dll based on the plplot 5.6.1-RC
> source code, and made some changes
> to let it have GD support(GD 2.0.33). It works fine as far as my test
> shows. I want to share it so that others will not
> need to repeat the work.
>   This is my first time to participate in such activities, I don't know
> the rules, and I'm not quite sure what I shall do
> to publish my work here. Could anybody kindly help me figure it out?
>   Thanks!

The usual way to donate code to the PLplot project is with a patch file.

The way to generate a patch file with Linux is

diff -Naur old_clean_tree new_clean_tree >patch.out

where old_clean_tree would be the result of a freshly unpacked 5.6.1
tarball, and new_clean_tree is the same except for your source code changes
but nothing else (i.e., nothing compiled in either tree).  In general you
should look over the patch file to make sure all the changed and new files
there are exactly what you want.

Then submit patch.out to the plplot_devel mailing list as an attachment
(preferred) or else submit it to
http://sourceforge.net/tracker/?group_id=2915&atid=302915. Either submission
method will give the core developers a chance to evaluate whether we want to
include your patch in PLplot or not or perhaps provide the same
functionality in a different way or whatever.

Good luck with the patch generation, and thanks for making an effort to
submit code to PLplot.

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); PLplot scientific plotting software
package (plplot.org); the Yorick front-end to PLplot (yplot.sf.net); the
Loads of Linux Links project (loll.sf.net); and the Linux Brochure Project
(lbproject.sf.net).
__________________________

Linux-powered Science
__________________________

[Plplot-general] Want to share a c++ wrapper plplot dll with libgd

[Ruwei L. <rw...@ph...>]

Hi, All,

    I've worked out a .Net c++ wrapper dll based on the plplot 5.6.1-RC 
source code, and made some changes
to let it have GD support(GD 2.0.33). It works fine as far as my test 
shows. I want to share it so that others will not
need to repeat the work.
   This is my first time to participate in such activities, I don't know 
the rules, and I'm not quite sure what I shall do
to publish my work here. Could anybody kindly help me figure it out?
   Thanks!

   Regards,
   Ruwei Liu

[Plplot-general] Building the java binding on Debian testing/unstable

[Rafael L. <rla...@us...>]

Here is a tip for those out there building the PLplot Java binding on
Debian testing or unstable systems (or any other derivative distribution,
like Ubuntu).  It is enough to install the java-gcj-compat-dev package
and use --with-java-home=/usr/lib/jvm/java-gcj.  This will become the
default method to build java packages in Debian soon.

On my unstable system, java-gcj-compat-dev depends currently on gcj-4.1
and libgcj7-dev, which belong to the recommended version of the GCJ
compiler. The symlink /usr/lib/jvm/java-gcj points correctly to the Java
home directory /usr/lib/jvm/java-1.4.2-gcj-4.1-1.4.2.0.

-- 
Rafael

[Plplot-general] PLplot 5.6.1 Debian packages available

[Rafael L. <rla...@us...>]

Debian (unstable) packages for PLplot 5.6.1 are available at the
apt-getable repository:

    http://people.debian.org/~rafael/plplot/

These packages will be uploaded to the Debian repository as soon as the
5.6.0-1 packages will enter unstable (the later are waiting at the NEW
queue for ftp-admin approval: http://ftp-master.debian.org/new.html).

Reports about eventual problems will be appreciated.

-- 
Rafael

Re: [Plplot-general] problems with using Dll in VC++.net

[Arjen M. <arj...@wl...>]

Ruwei Liu wrote:

> Hi, Arjen,
>
>    I was about to send you the source code and project file when I 
> found that my project was using
> C++ binding, and I tried to link it to the original plplotd.dll and a 
> new "plstream.dll" which I created using
> the source code (with minor change) to support C++ binding. So those 
> error messages may not due to
> plplotd.dll. It maybe caused by my "plstream.dll".  Is it the right 
> way to create a c++ binding dll for plplot
> just using the "plstream.cc" and "plstream.h"(change the declaration 
> of  "class plstream" to "class __declspec(dllexport) plstream")
> and link the project to "plplotd.dll"?  Thanks a lot!
>
>     I'm really really sorry for my mistake.  I got so dippy with 
> adding on C++ binding, GD lib and/or freetype
> to the Plplot library.

Hi Ruwei,

no need to apologise - building programs is a hazardous and complicated 
task. All the various
options for using libraries do not help making it easier.

Right now I do not have any experience using the C++ bindings under MSVC 
(more
precisely the sys/win32/msdev platform), as I do not program in C++ 
myself. It may be
worth looking at, but that is a task for the hopefully near future.

I intended to get freetype and GD working correctly for this release, 
but there are a
few wrinkles to iron out first.

Regards,

Arjen

Re: [Plplot-general] problems with using Dll in VC++.net

[Ruwei L. <rw...@ph...>]

Hi, Arjen,

    I was about to send you the source code and project file when I 
found that my project was using
C++ binding, and I tried to link it to the original plplotd.dll and a 
new "plstream.dll" which I created using
the source code (with minor change) to support C++ binding. So those 
error messages may not due to
plplotd.dll. It maybe caused by my "plstream.dll".  Is it the right way 
to create a c++ binding dll for plplot
just using the "plstream.cc" and "plstream.h"(change the declaration of  
"class plstream" to "class __declspec(dllexport) plstream")
and link the project to "plplotd.dll"?  Thanks a lot!

     I'm really really sorry for my mistake.  I got so dippy with adding 
on C++ binding, GD lib and/or freetype
to the Plplot library.

     Regards,
     Ruwei


    
   

Arjen Markus wrote:

> Ruwei Liu wrote:
>
>> Hi, all
>>
>>     I'm trying to use plplot library for a project. It works fine 
>> with the static library , but it fails to
>> link when I using the dynamic library. I got the following error 
>> message:
>>
>>         ------ Build started: Project: Tplot, Configuration: Debug 
>> Win32 ------
>>
>>    Linking...
>>    LINK : error LNK2020: unresolved token (0A0000B3) 
>> plD_dispatch_init_null
>>    LINK : error LNK2020: unresolved token (0A0000B4) 
>> plD_dispatch_init_gif
>>    LINK : error LNK2020: unresolved token (0A0000B5) 
>> plD_dispatch_init_jpeg
>>    LINK : error LNK2020: unresolved token (0A0000B6) 
>> plD_dispatch_init_png
>>    LINK : error LNK2020: unresolved token (0A0000B7) 
>> plD_dispatch_init_ljii
>>    LINK : error LNK2020: unresolved token (0A0000B8) 
>> plD_dispatch_init_ljiip
>>    LINK : error LNK2020: unresolved token (0A0000B9) 
>> plD_dispatch_init_xfig
>>    LINK : error LNK2020: unresolved token (0A0000BA) 
>> plD_dispatch_init_psc
>>    LINK : error LNK2020: unresolved token (0A0000BB) 
>> plD_dispatch_init_psm
>>    LINK : error LNK2020: unresolved token (0A0000BC) 
>> plD_dispatch_init_plm
>>    LINK : error LNK2020: unresolved token (0A0000BD) 
>> plD_dispatch_init_win3
>>    LINK : fatal error LNK1120: 11 unresolved externals
>>
>>    Tplot - 12 error(s), 0 warning(s)
>>
>>     I did put the "plplotd.lib", "plplotd.def", "plplotd.exp" and 
>> "plplotd.dll" to the project directory, link
>> to "plplotd.lib", and also defined "__PLDLL_H__", is there anything 
>> I'm still missing?
>
>
> Have you looked at the makefile for the examples with the DLL in 
> sys\msdev\win32\dexamples?
> These compile and link allright, so I would suggest you mimick the 
> compile and link options
> from these examples.
>
> If this does not work, can you send me the source code and the 
> makefile/project file, so that I
> can try and reproduce this on my machine? (I maintain the win32 device 
> and the MSVC
> build system)
>
> Regards,
>
> Arjen

Re: [Plplot-general] problems with using Dll in VC++.net

[Ruwei L. <rw...@ph...>]

Hi, Arjen

    Thanks for replying.
    Actually, examples with DLL won't compile on my computer. I got the 
following error messages when I tried to
make:

    Microsoft (R) Program Maintenance Utility Version 7.10.3077
    Copyright (C) Microsoft Corporation.  All rights reserved.

    No configuration specified.  Defaulting to x19d - Win32 Debug.
            if not exist "./x01d/" mkdir "./x01d"
            if not exist ".\x01d\Release/" mkdir ".\x01d\Release"
            cl.exe /nologo /ML /W3 /GX /O2 /D "NDEBUG" /D "NOBRAINDEAD"
    /D "WIN32" /
    D "_CONSOLE" /Fp".\x01d\Release/x01d.pch" /D "__PLDLL_H__" /YX
    /Fo".\x01d\Releas
    e/" /c ..\..\..\..\tmp\x01c.c
    x01c.c
    c:\downloads\plplot-5.6.1_RC1\tmp\plplot.h(617) : warning C4005:
    'API' : macro r
    edefinition
            c:\downloads\plplot-5.6.1_RC1\tmp\plplot.h(149) : see
    previous definitio
    n of 'API'
    c:\downloads\plplot-5.6.1_RC1\tmp\plplot.h(1886) : error C2373:
    'plhlsrgb' : red
    efinition; different type modifiers
            c:\downloads\plplot-5.6.1_RC1\tmp\plplot.h(1141) : see
    declaration of 'p
    lhlsrgb'
    c:\downloads\plplot-5.6.1_RC1\tmp\plplot.h(1889) : error C2373:
    'plrgbhls' : red
    efinition; different type modifiers
            c:\downloads\plplot-5.6.1_RC1\tmp\plplot.h(1315) : see
    declaration of 'p
    lrgbhls'
    ..\..\..\..\tmp\x01c.c(121) : warning C4013: 'plparseopts'
    undefined; assuming e
    xtern returning int
    NMAKE : fatal error U1077: 'cl.exe' : return code '0x2'
    Stop.

     I'll send you the project source later.

     Regards,
     Ruwei


Arjen Markus wrote:

> Ruwei Liu wrote:
>
>> Hi, all
>>
>>     I'm trying to use plplot library for a project. It works fine 
>> with the static library , but it fails to
>> link when I using the dynamic library. I got the following error 
>> message:
>>
>>         ------ Build started: Project: Tplot, Configuration: Debug 
>> Win32 ------
>>
>>    Linking...
>>    LINK : error LNK2020: unresolved token (0A0000B3) 
>> plD_dispatch_init_null
>>    LINK : error LNK2020: unresolved token (0A0000B4) 
>> plD_dispatch_init_gif
>>    LINK : error LNK2020: unresolved token (0A0000B5) 
>> plD_dispatch_init_jpeg
>>    LINK : error LNK2020: unresolved token (0A0000B6) 
>> plD_dispatch_init_png
>>    LINK : error LNK2020: unresolved token (0A0000B7) 
>> plD_dispatch_init_ljii
>>    LINK : error LNK2020: unresolved token (0A0000B8) 
>> plD_dispatch_init_ljiip
>>    LINK : error LNK2020: unresolved token (0A0000B9) 
>> plD_dispatch_init_xfig
>>    LINK : error LNK2020: unresolved token (0A0000BA) 
>> plD_dispatch_init_psc
>>    LINK : error LNK2020: unresolved token (0A0000BB) 
>> plD_dispatch_init_psm
>>    LINK : error LNK2020: unresolved token (0A0000BC) 
>> plD_dispatch_init_plm
>>    LINK : error LNK2020: unresolved token (0A0000BD) 
>> plD_dispatch_init_win3
>>    LINK : fatal error LNK1120: 11 unresolved externals
>>
>>    Tplot - 12 error(s), 0 warning(s)
>>
>>     I did put the "plplotd.lib", "plplotd.def", "plplotd.exp" and 
>> "plplotd.dll" to the project directory, link
>> to "plplotd.lib", and also defined "__PLDLL_H__", is there anything 
>> I'm still missing?
>
>
> Have you looked at the makefile for the examples with the DLL in 
> sys\msdev\win32\dexamples?
> These compile and link allright, so I would suggest you mimick the 
> compile and link options
> from these examples.
>
> If this does not work, can you send me the source code and the 
> makefile/project file, so that I
> can try and reproduce this on my machine? (I maintain the win32 device 
> and the MSVC
> build system)
>
> Regards,
>
> Arjen

Re: [Plplot-general] problems with using Dll in VC++.net

[Arjen M. <arj...@wl...>]

Ruwei Liu wrote:

> Hi, all
>
>     I'm trying to use plplot library for a project. It works fine with 
> the static library , but it fails to
> link when I using the dynamic library. I got the following error message:
>
>         ------ Build started: Project: Tplot, Configuration: Debug 
> Win32 ------
>
>    Linking...
>    LINK : error LNK2020: unresolved token (0A0000B3) 
> plD_dispatch_init_null
>    LINK : error LNK2020: unresolved token (0A0000B4) 
> plD_dispatch_init_gif
>    LINK : error LNK2020: unresolved token (0A0000B5) 
> plD_dispatch_init_jpeg
>    LINK : error LNK2020: unresolved token (0A0000B6) 
> plD_dispatch_init_png
>    LINK : error LNK2020: unresolved token (0A0000B7) 
> plD_dispatch_init_ljii
>    LINK : error LNK2020: unresolved token (0A0000B8) 
> plD_dispatch_init_ljiip
>    LINK : error LNK2020: unresolved token (0A0000B9) 
> plD_dispatch_init_xfig
>    LINK : error LNK2020: unresolved token (0A0000BA) 
> plD_dispatch_init_psc
>    LINK : error LNK2020: unresolved token (0A0000BB) 
> plD_dispatch_init_psm
>    LINK : error LNK2020: unresolved token (0A0000BC) 
> plD_dispatch_init_plm
>    LINK : error LNK2020: unresolved token (0A0000BD) 
> plD_dispatch_init_win3
>    LINK : fatal error LNK1120: 11 unresolved externals
>
>    Tplot - 12 error(s), 0 warning(s)
>
>     I did put the "plplotd.lib", "plplotd.def", "plplotd.exp" and 
> "plplotd.dll" to the project directory, link
> to "plplotd.lib", and also defined "__PLDLL_H__", is there anything 
> I'm still missing?

Have you looked at the makefile for the examples with the DLL in 
sys\msdev\win32\dexamples?
These compile and link allright, so I would suggest you mimick the 
compile and link options
from these examples.

If this does not work, can you send me the source code and the 
makefile/project file, so that I
can try and reproduce this on my machine? (I maintain the win32 device 
and the MSVC
build system)

Regards,

Arjen