[Plplot-devel] Progress on the tarball generation with cpack
On 2006-10-29 10:57-0800 Alan W. Irwin wrote: I am starting work today on one release blocker (a script and accompanying CBS changes to create a release tarball). I have just finished the first part of this which is to create a custom target called prebuild_dist to pre-build everything required for a distribution tarball and copy it to the source tree (if the build tree is different from the source tree). The next step (tomorrow) is to collect almost everything in the source tree into a tarball using cpack. Comparisons between a tarball created this way and previous tarballs should tell me whether there is anything significant that is missing from the cpack-created tarball. The final step (hopefully also tomorrow) is to arrange with our CBS to use those generated files in the source tree if they exist (e.g., if they come as part of the distribution tarball). Note, I have deliberately excluded the SWIG-generated files from everything above. This means SWIG will be an extra prerequisite to build PLplot (if you want the Python and Java interfaces), but SWIG is now widely deployed in binary versions on Mac OS X (via fink) and Linux. Furthermore, binaries are available for windows from the SWIG download area. Thus, I don't think this extra prerequisite would be a significant burden to our users. If any developer here feels strongly that SWIG-generated files should be prebuilt for our future distributions then (a) it should be possible (although non-trivial) for them to implement pre-building those files, and (b) more difficult but still possible to implement a way for users to actually use those pre-built files in their CBS build. The reason why (b) is difficult is CMake uses SWIG in a rather obfuscated way deep in a macro so a lot of that logic will have to be figured out and copied for the special case when SWIG-generated files are pre-built. If nobody wants to bother with this, then that might be a good thing since it is certainly a cleaner/simpler solution to ask our users to generate those files with SWIG themselves. 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 __ - Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnkkid=120709bid=263057dat=121642 ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] CMake build system status/release plans
Hi Orion: I am going to post my reply to your question on list because it is a good question...Alan On 2006-10-30 16:33-0700 Orion Poplawski wrote: Alan W. Irwin wrote: I think we should be mentally prepared for a PLplot development release in the very near future featuring our CBS. Cool. I'll probably start working on testing building rpms for Fedora Extras. One question, what release version will this likely be? 5.6.2? 5.7? 5.7.0 since it is the first development release beyond the 5.6.x series of stable releases. (Odd minor numbers are development releases while even minor numbers are stable releases.) Although this next release is a development release, I hope it will work well for the majority of our users. I want the first impressions of our new CMake build system to be good. BTW, to help insure a high-quality development release we need testing of our CMake Build System (CBS) using the CVS version of PLplot on as many platforms as possible by those most familiar with PLplot (i.e., those lurking on this list.) Thus, I hope that all of you will give our CBS a try on your favorite platform following the instructions at http://www.miscdebris.net/plplot_wiki/ For example, Orion, I don't think anybody has tried it with Fedora yet although it works well on my two Linux platforms (Debian stable and Ubuntu Dapper). 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 __ - Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnkkid=120709bid=263057dat=121642 ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
[Plplot-devel] My progress on our CBS for today
Today I discovered and fixed a large number of dependency issues with the documentation build. The worst of these was a real dependency of the plplot-symbols.txt build on libplplot which was being completely ignored in doc/docbook/src/CMakeLists.txt. The final issue I discovered was causing pdfjadetex and jadetex to be unnecessarily run when I re-ran the make command. That was greatly slowing down my work on one of the issues blocking release, and I completely solved the issue by replacing the old method (which was based on dependencies) by a shell-based method closely related to that used in our ABS to solve the same issue of deciding how many times to run pdfjadetex and jadetex. Normally, shell-based commands should not be used for CMake because of potential cross-platform issues, but in this case we are only allowing building the documentation on Linux so we are free to use the power of the shell. I have done a number of elementary tests of the documentation build and all appears well now. With these side issue now resolved, I should be able to get back to working on the two remaining issues which are blocking release. Sorry, Hazen, on the large delay in creating the promised solution of one of those blockers, the Tcl/Tk build errors for Mac OS X. I should have something for you to test tomorrow (Tuesday). 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 __ - Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnkkid=120709bid=263057dat=121642 ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] CBS testing on Fedora
On 2006-10-31 20:41+0100 Per Persson wrote: (I have only 2.9.x installed) I have now fixed the regex problem that was screwing up detection of 2.9.x (where x is a single digit) so please try again. Hazen has reported that octave (version 2.1.xx) works for him on Mac OS X so it's likely the regex fix will be all you need. I suggest that we use it [octave-config] to get octave information for octave.cmake. The problem with octave-config is that the list of -p options changes rather drastically with octave version so queries to octave-config for build information must be version specific. That's a real PIA so I have given up on octave-config. Another possibility (see my previous post about this) is to use the mkoctfile command (essentially an independent platform-specific octave build system) within our build system. That's the approach we use with our ABS, but I never liked it particularly since it means there is a foreign build system within our build system (either CBS or ABS) so you have to deal with the union of the bugs for both build systems. So far (except for minor problems such as the version number) the current simplified linking approach I am using to build the octave interface to PLplot is working both on the Linux (Debian stable and Ubuntu Dapper with results still to come from Orion on Fedora) and Mac OS X platforms (at least for Hazen), but if we run into a showstopper problem with that approach as we extend testing to other platforms, I am mentally prepared to fall back to mkoctfile. 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 __ - Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnkkid=120709bid=263057dat=121642 ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] make check with CBS?
On 2006-10-31 15:41-0700 Orion Poplawski wrote: There no longer appears to be a make check with the CBS system. How would I do the equivalent now? Use the -DBUILD_TEST=ON option to cmake (or ccmake) to set up the subsequent ctest command. This option is OFF by default since building all the examples substantially increases the build time. Follow the cmake (or ccmake) command with the usual make, then invoke ctest to do a complete check of all the examples built in the build tree. 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 __ - Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnkkid=120709bid=263057dat=121642 ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] CBS testing on Fedora
Geoffrey, your assumption is incorrect. As I have made clear from the first (and also in a recent post to this list), autotools will be deprecated (since nobody has stepped forward to support it any more) but available (including configure and plplot_libtool) in the tarball to ease the transition to our CBS. That said, since you and Maurice appear to be PLplot developers who intend to continue to use autotools for a while, I am going to rely on you to test our development releases to make sure that no autotools-related problems inadvertently creep in. Also, there are disadvantages for Lightspeed (or anyone else) to adopt our CBS either too early or too late. Some factors that should influence that judgement call about the adoption timing should be at least the following: * There are obvious advantages to waiting while you let early adopters find all but the most obscure bugs. Note, PLplot users tend to be early adopters with lots of downloads of development releases so I think our planned series of development releases should be enough to get rid of all but the most obscure bugs in our CBS. * Our CBS requires CMake to be installed on user's systems. Thus, there are advantages to delaying adoption until CMake is at least widely deployed on Linux distributions. However, that should not be too long now; Debian testing already has cmake-2.4.3 (the version what we require). Also, if you check the archives of the CMake mailing list, it is clear a large number of open-source projects are converting to CMake as I write this. Thus, the pressure will be strong for all distributions to follow Debian's lead if they haven't done so already. * The transition from our ABS to our CBS is actually not that difficult. I would be extremely surprised if it took a moderately experienced PLplot developer (i.e., one who has used our downloads before) more than a day (assuming they follow the steps what we suggest in the PLplot wiki). Those with more experience with PLplot should take substantially less than a day. * There are substantial developer advantages to our CBS (a configuration system that is easier to understand, maintain, and extend than our ABS). * Bit-rot will eventually creep into our ABS unless some developer with a strong need for it steps forward to maintain it. Also, as PLplot keeps getting extended with CBS, none of this new development will be accessible with our ABS. * There are substantial advantages to our CBS for both users and developers (faster configuration as well as the availability of our CBS for the large variety of different windows platforms). 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 __ - Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnkkid=120709bid=263057dat=121642 ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] OS-X and Tcl/Tk
On 2006-11-07 08:24+0100 Werner Smekal wrote: Hi, Have a look at include/plplot.h. It has a whole series of defined's that it checks. If any of those are true it #define's __unix. I suspect you just have to add the appropriate check for Mac OS X to that list. I found this website ( http://predef.sourceforge.net/index.php ) which is very useful if you need to know predefined macros for your compiler. Maybe this is of any help. Thanks, Werner. That looks like an extremely useful site. For example, it gives (http://predef.sourceforge.net/preos.html#sec17) the Mac OS X identification which should help Hazen test the idea above. 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 __ - Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnkkid=120709bid=263057dat=121642 ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
[Plplot-devel] Our next development release (or it's over to you Hazen... :-))
On 2006-11-11 16:13-0800 Alan W. Irwin wrote:
The two remaining things I need to do are to make a script that does what I
did by hand to create the tarball and also make some CBS changes so that
prebuilt files in the tarball will be used in the build when they are
available.
I am now done with an initial version of a script called make_tarball.sh
(that's a reuse of the filename of an old script of mine to create the
distribution tarball which has not been used in a long time.)
This new script adapts make-cvs-tarball.sh to our CBS build system. In
addition, I have made some simplifications (for example, I have dropped all
the fooling around with version so all the version changes must be committed
to CVS before invoking this script). So run
./make-cvs-tarball.sh -h
first to see what is possible. I have worked hard on that
self-documentation so that is probably all you need. N.B. don't expect
everything to be identical to make-cvs-tarball.sh. (When in doubt, I have
dropped some of the finesse of make-cvs-tarball.sh to create something more
straightforward/easy to understand.)
I have tested this script with the following invocation:
[EMAIL PROTECTED] ./make_tarball.sh \
-c -i /home/software/plplot_cvs/installcmake -o -DENABLE_java=OFF \
make_tarball.sh_out
The options concern some tests you can do with the created tarball.
-c says unpack the created tarball and build it with cmake and make
and test that result with ctest.
-i /home/software/plplot_cvs/installcmake specifies the install prefix
for the tarball test configure and runs make install after cmake, make,
and ctest.
-o cmake options allows you to specify extra cmake options for the
tarball test configuration. In my case, I must disable java
because of java bugs in Debian stable.
I have also tested the generated tarball by hand using the deprecated ABS
./configure; make; make check commands, and all seemed to be well.
Orion, I suggest you generate a tarball with this script, and then check
whether that tarball is ready to create an rpm for your RedHat distro
using our CBS.
Release status:
One nagging issue still outstanding is the run-time trouble for Tk-related
applications that Hazen discovered on Mac OS X. It would be nice to get
this fixed, but I don't think it is release critical (so long as we mention
it in the release notes) since this issue has probably always been with us.
Also, I have not yet made the changes so that our CBS system can take
advantage of prebuilt files in the tarball, but again I don't believe this
is release critical (and I may get to it before the release in any case).
My conclusion is our source tree is now ready for a development release
featuring our new CBS.
Hazen, some changes obviously need to be made to the release process to
avoid using our ABS (and thus, to test that our CBS is up to the task of
making a release). So if you ever find yourself ready to invoke either
cf/bootstrap.sh (except inside the above script to prepare files that are
going into the tarball) or ./configure during the release process, don't!
Instead, use the CBS alternative. For example, in the above script you will
see the following:
cmake -DCMAKE_VERBOSE_MAKEFILE=ON \
-DPREBUILD_DIST=ON ${DOC_ARG} \
-DWWW_USER=${WWW_USER} ../plplot cmake.out
followed later by
make prebuild_dist make_prebuild_dist.out
DOC_ARG defaults to -DBUILD_DOC=ON by default and -DWWW_USER=${WWW_USER}
transforms to -DWWW_USER=hbabcock by default. The make prebuild_dist
command builds the documentation and copies it back to the source tree
before the subsequent make package_source creates the source distribution
tarball with the appropriate cpack command.
Afterward, to install the identical documentation that was built for the
tarball to the website, you merely have to do the following commands by
hand:
cd /tmp/plplot-dist-prep/build_dir/doc/docbook/src
make www-install
Hazen, I think the timing of the next release (5.7.0) is now in your hands.
Obviously, you will need some extra time beyond what you normally spend on a
release to figure out and document the changed CBS-only procedure for
release that I have alluded to above. In particular,
README.Release_Manager_Cookbook is now pretty dis-jointed with add-ons of
everybody's different comments, and this would be an excellent opportunity
to turn that into a straightforward cookbook of how you prepare the release
with our CBS.
Let us know when/what you plan to do. Also, contact me off list if you run
into any release generation issues for our CBS.
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
Re: [Plplot-devel] SVG driver
On 2006-11-12 18:45-0800 Alan W. Irwin wrote: [...]Also, you might want to look at your area-fill functionality (which is used a lot for example 8). Area fills are not particularly well suited for vector operations so some optimization care may be needed for your area fills for this vector device. This probably isn't the reason why the rendering of example 8 is so much slower than example 1. I took a quick look at the code and the difference between unfilled and filled is apparently only a matter of setting the fill color. Thus, I guess the renderer is expected to do the filling with its own algorithm, and the svg.c device driver simply specifies the polygons that need to be filled by the renderer application. So I presume svg.c has no control over the efficiency of the renderer area filling algorithm unless the number of points describing the polygons are much higher than they really need to be. 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 __ - Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnkkid=120709bid=263057dat=121642 ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Our next development release (or it's over to you Hazen... :-))
On 2006-11-13 22:18-0500 [EMAIL PROTECTED] wrote: This week: I will try PLplot with Tcl/Tk from fink to see if I have any better luck. In response to your subsequent post, I am glad you now have Tcl/Tk working properly in all respects on Mac OS X. Thanks very much for all your hard work that made this result possible. The command-line parsing issue you just discovered is probably something you will want to debug for Mac OS X after the release, but it is nice to know you have narrowed down the problems to such a small part of PLplot now. [...] So, my target is a test release of 5.7.0 on the 26th. Sounds good to me. 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 __ - Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnkkid=120709bid=263057dat=121642 ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] new wxwidgets driver
On 2006-11-21 21:58+0100 Werner Smekal wrote: Hi, I updated the wxwidgets driver (hopefully it makes the 5.7.0 release) a great deal. Most important I improved the antialized part of the driver. It is now about 4 times faster and not much slower than the non antialized code any more. The antialized code can now handle filled polygons and also resizing works now. You can test the antialized driver with ./x08c -dev wxwidgets -drvopt antialized=1,text=0 That antialiased result for all plot lines looks really good and renders reasonably rapidly (~1 sec per page) even on my 600MHz (Ubuntu Dapper) box. I also tested on my Debian stable box where I don't have libagg installed. This new version of the device worked (except for the antialiasing which was unavailable because of its dependendence on libagg) there as well. Thanks, Werner, for this excellent work! There is one minor glitch you might want to address; when you hit q to quit out of the wxwidgets GUI, the rest of the pages (up to 7 of them for this example) get rendered before the exit. It would be better if that exit occurred immediately. It looks like this new version of the wxwidgets device will go into 5.7.1, but OTOH, that is apparently going to be released this weekend. 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] SWIG on Windows
On 2006-11-22 22:48+0100 Werner Smekal wrote: Hi, I managed to make SWIG work with plplot on Windows. * I downloaded the latest SWIG for Windows package: http://prdownloads.sourceforge.net/swig/swigwin-1.3.31.zip and unzipped it somewhere. * the created folder needs to get into the PATH variable than and the lib folder in the CMAKE_INCLUDE_PATH variable, e.g. in my batch files I have: set SWIGDIR=C:\DevZone\swigwin-1.3.29 set PATH=%SWIGDIR%;%PATH% set CMAKE_INCLUDE_PATH=%SWIGDIR%\lib cmake is than able to find SWIG and it can be used. I was also successful to compile the python bindings. That's marvelous news. but they are still some problems running the examples (python 2.4 on windows uses numpy not Numeric and so on). I assume that is NumPy rather than numpy. There is some confusion over the various names. numpy came out around python 1.5 but was eventually renamed Numeric, then Old Numeric. Numarray came out later, but was eventually renamed Old Numarray. Finally, NumPy came out (version 1.0 just released) which is easy for old-timers to confuse with the original numpy, but it supersedes all the others I have mentioned. Once NumPy is adopted by Debian (AFAIK that hasn't occurred yet) and other leading-edge distros, I would like to change our Numeric code (which started life as numpy code) over to NumPy. Apparently, that transition is easy to do. 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] How to use the installed PLplot libraries
On 2006-11-27 22:29+0900 hiroyasu yasuda wrote: Hello Arjen: - Are you using the latest release (PLplot 5.7.1) or version 5.6.1? I was using PLplot 5.7.0. yesterday. - Are you using the ./configure script to build the libraries or are you using the CMake build system? I followed 'INSTALL' document below: 3) With g95: [Fortran 95 (GCC) 4.2.0 20060512 (experimental)] ./configure --disable-wxwidgets --enable-f95 FC=gfortran F77=gfortran make FFLAGS=-Wl,-single_module FCFLAGS=-Wl,-single_module make install make check I suggest you use PLplot 5.7.1 and follow the CMake build system instructions at http://www.miscdebris.net/plplot_wiki. The instructions in INSTALL are for our deprecated autotools (./configure) build system which is no longer maintained. And please tell me how to compile fortran source with PLplot libraries ? Make sure you have pkg-config installed on your computer before you run cmake the first time. Then after make and make install look at $INSTALL_PREFIX/share/plplot5.7.1/examples/f77 (or f95). The Makefiles in those two directories show exactly how to build fortran 77 or fortran 95 executables using the pkg-config approach. 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Issue with 5.7.1 and h2def.py on Fedora
On 2006-11-27 14:26-0700 Orion Poplawski wrote: /usr/share/pygtk/2.0/codegen/h2def.py on Fedora is not executable. The following patch fixes. If there is a variable for the python executable, presumably you want to use that instead. Thanks, Orion. I have committed the fix (slightly modified) to CVS. 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
[Plplot-devel] Status of broadened examples and cmake-2.4.5 pre-release news
On 2006-12-01 16:37- Andrew Ross wrote: I've started implementing this [broadening the examples]. It has shown problems in example 20 on the java platform. I've also started on the octave version, which again seems to throw up some problems. I have run the install tree tests right now, and the errors in your broadened examples are now reduced to just the octave ones: Failed tests: x20c.m, x24c.m, x25c.m, x26c.m Interestingly, ctest succeeded for octave so these octave errors are triggered by the install tree but not the build tree or else some error flag is not being communicated to ctest by these failures. In other news, I have found that cmake-2.4.5-RC-4 builds PLplot correctly (including the build-tree examples and the full documentation) on both my Debian stable and Ubuntu dapper platforms with the disambiguate workaround in examples/c++/CMakeLists.txt removed. Previous RC's failed as did 2.4.4 (unless I did the disambiguate workaround) so this RC-4 is a big step forward, and gives cmake-2.4.5 the PLplot seal of approval. That seal of approval remark is mostly serious; I appear to be one of a very small group of developers to report a bug in 2.4.4, but that target/filename clash and one other bug was sufficiently serious that the cmake developers decided to make a 2.4.5 release in a hurry just with a limited number of fixes including the fix for my bug. Fortunately, they made RC's for their forthcoming 2.4.5 release. It turned out there was a bug in their fix to the other serious bug which PLplot was kind enough to find for them (in their RC-2). I think PLplot is an excellent test of cmake because we use almost the entire language under a wide variety of circumstances. Anyhow, RC-4 works well now, and I expect 2.4.5 to come out within a day or so in excellent shape to build 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
[Plplot-devel] Some beautiful looking gnuplot results
Please have a look at Plotting the spirograph equations with 'gnuplot' (http://linuxgazette.net/133/luana.html). The article gives all the mathematical background so I am wondering if anyone is inspired to try their hand at generating these beautiful curves with PLplot for example 27? 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
[Plplot-devel] Recent (c)test system changes
Hi Werner: I have fixed the introduced chmod issue and dispensed with the unconfigured DIRNAME logic. It all works fine on Linux with our CBS. Please let me know if these small changes have introduced any problems for you on windows. I will deal with the SH_EXECUTABLE stuff for autotools later today. (I hate going back into that world :-( ) 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Recent (c)test system changes
On 2006-12-02 13:58-0800 Alan W. Irwin wrote: I will deal with the SH_EXECUTABLE stuff for autotools later today. Done and tested. 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] cmake-2.4.5 is now the minimum version and summary of outstanding issues for PLplot/CMake.
On 2006-12-06 22:08-0500 [EMAIL PROTECTED] wrote: [...] Most people who have Perl do not have PDL installed, and the binding in the relevant Perl module (PDL::Graphics::PLplot) are, as the name implies, dependent on PDL. This module is maintained by a member of the PDL project and is distributed by as a part of PDL. That's the key issue. Rafael was the guy maintaining it as part of PDL, and he is not doing that any more. So if we don't find a replacement developer willing to work on the Perl/PDL interface to PLplot it will probably fall into disrepair. You have also argued that external development of the Perl/PDL interface was not too bad (and I agree), but I know there are some advantages to doing it as part of PLplot as well (such as the important issue of windows platform availability). Anyhow, the decision about external versus internal is really up to whoever decides to volunteer for this. If they decide to take over Rafael's work that is fine (although it would be a shame to lose windows availability). If they decide to do it as part of PLplot, I wish to point out that a lot of the work (the SWIG API file, bindings/swig-support/plplotcapi.i) has already been done. Thus adding Perl/PDL (or any of the other languages covered by SWIG) should be straightforward. 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] cmake-2.4.5 is now the minimum version and summary of outstanding issues for PLplot/CMake.
On 2006-12-08 18:13+0100 Werner Smekal wrote: Hi Hazen, I believe that the current interface was a joint effort between Doug Hunt and Rafael and that Doug is still maintaining the Perl/PDL interface on behalf of the PDL project. There is an ActiveState PDL package, though it seems a bit old. Has anyone tried to use that in combination with PLplot and PDL::Graphics::PLplot on Windows? I installed ActiveState Perl and the the PDL package, but have no idea how to test how PDL/Perl and plplot come together. My efforts here were not because I need this configuration, I just thought it's as easy as the python and java bindings, so I gave it a try. I just committed a change so that ctest actually tested the Perl/PDL interface. To use this set ENABLE_pdl=ON and BUILD_TEST=ON. The result (for CVS HEAD) is that the examples in the examples/perl directory are run using the pre-installed Perl/PDL/PLplot interface. Here are the results on my Debian stable system: [EMAIL PROTECTED] ctest --verbose --tests-regex perl Start processing tests Test project /home/software/plplot_cvs/HEAD/build_dir Constructing a list of tests Done constructing a list of tests Changing directory into /home/software/plplot_cvs/HEAD/build_dir/test 7/ 7 Testing examples_perl Test command: /bin/bash -c EXAMPLES_DIR=/home/software/plplot_cvs/HEAD/build_dir/examples\ SRC_EXAMPLES_DIR=/home/software/plplot_cvs/HEAD/plplot_cmake/examples\ PLPLOT_LIB=/home/software/plplot_cvs/HEAD/plplot_cmake/data/\ ./plplot-test.sh\ --front-end=perl Testing front-end perl PLplot library version: 5.3.1 Can't open pnm fileconverter error: at /usr/lib/perl5/PDL/IO/Pnm.pm line 77 -- Process completed Passed The following tests passed: examples_perl 100% tests passed, 0 tests failed out of 1 I got rid of that fileconverter error by excluding example 20. Apparently, it is having trouble finding lena.pgm. Note the ancient library version. That's because libplplot is a dependency of the Perl/PDL PLplot interface so Debian stable installed its (ancient) system version of libplplot to satisfy that dependency. Werner, please give this a try again. I am most interested whether your new experiment finds the PLplot interface in ActiveState Perl/PDL. 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Some minor debugging probably all that is required to obtain a binary release of PLplot for windows
On 2006-12-17 09:37+0100 Werner Smekal wrote: Alan W. Irwin wrote: There is also some indication on the CMake list that a solution to the current limitation for CPack will be found. One way or another I hope this showstopper for binary releases of PLplot is resolved within the next week or so (bearing in mind that our forthcoming source release will be in mid-January). To prepare for that resolution, I suggest those with windows platforms should try make install (which exercises all the INSTALL commands which are used by make package). Yes, I try make install regularily and it works without problems, all seem to be there. I would like to thank Jim Dishaw who pointed out another make package problem off-list which I have verified. If you completely remove your install directory and run the usual cmake and make followed by make package the install directory is populated. This unexpected side effect is especially nasty for the case when the install directory is owned by the root account. It turns out there is an easy work around that solves both the absolute path issue and the install directory population issue. On Linux, this is how I can now make a binary release by hand * rm -rf /staging/area * make install DESTDIR=/staging/area * pushd /staging/area * tar zcf plplot-5.7.1-Linux.tar.gz . * popd * mv /staging/area/plplot-5.7.1-Linux.tar.gz . * rm -rf /staging/area (/staging/area is any absolute or relative directory location which you don't mind overwriting.) Werner, if a windows equivalent of the above procedure works, then I suggest you put it into a script for convenience (until make package is fixed), and that should conveniently take care of the fundamental issue of how to make a binary release on windows. The only remaining binary-release issue then appears to be integrating external libraries into the build so that make install (for now and make package eventually) installs them in DESTDIR. Werner, I have suggested a low-impact way to deal with this remaining issue. What do you think? 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Ada bindings
Jerry wrote (earlier): I'd like to know if there is any interest in including these bindings with the PLplot distribution. Hi Jerry: Yes, there is some interest assuming you are willing to donate your code under our preferred license which is the LGPL. We would need e-mail stating that from you before we could accept your donation. Furthermore, it's important for you to give us enough detailed information to build the interface and examples for ourselves (see below) since I believe our current Ada expertise is pretty low. On 2006-12-20 17:46-0700 Jerry wrote: On Dec 20, 2006, at 4:50 AM, Werner Smekal wrote: Could you provide a cookbook what is needed to make the bindings work (for persons who never used ada)? I can provide a help file or readme although compiling sources and linking libraries is just the usual stuff. Command line usage requires compiling your main Ada program, assuming the sources are in your search paths. Compiling Ada is quite easy because every Ada package must specify everything it needs, thus simplifying the build process. Compiling a multi-file program is done by simply directing the compiler to compile the main program. The Ada compiler that is included with gcc is especially easy to invoke, as I understand (not having used any others). After compiling, you tell the linker where the PLplot library is and you're done. Werner had an excellent question here. We could probably figure out how to build the Ada interface from the general description you give above and reading some of the Ada documentation, but to make it easy for all our developers who have never used Ada (such as me) could you please provide a step by step cookbook of the exact commands (command-line much preferred over GUI commands) we should use to build the interface for those with access to your platform (OS X)? Once we had those, it should be straightforward to integrate those instructions into our CMake build system so that the interface could be easily built on all platforms. And would it also possible for you to port the examples which you can find in the examples directory to ada examples (using the thin bindings)? I would like to port the examples but I'm a one-person operation so this would take extra time, especially since I still struggle to decipher C. (The FORTRAN examples are easier for me to read so maybe I would use them as a starting point.) I would use the thick bindings, however, because I'm afraid that many Ada programmers would look at the examples written using the thin bindings and keep looking elsewhere for a plotter. For example, the argument lists for the thick binding are frequently shorter because it is not necessary to pass array dimensions. Does the PLplot distribution policy absolutely require that the examples be included with the binding? Again, Werner brought up an important point. Showing that the standard set of examples in the language of choice produces the same postscript results as the C examples is an excellent test of that language interface. Thus, an interface is not really considered to be first class until all the standard examples are implemented with results compared with the C results for the standard examples. That said, if your initial donation included just a few of the standard examples, that would encourage us to do the integration work with our CMake build system with the understanding that you would be willing to fill in more standard examples as time permitted. Furthermore, the exposure your interface got with PLplot might inspire others to help you out with that task even if they didn't know Ada to start since it should be straightforward to infer what the source code should be for additional Ada examples once we see the required language syntax in the examples that you supply. I suggest you use the f95 examples as the template for your initial subset of the standard examples in ada. That interface (like most of our high-level interfaces) has all redundant dimension information dropped so that sounds most compatible with your thick ada interface 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 __ - 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
Re: [Plplot-devel] 3rd party libraries
On 2006-12-30 11:44+0100 Werner Smekal wrote:
Hi,
since it's especially for Windows more and more complicated to have all
3rd party libraries included in the plplot library I would propose the
following solution.
We make an additional 3rd party library package, which contains the
source files, header files and a custom CMakeLists.txt of all libraries
we want to include (agg, freetype, libharu, cd, qhull,...). The
structure would look like this:
plplot
lib
--agg2.4
CMakeLists.txt
src
include
--cd1.3
etc.
The maintainance of this package would not be much effort, since we
wouldn't need to always include the newest version of a library - it's
just libraries known to work with plplot. And some libraries (qhull,cd)
aren't updated anyway.
Than we would need to change the Findxxx.cmake files a little bit:
e.g. FindCD.cmake:
find_path(CD_INCLUDE_DIR cd.h /usr/local/include /usr/include)
if(CD_INCLUDE_DIR)
[...]
else(CD_INCLUDE_DIR)
# library not found, see if user installed 3rd party package
find_path(CML_DIR CMakeLists.txt ${CMAKE_SRC_DIR}/lib/cd1.3)
if(CML_DIR)
# yes?, so we compile it on our own
subdirectory(CML_DIR)
set(CD_INCLUDE_PATH ...)
[...]
endif(CML_DIR)
endif(CD_INCLUDE_DIR)
[...]
This script won't work exactly like that, but you get the idea. This
would have several advantages:
1) Very easy than for Windows users
2) if the system provides the libraries, we don't compile it on our own
3) if the user provides newer libraries, we don't compile it on our own
4) we have complete control of the library (shared, static, debug etc.)
5) Do get most features of plplot, you have only to untar the 3rd party
package and you're ready
6) For users who don't use the 3rd party package there is no change at all
disadvantage:
1) Maintain the package, but I think it's really low effort as soon as
it is setup.
2) gd and wxwidgets can't be included that way.
What do you think?
If you are willing to take on the responsibility of preparing and
maintaining a separate 3rd-party library source package release (presumably
a zip file) which you would separately release at our SourceForge file
release area and which the the windows user would unzip on top of their
PLplot source tree for an integrated 3rd-party and PLplot build, then that
would be fine with me.
However, I have just been struck with an alternative idea that might be even
better; make the 3rd-party build completely independent of PLplot and
provide not only a 3rd-party source release but also a binary release (which
keeps the 3rd-party libraries out of the PLplot binary release, although
they would have a common install location). The reason why I suggest this
alternative is I am a strong believer in the Unix philosophy that separate
is beautiful and over-integration can cause problems. For example, with the
original integrated build proposal, the user can really screw up if they
unzip the 3rd-party source package into the wrong location.
Note, the idea of a separate 3rd-party build and binary release is only a
possibility that I want to draw to your attention, and if you (and Arjen)
feel the integrated 3rd-party and PLplot build approach and binary release
for windows users will be easier to support, than I am willing to go along
with it since you will be the ones doing that support. :-)
By the way, when you prepare the 3rd-party source release, I assume you will
personally download the 3rd-party source code version that you want to
include in the release rather than attempting to maintain that 3rd-party
source code as part of our CVS. However, its fine with me if you maintain
the CMake infrastructure (CMakeList.txt files, etc.) for the 3rd-party
software as part of our CVS (like you do now in cmake/external for some of
the 3rd-party libraries).
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
__
-
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.phpp=sourceforgeCID=DEVDEV
___
Plplot-devel mailing list
Plplot-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/plplot
Re: [Plplot-devel] pdf driver
On 2006-12-30 23:20+0100 Werner Smekal wrote: Hi Alan, Werner, let me know if you need any more information to duplicate this pdf device build error on your Linux box with libharu-2.0.8. I think I fixed it, though I haven't been to Linux yet. I'll do that tomorrow. If you have time, you can give it a try again. Thanks, Werner. That was a good catch. The Linux build error is now gone, and the only apparent build issue remaining is the following warning: /home/software/plplot_cvs/HEAD/plplot_cmake/drivers/pdf.c:162: warning: passing arg 1 of HPDF_New' from incompatible pointer type. However, that did not seem to affect the results with this device. I installed and did a superficial install-tree test using c/x01c -dev pdf -o test.pdf When the resulting test.pdf file was viewed with the gv application, there appeared to be an overgenerous bounding box, but otherwise the rendering looked okay including the black background that is supposed to be there. A similar test with the xpdf pdf viewer showed the same bounding-box issue and a white (!) background. This might be an issue with how -dev pdf creates the background colour of the plot because if I use -dev psc and transform the result to pdf using ps2pdf, xpdf renders that transformed result with the correct black background. So it appears xpdf can render the background colour correctly, but it is sensitive to exactly how that background colour is created. Anyhow, it seems your new pdf device is a good start, and I suggest you will want to include this device as an option in plplot-test.sh so we can easily generate pdf's for all install-tree examples as a test of -dev pdf. 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] [Plplot-cvs] plplot/bindings/f77 CMakeLists.txt, 1.14, 1.15 sc3d.c, 1.15, 1.16 sccont.c, 1.13, 1.14 scstubs.c, 1.44, 1.45
On 2007-01-08 08:39+0100 Werner Smekal wrote: [...] For example, gcc-4.1.0 was released almost a year ago so I don't understand why the MinGW developers are not releasing an official version of gcc-4.x yet. Perhaps it will be coming soon? Either there are major problems or the MinGW project is dying. Which would be a pity. MinGW gcc 3.4.5 is actually not a bad compiler at all, quite fast and already has precompiled header support. They seem to try to get everything working (all languages) - since djgpp is already at version 4.1 with it's c/c++ compilers. I have done some searching on the mingw-users list, and my above remarks were probably too severe. One of the MinGW developers is actively working on gcc-4.x, but for some reason it is taking him rather long to get out even a development release. The MinGW developer may well be a perfectionist. Whatever the reason for the delay in the official MinGW release of gcc-4.x, the MinGW project is extraordinarily successful and there is no fear that they are dying. Their mailing list has huge traffic, and they are usally in the top 25 of the most actively downloaded projects on SourceForge. Everybody loves a free (in both senses) compiler! Anyway, I'll look into gfortran when time permits. Good. gfortran is good enough for our PLplot needs on Linux so that MinGW binary put together by the GNU gfortran project is certainly worth trying. According to http://gcc.gnu.org/wiki/GFortranBinariesWindows the installer seems to give you quite a bit of flexibility about how/where you install gfortran. 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] cmake itcl/itk
On 2007-01-09 06:27-0600 Maurice LeBrun wrote: A ghostly figure emerges from the mist.. Has anyone gotten the cmake build to auto-detect itcl itk? Yes, itcl and itk were detected for my two systems without problems. I didn't have to do anything special because I am using system versions that are installed in a standard location. For non-standard locations you have to help CMake find things by setting some environment variables, see below. I'm trying my first cmake build (ok feel free to rag on me for being such a laggard :), and those two were inexplicably not detected in my usual prefix area that contains all my tcl/tk/itcl/itk builds. Try setting CMAKE_INCLUDE_PATH and CMAKE_LIBRARY_PATH appropriately. http://www.miscdebris.net/plplot_wiki/index.php?title=Main_Page has one particular example of how you can set those environment variables whenever you are using software that is not installed in a standard location. 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] cmake itcl/itk
On 2007-01-11 03:10-0600 Maurice LeBrun wrote:
Alan W. Irwin writes:
On 2007-01-09 06:27-0600 Maurice LeBrun wrote:
I'm trying my
first cmake build (ok feel free to rag on me for being such a laggard :),
and
those two were inexplicably not detected in my usual prefix area that
contains all my tcl/tk/itcl/itk builds.
Try setting CMAKE_INCLUDE_PATH and CMAKE_LIBRARY_PATH appropriately.
That worked, thanks. But I'm surprised that:
$(CMAKE_INSTALL_PREFIX)/include
and
$(CMAKE_INSTALL_PREFIX)/lib
are not automatically at the front of CMAKE_INCLUDE_PATH CMAKE_LIBRARY_PATH
by default. Our build systems have had that property for an eternity. Would
it be hard to add?
I was unaware that our previous build system had that capability. I
certainly have no objection to adding that capability to our cmake build
system, but it will take some experiments to see whether it is possible to
do that from inside a cmake environment.
According to the documentation you can get with the cmake --help-full
command there is a -D option equivalent of CMAKE_INCLUDE_PATH and
CMAKE_LIBRARY_PATH that is used as a second choice if the environment
variable form is not set. The -D option form implies that
CMAKE_INCLUDE_PATH and CMAKE_LIBRARY_PATH are variables that can be re-set
from within cmake. But you should check that.
The way you do that is inside cmake/modules/plplot.cmake try
message(original CMAKE_INCLUDE_PATH = ${CMAKE_INCLUDE_PATH})
if(CMAKE_INCLUDE_PATH)
set(CMAKE_INCLUDE_PATH ${CMAKE_INSTALL_PREFIX}:${CMAKE_INCLUDE_PATH})
else(CMAKE_INCLUDE_PATH)
set(CMAKE_INCLUDE_PATH ${CMAKE_INSTALL_PREFIX})
endif(CMAKE_INCLUDE_PATH)
message(revised CMAKE_INCLUDE_PATH = ${CMAKE_INCLUDE_PATH})
and similarly for CMAKE_LIBRARY_PATH
Even if the revised CMAKE_INCLUDE_PATH and CMAKE_LIBRARY_PATH output by the
second message command are what you want, you should also check whether
resetting them this way has the desired effect on the various find commands.
If your experiments with the variable forms work out, then you should add
additional stanzas to do the equivalent for the environment variable forms.
For example, I notice in the documentation you can set environment variables
from within cmake using
SET(ENV{CMAKE_INCLUDE_PATH}
and you can read environment variables using, e.g., ENV CMAKE_INCLUDE_PATH.
If your experiments don't work, then we won't be able to supply this feature
in general, but the individual can always use a shell script to set
CMAKE_INSTALL_PREFIX, CMAKE_INCLUDE_PATH, and CMAKE_LIBRARY_PATH in the way
that they like.
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
__
-
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.phpp=sourceforgeCID=DEVDEV
___
Plplot-devel mailing list
Plplot-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Static libraries on Windows
On 2007-01-11 08:45+0100 Arjen Markus wrote: I checked the entire directory structure: there is only one file matching cygfreetype*.dll and that is cygfreetype-9.dll. Hi Arjen: One way cygfreetype-6.dll could still be interfering is if one of the system libraries required by PLplot is out of date so that it is linked to cygfreetype-6.dll rather than cygfreetype-9.dll. The freetype library is linked by a lot of different system libraries so the potential for such inconsistencies is high. I am not familiar with Cygwin but the Debian analogy I am using here is if some Debian user was attempting to use a mixture of Debian stable and unstable they would run into trouble because of library version inconsistencies. Anyhow, an upgrade to a self-consistent version of Cygwin may be the solution in your 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); 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
[Plplot-devel] CMake-2.4.6 has just been released
I tried CMake-2.4.6 on my Debian stable system, and it appears to work as well as 2.4.5. I suggest you try 2.4.6 for yourselves to make sure it is fine on all platforms. We don't want to get caught by surprise by some introduced platform incompatibility in 2.4.6 since that version is now the only advertised 2.4.x download available for new CMake users. However, I plan to leave the minimum version at 2.4.5 because I don't see any compelling new feature in 2.4.6 (such as one that will remove cmake bug workarounds for us). Just to review, those key workarounds are still the following: (1) make package still does not work so we still must use the make install DESTDIR=whatever command followed by the appropriate tar or zip command to make a binary package. (2) We still must continue to include Arjen's Windows-df.cmake and Windows-f90.cmake modules in cmake/modules/Platform since they have not been accepted (for unknown reasons) in 2.4.6. (3) We must replace use by use_ for all f77 comment lines in bindings and examples because CMake has a bug which parses those comment lines for use commands to determine module dependencies. This is cmake bug 3109. (4) We must create a file with arbitrary contents called plplot.mod.proxy in the top-level directory to satisfy a cmake issue which completely screws up fortran 95 module dependencies even when there are no use commands being improperly parsed from fixed-form comments. This is cmake bug 3984. I plan to follow up on (1) so that the trivial fix (using make install with DESTDIR internally in cmake) will be in the next version of cmake, and I hope Arjen does the same for issue (2). Arjen was actually involved in the final decision about how issue (3) (bug 3109) should be fixed, but something went awry with the one-line change (the parser went into an infinite loop according to the CMake developer) so he gave up. Arjen, would you be willing to re-ignite this issue with a test to see why the one-line fix did not work? It would be great to get this resolved since I would like to see those ugly forced changes in the comment lines removed. Andrew (Ross) tells me that issue (4) (bug 3984) is actually going to be quite complicated to fix so it may be a while before we can do away with that workaround. But getting rid of the first three workarounds in the next release of CMake would be a most worthwhile goal. 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Typo in bindings/f95/sfstubs.h and bindings/f77/sfstubs.h
On 2007-01-11 15:27-0700 Orion Poplawski wrote: Orion Poplawski wrote: Interestingly enough, GNU f95 aborts with an illegal instruction on this code, otherwise I'm not sure it would have been caught. Hmm, even with the changes, I still get an illegal instruction with the latest Fedora Development f95. Hi Orion: Can you please give a lot more details? Below I pose a series of questions that come to mind, but don't hold back any extra information you think might be relevant beyond these questions. If that illegal instruction occurs during compilation, what file is being compiled when it occurs? If that illegal instruction occurs during excecution of compiled code, what executable are we talking about and how was it compiled? Is it just one of the standard examples, or does the problem occur for all standard examples. If it was an example, was it built in the build tree or install tree? What is the result of the f95 --version command? On my Ubuntu dapper system I have the following results: [EMAIL PROTECTED] f95 --version GNU Fortran 95 (GCC 4.0.3 (Ubuntu 4.0.3-1ubuntu5)) Copyright (C) 2006 Free Software Foundation, Inc. GNU Fortran comes with NO WARRANTY, to the extent permitted by law. You may redistribute copies of GNU Fortran under the terms of the GNU General Public License. For more information about these matters, see the file named COPYING For that version of gfortran, there is no problem like what you are describing. Can you install gcc-4.0.x versions of the gcc compilers for Fedora to see if that makes a difference for you? 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
[Plplot-devel] PLplot dependencies
On 2007-01-11 17:03-0700 Orion Poplawski wrote: [...]plplot is hard because it touches nearly every compiler/language/library. Orion, have you just coined a new mantra for our development team? :-) Seriously, I couldn't agree more with your statement. A full-blown PLplot build is an excellent test that almost everything has been properly installed/is working correctly on your system. It does make life difficult if you are on the cutting edge, as with Fedora development. It also has made creating our CBS (and our ABS before that) an interesting challenge. Nevertheless, I fundamentally don't mind all the PLplot dependencies because most are optional, and each added PLplot option adds a lot of value. Thus, the user has a chance to build a minimal plplot that is really safe but rather uninteresting (say no bindings other than C and only the ps device driver) or he can add a lot of value to PLplot (in terms of language bindings and device drivers) assuming all the additional dependencies are installed correctly and working properly. Outside some cutting-edge situations, the dependency issue is actually not much of an issue on Linux, and I think the same is true on Mac OS X (in part due to fink). On windows, I am told dependencies are an important issue which is why I am delighted that Werner and Arjen are trying to help solve it with their 3rd-party and PLplot binary releases. 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] cvs to svn conversion
Hi Werner. Thanks for creating the conversion cookbook. On 2007-01-13 00:19+0100 Werner Smekal wrote: I ran through the source forge documentation about the cvs to svn conversions and I tested practically all steps. I made a copy of the whole repository, converted it, and uploaded it to my server, but stopped the import since it just took too long (it finished 256 revisions of 7586) - you can see this historic plplot revision here: http://svn.miscdebris.net/plplot/trunk/plplot/ . That's already an interesting result. Just have a look at what is (not) in bindings to get a feel for what PLplot was like in that early (4.99c) era. Is this long import process the step that takes so long in the official SourceForge process? They mention 24-hours or more to complete the task. Anyway, below you'll find the cookbock for this process. Rather straight forward. We also need to agree, when to do it and how. My proposal would be to do it after the next release. Nobody should commit than something for some days, until it is imported into the sf-svn repository and tested/compared. If the conversion was done satisfactorily we should than use svn exclusively. What do you think about that? Before we do the conversion to subversion at SourceForge, I think we should at least provide an opportunity for PLplot developers to get familiar with svn. Could you (or somebody else) make a complete test PLplot subversion repository available for us to play with? Alternatively, could you add instructions (probably about the exact import step that must be done) to your cookbook about how each of us could set up our own local test PLplot subversion repository? Some of the tests I would like to do for such a test repository are whether we could reproduce a number of our historical source releases and our complete historical ChangeLog (currently created by cvs2cl from our CVS repository) from a test subversion repository. I think such tests are essential because I have already had one bad experience with our current CVS repository; it turned out to be completely broken for branches until a lot of maintainence (re-deleting a whole bunch of files which kept rising from the dead) was done. So there still may be something non-standard in our CVS that will cause the conversion process to give bad results for some epochs. Reproduction of some of our historical tarball releases and our complete ChangeLog would help to satisfy me on this issue. In answer to your question, I think we should go ahead with the conversion to subversion, but only after the next stable release (i.e., 5.8.0). That gives the current shakedown process for our new CBS (i.e., the current on-going set of 5.7.x development releases) time to complete without interference from a repository format switch. If there is a well-known deadline for the conversion after the 5.8.0 release, that should give plenty of lead time (see below) for our developers to play with subversion and do all the tests that they think our necessary. Furthermore, if we do the conversion post-5.8.0, then that allows us to use the 5.9.x development cycle in part for working out any script issues (remember, many of our release scripts currently depend on cvs). By the way, I am not sure when 5.8.0 will be happening, but my best guess is probably something like 4-6 months from now. For the forthcoming 5.7.2 development release we are going to deploy something entirely new (binary releases for windows users), and presumably it will take us a while to respond to feedback from our windows users on those releases and any additional CBS and code issues they might find. So I wouldn't be surprised if we need at least a 5.7.3 development release (both source and binary) and probably a release candidate release for 5.8.0 before we finally go ahead with 5.8.0 final. 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Typo in bindings/f95/sfstubs.h and bindings/f77/sfstubs.h
On 2007-01-15 08:49+0100 Arjen Markus wrote: Well, to cut a long story short: we will use unique names - period. OK. I noticed you didn't make the commit so I did. I didn't test the change, but I think it should be okay. 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Problem with octave.cmake
On 2007-01-11 12:21-0700 Orion Poplawski wrote:
Getting the following cmake error:
-- PLPLOT_OCTAVE_DIR = /usr/share/plplot_octave
CMake Error: Error in cmake code at
/usr/src/redhat/BUILD/plplot-5.7.1/cmake/modules/octave.cmake:230:
Current CMake stack:
/usr/src/redhat/BUILD/plplot-5.7.1/CMakeLists.txt;/usr/share/cmake/Modules/CMakeCInformation.cmake;/usr/share/cmake/Modules/CMakeCXXInformation.cmake;/usr/src/redhat/BUILD/plplot-5.7.1/cmake/modules/plplot.cmake;/usr/src/redhat/BUILD/plplot-5.7.1/cmake/modules/fortran.cmake;/usr/src/redhat/BUILD/plplot-5.7.1/cmake/modules/octave.cmake
RegularExpression::compile(): Unmatched [].
RegularExpression::compile(): Error in compile.
-- OCTAVE_M_DIR =
-- OCTAVE_OCT_DIR = /usr/share/plplot_octave
I think this is the proper fix:
diff -u -r1.11 octave.cmake
--- plplot/cmake/modules/octave.cmake 6 Nov 2006 23:28:32 - 1.11
+++ plplot/cmake/modules/octave.cmake 11 Jan 2007 18:58:37 -
@@ -227,7 +227,7 @@
${NUMERICAL_OCTAVE_TESTING_MINIMUM_VERSION}
)
# Replace the OCTAVE_PREFIX with the PLplot prefix in OCTAVE_M_DIR
- string(REGEX REPLACE
+ string(REPLACE
${OCTAVE_PREFIX}
${CMAKE_INSTALL_PREFIX}
OCTAVE_M_DIR
Hi Orion:
I agree REGEX REPLACE is overkill in this case so I have made the above
change, tested it, and committed it.
However, although REGEX REPLACE is overkill, I do not understand why the
REGEX REPLACE failed in your case with a
STRING sub-command REGEX, mode REPLACE failed to compile regex /usr.
error message. It's always worked fine for me on both my Debian stable and
Ubuntu dapper platforms and nobody else has reported problems with it. An
ordinary string like /usr should work fine as a regex so I suspect some
deeper trouble exists on your system (perhaps with the compilation of cmake
itself using the cutting-edge/unstable compiler on your 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
__
-
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.phpp=sourceforgeCID=DEVDEV
___
Plplot-devel mailing list
Plplot-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/plplot-devel
[Plplot-devel] Minor vertical alignment issue for symbols with -dev psttfc
Hi Andrew (Ross): Will you have a look at the vertical alignment of the symbols in example 1 for -dev psttf? On my two systems (Debian stable and Ubuntu Dapper) and also on Werner's MinGW windows system (where he recently got -dev psttfc to work), the symbols used to plot the line in the first two panes are all plotted one symbol radius above the line rather than centred on it. Thanks in advance for your help solving this issue. 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] PLplot 5.7.2 is now available
Hi Hazen: I have downloaded plplot-5.7.2.tar.gz, checked its signature, built it, tested both the build tree and install tree examples. All seems well on both Debian stable and Ubuntu Dapper except for the same issue recently reported by Werner for MinGW; for both examples 7 and 23 on Ubuntu -dev psttfc gives the following error message: terminate called after throwing an instance of 'std::runtime_error' what(): Error returned from FT_Load_Glyph Aborted Debian standard does not have these errors nor does -dev png on Ubuntu (which eliminates missing fonts or something wrong with libfreetype [used by both devices] as the cause). Does anybody else see this issue? Hazen, there seems to be a lot of interest in these development releases. Your previous effort (5.7.1) resulted in 1000+ downloads in two months. Thanks for your continuing release efforts. 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Java compile problems
On 2007-01-17 11:51-0700 Orion Poplawski wrote: Orion Poplawski wrote: I'm getting the following trying to compile on Fedora Core 6 with the GNU java compiler: [ 15%] Generating plplot/core/plplotjavacJNI.class cd /builddir/build/BUILD/plplot-5.7.1/fedora/bindings/java /usr/bin/javac -classpath /builddir/build/BUILD/plplot-5.7.1/fedora/bindings/java /builddir/build/BUILD/plplot-5.7.1/fedora/bindings/java/plplotjavacJNI.java -d /builddir/build/BUILD/plplot-5.7.1/fedora/bindings/java -- 1. ERROR in /builddir/build/BUILD/plplot-5.7.1/fedora/bindings/java/plplotjavacJNI.java (at line 78) public final static native void PLGraphicsIn_type_set(long jarg1, PLGraphicsIn jarg1_, int jarg2); PLGraphicsIn cannot be resolved to a type [... lots more snipped ...] Any ideas? Maybe PLGraphicsIn.java needed to be compiled first? Looks like a circular dependency - plplotjavacJNI depends on PLGraphicsIN and vice versa. I'm no java person but it looks like you need an include file or forward class declaration in the source files. Hi Orion: I saw your recent post, but I am answering your earlier message since that includes the error message you got with Fedora Core 6. My apologies for taking so long to respond. Both plplotjavacJNI.java and PLGraphicsIN.java are files that are generated by swig. I just checked, and plplotjavacJNI.java generated on my Ubuntu Dapper system with swig-1.3.27 has no type PLGraphicsIn for any of the arguments. For example, in replacement for the above line it uses the following: public final static native void PLGraphicsIn_type_set(long jarg1, int jarg2); so there is no generated circular dependency with the Ubuntu Dapper version of swig. So my guess is you have run into a problem with the version of swig that Fedora has adopted. Just in case Fedora have made some bad patches to the latest swig (1.3.31) or are using a bad earlier version, could you try to replicate the problem with vanilla swig downloaded directly from the swig site? If you get the same error with latest vanilla swig, let me know, and I will try and reproduce it here, see if we have to change our swig configuration for that version, etc. (IIRC, once or possibly twice before, the swig 1.3.x series introduced an incompatible change that required changing our swig configuration, but this hasn't happened for a long time.) If you cannot reproduce the issue with the latest vanilla swig, then agitate with fedora to use that version rather than the one they have adopted. 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] PLplot 5.7.2 is now available
On 2007-01-22 22:05-0500 [EMAIL PROTECTED] wrote: On Jan 22, 2007, at 1:48 AM, Alan W. Irwin wrote: Thanks for your continuing release efforts. No problem. A few questions came up in the process of generating this release: (1) In our Cookbook at the make_tarball.sh stage we mention a /a/ install/directory created for testing purposes. This did not seem to get populated, so I unpacked the tarball to run the tests. Is this a problem with the script? Or is the Cookbook wrong? The cookbook instructions say to ./scripts/make_tarball.sh -c -i /a/install/directory 21 | tee build.log I am positive I tested this, and all was well. Also, I have looked at the script and cannot see anything that would cause the -i option not to work. (To get some documentation of the script, run ./scripts/make_tarball.sh -h ) However, screwups occur so please try the script again and send me the exact command you used, the resulting *.out files (in the script build directory, /tmp/plplot-dist-prep/build_dir), and most importantly the build.log file. (2) Does anyone know of an easy way to force a computer to time sync using ntp or equivalent? I've installed the debian ntp package but it doesn't seem to be doing anything. I'd like to add this to the Cookbook since it is often an issue with my computer. If you were running a Linux computer all the time, I highly recommend using ntp which is great for maintaining system time to the nearest 10 ms or so indefinitely (despite computer hardware clocks trying to get out of synch by typically 5-10 seconds per day). However, if you just want to boot into Debian set the time, do your release, then quit, then ntpdate (which sets the system time accurately once, but does not maintain it) is probably what you want to use. For ntpdate use man ntpdate. For ntp the instructions are a little more complicated. It has been more than two years since I installed ntp on my Debian stable system, and it has worked fine ever since. From what I can remember, you have to edit /etc/ntp.conf to provide it with some ntp servers to get it to work. I used server pool.ntp.org server pool.ntp.org server pool.ntp.org to access three separate ntp servers from the pool. After you adjust your /etc/ntp.conf file you must do the following two commands to stop and start the service so that the new versions of /etc/ntp.conf will be read. /etc/init.d/ntp-server stop /etc/init.d/ntp-server start If your system clock is badly out (hours or days), it will take ntp an equivalent time or much longer to adjust it. So you may want to use the ntpdate command to set your system time accurately before you run /etc/init.d/ntp-server stop/start combination to maintain the accuracy of that system time indefinitely. There is a command you can use to watch what ntp is doing as it adjusts the clock. I have forgotten what it is, but it is documented in the extensive ntp html documentation. (You can also watch your clock, but it's more fun to watch paint dry since it generally takes quite a while for ntp to slowly adjust the time to be accurate to the nearest second, and a bit longer to get its ultimate accuracy, 10 ms or so.) BTW, the reason ntp is so slow to make large time adjustments is it wants to be sure it doesn't disrupt anything time-dependent on the system (such as a Makefile in the middle of execution). ntpdate is time-disruptive, but it is meant to be run by hand when you are pretty sure nothing time-sensitive is being run on your system. (3) How long should we keep links to old versions of PLplot on our main webpage? Currently we have Release Notes back to 5.1.0. I think the latest stable release (5.6.1) and the latest development release (5.7.2) should be sufficient. They should be labelled on the web page as latest stable and latest development. 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Java compile problems
I have tried vanilla swig 1.3.31, and I confirm the problem. Orion, I had a look at the swig documentation you recommended, and I must say I don't understand enough Java to be able to follow your conclusion that the -nopgcpp command-line option might help us out. However, I do know how to try such command-line options (you later found this for yourself), and the result was with -nopgcpp, java built and ctested fine with vanilla swig 1.3.31. However, this option does cause trouble for my system version of swig (1.3.27). Therefore, I have committed some substantial changes to do some swig version checking. In particular, I have temporarily (until CMake adopts it) added an experimental FindCMake.cmake module to PLplot which is just more reliable than the current CMake version of this module, and which also provides increased functionality such as returning SWIG_VERSION. (A side effect of this new module is the vanilla version of swig-1.3.31 can easily be tried out. All you have to do is put the new version on your PATH.) The result of the changes is swig versions older than 1.3.22 are not accepted (set(SWIG_FOUND OFF)), and the -nopgcpp command-line option is only used for swig versions = 1.3.30 (I chose this limit according to one of the tests that you did). Could you please try the CVS version of PLplot now? Andrew (Ross), could you please review my solution (in particular, can you explain the necessity of -nopgcpp for newer versions of swig?) Thanks, Orion, for all your investigations of this issue. 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] PLplot build fails in FindSWIG.cmake
On 2007-01-30 09:50- Andrew Ross wrote: Aaargh! I've discovered the source of the error. Somewhere between 1.3.24 (default version on Debian stable) and 1.3.28 (version on Ubuntu edgy) the behaviour of swig changed so the output of swig -version was directed to standard output rather than standard error. This explains the blank output for SWIG_version_output on Debian stable. 1.3.27 on Ubuntu dapper is still directed to stdout (which is why I didn't see this error). The change from stdout to stderr that you discovered is ugly indeed. Modification commited to CVS to check ERROR_VARIABLE if OUTPUT_VARIABLE is blank. Now works again for me. Thanks, Andrew, for sorting this out. Note, I imported this version from the CMake bug system. From cvs log FindSWIG.cmake : revision 1.1 date: 2007/01/23 21:21:25; author: airwin; state: Exp; MUCH better FindSWIG.cmake module downloaded from http://public.kitware.com/Bug/bug.php?op=showbugid=4145 (click on FindSWIG.cmake-4) I just checked, and FindSWIG.cmake-4 continues to be the last version there. Were you going to follow up with your FindSWIG.cmake-5 to that location (plus an explanation of the horrible stderr change) or would you like me to do it? 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] PLplot build fails in FindSWIG.cmake
On 2007-01-30 17:49- Andrew Ross wrote: Alan - change is the other way round. stderr on old version, stdout on new version. Stdout seems much more sensible. Sorry for my initial confusion. I confirm your Debian stable results on my system. I don't understand how this logic worked at all before I made my change to TRANSFORM_VERSION to make it (I thought) more robust. There are some more aspects of this I am pursuing such as improperly nested if statements in both the old and your modified versions FindSWIG.cmake, and a better response from TRANSFORM_VERSION when it gets an empty version string. Anyhow, give me some time to sort this out before we upload the new FindSWIG.cmake to the CMake bug 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] PLplot build fails in FindSWIG.cmake
On 2007-01-30 11:15-0800 Alan W. Irwin wrote: There are some more aspects of this I am pursuing such as improperly nested if statements in both the old and your modified versions FindSWIG.cmake, and a better response from TRANSFORM_VERSION when it gets an empty version string. OK, all sorted (including a simpler fix for the stderr versus stdout question that works on both my Debian stable and Ubuntu Dapper platforms). Andrew (Ross), will you please check the latest? If you are happy with the latest FindSWIG.cmake, then I will follow up with the CMake bug 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] wxWidgets bindings and other changes
On 2007-02-05 00:42+0100 Werner Smekal wrote: I also have a small demo application - Alan, where should this file go (together with a cmake file), or should I keep this separate (e.g. in the wiki)? If it is analogous to the gnome canvas case, then I believe it should go in the appropriate examples directory. For example, in the gnome canvas case there are demos in examples/c and examples/python (See c/README.plplotcanvas and python/README.plplotcanvas). Since those demos are interactive, they are not used as part of ctest, and the compiled language ones are not built in the build tree (no point since ctest cannot use them, and it is probably too much trouble to set up on the environment stuff needed for the build tree to actually run them there). However, the demos can straightforwardly be run from the install tree. For example, the compiled language ones are built in the install tree (see examples/c/Makefile.examples.in). Thanks, Werner, for this wxwidgets interface to PLplot. I am looking forward to playing with your demos, and I am sure the wxwidgets world will be most interested in what you have done to make PLplot accessible from wxwidgets. 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 __ - Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier. Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnkkid=120709bid=263057dat=121642 ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] FindPLplot.cmake
On 2007-02-05 09:59+0100 Werner Smekal wrote: Hi, we are using cmake for building plplot, but what about if you want to use plplot for your own programs using cmake? Sure, you use FindPLplot.cmake which has yet to be written. I attach a first version which should work for all platforms, but needs to be changed to use pkg-config for Linux and OSX. Also I didn't include all libraries (plplot_canvas, etc). I think we should somehow put it in cvs and make it available to the cmake team (hopefully they include it). Where should I put it? cmake/external wouldn't be the right place and cmake/modules would also be wrong. Who could help me (Alan :) to make the script perfect? I think the obvious place is the cmake directory itself so I have committed it there. Although cmake/modules and cmake/external are reserved for modules that are used in the PLplot build, we don't need to enforce that limitation for the cmake directly itself. I suggest you use the pkg_check_modules macro when you add the pkg-config capabilities to FindPLplot.cmake. My experience with libLASi shows the pkg_check_modules macro (in FindPkgConfig.cmake which is currently in CMake cvs) is far superior to the deprecated (in CMake cvs at least) pkgconfig macro which is why I intend to use pkg_check_modules for all new work from now on. I also intend to convert PLplot to use pkg_check_modules so that is why I have recently committed (as a temporary measure until FindPkgConfig.cmake is released as part of a normal CMake release) FindPkgConfig.cmake into cmake/modules. The FindPLplot.cmake users will also have to take a temporary measure like that, but it will be worth it. Thanks very much for thinking of making a FindPLplot.cmake module. It should prove most helpful to other projects with CMake build systems. 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 __ - Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier. Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnkkid=120709bid=263057dat=121642 ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Errors with wxwidgets bindings
On 2007-02-06 12:27-0800 Alan W. Irwin wrote:
However, when I tried the same build on Debian Dapper (one earlier version
than Andrew's Debian Edgy) I got similar errors to Andrew's. Nevertheless,
the good Debian stable result shows you are close to having everything
working on all Linux platforms.
Since your last commits, all seems to be well on Debian Dapper now with
the build and install. However, I could not get wxPLplotDemo.cpp to
build in the installed examples tree because the compilation flags have
not been configured properly in plplot(d)-wxwidgets.pc. I also suspect
there may be a unicode problem for that example as well because of strange
characters in the compiler error output such as
/usr/include/wx-2.6/wx/event.h: In copy constructor â:
/usr/include/wx-2.6/wx/event.h:922: error: class â does not have any field
named â
BTW, to attempt to build the example in the install tree I tried the
following command (wrapped for e-mail) following the pattern I got from the
first example, x01:
/usr/bin/c++ wxPLplotDemo.cpp -o wxPLplotDemo -Wl,-rpath
-Wl,/home/software/plplot_cvs/installcmake/lib:/home/software/autotools/install/lib
`PKG_CONFIG_PATH=/home/software/plplot_cvs/installcmake/lib/pkgconfig
pkg-config --cflags --libs plplotd-wxwidgets`
On your Linux box, the rpath and PKG_CONFIG_PATH values will be different, but
you get the idea.
To sum up the current problem, the configured plplotd-wxwidgets.pc file
has
Cflags: -I${includedir}
and, of course, you need a lot more information there in order to have a
successful, wxwidgets-related compilation.
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
__
-
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier.
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnkkid=120709bid=263057dat=121642
___
Plplot-devel mailing list
Plplot-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Docbook/Documentation Update
On 2007-02-07 21:34-0500 [EMAIL PROTECTED] wrote: Some questions / issues with our current docbook manual: (1a) Chapter 23 (i.e. how to install). Should we keep this and update it? Refer users to the INSTALL file instead? To the wiki? (1b) Chapter 4, how to deploy. See questions for 1a. (2) We seem to be missing a chapter on Java. Any Java experts have any interest in providing at least a quick overview? (3) The Python chapter is, well, terse probably dated. Any Python experts willing to bring this up to date, including (a) a brief overview of the bindings and (b) a simple example? (4) The C language chapter claims to be old in need of updating. Any ideas about what specifically might be wrong? Do we no longer recommend FLOAT, INT and PROTO? Do we no longer support non-standard C compilers? Notes are needed about PLBOOL PLUNICODE? (5) Chapter 10, The plstream C++ interface. This is stable now and hasn't changed for a long time? Is there an example to reference that demonstrates how to use it? (6) and the chapter on fortran 95 is essentially copied (with many errors in detail as a result) from the fortran 77 chapter simply as a placeholder for now. With regard to your 1a and 1b questions, I think our docbook documentation should stand completely on its own as the definitive source of our documentation. So once our wiki settles down (and I think we have reached that point or are pretty close) I think we should copy a lot of that material to our docbook documentation. With regard to your further questions, note that for python, perl(?), java, fortran 95, and C++ we actually deploy an alternative interface where the redundant dimensions are dropped (for backwards compatibility the C++ interface also has an interface similar to the old C-like interface as well). Perhaps the best procedure here is to add a chapter closely following our existing API chapter, but with the details changed for all the dropped redundant dimensions. Then you could add short documentation for each of the above languages that simply refers to the alternative api chapter for the details. Unfortunately, I will not be able to work further on that alternative API documentation idea until my responsibility for the svn testing and conversion is done, but hopefully somebody else will have the time to move forward with that idea in the meanwhile. 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 __ - Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier. Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnkkid=120709bid=263057dat=121642 ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
[Plplot-devel] Orion's latest non-Tcl/Tk patches
Hi Orion: New subject since this subset of your patches have nothing to do with Tcl/Tk The discussion of your remaining (i.e., Tcl/Tk) patches should continue in the original thread. On 2007-02-12 17:28-0700 Orion Poplawski wrote: [...] plus a couple extra patches. plplot-5.7.2-sitearch.patch: The python stuff should be installed in an arch specific directory. Done. Thanks for spotting this. plplot-5.7.2-multilib.patch: Fixes up some multilib issues in Fedora. I don't think either variable is needed with a properly installed pkg-config. No action (as yet). Both PKG_CONFIG_ENV and RPATHCMD are essential for non-standard install locations for PLplot (which we have supported forever, and which I use all the time). They should not interfere with anything you do on Fedora. If they do interfere, let me know details, and we can work something out to deal with both the usual non-standard install location needs and your Fedora needs. Note, if you don't want to see any rpath imposed at all for the installed example builds, then use -DUSE_RPATH=OFF. 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 __ - Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier. Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnkkid=120709bid=263057dat=121642 ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Tcl 8.5 (and more)
On 2007-02-12 17:28-0700 Orion Poplawski wrote: Maurice LeBrun wrote: Orion Poplawski writes: Could the Tcl interface folks take a look at trying to compile plplot against Tcl 8.5? It's been put into Fedore Development and plplot compile fails with: .. I may be able to look into it over the weekend. If someone beats me to it, I won't complain tho. :) Here's what I've got so far [...] plplot-5.7.2-itcl.patch: itcl/itk are in /usr/lib/itclversion/ and /usr/lib/itkversion in Fedora. Applied. I just learned something new about CMake. If you specify PATH_SUFFIXES it also searches the non-suffixed directory (which is necessary for my two platforms). I just tested this on my Debian stable platform to make sure. So Maurice, that just leaves Orion's plplot-5.7.2-tcl85.patch for you to evaluate. 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 __ - Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier. Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnkkid=120709bid=263057dat=121642 ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Python API
On 2007-02-18 22:36-0500 [EMAIL PROTECTED] wrote: Hello, In the process of working my way through updating the documentation I've noticed that there are a reasonable number of functions that I cannot find in the Python API. Is this because they are really missing? Or am I not looking in the right place? My approach has been to grep for the function name in the bindings/python directory. Example (missing) functions include plfill3, plflush, plgcol0 and plgcolbg. Both Python and Java API are implemented in the file bindings/swig-support/plplotcapi.i. Those API's are probably our third-most complete ones after C and C++. 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Python API
On 2007-02-19 17:44-0500 [EMAIL PROTECTED] wrote: On Feb 18, 2007, at 11:54 PM, Alan W. Irwin wrote: On 2007-02-18 22:36-0500 [EMAIL PROTECTED] wrote: Hello, In the process of working my way through updating the documentation I've noticed that there are a reasonable number of functions that I cannot find in the Python API. Is this because they are really missing? Or am I not looking in the right place? My approach has been to grep for the function name in the bindings/python directory. Example (missing) functions include plfill3, plflush, plgcol0 and plgcolbg. Both Python and Java API are implemented in the file bindings/swig-support/plplotcapi.i. Those API's are probably our third-most complete ones after C and C++. Thanks! Now how do I figure out the redacted form for Python? Using the function plbin as an example, for java I can go: grep plbin bindings/java/* and I get: java/PLStream.java:plplotjavac.plbin(x, y, center); From which I infer that the first argument, nbin, has been dropped. If I look in bindings/swig-support/plplotcapi.i to try and see what Python will do I see: %feature(autodoc, Plot a histogram using x to store data values and y to store frequencies.) plbin; void plbin(PLINT n, PLFLT *Array, PLFLT *ArrayCk, PLINT center); Which suggests that Python has kept the first argument. Actually not. Both the python and java interfaces are generated using the information in bindings/swig-support/plplotcapi.i, but bindings/python/plplotcmodule.i and bindings/java/plplotjavac.i do additional platform-dependent processing. In both those latter files look for with preceding count. The lines just after that comment show SWIG how to transform the pattern PLINT n, PLFLT *Array, into one array argument for both python and java. Through such tricks, the python and java redundant array dimension information is all removed. For your documentation efforts, I suggest you assume that all dimension information dropped by swig for java is also dropped for python. It is a bug in how the argument patterns are handled in bindings/python/plplotcmodule.i and bindings/java/plplotjavac.i if that turns out not to be the case. N.B. where the python interface actually differs from java is the way information is returned for output arguments. In python the output arguments are completely removed from the argument list, and instead the output arguments are returned as the function value (which is returned as a list of output arguments). For the java case, the output arguments stay in the argument list. For example, look for how plg* functions are used in examples/java versus examples/python and assume that same treatment is used for both languages for all output variables. (Again it is an interface bug if that proves not to be the 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); 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Ada binding attached
On 2007-02-23 22:37-0700 Jerry wrote: Please consider the attached tarball my contribution of an Ada binding to the PLplot project. This code and the other files in the attachment may be licensed under the LGPL. The binding is a very very nearly complete implementation of the public API plus a partial implementation of the C-only API. It includes Example 1 as a demo and an extensive README in HTML form. The only part of the public API which needs attention deals with scaling data for things like contour plots. I'm not quite sure what is customary in stating the legal notices and documentation and copyright notices within the code, so perhaps someone can provide a pointer or two. Please don't be offended if names don't appear correctly in the present offering as it is not my intention to overlook them. Comments are welcomed. Thanks, Jerry, for this donation of an Ada interface to PLplot and the examples. I need some help from you to get this integrated into PLplot. I have read PLplot.sh, but I am a little lost since you only give how to build the example using ada_plplot.adb as the only Ada source. I believe this automatically builds the first example and thick and thin interface libraries as needed, but I need to be able to do that separately and explicitly, and I also need to know how to install the thick and thin interface libraries. So the basic questions I would like you to answer are how do you do the following with the gnatmake command: * Build the thick and thin interface libraries in the build tree. * Install those libraries in the install tree (which should involve relinking to the core PLplot library which will also be installed in the install tree, but this step should not involve recompilation). * Build an ada example which links to those libraries in (a) the build tree and (b) in the install tree. N.B. just give the cookbook that works for you on Mac OS X for the 6 steps (2 steps per asterisk) above (please be exact because I am unfamiliar with the gnatmake command), and I should be able to infer the equivalent on Linux with its different libplplot and libgcc_s names. Once I can get the Linux build and install to work for the interface libraries and example, then I should be able to automate everything (both for Linux and Mac OS X) with our CMake build system. BTW, to integrate the Ada interface and examples into PLplot, I plan to put the thick and thin interface library source files; plplot.adb, plplot.ads, plplotthin.adb, and plplotthin.ads; into bindings/ada and the example source files; ada_plplot.adb, x01.adb x01.ads into examples/ada. 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
[Plplot-devel] Vastly improved configuration of Ada interface and examples for PLplot. Please test!
I am pretty satisfied with the CMake build system improvements I have just checked in to substantially refine how we deal with the Ada interface and examples. Here is what I have done. * I have implemented a limited Ada language support for CMake. As a result of this effort, add_library should now work cross-platform, but add_executable is not available since it needs a change to the core of CMake to do the required gnatbind and gnatlink. * I have substantially refined the configuration and taken advantage of the new Ada language support so that the build of libplplotada and the build of the example (still done with a custom rule rather than add_executable because of above limitation in the language support ) should just work cross-platform. Caveats: * GNAT compiler package has to be installed (e.g., gnat-4.0 package on Ubuntu Dapper). GNAT is an add-on package for gcc so you must ensure that the version of GNAT that you install is consistent with the gcc version that you have installed. * ENABLE_ada is OFF by default (and we should continue with that default until this interface gets a lot more testing) so must use -DENABLE_ada=ON cmake option to build the libplplotada library. * On really old systems (e.g., Debian stable but not Ubuntu Dapper) it is difficult to find libgnat so you may have to set, e.g., export CMAKE_LIBRARY_PATH=/usr/lib/gcc-lib/i486-linux/3.3.5/adalib to give CMake a strong hint where to find libgnat. * Must use -DBUILD_TEST=ON cmake option so that the make command in the build tree builds the ada example. * Must set, e.g., the following environment variables to run the example that was built in the build tree (these have to be set for any example in the build tree, but normally you don't have to worry about this because ctest takes care of it for you): export EXAMPLES_DIR=/home/software/plplot_cvs/HEAD/build_dir/examples export SRC_EXAMPLES_DIR=/home/software/plplot_cvs/HEAD/build_dir/../plplot_cmake/examples export PLPLOT_LIB=/home/software/plplot_cvs/HEAD/build_dir/../plplot_cmake/data Once those caveats are satisfied, then from the top-level of the build tree ldd -r bindings/ada/libplplotadad.so and ldd -r examples/ada/ada_plplot give good results after the cmake and make commands are executed showing all libraries have been properly found and all symbols resolved. Furthermore, examples/ada/ada_plplot -dev whatever gives good results (so far on Debian stable and Ubuntu Dapper). ToDo: (1) Lots of testing on a variety of platforms. Jerry (and other interested developers), please give it a try, and let me know whether everything works well on your platform of choice. (2) Set up ctest so that the ada example will be executed as a result of the ctest command. This refinement is not essential for testing so long as you remember to set the EXAMPLES_DIR, SRC_EXAMPLES_DIR, and PLPLOT_LIB environment variables to the proper locations for everything. (3) Fill in the rest of the standard examples implemented in Ada. (4) Once a substantial number of standard examples have been implemented and this Ada binding has substantially matured, configure the install. The big question there is where to install libplplotada and its accompanying Ada source files and *.ali files. (N.B. Ada libraries usually have their own directory filled with the library and the source files [at least the spec files but possibly body files as well] and *.ali files.) I will take responsibility for (2) fairly soon, and (4) in the fullness of time. I may chip in with some help on (3) as well once the pattern of how you program in Ada has been established by the implementation of a few more standard examples. 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Vastly improved configuration of Ada interface and examples for PLplot. Please test!
On 2007-02-27 16:25-0800 Alan W. Irwin wrote: ToDo: (2) Set up ctest so that the ada example will be executed as a result of the ctest command. This refinement is not essential for testing so long as you remember to set the EXAMPLES_DIR, SRC_EXAMPLES_DIR, and PLPLOT_LIB environment variables to the proper locations for everything. Done. (3) Fill in the rest of the standard examples implemented in Ada. I added the standard example 10 just to see if I was now an expert Ada programmer. The answer is probably not... :-) N.B. I have renamed the previous ada_plplot.adb to x01a.adb (the main routine that withs in x01.ad?), and the new standard example is contained in the files x10a.adb, x10.adb, and x10.ads. Jerry, to do build-tree tests of the Ada interface (after make is executed) for the two examples in CVS now all you have to do is type ctest --tests-regex ada The resulting test/x01a.ps was identical to test/x01c.ps (generated with a prior run) verifying at least the thick Ada interface API exercised in this test is fine. However, x10c.ps was not identical to x10c.ps. In the Ada result the plot boxes are made with tick marks that should not be there according to the xopt and yopt command strings bc. Jerry, could you please look at the new example 10? I wrote it almost entirely in the thin interface. I notice that the C version of plbox is called by Box_Around_Viewport in the thick interface in a different way then the thin interface plbox Ada routine calls the C version so there may be a systematic thin API problem that is revealed by this example. Also, I think it is important to emphasize the parallel nature of the standard examples for every language so I strongly prefer that we provide thin interface (e.g., plbox rather than Box_Around_Viewport) standard examples. That, of course, does not preclude additional examples written in the thick interface with the more descriptive names for the API. Also, could you provide plparseopts in the thin interface? Virtually every language interface has plparseopts defined (normally with just one argument). That would have allowed me to write x10.adb completely in the thin interface. 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Vastly improved configuration of Ada interface and examples for PLplot. Please test!
On 2007-02-28 19:34-0700 Jerry wrote: So--I recommend just renaming all the example files to x??a.adb and letting them run as standalones, and eliminating the x??.adb and x??.ads naming level completely. That was my intention originally but I see that I forgot to adopt that naming convention. I can take care of this. Hi Jerry: I have just tried this, but it does not work for me. gnatmake compiles x01.adb without problems, but it did not bind or link it. (I tried this experiment for the original source code you sent me before the reorganization that occurred in the CVS version just to be sure the problem was not an artifact of that reorganization) [EMAIL PROTECTED] gnatmake x01.adb gcc-3.3 -c x01.adb gcc-3.3 -c plplot.adb gcc-3.3 -c plplotthin.adb If I attempt to force it with the -b option, gnatmake generates the following error message: [EMAIL PROTECTED] gnatmake -b x01.adb gnatbind -x x01.ali error: x01.ali does not contain a unit that can be a main program gnatmake: *** bind failed. You should be able to easily verify this problem for yourself. If there is some simple change that can be made to x01.adb so that it is suitable for a main program please send it to me. I too would prefer to eliminate the separate driver programme for each example if that is possible. Two files per example is preferred to three files per example. 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Vastly improved configuration of Ada interface and examples for PLplot. Please test!
On 2007-03-01 09:40-0800 Alan W. Irwin wrote: On 2007-02-28 19:34-0700 Jerry wrote: So--I recommend just renaming all the example files to x??a.adb and letting them run as standalones, and eliminating the x??.adb and x??.ads naming level completely. That was my intention originally but I see that I forgot to adopt that naming convention. I can take care of this. Hi Jerry: I have just tried this, but it does not work for me. gnatmake compiles x01.adb without problems, but it did not bind or link it. [...]Two files per example is preferred to three files per example. Hi Jerry: Never mind, I figured it out. (I had to remove the package body statements from the standalone versions). A side benefit is a spec file is not needed. Thus, we are now reduced to one standalone source file (x??a.adb) per example. I just committed all my changes to CVS. ctest --tests-regex ada works without run-time errors and creates postscript results (x01a.ps and x10a.ps) in the test subdirectory. x01a.ps is identical to x01c.ps (the equivalent C result) and similarly for x10a.ps and x10c.ps. However, that was because I replaced -plbox(bc, 0.0, 0, bc, 0.0, 0); +Box_Around_Viewport(TUB(bc), 0.0, 0, TUB(bc), 0.0, 0); If I use plbox rather than Box_Around_Viewport, then the result has extraneous tick marks. I don't understand the cause of this problem. The same example has a call to plptex with another character array, and that works without problems. I have also carefully compared how plptex and plbox are set up in plplotthin.ads, and I cannot spot anything. I also tried the TUB(bc) in the argument list for plbox, but the compiler complained about something being inconsistent before erroring out so I dropped it. This marks the end of the initial intense phase of my getting Ada supported under CMake for PLplot. We have made a good start, and the way forward is clear. * All developers should check the Ada aspects of the CMake build system on their platform of choice. Simply install the gnat extension to gcc (the gnat-4.0 package does this on Ubuntu Dapper), use the -DENABLE_ada and -DBUILD_TEST=ON cmake options, and run ctest --tests-regex ada after the make command. * Jerry needs to sort out the existing issues (plparseopts and plbox) with the thin API before we move forward with filling in the rest of the standard examples in the thin API (which may well turn up other API issues). 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Vastly improved configuration of Ada interface and examples for PLplot. Please test!
On 2007-03-03 21:18+0100 Werner Smekal wrote: Hi Alan, That's a surprising result. What happens for the equivalent PLplot C core library build? I assume the extensions for the -o option will be .obj (like above), that gcc will build those objects without complaint, and in the next step (doing the special linking with gcc to create the shared library) gcc will happily link the *.obj files into the shared library. Yes, gcc has actually no problems with the .obj extension. But gcc seems to call gnat1 to compile the ada file and though I found a description of gnat for Windows, where it is explained that the -o option should be used to set the output filename with the ending .obj or .o, it seems there is a bug, that prevents gnat1 from accepting .obj files. Since normally someone uses gnatmake this was no problem so far? I thought gnatmake might be the solution, but I have just done a series of exhaustive tests, and there is no way that it works correctly with the -o option or the -cargs -o options. (In the first case, the -o option is not passed to the compiler. In the latter case it calls the compiler with the correct -o option, but then it checks afterward for the object files it expects in the directory where it was invoked, doesn't find them, and errors out.) I have concluded that gnatmake cannot be used for compiles where you want control of the path+name of the resulting object file. What happens when you run gnatmake testadacompiler.adb where that file consists of the following simple Ada source? procedure testAdaCompiler is begin null; end; On my Debian stable system it simply generates the following commands: [EMAIL PROTECTED] gnatmake testadacompiler.adb gcc-3.3 -c testadacompiler.adb gnatbind -x testadacompiler.ali gnatlink testadacompiler.ali Also testadacompiler.o is produced. Does this test produce *.o or *.obj output on MinGW? The point of this test is I am wondering if gnatmake will generate a *.obj file by default, and if it uses special flags on the gcc command to make that happen. From my results above we cannot use gnatmake to compile when we need to control the output path+name, but if this test lets us know the correct compile options for gcc to generate a *.obj, then we can use those flags plus the ordinary gcc -o option to get the desired result on MinGW. 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Vastly improved configuration of Ada interface and examples for PLplot. Please test!
On 2007-03-05 04:10-0700 Jerry wrote: It sounds like this might be a question for comp.lang.ada if you guys don't get it cracked soon. That's a good idea for the future. However, what we need most now is some comprehensive platform testing. The current form of experimental CMake support for Ada works fine on all Debian and Ubuntu platforms where it has been tested. It does have filename extension problems on MinGW, but at most those problems will affect Cygwin and MinGW, and they might be isolated to just MinGW. To get a comprehensive picture of the overall problems we face with the GNAT version of Ada, we need testing not only on Cygwin, but also additional Linux platforms and Mac OS X. Jerry, I especially urge you to move forward with your Mac OS X testing since that is the platform where you developed the Ada interface, and I want to be sure that our CMake support of that interface is working fine there for you. If after reading http://sourceforge.net/cvs/?group_id=2915 and http://www.miscdebris.net/plplot_wiki/ you still have some questions about or problems with CVS or our CMake build system on Mac OS X, please don't hesitate to post them to this 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); 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] ada bindings mingw fix (?)
On 2007-03-05 23:03+0100 Werner Smekal wrote: Hi Alan, I just found an easy fix for the object file ending problem. In CMakeAdaCompiler.cmake.in change the last lines to: IF(UNIX OR MINGW) SET(CMAKE_Ada_OUTPUT_EXTENSION .o) ELSE(UNIX OR MINGW) SET(CMAKE_Ada_OUTPUT_EXTENSION .obj) ENDIF(UNIX OR MINGW) If you think this workaround is ok, I will submit these changes. Good catch, Werner. Please go ahead immediately with that commit. That leaves just the executable suffix issue. I think the thing to do there is to play with SET_TARGET_PROPERTIES (in ada/examples/CMakeLists.txt) until you get something that works. For example, it appears you have control of the output name and/or the SUFFIX. If you can get a build of x01 that way, then to actually get it to execute on MinGW and Cygwin I understand (from informed opinion on the CMake list, :-) ) that you will have to copy x01 to x01.exe. I think the easiest way to do that is with a cmake -e copy command in ada/examples/CMakeLists.txt using the add_custom_command signature that modifies a target result. Anyhow, please play with this executable suffix issue and commit it if you can get something to work on MinGW. I will clean up afterward if necessary. 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
[Plplot-devel] university students with plsmem problem (fwd)
Does anybody have a good simple example of using plsmem? See request that was sent to me below. I encourage you to reply directly to Raul, but if you do so, please CC me or this mailing list since I would like to know the answer to the question as well. 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 __ -- Forwarded message -- Date: Mon, 05 Mar 2007 21:42:48 -0800 From: elpillo [EMAIL PROTECTED] To: [EMAIL PROTECTED] Subject: university students with plsmem problem Message body follows: hello irwin, iam a university student, my team and I are working on a proyect to plot with you library, but we have a problem that we dont know how to solve with the plsmem function, on the Web we havent found any explicit example of who it works, because we want to put the ploting on a buffer and then loaded from cairo and paintied. ive you could give us any advise or a good working example we will apreciated a lot. my email es [EMAIL PROTECTED] thanks a lot. greatigs Raul -- This message has been sent to you, a registered SourceForge.net user, by another site user, through the SourceForge.net site. This message has been delivered to your SourceForge.net mail alias. You may reply to this message using the Reply feature of your email client, or using the messaging facility of SourceForge.net at: https://sourceforge.net/sendmessage.php?touser=1735924 - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
[Plplot-devel] New DEFAULT_NO_BINDINGS option
I have made changes to the CMake build system so that if you set DEFAULT_NO_BINDINGS ON (assuming you have removed the cache file), then no bindings will be enabled by default other than the C one and any that you specifically turn on. Note there is a DEFAULT_NO_DEVICES option that plays a similar role for the devices. Thus, combining these two flags and picking just the combination of the binding and device you want to test can give you extremely fast builds and therefore quick testing turnaround. 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Vastly improved configuration of Ada interface and examples for PLplot. Please test!
On 2007-03-07 04:19-0700 Jerry wrote: [...]so Tuesday I figured out the Cmake stuff and now have my example (not an official plplot example but more complicated) working with 5.7.2. So I believe I'm ready to fix the things that Alan found a few days ago and then generate some more example files. I am very pleased to hear the CVS version of the Ada interface to PLplot is working for you on Mac OS X, and I look forward to your patches for improving the Ada interface and extending the examples. 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] ada bindings mingw fix (?)
On 2007-03-13 11:29+0100 Werner Smekal wrote: Hi Alan, Ada bindings and example work now for the MinGW environment. Apart from the fix below I only needed a small change - instead of TARGET.adb I used TARGET_BASE.adb, since TARGET_BASE is the target name without suffix. I also checked that changes in Linux, so these changes should have no impact on Linux and Mac OSX. Thanks, Werner. That's a nice solution. Out of curiosity where did you find TARGET_BASE documented? Also, thanks for implementing example 12 in Ada and discovering that the thin interface does not (yet) terminate the C strings with a null (which should help Jerry find a good solutions for this issue). BTW, this morning I was in off-list contact with Jerry about sorting out a CMAKE_LIBRARY_PATH issue for libgnat. I believe that should be the last CMake issue for him. This means as of today he should finally be in good position to start dealing with the backlog of Ada interface issues that has been discovered. 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Vastly improved configuration of Ada interface and examples for PLplot. Please test!
On 2007-03-14 17:09-0600 Orion Poplawski wrote: I'm having trouble on Fedora 6 and development, which use gnat 4.1.1 and 4.1.2. First off, libgnat appears to be /usr/lib/libgnat-4.1.so. Also, it appears to be picking up my CFLAGS and passing that to the gnatmake line: I have just committed tentative fixes for both issues. To deal with the first issue our build system now first looks for gnat then for gnat-4.1. (Version numbers in the root of library names cause on-going build system maintenance issues, but I guess we are stuck with this.) To deal with the second issue, I have used the -cargs option for gnatmake (which should just pass on the content of FLAGS to the gcc compiler and not the Ada binder or linker). These changes work on Debian stable. Please test on all platforms. 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Ada fixups for our CMake build system
On 2007-03-14 18:20-0700 Alan W. Irwin wrote: I have just now finally figured out how to use the first form of TRY_COMPILE for the Ada language testing. This makes the Ada language test similar to any add_executable with Ada. Thus, any changes to the Ada build rules (such as the one I just put in for the http://macada.org/ users [e.g., Jerry] on Mac OS X) automatically get used now for the Ada language test. I believe (since I am an optimist) that this should be the last change Jerry needs so he can build the Ada interface to PLplot and run the Ada examples with ctest on his Mac OS X/macada.org platform. But if something more is needed, I am confident we will be able to figure it out. Well, I was too optimistic. :-) Jerry continued to have problems, but he kindly sent me a really clear script of exactly what he did + full cmake.out results which eliminated all sorts of possibilities and eventually lead me to look in minute detail at what happens on Linux. If you force verbose output for the TRY_COMPILE step it turns out the TRY_COMPILE is still not working correctly! In the Linux case it manages to escape the problem and return success even though a nonsensical command was used rather than a test of an Ada build. On Mac OS X it completely fails to escape the TRY_COMPILE problem and returns a failure. For now, I have temporarily replaced this non-working test with something innocuous that should always return success on all platforms. After all, testing the Ada builder in a complicated (and poorly documented!) TRY_COMPILE environment is a refinement, and for now it is probably better to let Ada errors show up under an ordinary make or ctest environment instead where everything is much better understood. Jerry, please let me know whether this simplifying change allows you to proceed further. Also, I wish to thank you publicly for giving me especially clear test results which helped me to discover this on-going problem with testing the Ada builder under a TRY_COMPILE environment. 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Ada fixups for our CMake build system
On 2007-03-15 12:06-0700 Alan W. Irwin wrote: Jerry continued to have problems, but he kindly sent me a really clear script of exactly what he did + full cmake.out results which eliminated all sorts of possibilities and eventually lead me to look in minute detail at what happens on Linux. If you force verbose output for the TRY_COMPILE step it turns out the TRY_COMPILE is still not working correctly! In the Linux case it manages to escape the problem and return success even though a nonsensical command was used rather than a test of an Ada build. On Mac OS X it completely fails to escape the TRY_COMPILE problem and returns a failure. For now, I have temporarily replaced this non-working test with something innocuous that should always return success on all platforms. After all, testing the Ada builder in a complicated (and poorly documented!) TRY_COMPILE environment is a refinement, and for now it is probably better to let Ada errors show up under an ordinary make or ctest environment instead where everything is much better understood. Jerry, please let me know whether this simplifying change allows you to proceed further. Jerry reported further progress off-list, we had several more iterations, and now I can report he can do a cmake, make, and make install without problems for CVS HEAD. Furthermore, his own non-cmake build of some of his private examples now works with the latest (installed?) libplplot library for the first time. So that is all good progress, but there are still the following issues. (1) Jerry's successful test reported above of make install is certainly a worthwhile test of the install of non-Ada PLplot software, but it is not relevant to the install of Ada PLplot software since I am waiting to implement that until after the build-tree Ada issues (e.g., ctest, see below) are solved. (2) ctest does not work with the Ada examples on Mac OS X for Jerry. Apparently make builds the Ada executables without obvious errors, but the build-tree libplplotadad library cannot be found at run time. Does anybody here know the Mac OS X equivalent of the Linux LD_LIBRARY_PATH? (It was mentioned a year or two ago on list, but I have forgotten its name.) Setting that environment variable may be required before the x01a, etc., Ada examples will work in the build tree. Hazen, I have an important question for you. What are your current results (after cmake -DBUILD_TEST=ON and make are run) for ctest --verbose --tests-regex c ? Do you have to set any environment variables (see question above about the Mac OS X equivalent of LD_LIBRARY_PATH) to get the build-tree C examples run by the above ctest invocation to work? If you have no problems with ctest for the C examples, and Jerry does (not proved yet one way or the other), then that might give a clue about what is going wrong for his attempt to run Ada examples that were built in the build tree. 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] pstex driver
On 2007-03-20 16:17-0400 Miguel de Val Borro wrote: Thanks for the advice. The pstex driver can be built with the cmake option -DENABLE_DYNDRIVERS=OFF. I have run all the example programs in C, C++, Fortran and Python for the pstex driver and get a segmentation fault for all of them. I confirm that. Nevertheless, since you have showed interest, you have inspired me to make some more corrections to the CMake infrastructure for this device. So now it builds fine both for -DENABLE_DYNDRIVERS=OFF and ON. (The trick in the -DENABLE_DYNDRIVERS=ON case was to add the ps.c source code to the compilation of the plug-in since the pstex.c code is parasitic on the ps.c code [i.e., it uses functions defined by ps.c]). Also, if I invoke ldd -r drivers/pstex.so, I get no missing libraries or undefined symbols. So the linking of the plug-in for the -DENABLE_DYNDRIVERS=ON case seems perfect. However, for both -DENABLE_DYNDRIVERS=OFF and ON you get segfaults with any example. Miguel, if you care to debug that further, valgrind should be a big help. I had a quick look with our simplest examples (x10c in the install tree), and it appears something is not defined (invalid read). I imagine there was some minor global change we made for all devices, and we forgot to make the change for pstex.c. CVS HEAD is quite stable now so I suggest you use that if you want to take advantage of my pstex CMake infrastructure changes. 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] pstex driver
On 2007-03-21 22:58+0100 Werner Smekal wrote:
Hi,
the segmentation fault was due the fact, that the pstex driver never
told the plplot core that it can't handle unicode strings, but plplot
just anticipated that. I fixed the code in cvs and it works for me now
in Kubuntu.
Should work for you as well.
Thanks for the bug report.
Thanks for your fix, Werner. Good catch! ctest --verbose --tests-regex
pstex works for me now (on Debian stable). It appears this device driver
is generally working now. However, I noticed some obvious scale errors in
my experiments with it (see below) which tweaked my memory of a fix
recommended by Bill Paxton more than two years ago. When I tried the fix
(now possible because the device is working again), the scale problems
disappeared so I have just now committed Bill's fix.
It appears all this device needs now is some documentation of exactly how to
use its results. Here is a first attempt at that, but I hope somebody
improves on this and transcribes it to doc/docbook/src/drivers.xml since
I am really short of time.
The results from this device are two files. One is an EPS file (e.g.,
x01c.eps) of the plots without the characters, and the other is a file of a
latex fragment (e.g., x01c.eps_t) which imports that EPS file and uses latex
commands to render the characters in a way that is perfectly aligned with
the EPS file.
You may use the generated files with the following latex code:
x01c.tex*
\documentclass{article}
\usepackage[dvips]{graphicx}
\begin{document}
\input{x01c.eps_t}
\end{document}
x01c.tex*
Then you generate a postscript file from this using the commands:
latex x01c.tex
dvips -f x01c.dvi x01c.ps
The results look good (aside from an obvious bounding-box problem) and should
be useful for Latex enthusiasts.
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
__
-
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.phpp=sourceforgeCID=DEVDEV
___
Plplot-devel mailing list
Plplot-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/plplot-devel
[Plplot-devel] Changes to the Ada thick and thin interfaces and examples.
Yesterday (Wednesday) marked a lot of minor changes to the Ada interfaces and examples. All the details are in the CVS commit messages which I urge those who have an interest in Ada to review. Jerry started the ball rolling with a patch Wednesday morning which made a lot of corrections. After I committed his patch, I followed up with changes to x12a to make it consistent with the other thin interface examples. Comparison with the equivalent C result showed a bug in the thin version of plfill which I fixed by brute force. (Jerry should review that fix since I don't really know what I am doing with Ada. :-) ) I then replaced x01a with a version that used the thin interface and moved the old version to xt01a (where the t means the thick interface is used by the example). ctest on xt01a, x01a, x10a, and x12a reveal no differences with the corresponding C results so all seems to be well so far. However, before adding any more thick or thin examples, I have suggested to Jerry off-list some changes I hope he makes to the API, but we will see what he decides about those issues. If he makes a quick decision, we might get in one more round of changes before the release this weekend. Another continuing issue is the CMake build system infrastructure for Ada. I noticed some weird results for it today. The source code dependencies were not working correctly (so you had to start with an empty build tree each time to get changes compiled for plplot*.ad?) Also, if you did make clean the linking command subsequently disappeared (???) unless you started all over with an empty build tree. So the current advice is always start with an empty build tree for libplplotada source code changes. Strangely enough this was not necessary for example source code changes where the dependencies were working properly. I don't have time to tackle these issues at the moment. I think the Ada interface and example build system is adequate for now (if you don't expect too much from it and use the workaround --- start with empty build tree --- that I mentioned), and I plan to sort out these issues substantially after the current release. 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] cvs to svn conversion
On 2007-03-21 00:46-0700 Alan W. Irwin wrote: Tomorrow I plan to compare selected CVS and SVN tagged releases for lbproject, upload to SF, and do final checks of the SF result to complete the lbproject conversion to subversion. Done. Almost all of the conversion is handled by two high-level scripts, and I am completely satisfied by the new lbproject svn results at SourceForge. On to yplot 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
[Plplot-devel] More on repository formats
Arjen asked a good question on plplot-general about incompatibilities between the various Subversion repository formats. To follow up on that here, if you use the file:/// access method rather than the normal https:// method (where an apache module is interfacing), then you do have to worry about whether your client Subversion software is consistent with the repository format. That is, if I use the rsynch method to download our Subversion repository (which is in the format understood by the SF version of Subversion), then both my Debian stable Subversion client and Ubuntu Dapper Subversion client are too old to understand that repository using the direct file:/// method. However, if I build the latest stable release (version 1.4.3) of Subversion (which I built and used for my tests in any case rather than the dated system versions that I had), then it understands that format perfectly. In fact, if I compare the SF repository as downloaded by rsync before I made any commits, and the repository that I locally built, they agree almost exactly (except for dates, a one-line uuid file, the path to svnlook, and the hook script to generate mail on plplot-cvs whenever a commit is done). This agreement makes sense because both repositories were built with the svnadmin load command using the same two svn dumpfiles. The only difference between them would be due to a difference between my local svnadmin software and the SF version, but apparently the vanilla version-1.4.3 of Subversion (which is what I have) is close enough to the SourceForge version so the two repositories that were built agree in detail. By the way, this good agreement saved me from having to do some final tests I was planning using detailed comparisons between various checkouts and generated ChangeLogs of trunk, tags, and branches of the local versus SF SVN repositories. 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 __ - 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.phpp=sourceforgeCID=DEVDEV ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
[Plplot-devel] Cairo-related devices and interactive capabilities of PLplot
Hi Hazen, As you suggested, I am going to put this thread on plplot_devel because it will probably be of interest to the other developers there. On 2007-05-16 23:18-0400 Hazen Babcock wrote: On May 16, 2007, at 12:30 AM, Alan W. Irwin wrote: [...]Speaking of which, do you have any near-term plans for xwinttf or the further Cairo devices you proposed? Near term? I'd like to finish this project up in the next few months and hopefully I can find the time. My plan is to put xwinttf and all the other Cairo drivers into one file, probably cairo.c, as was done for the GD driver. Note, the proposed cairo-based devices would belong both to the PLplot display device driver family (xwinttf and others?) and to the PLplot output file device driver family (postscript, and others). For a discussion of those two families of devices see http://plplot.sourceforge.net/docbook-manual/plplot-html-5.7.3/x-drivers.html and http://plplot.sourceforge.net/docbook-manual/plplot-html-5.7.3/output-drivers.html. Since display device drivers potentially have interactive considerations to take into account (how to move from one page to the next, full cursor capability [see below], and other GUI considerations) and output device drivers do not, I suggest you split the proposed cairo-based set of devices into two separate drivers (one for display devices and one for file output devices) with two separate files for the source. I'm not sure how to proceed with the xwinttf driver regarding the interactivity issues that you mentioned. We could make it behave the same as the xwin driver, i.e. only use one window and keep clearing after each plot, but this is not ideal for interactive environments, such as Perl / PDL where it is nice to be able to keep a couple plotting windows open, even if you might not be able to update them. On OS-X the operating system takes care of keeping them updated when you cover them with another example or whatever, so they don't even need to handle refresh events. I've come to believe (perhaps incorrectly?) that one of the major shortcomings of PLplot is that it is difficult to use in an interactive environment. The library only supports output to one stream at a time and even if it did support multiple streams the stream is an internal structure in the library, you can't even get an generic pointer to it so that you could cache it, draw something else, and go back to drawing the original plot. Note, you can randomly switch between various streams using the plsstrm command (see the 14th example). And the device driver interface is written generally enough (with an escape function) so that devices can have a tremendous amount of capability. However, your general comments on interactive PLplot struck a nerve with me. In my opinion where PLplot does very poorly on the interactive front is the severely limited interactive GUI devices that have been implemented so far. For example, the -dev gcw GUI allows selection of pages to display, and that is about it. -dev tk is our most extensively developed GUI for PLplot at the moment. It has zooming capability, a colour pallette adjustment, and perhaps even cursor capability. (I know that -dev xwin has cross-hairs [or cursor] capability. However, I have forgotten how to fire it up and IIRC, the world coordinates of the cross-hairs are printed out rather than returned to the calling routine which is sort of useless. I think -dev tk may have the same limited cursor capability as well). In my opinion what is truly needed for interactive plotting capability is full cursor capability where the world coordinates of the cursor when a keyboard key was struck _and_ the identity of that key are returned from the device to the PLplot library and then back to the routine calling the PLplot library. Currently no PLplot API exists for returning full cursor data, but it should be straightforward to extend the API appropriately with plgcursor. Such a function should make an interactive device wait until the user struck a key and then the function would return the x, y, (and z?) world coordinates of the cursor position when the key was struck as well as the identity (X keycode) of the key that was struck. I well recall a spectrum analysis programme I used extensively in the early 80's whose plotting device drivers had that full cursor capability which allowed an application that used the library to have a very large menu of possible transformations or measurements of the spectrum being plotted depending on where the user placed the cursor and what key was struck. I have a potential interactive research software project that would require exactly such an approach. I have put off the project in favour of other research projects for a number of reasons. But one of those reasons is I am currently not allowed to implement that project software with PLplot because no PLplot GUI devices currently have the full cursor capability that I
Re: [Plplot-devel] Road to our next stable release
On 2007-05-21 00:34-0400 Jim Dishaw wrote: Alan W. Irwin [EMAIL PROTECTED] writes: My own feeling is we should do at least one more development release, 5.7.4, ... and a new windows device driver (Jim Dishaw). I defintely concur about having one more developmental release. I will not have the windows driver release quality until late August. I'm in the midst of finishing my dissertation and preparing for my defense and the only work I do on plplot is to add a necessary feature (you should see the debugging messages spewing out of the new driver...). So, I would prefer that the new windows driver makes it debut in a developmental version rather than a stable release. Jim, thanks for that update on the current experimental status of your windows device driver and also your currently severe time constraints. I should mention that your project as well as the rest of those that I listed are all in the would-be-nice category, and I don't think we should delay the proposed 5.7.4 and 5.8.0 releases for any of them. The reason why I feel this way is we have done a lot of work in the current series of development releases, and it would be nice to make that work available as a stable release as soon as possible so the conservative subset of our users that wait for stable releases will benefit from our work sooner rather than later. The only reason why I think we need one more development release before that stable release is to to get any kinks out of the release scripts and procedures caused by our conversion to subversion. Of course, depending on Hazen's time constraints that delay in the stable release might give any of the projects I mentioned (including your own) a chance to get into the last development release before the stable release. Note, your code does not have to be perfect to get into PLplot. I expect, instead, that Werner, Arjen, and Andrew (the core developers in a position to evaluate a windows device) will instead just want to make sure your windows device has some unique potential that does not exist in current windows devices. Once that uniqueness test is passed, I presume they would be willing to commit experimental device code donated by you with the device turned off by default so the experimental code only builds if specifically requested by the user. (Note, that is how I am treating Jerry Bauck's experimental Ada interface code that he has been working on recently.) But if you decide to contribute your code for evaluation by our core windows developers after 5.8.0, that is fine as well. 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 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 __ - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Subversion-related changes for our next release.
On 2007-05-22 23:11-0400 Hazen Babcock wrote: On May 22, 2007, at 2:25 AM, Alan W. Irwin wrote: The rest of the release build process will then be done working off the contents of this directory. When the release checks out the RM will commit this directory to Sourceforge. I assume you should commit it when you first create it with the svn copy command and after updates which should be rarely necessary since you are working with a copy of trunk that presumably has been thorougly tested by our core developers. The only exception to this no updates for tags rule I can think of is if you forget some part of the release process (e.g., updating a release file) that you catch before you release the tarball at SourceForge. In that special case the additional commit required before running the make_tarball.sh script would be quite fast since typically only one file will be changed. By the way, I think you forgot to comment about the tag commit step in your README.Release_Manager_Cookbook changes. My approach was to keep the tag local until the release was actually released on Sourceforge. The previous revision of make_tarball.sh cvs exported from the remote repository (not a local filesystem) with no problems at all so that is the approach I am starting with for svn. (I am being conservative here because we are all pretty much svn newbies.) Later, if your svn export experiments show there is a better way to do everything locally until the last moment before release, then bash scripting is so easy you will probably find it straightforward to make the script changes you want, and I certainly encourage you to do that independently. (Our scripts belong to all of us so anyone and especially the release manager should feel free to change them.) However, I would be glad to have a look at your script changes and help you debug them if you feel you need that moral/practical support. 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 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 __ - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Problem with ada interface example code
On 2007-05-24 15:25-0600 Orion Poplawski wrote: In plplot-5.7.3/examples/ada/x12a.adb, the sprintf function is declared and imported as follows: procedure Sprintf( buffer : out char_array; format : in char_array; variable: in PLFLT ); pragma Import(C, Sprintf, sprintf ); This is incorrect because sprintf's declaration is sprintf (char *, const char *, ...) (varargs) not sprintf (char *, const char *, float). This happens to work on i386 because of the ABI happens to be the same. It isn't on x86_64 and so crashes. See https://bugzilla.redhat.com/bugzilla/process_bug.cgi#c1 and the http://gcc.gnu.org/onlinedocs/gnat_ugn_unw/Calling-Conventions.html note about varargs for more info. If you really need to use sprintf you'll need to write a write a sprintff(char *, const char *, float) wrapper to sprintf and interface to that. Likewise for any other arguments used (currently just the one). Not sure if any plplot routines use varargs arguments, but if so the current ada imports are probably incorrect. Thanks, Orion, for your report. To answer Orion's last comment first, I looked in src/*.c for ... and there doesn't seem to be any libplplot routines with a variable number of arguments so I don't think that is going to be an issue. To answer Orion's second-last comment, I think we should look for an Ada alternative to Importing the C sprintf function because of the difficulties you have found with that method. Virtually every language has a native way to transform from integer, floating-point etc. representations, to strings. Jerry, are you aware of the normal way to do this in Ada? N.B. this has nothing to do with API questions or our Ada interface and is merely a question about how to transform from integer and floating point representations to strings in Ada since that capability is needed for the Ada implemention of example 12 and several other of our examples. 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 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 __ - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] CMake and python bindings
Hi Andrew: Welcome back! :-) More below. On 2007-05-25 08:27+0100 Andrew Ross wrote: I've (finally) found some time to do some work on plplot again. I've noticed for a while that the python bindings were failing ctest on my debian and ubuntu systems, but since I don't use python I'd not looked further. Turns out that the problem is with multiple versions of python on the same machine. My default python /usr/bin/python is a symlink to python2.4, however FindPythonInterp looks for specific versions (python2.5 python2.4 ... ) before looking for python. This means that it picks up python2.5 as the interpreter. Similarly for the locations of the python libraries. When I come to run ctest the scripts call /usr/bin/python as the interpreter, but this is the wrong version and I get a seg fault. Has anyone else run into this problem? No. But my default python is 2.3 while CMake picks up the 2.4 version on my Debian sarge system. It's possible 2.3 and 2.4 are more compatible than 2.4 and 2.5. I can see 3 alternative routes to solve this. 1) Set all the python related cmake variables by hand (a little tedious). 2) Alter the python scripts so you can substitute in the name of the interpreter that was actually used to build plplot. 3) Alter the cmake modules so you can specify a prefered python version which will then propagate through when it comes to looking for the interpreter and the libraries. Only 2) would fix ctest out of the box which is generally a good thing for new users. Why am I suddenly interested in python? Well, I've been trying to port the debian packages to the new CBS so we can continue to support Debian and Ubuntu. This includes the python-plplot package. This actually involves building the bindings for multiple versions of python, which could potentially be a little tricky with the cmake system as-is. Option 3) would probably help. More on the debian packages later though. I have just included PYTHON_EXECUTABLE in our summary, and I included PYTHON_INSTDIR in our summary a few days ago. These data should help a little bit in sorting this issue out. Whatever PYTHON_EXECUTABLE is chosen, I think we will always want the python scripts to be compatible with it so I think you should go ahead with (2) (which, of course, means configuring all our python scripts). With regard to your last paragraph, all I know is python packaging in Debian is really tricky. So be careful to follow their most recent set of python packaging rules (which change from time to time as they gain more experience with migrations between python versions). 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 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 __ - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
[Plplot-devel] USE_RPATH now honored when installing the various PLplot libraries
I have made widespread changes in our build system so that USE_RPATH is now honored when installing the various PLplot libraries. (USE_RPATH has always been honored when building the installed examples.) I have tested both the default (USE_RPATH=ON) and also -DUSE_RPATH=OFF, and I get the expected results. In both cases the ctest tests (the build-tree tests) work. In the first case, make install has lots of rpath link commands, and you don't need to specify LD_LIBRARY_PATH in order for the install tree examples to work. In the second case, make install generates no rpath link commands (no rpath is required by some packaging systems), and you must specify LD_LIBRARY_PATH to get any of the installed examples to work (assuming you have installed PLplot in a non-standard location). Those who are doing PLplot packaging should try -DUSE_RPATH=OFF to see if it makes your rpath life a little easier. Valery, this still leaves the issue of the trailing : for rpath that you detected for your particular combination of software packages. You should be able to work around it now by specifying -DUSE_RPATH=OFF, but I would also appreciate it if you pursue the trailing : rpath issue some more for the default (-DUSE_RPATH=ON) case so we can find out what is going wrong on your particular platform for 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); PLplot scientific plotting software package (plplot.org); 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 __ - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Ada bindings and examples
On 2007-06-01 15:18-0700 Jerry wrote: [...]To summarize: The plan is to make a quasi-third Ada binding which is exactly like the current thick binding except that the subroutine names are the traditional PLplot names. There will be two example files for each example, one written to the thick binding and one written to the new binding. The naming conventions are e.g. for Example 1, xt01a.adb is written for the thick binding and x01a.adb is written to the new binding. Coding for the Ada examples should use Ada constructs and functions as needed. (See the other note from me today relating to the sprintf problem from a few days ago.) Once an example is complete for one binding, the corresponding example can be automatically generated by a series of string substitutions for subroutine names on the source--that's how I plan to generate the new binding as well. Sounds like an excellent plan. 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 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 __ - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Cairo Driver Family
On 2007-06-03 18:20-0400 Hazen Babcock wrote: Hello, I've added the Cairo driver family to PLplot. This includes: xcairo (formally known as xwinttf). pdfcairo (PDF) pscairo (PS) svgcairo (SVG) pngcairo (PNG) memcairo (MEM) Known issues: 1. If you have Cairo but not X Windows your build will likely fail since I still haven't solved the pre-processor directive problem I mentioned Yesterday. 2. The svgcairo driver does not display text properly. I'm not really sure what to do about this one since it seems to be an issue with the cairo library. 3. The pngcairo driver does not support multiple pages. You only get one page, which is the last page that you drew. This could be fixed by having the driver generate multiple files, say my-graph.png, my- graph-2.png, etc... I don't think the PNG format itself supports multiple page output. That is correct. The way this is taken care of for the png device (as well as jpeg and gif) for gd.c is to use the normal PLplot familying facility (see http://plplot.sourceforge.net/docbook-manual/plplot-html-5.7.3/output-devices.html#familying). It's a bit tricky to set up, but if you follow what is done for gd.c, you should be fine. I thought you might have inadvertently set up familying already simply by following what is done in gd.c, but apparently not since the -fam command-line option only gives the same last page of example 8 that you get without the option. 4. The memcairo driver is untested. It appears that a side effect of a call to plsmem is to force PLplot to use the mem driver. An attempt to overide this behavior by calling plsdev(memcairo) after the plsmem call but before plinit was unsuccessful. 5. Plots output by the pscairo driver do not fit the page, i.e. there is a largish white margin on two sides of a plot. All the drivers are off by default so you will have to turn them on to test them, if you are so inclined. I didn't bother with memcairo, but I built all the rest with no build errors on Ubuntu Dapper. I have only done the most superficial tests of the built results, but I immediately noticed two issues on Linux beyond what you mentioned. (1) the 'ldd -r cairo.so' test showed the following problems: undefined symbol: cairo_ps_surface_create (./cairo.so) undefined symbol: cairo_pdf_surface_create (./cairo.so) undefined symbol: cairo_svg_surface_create (./cairo.so) It's fairly urgent to solve these problems since it excludes any testing of the pscairo, pdfcairo, or svgcairo devices on Linux. (2) ./x08c -dev xcairo shows only the first page of that example. Hazen, I am really happy you are developing all these promising cairo-related devices. 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 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 __ - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Cairo Driver Family
On 2007-06-04 21:17-0400 Hazen Babcock wrote: On Jun 3, 2007, at 7:30 PM, Alan W. Irwin wrote: (1) the 'ldd -r cairo.so' test showed the following problems: undefined symbol: cairo_ps_surface_create (./cairo.so) undefined symbol: cairo_pdf_surface_create (./cairo.so) undefined symbol: cairo_svg_surface_create (./cairo.so) It's fairly urgent to solve these problems since it excludes any testing of the pscairo, pdfcairo, or svgcairo devices on Linux. Maybe this is because I have Cairo v1.2.6 and debian testing is v1.2.4? According the cairo docs these functions have been in the library since 1.2. Actually, my platform is Ubuntu Dapper (released almost exactly a year ago) which has version 1.0.4 of the cairo library. So that explains the missing symbols for my platform. Note, _some_ Ubuntu Dapper platforms are going to be around for a long time since Canonical plans to support it for 3 years on the desktop (and 5 years on the server). However, I am not sure the majority of desktop users will actually use Ubuntu Dapper that long since even a one-year old Linux desktop is already way out of date thanks to the huge pace of change for Linux development. (A case in point is this already old libcairo version for Ubuntu Dapper). For example, I plan to move to Debian testing as soon as I can find the time to do the install. That's the best background I can give you about how relevant libcairo-1.0.4 will be in the future, and I will let you be the judge whether you want to support this older version of libcairo or not. I hope to install debian testing some time this summer after I finish my current research project, and after that old libcairo versions won't be an issue for me. (2) ./x08c -dev xcairo shows only the first page of that example. Hm. On OS-X I get a bus error. It is interesting that it works fine if you just type ./x08c and then choose the driver. ./x08c -dev xcairo now actually pretty much works for me. I finally figured out that you should hit CR in the xterm where that command-line is typed and not the GUI (at least when threading is not turned on). Now that I know how to run it, ./x08c -dev xcairo shows some issues. (1) The 3D labelling needs work on the transformation to determine the rotation and slew of each letter. I thought there was now standard code in the plplot library for that? Or is it a bug in that standard code? (2) Memory management issues Here is the associated error message (after I hit one additional CR after the last page of example 8 is displayed) *** glibc detected *** double free or corruption (!prev): 0x08050358 *** Unless there is some internal problem in the Ubuntu Dapper libraries (this is possible since these are old versions of each library) it sounds like your code might be trying to free something that has already been freed. 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 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 __ - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Cairo Driver Family
On 2007-06-06 20:26-0400 Hazen Babcock wrote: On Jun 4, 2007, at 10:17 PM, Alan W. Irwin wrote: (2) Memory management issues Here is the associated error message (after I hit one additional CR after the last page of example 8 is displayed) *** glibc detected *** double free or corruption (!prev): 0x08050358 *** Unless there is some internal problem in the Ubuntu Dapper libraries (this is possible since these are old versions of each library) it sounds like your code might be trying to free something that has already been freed. I'm not seeing this in OS-X. If you are so inclined, could you go to the sub-routine plD_tidy_xcairo(), line 783, in cairo.c and comment out the call to plD_tidy_cairo() and/or the call to free() on line 796 and let me know which one (if either) makes this message go away? Commenting out just plD_tidy_cairo() on line 783 makes no obvious difference and I still get the above error message. Commenting out just free() on line 796 gets rid of the above error message. Commenting out both lines also seems to get rid of the above error message, but I assume that is not the correct fix since the free seems to be the culprit. 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 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 __ - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Cairo Driver Family
On 2007-06-06 18:48-0700 Alan W. Irwin wrote: On 2007-06-06 20:26-0400 Hazen Babcock wrote: On Jun 4, 2007, at 10:17 PM, Alan W. Irwin wrote: (2) Memory management issues Here is the associated error message (after I hit one additional CR after the last page of example 8 is displayed) *** glibc detected *** double free or corruption (!prev): 0x08050358 *** Unless there is some internal problem in the Ubuntu Dapper libraries (this is possible since these are old versions of each library) it sounds like your code might be trying to free something that has already been freed. I'm not seeing this in OS-X. If you are so inclined, could you go to the sub-routine plD_tidy_xcairo(), line 783, in cairo.c and comment out the call to plD_tidy_cairo() and/or the call to free() on line 796 and let me know which one (if either) makes this message go away? Commenting out just plD_tidy_cairo() on line 783 makes no obvious difference and I still get the above error message. Commenting out just free() on line 796 gets rid of the above error message. Commenting out both lines also seems to get rid of the above error message, but I assume that is not the correct fix since the free seems to be the culprit. BTW, we worked hard in the old days to get both the memory management and familying issues correct for gd.c. So if you follow that general pattern (albeit with libcairo specifics), you should be okay. 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 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 __ - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
[Plplot-devel] Dashboard results now possible with Plplot
I followed the directions at http://www.cmake.org/Wiki/CMake_Testing_With_CTest for enabling dashboard testing of PLplot, and they appear to work. The only CMake build system change was to change our top-level CMakeLists.txt file to replace ENABLE_TESTING() by INCLUDE(CTest). After that change, the results of ctest -D Experimental were as follows: Site: chickadee Build name: Linux-c++ Create new tag: 20070630-1606 - Experimental Start processing tests Configure project Each . represents 1024 bytes of output .. Size of output: 5K Build project Each symbol represents 1024 bytes of output. '!' represents an error and '*' a warning. ...*.. Size: 50K .. Size: 100K .. Size of output: 109K 0 Compiler errors 2 Compiler warnings Test project /home/software/plplot_cvs/HEAD/build_dir 1/ 10 Testing examples_c Passed 2/ 10 Testing examples_cxx Passed 3/ 10 Testing examples_f77 Passed 4/ 10 Testing examples_octave Passed 5/ 10 Testing examples_python Passed 6/ 10 Testing examples_tcl Passed 7/ 10 Testing examples_ada Passed 8/ 10 Testing examples_png Passed 9/ 10 Testing examples_svg Passed 10/ 10 Testing examples_plmeta Passed 100% tests passed, 0 tests failed out of 10 Performing coverage Cannot find any coverage files. Ignoring Coverage request. Submit files (using http) Using HTTP submit method Drop site: http://public.kitware.com/cgi-bin/HTTPUploadDartFile.cgi Uploaded: /home/software/plplot_cvs/HEAD/build_dir/Testing/20070630-1606/Build.xml Uploaded: /home/software/plplot_cvs/HEAD/build_dir/Testing/20070630-1606/Configure.xml Uploaded: /home/software/plplot_cvs/HEAD/build_dir/Testing/20070630-1606/Test.xml Using HTTP trigger method Trigger site: http://public.kitware.com/cgi-bin/Submit-Random-TestingResults.cgi Dart server triggered... Submission successful By default, the dashboard results were sent to the Kitware public dashboard which are collected at http://public.kitware.com/Public/Dashboard/20070630-0100-Nightly/Dashboard.html, but it appears we will have to wait until 1:00 (UTC) to actually see the above results in html form. Since this submission of ctest results to the Kitware public dashboard appears to work, I encourage everyone here to try such submission following the instructions at http://www.cmake.org/Wiki/CMake_Testing_With_CTest. ctest -D Experimental is the simplest thing you can do, but it appears the instructions show how to set up an automatic nightly build from the svn version of PLplot which should give us some much-needed comprehensive automatic testing of the PLplot code once each day on the wide variety of platforms that are accessible to us as a group. 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 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 __ - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Dashboard results now possible with Plplot
On 2007-06-30 09:50-0700 Alan W. Irwin wrote: By default, the dashboard results were sent to the Kitware public dashboard which are collected at http://public.kitware.com/Public/Dashboard/20070630-0100-Nightly/Dashboard.html, but it appears we will have to wait until 1:00 (UTC) to actually see the above results in html form. The results are there now. Look for chickadee (the name of my build computer). There is no obvious identification in terms of PLplot, and if a number of us contributed the results would be scattered all over. Thus, what we really need to do is set up our own Dart 2 server (see http://public.kitware.com/Dart/HTML/Index.shtml for instructions) to collect PLplot build and ctest results. I don't think a dart server is possible on our SF website since dart requires xalan and java and xalan is certainly not installed on shell.sf.net. Do we have any volunteers with access to a web server box who would be willing to set up a dart server for 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 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 __ - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] [Fwd: Build Error (Job 31606): plplot-5_7_3-2_fc7 on fedora-development-extras]
On 2007-07-02 15:00-0600 Orion Poplawski wrote: Alan W. Irwin wrote: [...]I tried make -j2 on a fully configured system for revision 7687 on Ubuntu Dapper, and I got good results:[...] Please try make -j2 yourself on revision 7687 to confirm that the problem has now been solved. I still got an error, though in a different place: [122%] Building C object src/CMakeFiles/plplotd.dir/plvect.o /usr/bin/gcc -Dplplotd_EXPORTS -O2 -g -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=4 -m32 -march=i386 -mtune=generic -fasynchronous-unwind-tables -fPIC -I/builddir/build/BUILD/plplot-5.7.3/include -I/builddir/build/BUILD/plplot-5.7.3/fedora -I/builddir/build/BUILD/plplot-5.7.3/fedora/include -DHAVE_CONFIG_H -o src/CMakeFiles/plplotd.dir/plvect.o -c /builddir/build/BUILD/plplot-5.7.3/src/plvect.c Building C object src/CMakeFiles/plplotd.dir/plvpor.o /usr/bin/gcc -Dplplotd_EXPORTS -O2 -g -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=4 -m32 -march=i386 -mtune=generic -fasynchronous-unwind-tables -fPIC -I/builddir/build/BUILD/plplot-5.7.3/include -I/builddir/build/BUILD/plplot-5.7.3/fedora -I/builddir/build/BUILD/plplot-5.7.3/fedora/include -DHAVE_CONFIG_H -o src/CMakeFiles/plplotd.dir/plvpor.o -c /builddir/build/BUILD/plplot-5.7.3/src/plvpor.c /builddir/build/BUILD/plplot-5.7.3/src/plvect.c: In function 'plfvect': /builddir/build/BUILD/plplot-5.7.3/src/plvect.c:171: warning: 'a_y' may be used uninitialized in this function Linking C shared library libplplotd.so cd /builddir/build/BUILD/plplot-5.7.3/fedora/src /usr/bin/cmake -P CMakeFiles/plplotd.dir/cmake_clean_target.cmake cd /builddir/build/BUILD/plplot-5.7.3/fedora/src /usr/bin/cmake -E cmake_link_script CMakeFiles/plplotd.dir/link.txt --verbose=1 /usr/bin/gcc -fPIC -O2 -g -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=4 -m32 -march=i386 -mtune=generic -fasynchronous-unwind-tables -shared -Wl,-soname,libplplotd.so.9 -o libplplotd.so.9.3.0 CMakeFiles/plplotd.dir/pdfutils.o CMakeFiles/plplotd.dir/plargs.o CMakeFiles/plplotd.dir/plbox.o CMakeFiles/plplotd.dir/plcont.o CMakeFiles/plplotd.dir/plcore.o CMakeFiles/plplotd.dir/plctrl.o CMakeFiles/plplotd.dir/plcvt.o CMakeFiles/plplotd.dir/pldeprecated.o CMakeFiles/plplotd.dir/pldtik.o CMakeFiles/plplotd.dir/plfill.o CMakeFiles/plplotd.dir/plfreetype.o CMakeFiles/plplotd.dir/plhist.o CMakeFiles/plplotd.dir/plimage.o CMakeFiles/plplotd.dir/plline.o CMakeFiles/plplotd.dir/plmap.o CMakeFiles/plplotd.dir/plot3d.o CMakeFiles/plplotd.dir/plpage.o CMakeFiles/plplotd.dir/plsdef.o CMakeFiles/plplotd.dir/plshade.o CMakeFiles/plplotd.dir/plstdio.o CMakeFiles/plplotd.dir/plstripc.o CMakeFiles/plplotd.dir/plsym.o CMakeFiles/plplotd.dir/pltick.o CMakeFiles/plplotd.dir/plvpor.o CMakeFiles/plplotd.dir/plwind.o CMakeFiles/plplotd.dir/plbuf.o CMakeFiles/plplotd.dir/plgridd.o CMakeFiles/plplotd.dir/plvect.o -L/builddir/build/BUILD/plplot-5.7.3/fedora/lib/csa -L/builddir/build/BUILD/plplot-5.7.3/fedora/lib/nn -lltdl -ldl -lm -lcsirocsa -lcsironn -lfreetype -lm -lqhull -Wl,-rpath,/builddir/build/BUILD/plplot-5.7.3/fedora/lib/csa:/builddir/build/BUILD/plplot-5.7.3/fedora/lib/nn gcc: CMakeFiles/plplotd.dir/plvpor.o: No such file or directory gmake[5]: *** [src/libplplotd.so.9.3.0] Error 1 gmake[5]: Leaving directory `/builddir/build/BUILD/plplot-5.7.3/fedora' Wonder if we're bumping into cmake issues at this point... I just got good make -j2 results for latest SVN trunk of PLplot and CMake-2.4.7-RC-11 on my Debian sarge platform so I cannot verify the problem here. Of course, I have a very different tool chain than you (let's face it, my tool chain is obsolete compared to your cutting-edge Fedora one). Also, my system is a uniprocessor entry-level 2.4GHz machine bought used several years ago. Your timings are probably very different as a result of these major software/hardware differences so the bug could be triggered on your system and by accident be avoided on my system. All that said, one thing that is bothering me about the above report is the linking step deals with the object files in the same order that they are specified in src/CMakeLists.txt with plvpor being referenced a number of files from the end of the list. But the plvpor compilation seems to occur as the last compilation. However, that is followed by a plvect.c warning message. The only thing I can think of to explain that combination is you elided some of the output to save bandwidth, but please confirm that and also please confirm that the plvpor.c compilation was initiated before the compilations of plwind.c plbuf.c plgridd.c plvect.c were initiated. Note the lines of source code in those files: wc -l plvpor.c plwind.c plbuf.c plgridd.c plvect.c 461 plvpor.c 232 plwind.c 1230 plbuf.c 894 plgridd.c 309
Re: [Plplot-devel] TeXinfo patch
On 2007-07-02 16:47-0600 Orion Poplawski wrote: Argh, looks like there is a bug in docbook2X 0.8.7 that causes --to-stdout to be ignored, so that doesn't work. Looks like this has been around for a while so make sure you aren't ending up with an empty plplotdoc.texi in addition to the_plplot_plotting_library.texi. But, I think I've figured out how to fix this in the xml source. What I don't like it that it changes the title but I haven't found a way around this yet. Not sure what else it affects. In any case, this generates a proper .info file now. --- plplotdoc.xml.in(revision 7739) +++ plplotdoc.xml.in(working copy) @@ -237,7 +237,14 @@ book id=plplot-docbook bookinfo -titleThe PLplot Plotting Library/title +titlePLplot/title +titleabbrev role=texinfo-fileplplotdoc/titleabbrev +abstract role=texinfo-node + paraThe PLplot Plotting Library/para +/abstract +subjectset scheme=texinfo-directory + subjectsubjecttermScientific Visualization/subjectterm/subject +/subjectset abbrevDocBook/abbrev subtitleProgrammer's Reference Manual/subtitle releaseinfoVersion 5.0/releaseinfo These could go anywhere in the bookinfo section. Watch the line wrap above. I too am concerned with the title change. That apparently applies to the whole documentation (see first line of http://plplot.sourceforge.net/docbook-manual/plplot-html-5.7.3/) so I don't think that overall title change is acceptable. What happens if you leave titleThe PLplot Plotting Library/title as is but make the rest of your changes? I assume titleabbrev role=texinfo-fileplplotdoc/titleabbrev completely determines the output file name. Thus, what issue is solved by changing the overall title at this point? If it is not a showstopper issue for you (like the filename issue is), then I am inclined to go with your changes except keeping the overall title as is. 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 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 __ - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Ada bindings: user no longer needs to incorporate the thin binding
On 2007-07-03 04:10-0700 [EMAIL PROTECTED] wrote: Hi Alan, Here's another patch. Changes: Modified all Ada bindings so that user programs (and thus the PLplot examples) no longer need to express a dependence on the thin binding. Thus, user programs and examples no longer need to with or use PLplot_Thin, and the namespace collision described in the last patch no longer exists (for the user). Moved some constant declarations out of the thin binding into the other bindings to improve visibility for the user and to remove the need to reference PLplot_Thin. There is now some duplication between the higher-level bindings PLplot and PLplot_Traditional and the lower-level PLplot_Thin in these constants, but the user doesn't see the duplication since there is no need to look at PLplot_Thin. Added two new files to the bindings directory, plplot_auxiliary.adb and plplot_auxiliary.ads. This was necessary in order to make the PLplot_Thin bindings disappear to the user. These files contain a few basic declarations used by all of the bindings (which were formerly exported by PLplot_Thin) and some utility procedures used by several of the examples. Hi Jerry: I committed your Ada patch. I also checked (on my Debian sarge platform) that it built from scratch without problems and produced identical example results to the C examples. Thanks for your good work. 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 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 __ - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Ada bindings: user no longer needs to incorporate the thin binding
On 2007-07-03 07:11-0700 Alan W. Irwin wrote: Added two new files to the bindings directory, plplot_auxiliary.adb and plplot_auxiliary.ads. This was necessary in order to make the PLplot_Thin bindings disappear to the user. These files contain a few basic declarations used by all of the bindings (which were formerly exported by PLplot_Thin) and some utility procedures used by several of the examples. Hi Jerry: I committed your Ada patch. After committing your patch I tried one further local change; removing all references to PLplot_Auxiliary from the examples because both the thick and traditional examples refer to Ada interfaces which in turn refer to PLplot_Auxiliary. However, that lead to the following type of error for both thick and traditional examples for both example 1 and example 12 (example 10 is so simplistic it doesn't run into the problem): xthick01a.dir/xthick01a.o xthick01a.adb:39:14: Real_Vector is not visible (more references follow) xthick01a.adb:39:14: non-visible declaration at plplot_auxiliary.ads:26 xthick01a.adb:39:26: incorrect constraint for this kind of type xthick01a.adb:43:11: String_80 is not visible xthick01a.adb:43:11: non-visible declaration at plplot_auxiliary.ads:68 I presume with the appropriate namespace commands in plplot_traditional.ad[bs] and plplot.ad[bs] all the PLplot_Auxiliary definitions (e.g., String_80, etc.) could be made visible to the examples without having to explicitly mention PLplot_Auxiliary in the examples. In sum, there should be no fundamental need to mention PLplot_Auxiliary in the examples but this appears to be currently required because of namespace issues. Getting such namespace concerns properly dealt with is always a bit of a struggle in any given language, but I hope you are able to figure this out for Ada with help from your Ada contacts/references. 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 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 __ - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] TeXinfo patch
On 2007-07-03 14:14-0600 Orion Poplawski wrote: Alan W. Irwin wrote: [...] I am inclined to go with your changes except keeping the overall title as is. Here's what I'm trying to produce, an info entry that looks like: * PLplot: (plplotdoc). The PLplot Plotting Library. rather than the current: * The PLplot Plotting Library: . ??? For comparison, most info entries look like: * URL: (url). URL loading package. The first item is the node name, the second is the filename, the third is the description. Without changing the title, I get: * The PLplot Plotting Library: (plplotdoc). The PLplot Plotting Library The problem with that is that info plplot won't bring you to the plplot page. I have committed the following change: Index: plplotdoc.xml.in === --- plplotdoc.xml.in(revision 7740) +++ plplotdoc.xml.in(working copy) @@ -238,6 +238,13 @@ book id=plplot-docbook bookinfo titleThe PLplot Plotting Library/title +titleabbrev role=texinfo-fileplplotdoc/titleabbrev +abstract role=texinfo-node + paraThe PLplot Plotting Library/para +/abstract +subjectset scheme=texinfo-directory + subjectsubjecttermScientific Visualization/subjectterm/subject +/subjectset abbrevDocBook/abbrev subtitleProgrammer's Reference Manual/subtitle releaseinfoVersion 5.0/releaseinfo i.e., your patch with the overall title left as is. But I am not getting the menu entry * The PLplot Plotting Library: (plplotdoc). The PLplot Plotting Library that is expected from your above notes. Instead, when I execute info plplotdoc.info the detailed node listing under Introduction is as follows: * The PLplot Plotting Library:: * Getting a Copy of the PLplot Package:: ... Am I looking in the wrong place in the menu or is this the right place, but my old software is producing a simplified menu result? Please check this commit (and the prior one) to make sure at least the filename aspect of it finally works for you. 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 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 __ - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Error in gd.c when compiling plplot-5.7.3 on Debian/Etch
On 2007-07-04 18:01+0200 Dominique MICOLLET wrote: Hello, When I compile on a genuine Debian/Etch I get the following error gd.c: In function 'plD_init_png_Dev': gd.c:432: error: 'smooth_text' undeclared (first use in this function) gd.c:432: error: (Each undeclared identifier is reported only once gd.c:432: error: for each function it appears in.) which I supersede by moving the declaration of smooth_text outside the #ifdef..#endif blocks at lines 368 and 562 : static int smooth_text=1; #ifdef HAVE_FREETYPE static int freetype=1; #static int smooth_text=1; FT_Data *FT; #endif I guess it is not the good way... Any advice ? I am not sure, but I think the above problem has been fixed in the svn version. Could you try that please? It does change from one day (or week) to the next, but we almost always leave it in a stable state. Instructions for retrieving the svn version at http://sourceforge.net/svn/?group_id=2915 . Instructions for building (using the preferred CMake build system) at http://www.miscdebris.net/plplot_wiki/ . 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 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 __ - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Cookbook for implementing the remaining Ada examples
On 2007-07-04 09:12-0700 Alan W. Irwin wrote: If you want to help with this effort of implementing the traditional Ada examples, please let the list know exactly which ones you are working on so there is no duplication of effort. For example, I am currently working on x02a.adb. Which is now done. It took a bit of thrashing around (I didn't realize Ada rounded when converting from floating point to integer) to get exact agreement with the result for example 2 created with the C implementation, but eventually I got there. One thing revealed by this example is our Ada bindings were much too conservative about the allowed range of plcol0 indices. I changed that from 0..15 to the much more reasonable 0..255. Example 2 uses 116 different cmap0 indices, but some users may want more. I chose to limit the range in the bindings to 0..255 since that corresponds to 24 bits total for r, g, and b and is more than adequate to handle example 2. Jerry, please review and correct any strange things I did in Ada for x02a.adb. 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 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 __ - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Need help with OS X compilers
On 2007-07-13 00:31-0400 Hazen Babcock wrote: On Jul 6, 2007, at 8:27 AM, Jerry wrote: I hope there is another Mac user on the list. Here is the output from the 4.3 compilers which include Ada, C and Objective C, which output looks the same as it did when I was using the 4.2 compiler which included only Ada and C. (Ignore the fact that the path to the compiler is /usr/local/ada-4.2/bin/gcc because I just dumped the 4.3 compilers in there for a quick and dirty attempt at fixing the problem. [ 45%] Building C object drivers/CMakeFiles/aqt.dir/aqt.o /usr/local/ada-4.2/bin/gcc -Daqt_EXPORTS -fPIC -I/Users/jerry/ Documents/Programs/Ada/Code/Bindings/PLplot/plplot_svn/plplot/include -I/usr/local/plplot_build_dir -I/usr/local/plplot_build_dir/include -DHAVE_CONFIG_H -ObjC -o drivers/CMakeFiles/aqt.dir/aqt.o -c /Users/ jerry/Documents/Programs/Ada/Code/Bindings/PLplot/plplot_svn/plplot/ drivers/aqt.c cc1: error: invalid option argument '-ObjC' make[2]: *** [drivers/CMakeFiles/aqt.dir/aqt.o] Error 1 make[1]: *** [drivers/CMakeFiles/aqt.dir/all] Error 2 make: *** [all] Error 2 I've installed macada and I've managed to get this far on my OS-X box. I was hoping to be able to tell cmake (using the ccmake facility) to use /usr/bin/gcc for the C compiler. However when I set the CMAKE_C_COMPILER = /usr/bin/gcc and then configure it always gets flipped back to /usr/local/ada-4.3/bin/gcc. My suspicion is that we could get both Ada and Aquaterm to work if we could somehow suppress the behavior. Ideas? How do you set CMAKE_C_COMPILER? I have never used that CMake variable, but my interpretation of http://www.cmake.org/Wiki/CMake_Useful_Variables#Compilers_and_Tools is you can only set it using a -D option to cmake which will only work if you haven't cached anything before (i.e., you are starting fresh with no CMakeCache.txt file in your build directory). Anyhow, I assume the flipping back to /usr/local/ada-4.3/bin/gcc is cause by a cached value. Another way to set the C compiler (and associated flags) which I know works, is to set the CC environment variable (as also discussed at the above site) before invoking cmake _with no cache file_. Similarly, to set the Ada compiler, specify the ADA environment variable (see the first few lines of cmake/modules/CMakeDetermineAdaCompiler.cmake). From further along in that file you can see that it is always assumed that the Ada executable builder is going to be called gnatmake, and gnatmake is looked for in the same directory as the specified Ada compiler. Anyhow, I believe that setting the CC and ADA environment variables before running cmake with no CMakeCache.txt file in your build directory should give you full and separate control of the C compiler, the Ada compiler, and the Ada executable builder. Let me know if that is not the 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); PLplot scientific plotting software package (plplot.org); 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 __ - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Need help with OS X compilers
On 2007-07-13 22:10-0400 Hazen Babcock wrote: On Jul 13, 2007, at 1:32 AM, Alan W. Irwin wrote: On 2007-07-13 00:31-0400 Hazen Babcock wrote: I've installed macada and I've managed to get this far on my OS-X box. I was hoping to be able to tell cmake (using the ccmake facility) to use /usr/bin/gcc for the C compiler. However when I set the CMAKE_C_COMPILER = /usr/bin/gcc and then configure it always gets flipped back to /usr/local/ada-4.3/bin/gcc. My suspicion is that we could get both Ada and Aquaterm to work if we could somehow suppress the behavior. Ideas? How do you set CMAKE_C_COMPILER? I was trying to set it using the cmake text GUI interface, ccmake. I have never used that CMake variable, but my interpretation of http://www.cmake.org/Wiki/CMake_Useful_Variables#Compilers_and_Tools is you can only set it using a -D option to cmake which will only work if you haven't cached anything before (i.e., you are starting fresh with no CMakeCache.txt file in your build directory). Anyhow, I assume the flipping back to /usr/local/ada-4.3/bin/gcc is cause by a cached value. Another way to set the C compiler (and associated flags) which I know works, is to set the CC environment variable (as also discussed at the above site) before invoking cmake _with no cache file_. Similarly, to set the Ada compiler, specify the ADA environment variable (see the first few lines of cmake/modules/ CMakeDetermineAdaCompiler.cmake). From further along in that file you can see that it is always assumed that the Ada executable builder is going to be called gnatmake, and gnatmake is looked for in the same directory as the specified Ada compiler. Anyhow, I believe that setting the CC and ADA environment variables before running cmake with no CMakeCache.txt file in your build directory should give you full and separate control of the C compiler, the Ada compiler, and the Ada executable builder. Let me know if that is not the case. This worked, thanks! I was able to get both ADA and Aquaterm to compile by doing the following (assuming macada standard install): (1) Set the environment variable CC to /usr/bin/gcc. (2) Set the environment variable CXX to /usr/bin/g++. (3) Run cmake. (4) Run ccmake and enable ADA. (5) Using ccmake, set the following: (it takes a few iterations to get it to show you all these variables) CMAKE_Ada_COMPILER /usr/local/ada-4.3/bin/gcc GNAT_EXECUTABLE_BUILDER /usr/local/ada-4.3/bin/gnatmake GNAT_LIB /usr/local/ada-4.3/lib/gcc/powerpc-apple-darwin8/4.3.0/adalib/libgnat.dylib I am glad you got this to work, but I always feel it is a tricky business to use ccmake to set variables because of caching issues (as indicated by all the retries that were necessary). To make life simpler, why not set the ADA environment variable like I suggested (e.g., in this case to /usr/local/ada-4.3/bin/gcc)? That should automatically give you GNAT_EXECUTABLE_BUILDER as a bonus. I also suggest you set the CMAKE_LIBRARY_PATH environment variable so that GNAT_LIB will automatically be found. In sum, use the environment variables (I have a file that sets this up for me in my case) to make life easy. I can't get the ADA examples to work though, it looks like the library dependencies are not properly specified: otool -L ./x01a ./x01a: libplplotadad.0.0.0.dylib (compatibility version 0.0.0, current version 0.0.0) /Users/hbabcock/Documents/OpenSource/PLplot/plplot-CBS-3/src/ libplplotd.9.dylib (compatibility version 0.0.0, current version 0.0.0) /usr/local/lib/libltdl.3.dylib (compatibility version 5.0.0, current version 5.4.0) /usr/lib/libSystem.B.dylib (compatibility version 1.0.0, current version 88.1.10) /Users/hbabcock/Documents/OpenSource/PLplot/plplot-CBS-3/lib/ csa/libcsirocsa.0.dylib (compatibility version 0.0.0, current version 0.0.0) /Users/hbabcock/Documents/OpenSource/PLplot/plplot-CBS-3/lib/ nn/libcsironn.0.dylib (compatibility version 0.0.0, current version 0.0.0) /usr/local/lib/libqhull.5.dylib (compatibility version 6.0.0, current version 6.0.0) libgnat-4.3.dylib (compatibility version 0.0.0, current version 0.0.0) /usr/local/ada-4.3/lib/libgcc_s.1.dylib (compatibility version 1.0.0, current version 1.0.0) otool -L libplplotadad.dylib libplplotadad.dylib: libplplotadad.0.0.0.dylib (compatibility version 0.0.0, current version 0.0.0) /Users/hbabcock/Documents/OpenSource/PLplot/plplot-CBS-3/src/ libplplotd.9.dylib (compatibility version 0.0.0, current version 0.0.0) libgnat-4.3.dylib (compatibility version 0.0.0, current version 0.0.0) /usr/local/lib/libltdl.3.dylib (compatibility version 5.0.0, current version 5.4.0) /usr/lib/libSystem.B.dylib (compatibility version 1.0.0, current version 88.1.10) /Users/hbabcock/Documents/OpenSource/PLplot/plplot-CBS-3/lib/ csa
[Plplot-devel] Exclusions from the sys tree in our next release tarball
The command
find sys -type d |grep -v .svn
reveals the following subdirectories of sys (edited to get rid of redundancy):
sys/mac
sys/win32
sys/os2
sys/unix
sys/dos/msc
sys/dos/djgpp
sys/dos/bcc
sys/dos/emx
sys/win-tk
My judgement is sys/mac, sys/os2, sys/unix, sys/dos/msc, sys/dos/bcc,
sys/dos/emx, and sys/win-tk are of historical interest only and simply
confuse release tarball users so I have eliminated those directories from
our next release tarball by modifying the list of regex's contained in
CPACK_SOURCE_IGNORE_FILES in our top-level CMakeLists.txt. That just leaves
sys/dos/djgpp and sys/win32 from the sys tree that will currently be
included in the tarball.
I assume Andrew Roach wants to continue to support DJGPP (since there is no
CMake-based alternative) so sys/dos/djgpp should stay in unless we hear
differently from him.
Important question for Arjen (the historical maintainer of sys/win32): Do
you want to drop sys/win32 from the next release tarball now that most
(all?) windows users are using the more powerful CMake alternative?
Please go ahead and put
^${PROJECT_SOURCE_DIR}/sys/win32
in the CPACK_SOURCE_IGNORE_FILES list if your decision is to drop sys/win32.
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 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
__
-
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now http://get.splunk.com/
___
Plplot-devel mailing list
Plplot-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] MinGW/Ada testers needed.
On 2007-07-28 00:09+0200 Werner Smekal wrote: Hi Alan, I actually tested ADA with MinGW already and made the examples run (as far as I remember) - look at the plplot email communication around 13th of March 2007. I didn't test the new examples but I don't expect much troubles though. Anyway I will test it soon, since I have all installed already. Hi Werner: Sorry for forgetting about your previous MinGW check, but in any case there have been very large Ada changes since then so your new check will be most helpful. Once your new Ada/MinGW check is complete, it would be great if you put a line into http://www.miscdebris.net/plplot_wiki/index.php?title=Overview_of_the_status_on_Windows reporting the result. 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 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 __ - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Cairo Driver Family
On 2007-06-04 21:17-0400 Hazen Babcock wrote: On Jun 3, 2007, at 7:30 PM, Alan W. Irwin wrote: I have only done the most superficial tests of the built results, but I immediately noticed two issues on Linux beyond what you mentioned. (1) the 'ldd -r cairo.so' test showed the following problems: undefined symbol: cairo_ps_surface_create (./cairo.so) undefined symbol: cairo_pdf_surface_create (./cairo.so) undefined symbol: cairo_svg_surface_create (./cairo.so) It's fairly urgent to solve these problems since it excludes any testing of the pscairo, pdfcairo, or svgcairo devices on Linux. Maybe this is because I have Cairo v1.2.6 and debian testing is v1.2.4? According the cairo docs these functions have been in the library since 1.2. Hi Hazen: I am reviving this thread because I got some bad news today about libLASi (the library used by the psttf device). It has no support for font hinting. What that means is letters are misaligned relative to the pixel grid of the display device causing letters to have inconsistent shapes if antialiasing is turned off or letters having a muddy appearance with antialiasing turned on. A lot of users don't mind such issues so psttf is still going to be useful for them, but for my needs I prefer the hinting of characters to be done correctly so this evening I started evaluating the pscairo alternative for making postscript plots. I built libcairo-1.4.10 for myself since that oldstable Debian did not even have libcairo. 19 of 124 tests failed as a result of make check. I think that is an expected result since the libcairo developers probably do not have access to such an old set of system libraries for testing purposes (and probably wouldn't want to support such old versions in any case). I didn't see anything wrong in the failed tests that seemed relevant to postscript so I forged ahead. I used the latest pango stack that I had also built for myself on that machine. For the PLplot build, I built all pango-related devices except for the memory one (which I don't know how to test). All seemed well in the build after I removed the duplicates from the X11_INCLUDE_DIR list (see latest commit) that were occurring at least on my Linux system. Now for some issues I discovered with the pscairo results. (1) The bounding box is not correct. (I believe that is a known issue which you plan to fix.) (2) At least for single pages I planned to get around the bounding box issue using -dev pscairo -o - | ps2eps -l -q --ignoreBB filename.eps. but stdout (at least when invoked by -o - which is the PLplot standard for specifying stdout) does not work. In fact, if you try it with -dev pscairo you will simply get output to the filename -. (Files with that name are tough to get rid of unless you remember the special construct rm -- -.) I can get around this stdout issue for now by the following clumsy construct to fix the bounding box -dev pscairo -o temp.ps ; cat temp.ps |ps2eps -l -q --ignoreBB filename; \ rm -f temp.ps (3) There are still minor 3D plot labelling problems. For example, if you look at example 8 the y-axis labels (and possibly the z-axis labels) are oriented properly (Hazen, thanks for fixing that issue!), but they have a size that is too large. To see this I suggest you visually compare with the same plot generated with -dev psc. Now for the good news about pscairo hinting. It's perfect! Without antialiasing turned on in gv, the letters have consistent shapes because they are consistently aligned with the display device pixel grid. And once you turn on antialiasing the result looks exceptionally clean and clear. Of course you pay for the much cleaner looking pscairo results by somewhat longer plot generation times (e.g., 0.456s versus 0.359s for psttfc) and somewhat larger (e.g., 64K versus 48K for psttfc) compressed postscript file sizes. (These speed/size comparisons between pscairo and psttfc were done for one of my 3D research plots that calls plsurf3dl with options MAG_COLOR | BASE_CONT | SURF_CONT.) The relatively modest speed loss and size gain for pscairo aren't that important for my research plots, and I really like the look of the well-hinted text label results that you get with this device. Thus, I am really happy pscairo exists and my thanks to Hazen for making this possible. 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 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-devel] Cairo Driver Family
On 2007-08-05 12:02-0400 Hazen Babcock wrote: On Aug 3, 2007, at 3:54 AM, Alan W. Irwin wrote: On 2007-08-02 23:04-0700 Alan W. Irwin wrote: Now for some issues I discovered with the pscairo results. (1) The bounding box is not correct. (I believe that is a known issue which you plan to fix.) Is the problem that the bounding box does not match the size of the plot? Yes. That the plots bleed over onto subsequent pages? No. Do you have a example that demonstrates the undesired behavior? Example 10. I ran the following commands: ./x10c -dev psc -o test.psc ./x10c -dev psttfc -o test.psttfc ./x10c -dev pscairo -o test.pscairo Here are the bounding boxes from those results: [EMAIL PROTECTED] grep -i bounding test.ps* test.psc:%%BoundingBox: 32 32 572 752 test.pscairo:%%BoundingBox: 0 0 720 540 test.pscairo:%%PageBoundingBox: 0 0 720 540 test.psttfc:%%BoundingBox: 32 32 572 752 test.psttfc:%%HiResBoundingBox: 32 32 572 752 You can see devices psc and psttfc agree (and also the gv viewer confirms the bounding boxes are correct for those cases), but for the pscairo case the bounding box is correct except that the x and y coordinates of the bounding box have been switched (which causes lots of troubles for the gv viewer with part of the plot cut off). It appears pscairo always (I have checked other examples as well) sets the same bounding boxes regardless of size of the plot. For now, if you swap the x and y bounding boxes you will most likely always get it too big, but at least you won't cut off part of the plot (which is what is happening for most of the examples now with swapped x and y bounding box values). Later as a much lower priority you may want to revisit this issue and set an exact bounding box to take care of some of our example cases (such as example 3) where the plot is actually smaller than 0 0 540 720. When playing with this I found issue number 5 which is (5) pscairo ignores the -portrait flag and the -ori flag gives strange looking results with part of the plot (the titles) oriented to a new direction and the rest of the plot ignoring -ori altogether. As far as I am concerned this is a low priority, but I bring it up because it will obviously affect the bounding box once you get -portrait and -ori implemented. (2) At least for single pages I planned to get around the bounding box issue using -dev pscairo -o - | ps2eps -l -q --ignoreBB filename.eps. but stdout (at least when invoked by -o - which is the PLplot standard for specifying stdout) does not work. In fact, if you try it with -dev pscairo you will simply get output to the filename -. (Files with that name are tough to get rid of unless you remember the special construct rm -- -.) I can get around this stdout issue for now by the following clumsy construct to fix the bounding box -dev pscairo -o temp.ps ; cat temp.ps |ps2eps -l -q --ignoreBB filename; \ rm -f temp.ps I was not aware of this convention, but I'll see if that is possible. I have never looked at these internals before, but it appears to be straightforward. When your device calls plOpenFile it looks at the filename that has been set up, and if it is - it sets pls-OutFile = stdout; pls-output_type = 1; It appears to be the responsibility of the file device driver to use OutFile (see ps.c where OF is #defined to by pls-OutFile), but currently cairo.c closes that file and uses FileName instead which is at least part of the source of the trouble. Currently no device other than plmeta uses pls-output_type logic (which is only set for filename of - case in plOpenFile), but for the cairo case where you are dealing with both interactive and file devices it might be useful. (3) There are still minor 3D plot labelling problems. For example, if you look at example 8 the y-axis labels (and possibly the z-axis labels) are oriented properly (Hazen, thanks for fixing that issue!), but they have a size that is too large. To see this I suggest you visually compare with the same plot generated with -dev psc. Is this just for the 3D labels? Yes. I see no problems like this for example 1 (which is just 2D). But if you compare -dev psc and -dev pscairo results for example 8, you will see y-axis and the left-hand z-axis strings are stretched in their vertical directions and similarly for the tick mark labels for both the y axis and left-hand z axis, but the x axis and right-hand z axis (both title and tick labels) are fine. I'm not sure why 2D labels would be differently sized than 3D labels. Perhaps there is some additional logic for the 3D case that needs fixing for the y and left-hand z axis? The x- and right-hand z-axis logic seems fine. Just discovered another pscairo issue. (4) Text clipping has not yet been implemented. (The first page of example 9 shows this issue as well as some of my research plots.) Do you mean clipping the text at the edge of the plotting area? Yes. Look
Re: [Plplot-devel] Cairo Driver Family
On 2007-08-11 21:05-0400 Hazen Babcock wrote: On Aug 8, 2007, at 5:27 PM, Alan W. Irwin wrote: Just to review the status here, Hazen has fixed the stdout issue (issue 2), and a resolution problem (issue 6 discussed off list with Hazen). Here are the remaining issues that I am aware of. (1) The bounding box is still not quite correct. Hazen has fixed the major problem (swapping of X and Y coordinates of the bounding box) so at least the bounding box is no longer smaller than the actual plot either coordinate. What remains to be done is to shrink the somewhat too large size of the current bounding box to the actual extent of the plotting data for each individual case. This is a standard postscript issue so libcairo may have a solution for it (i.e., returning the maximum x and y ranges of the plotting surface that was used), but what we have now is acceptable for most uses. The cairo driver plots look the same across different types of output. I don't think this will be easy to change since this consistency is one of the goals of the cairo project. OK. Good point. (3) There are still minor 3D plot labelling problems. For example, if you look at example 8 the y-axis labels (and possibly the z-axis labels) are oriented properly but their size is too large. To see this I suggest you visually compare with the same plot generated with -dev psc. (4) Text clipping has not yet been implemented. (The first page of example 9 shows this issue.) I'm pretty sure that none of the drivers I've nominally been responsible for (aqt, svg and cairo) implement text clipping. I'll see what I can do... (5) pscairo ignores the -portrait flag and the -ori flag gives strange looking results with part of the plot (the titles) oriented to a new direction and the rest of the plot ignoring -ori altogether. The driver itself ignores both flags. Unlike what I said before, it turns out I do need the -portrait option for one of my research plots so that is a showstopper for my particular case. From a general perspective, though, issues 3 (which by some chance does not affect me for the 3D angles I happen to be using for my research plots) and issues 4 are probably more important than the -portrait issue. Issue (1) should have a much lower priority since it is in the would be nice category. Should I continue to address these issues rather than the next release? Whatever course is more efficient for you is probably the one you should follow. Whatever you decide is fine with me. 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 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 __ - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Using OS X Skim to view PLplot PS files
On 2007-08-14 04:53-0700 Jerry wrote: With the help of the Skim list (and one Christiaan Hofman), we might have figured out why Skim (the PDF-PS reader) isn't reloading changed files. Recall that the initial motivation for automatic reloading was use Skim as a viewer for TeX. Since TeX can write to the file over some period of time, they look for %%EOF to determine if the file is fully written. PLplot does not terminate a PS file that way. Here is his comment: On Aug 14, 2007, at 4:42 AM, Christiaan Hofman wrote: That explains it. We check for %%EOF to find out if the file has been fully written. Otherwise we have no way to know if a file is complete (and tex writes incrementally, so loading too earl;y can lead to a crash). Christiaan Andrew, I have just discovered that -dev psttfc and -dev pscairo put the %%EOF postscript command at the end of the postscript file and -dev psc does not. Could you fix that psc problem, please? Once you do so, I am virtually positive that Jerry's troubles with Skim will be over. Jerry, to answer your previous question about which device you are using, the only ways you can choose the device is with the command-line -dev option or interactively from a menu, or with the plsdev command. So you must be doing one of those with either the psc, psttfc, or pscairo devices. The latter two require extra libraries which you probably don't have so I am virtually positive you are using the psc device on those grounds alone, and the lack of %%EOF for that device which I have asked Andrew to fix pretty much clinches it. 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 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 __ - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
[Plplot-devel] -DHAVE_NUMPY=OFF does not work for svn trunk
Hi Andrew: I just tried the following cmake options with a fresh svn update: -DBUILD_TEST=ON -DHAVE_NUMPY=OFF cmake (starting from an empty build tree) and make results seemed fine (although I can pass them on to you if you need them). However, ctest had a python/Numeric error as a result of these options. Here is the complete message. [EMAIL PROTECTED] ctest --verbose --tests-regex 'python' Start processing tests Test project /home/software/plplot_cvs/HEAD/build_dir Constructing a list of tests Done constructing a list of tests Changing directory into /home/software/plplot_cvs/HEAD/build_dir/test 5/ 11 Testing examples_python Test command: /bin/bash -c EXAMPLES_DIR=/home/software/plplot_cvs/HEAD/build_dir/examples\ SRC_EXAMPLES_DIR=/home/software/plplot_cvs/HEAD/plplot_cmake/examples\ PLPLOT_LIB=/home/software/plplot_cvs/HEAD/plplot_cmake/data/\ ./plplot-test.sh\ --device=psc\ --front-end=python Testing front-end python Traceback (most recent call last): File /home/software/plplot_cvs/HEAD/build_dir/examples/python/x01, line 36, in ? import xw01 File /home/software/plplot_cvs/HEAD/build_dir/examples/python/xw01.py, line 3, in ? from plplot_py_demos import * File /home/software/plplot_cvs/HEAD/build_dir/examples/python/plplot_py_demos.py, line 4, in ? from RandomArray import random ImportError: No module named RandomArray -- Process completed ***Failed 0% tests passed, 1 tests failed out of 1 The following tests FAILED: 5 - examples_python (Failed) Errors while running CTest This bug in our traditional Python/Numeric interface may affect 5.7.4 as well so there is probably some urgency to finding the fix. 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 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 __ - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
Re: [Plplot-devel] Cairo Driver Family
On 2007-08-15 21:41-0400 Hazen Babcock wrote: On Aug 8, 2007, at 5:27 PM, Alan W. Irwin wrote: (5) pscairo ignores the -portrait flag and the -ori flag gives strange looking results with part of the plot (the titles) oriented to a new direction and the rest of the plot ignoring -ori altogether. It looks like the ps driver supports this option but the xwin driver does not. Should all of the cairo drivers have a portrait option? Or is this only considered useful for file devices? Actually, I think it is only relevant to devices such as pscairo (and possibly pdfcairo?) which are historically related to paper output. I guess some LCD display devices have a portrait mode as well (where the whole LCD is twisted by 90 degrees to make it tall and skinny). When you create a plot in portrait mode with the ps driver the x and y pages sizes are not interchanged so the plot comes out looking a little strange, i.e. somewhat tall and narrow. Is this the Right Thing To Do? Yes, that is the whole point. The usual postscript landscape mode that PLplot produces is good for plots which need more space in the X direction, and all our examples are designed with that constraint in mind. But suppose you reorganize example 1 (say) so that the four subplots are stacked in a vertical column so you need more Y space than X space. (For my research plot that needs portrait mode, my subplots on the page are stacked in just such a vertical column.) Then portrait mode (by definition tall and narrow) gives best results for that tall and narrow case. But I do agree that our landscape examples do look badly proportioned (just as expected) in portrait mode. Or should portrait mode simply be a 90 degree rotation of the plot? That is part of it as you can see from ps.c, but you also need to set pls-freeaspect = 1; which acts the same as if the user had specified -freeaspect on the command line. I suggest you get the -ori option working first (so plotted results rotate properly). Once -ori 0, -ori 1, -ori 2, etc., work, then for fun try -ori 0.3 (i.e., non-integer -ori values). You will find a long-standing PLplot bug for that case which doesn't give correct plotted results (the rectangular viewport gets sheared into a parallogram) for non-integer -ori values if the aspect ratio is different from unity. Since the problem disappears for a unity aspect ratio, I think the x length is being taken rather then the appropriate y length or vice versa somewhere deep in the rotation code. But nobody has ever been able to track down the incorrect logic (which is, of course, a very low priority since integer -ori values are the ones that tend to be used). Once -ori works for integer values, then you should confirm that the -a option works as well. In other words, your device should give correct results for different aspect ratios passed to the device from the core PLplot library. After that confirmation, portrait mode should be a snap since pls-freeaspect = 1 tells the PLplot core library to adjust the aspect ratio appropriately (i.e., tall and skinny) for portrait mode. I hope I have clarified things for you. 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 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 __ - This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now http://get.splunk.com/ ___ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
