From: Maurice L. <ml...@us...> - 2002-12-26 20:03:23
|
Update of /cvsroot/plplot/plplot/src In directory sc8-pr-cvs1:/tmp/cvs-serv14138 Modified Files: plcore.c Log Message: Added a cast from (void *) to (PLDispatchInit) to satisfy the picky IRIX C compiler. |
From: Alan W. I. <ai...@us...> - 2003-01-15 15:56:44
|
Update of /cvsroot/plplot/plplot/src In directory sc8-pr-cvs1:/tmp/cvs-serv32270/src Removed Files: base.c convert.c genlin.c move.c pladv.c plaxes.c plbeg.c plbin.c plbox3.c plccal.c plcntr.c plcol.c plconf.c plctest.c plend.c plenv.c plerr.c plexit.c plfont.c plform.c plgra.c plgrid3.c plgspa.c pljoin.c pllab.c pllclp.c pllsty.c plmesh.c plmtex.c plnxtv.c plpat.c plpsty.c plptex.c plrot.c plside3.c plstar.c plstream.c plstring.c plstyl.c plsvpa.c plt3zz.c plvasp.c plvpas.c plvsta.c plw3d.c plwid.c plxybx.c plzbx.c string.c Log Message: Attempt to fix state information on zombied files (those in attic with state exp rather than state dead). |
From: Rafael L. <rla...@us...> - 2003-02-02 14:41:02
|
Update of /cvsroot/plplot/plplot/src In directory sc8-pr-cvs1:/tmp/cvs-serv498/src Modified Files: plcore.c Log Message: $LIB_TAG is no more added to the filenames of drivers objects <driver>_drv.so. Instead, only the drivers directory gets its name changed ($LIB_DIR/$DATA_DIR/../drivers$LIB_TAG/). To make things even more maintainable and robust, a new AC_SUBST variable has been created in configure.ac: DRV_DIR. It is used in src/plcore.c instead of the hard-code string "../drivers". Of course, this changes does not preclude simultaneous installation of both single and double precision version of PLplot (although I wonder who on earth besides Maurice needs that!) Besides being totally transparent to both users and developers, this changes has a beneficial side effect: it reduces the complexity of the build scheme of driver objects and drivers/Makefile.am becomes 20% smaller. This change has been quickly tested in all the combinations of with- vs. without-double and enable- vs. disable-dyndrivers. More extensive tests need to be done, though. |
From: Rafael L. <rla...@us...> - 2003-02-04 21:44:41
|
Update of /cvsroot/plplot/plplot/src In directory sc8-pr-cvs1:/tmp/cvs-serv30070/src Modified Files: plcore.c Log Message: Change the way the information about the devices provided by the dyndrivers are obtained. Instead of using the cache file drivers.db, the information is provided by the modules themselves. Extensive information about this change will be sent to plplot-devel shortly. |
From: Alan W. I. <ai...@us...> - 2003-02-05 14:34:16
|
Update of /cvsroot/plplot/plplot/src In directory sc8-pr-cvs1:/tmp/cvs-serv23875 Added Files: plbuf.c Log Message: Moved this file from ../drivers (version 1.36) as per discussion on list. It is not a driver in exactly the same sense as the rest of the drivers in ../drivers, and it's compiled object goes directly into libplplot (unlike the rest of the drivers) so this directory (src) where libplplot is created seems the better location. |
From: Alan W. I. <ai...@us...> - 2003-02-05 14:36:22
|
Update of /cvsroot/plplot/plplot/src In directory sc8-pr-cvs1:/tmp/cvs-serv26037 Modified Files: Makefile.am Log Message: Tweak to accommodate move of plbuf.c from ../drivers to this directory (src). |
From: Rafael L. <rla...@us...> - 2003-02-05 20:17:56
|
Update of /cvsroot/plplot/plplot/src In directory sc8-pr-cvs1:/tmp/cvs-serv23733/src Modified Files: plcore.c Log Message: Changed DEVICE_INFO_<driver> to plD_DEVICE_INFO_<driver> as per Maurice's suggestion. |
From: Rafael L. <rla...@us...> - 2003-02-07 08:38:59
|
Update of /cvsroot/plplot/plplot/src In directory sc8-pr-cvs1:/tmp/cvs-serv29605/src Modified Files: plcore.c Log Message: Commented out the call to lt_dlclose, since this is apparently the source of the segmentation fault problem. Actually, it is not lt_dlclose itself that causes the error, but the fact that there are memory leaks elsewhere in the code. When lt_dlclose is called, those memory management problems make surface. Notice that in Geoffrey's original dyndrivers code dlclose was never called, so the bug never manifested itself. (In other words: the bug was already in the previous released version of PLplot, and have not been introduced by my recent changes.) Of course, this is an interim solution. This should make PLplot work properly again, though. |
From: Rafael L. <rla...@us...> - 2003-02-11 07:37:18
|
Update of /cvsroot/plplot/plplot/src In directory sc8-pr-cvs1:/tmp/cvs-serv20574/src Modified Files: plcore.c Log Message: This is a first try on the new design to get device information dynamically from <driver>.rc files placed in the drivers directory. Those files are generated at build time by get-drv-info and transfered to the drivers directory at install time. One good thing about this design is that it avoids dlopening of driver modules and consequently the unnecessary loading of libraries. I did minimal tests with this new code, but there should be hidden bugs. Also, this stuff is pretty uncommented and there is almost no error sanity code. I will take care of these problems, eventually. |
From: Rafael L. <rla...@us...> - 2003-02-19 17:35:01
|
Update of /cvsroot/plplot/plplot/src In directory sc8-pr-cvs1:/tmp/cvs-serv16832/src Modified Files: Makefile.am Log Message: Changed the instances of @VARIABLE@ to $(VARIABLE) in all Makefile.am's, as I proposed in plplot-devel. This huge change has been successfully tested with my automatic Debian packages generator. Hope that nothing is broken now. If you are wondering how I got it so quickly and so reliably, here is the simple shell command that does the trick: perl -pi -e 's/@([-_a-zA-Z0-9]+)@/\$($1)/g' `find . -name Makefile.am` [I love Perl.] |
From: Rafael L. <rla...@us...> - 2003-02-20 11:17:45
|
Update of /cvsroot/plplot/plplot/src In directory sc8-pr-cvs1:/tmp/cvs-serv5695 Modified Files: plcore.c Log Message: Joao has continuously complained that with the current PLplot in CVS it is not possible to use the library without doing make install. He is right. One of the reasons for this is that the driver modules are looked for only under the absolute path DATA_DIR, which is actually a installation path. I changed the code of plInitiDispatchTable in plcore.c to load the drivers from the directory pointed by the env var PLPLOT_DRV_DIR. If this env var is not defined, it uses the default DATA_DIR. This is a interim solution that will allow test of PLplot (at least the access to driver modules) without a full install. There is also the problems of data files and fonts. Some of these used to be accessed in the past via the PLPLOT_LIB environment variable. This has to be sorted out and I would like to find a better global solution for this problem of dual access (devel vs. install locations). Another change in the code was the inclusion of some tests and plabort calls. |
From: <jc...@fe...> - 2003-02-20 15:39:24
|
On Thursday 20 February 2003 11:17, Rafael Laboissiere wrote: | Update of /cvsroot/plplot/plplot/src | In directory sc8-pr-cvs1:/tmp/cvs-serv5695 | | Modified Files: | =09plcore.c | Log Message: | | Joao has continuously complained that with the current PLplot in CVS | it is not possible to use the library without doing make install. He | is right. One of the reasons for this is that the driver modules are | looked for only under the absolute path DATA_DIR, which is actually a | installation path. | | I changed the code of plInitiDispatchTable in plcore.c to load the | drivers from the directory pointed by the env var PLPLOT_DRV_DIR. If | this env var is not defined, it uses the default DATA_DIR. This is a | interim solution that will allow test of PLplot (at least the access | to driver modules) without a full install. | | There is also the problems of data files and fonts. Some of these | used to be accessed in the past via the PLPLOT_LIB environment | variable. This has to be sorted out and I would like to find a | better global solution for this problem of dual access (devel vs. | install locations). In my sources, I use a PLplot global variable called build_tree that=20 says if the program being run was invoked inside the build tree(*); if=20 it was , then the path of the searched file in the build tree is used,=20 else the usual search path is used. This works fine for most drivers,=20 fonts and map files. The tk driver is an exception, as it needs to find plserver, and tcl/tk=20 also needs its auto_path to be setup; in my sources this is also=20 working fine.=20 However, I don't like the (tk) solution, as it is not general. There is also the problem of having a global variable, build_tree, which=20 is ugly. Joao *-build_tree is set getting the current working dir, getcwd(), and=20 comparing with BUILD_TREE, which is setup at configure time and is=20 defined in config.h=20 | | Another change in the code was the inclusion of some tests and | plabort calls. | | | | | ------------------------------------------------------- | This SF.net email is sponsored by: SlickEdit Inc. Develop an edge. | The most comprehensive and flexible code editor you can use. | Code faster. C/C++, C#, Java, HTML, XML, many more. FREE 30-Day | Trial. www.slickedit.com/sourceforge | _______________________________________________ | Plplot-cvs mailing list | Plp...@li... | https://lists.sourceforge.net/lists/listinfo/plplot-cvs |
From: Rafael L. <rla...@us...> - 2003-02-21 10:31:19
|
Update of /cvsroot/plplot/plplot/src In directory sc8-pr-cvs1:/tmp/cvs-serv30505 Modified Files: plcore.c Log Message: Oops, I left a debugging printf behind. Fixed now. |
From: Rafael L. <rla...@us...> - 2003-02-21 18:22:37
|
Update of /cvsroot/plplot/plplot/src In directory sc8-pr-cvs1:/tmp/cvs-serv11039 Modified Files: plcore.c Log Message: Make plLoadDriver honor the PLPLOT_DRV_DIR environment variable when loading drivers, in the same way plInitDispatchTable does when scanning the drivers directory. |
From: Rafael L. <rla...@us...> - 2003-02-23 11:25:30
|
Update of /cvsroot/plplot/plplot/src In directory sc8-pr-cvs1:/tmp/cvs-serv22607/src Modified Files: Makefile.am Log Message: Completed the elimination of with_double conditionals in the Makefile.am's with regard to LIB_TAG. In my first try, I have used $(LIB_TAG) in Autonmake/Libtool variable names (like libplplot$(LIB_TAG)_la_SOURCES). That did not work. However, I was glad to discover that the last version of Automake accepts the construct @VAR@ in its variable names. I used then @LIB_TAG@ where it was appropriate. I know, I recently exhorted the developers to use $(VAR) instead of @VAR@ in Makefile.am. However, since this is not allowed in Automake variable names and since there is a reduction of 80 lines total (improving robustness and maintainability), I will open the exception to @LIB_TAG@. |
From: Rafael L. <rla...@us...> - 2003-02-24 00:32:15
|
Update of /cvsroot/plplot/plplot/src In directory sc8-pr-cvs1:/tmp/cvs-serv6711/src Modified Files: plcore.c Log Message: Added Doug Hunt's mem driver. |
From: Rafael L. <rla...@us...> - 2003-02-24 17:40:10
|
Update of /cvsroot/plplot/plplot/src In directory sc8-pr-cvs1:/tmp/cvs-serv31183/src Modified Files: plfreetype.c Log Message: Added "#define FT_Data _FT_Data_" to this files, otherwise they won't compile with freetype6. It seems that the FreeType library already defines the symbol FT_Data. |
From: Rafael L. <rla...@us...> - 2003-02-25 10:42:17
|
Update of /cvsroot/plplot/plplot/src In directory sc8-pr-cvs1:/tmp/cvs-serv29608/src Modified Files: Makefile.am Log Message: This is a revival of the --with-rpath option to configure, which has been ripped out from sysloc.in when the AT branch has been created. I need this to build Lintian clean packages in Debian. Setting -rpath may be a handy thing for users who do not have root access, but enforcing it by default is not a Good Thing to do. In particular, Debian Policy enforces the non-use of rpath for upgradability reasons. This changes is transparent and harmless, and the PLplot build will happens as before, since with_rpath is set to "yes" by default. Please, do not undo this change in the future! |
From: Joao C. <jc...@us...> - 2003-03-01 18:27:38
|
Update of /cvsroot/plplot/plplot/src In directory sc8-pr-cvs1:/tmp/cvs-serv2796/src Modified Files: plctrl.c Log Message: Moved from "lib" dir to "data" dir |
From: Joao C. <jc...@us...> - 2003-03-01 18:52:41
|
Update of /cvsroot/plplot/plplot/src In directory sc8-pr-cvs1:/tmp/cvs-serv14849 Modified Files: plctrl.c Log Message: Oops, reverting to version 1.70 |
From: Joao C. <jc...@us...> - 2003-03-02 01:43:43
|
Update of /cvsroot/plplot/plplot/src In directory sc8-pr-cvs1:/tmp/cvs-serv14872/src Modified Files: Makefile.am Log Message: Add plgridd.c that implements plgriddata(). Attempt to specify the dependencies of libplplot on libnn amd libcsa. |
From: Joao C. <jc...@us...> - 2003-03-02 01:47:32
|
Update of /cvsroot/plplot/plplot/src In directory sc8-pr-cvs1:/tmp/cvs-serv16107/src Modified Files: plvpor.c Log Message: plenv(): The just argument now supports a square plot box. The axis argument now supports a *3 specification that draws minor tick marks. |
From: Joao C. <jc...@us...> - 2003-03-02 02:19:40
|
Update of /cvsroot/plplot/plplot/src In directory sc8-pr-cvs1:/tmp/cvs-serv24554/src Added Files: plgridd.c Log Message: Add plgridd.c that implements plgridadata() |
From: Joao C. <jc...@us...> - 2003-03-02 23:03:28
|
Update of /cvsroot/plplot/plplot/src In directory sc8-pr-cvs1:/tmp/cvs-serv2038/src Modified Files: plgridd.c Log Message: Surround nn and csa code with #ifdef HAVE_QHULL and #ifdef PL_DOUBLE |
From: Joao C. <jc...@us...> - 2003-03-03 02:28:35
|
Update of /cvsroot/plplot/plplot/src In directory sc8-pr-cvs1:/tmp/cvs-serv1130/src Modified Files: plgridd.c Log Message: Use c_plgriddata() instead of plgriddata() Enable the use of nn/csa even if plplot was configured with floats. Verify that Qhull was compiled with doubles. |