xemacs and xemacs-sumo install/delete warnings and errors
Getting some errors and warning when pkg_add'ing/pkg_delete'ing the xemacs xemacs-sumo packages. xemacs seems to work fine though. Not sure if this is the type of testing reports that you need. teletron1:/home/dangt$ sudo pkg_add -i xemacs xemacs-sumo Ambiguous: xemacs could be xemacs-21.4.19p0 xemacs-21.4.19p0-mule Choose one package 0: None 1: xemacs-21.4.19p0 2: xemacs-21.4.19p0-mule Your choice: 1 Ambiguous: xemacs-sumo could be xemacs-sumo-21.20060510 xemacs-sumo-21.20060510-mule Choose one package 0: None 1: xemacs-sumo-21.20060510 2: xemacs-sumo-21.20060510-mule Your choice: 1 xemacs-21.4.19p0: complete install-info: warning: no info dir entry in `/usr/local/lib/xemacs/xemacs-packages/info/tempo.info' install-info: warning: no info dir entry in `/usr/local/lib/xemacs/xemacs-packages/info/tm-view-en.info' xemacs-sumo-21.20060510: complete --- xemacs-21.4.19p0 --- XEmacs is pretty useless as-is. It is just an engine. You need to install modules to provide your users full functionality. There is two methods to achieve this : 1 - install the xemacs21-sumo package. It contains all the modules currently supported by the XEmacs team. It's fast, easy and your users won't ever bother you again about this. On the other hand the package is quite huge.2 - you have to install by hand every XEmacs' modules as described in the XEmacs documentation. (Search for packages in the info tree.) There are a lot of interdependencies between XEmacs' modules and you may end up with everything installed just as method 1 would have (without the hassle). Note: XEmacs modules are named packages by the XEmacs team, the term module is used here just to limit possible confusion with OpenBSD packages. teletron1:/home/dangt$ sudo pkg_delete xemacs-sumo xemacs Password: xemacs-21.4.19p0: complete install-info: warning: no entries found for `/usr/local/lib/xemacs/xemacs-packages/info/tempo.info'; nothing deleted install-info: warning: no entries found for `/usr/local/lib/xemacs/xemacs-packages/info/tm-mh-e-en.info'; nothing deleted install-info: warning: no entries found for `/usr/local/lib/xemacs/xemacs-packages/info/tm-view-en.info'; nothing deleted install-info: warning: no entries found for `/usr/local/lib/xemacs/xemacs-packages/info/tm-vm-en.info'; nothing deleted xemacs-sumo-21.20060510: complete Clean shared items: complete --- xemacs-21.4.19p0 --- You should also remove /usr/local/info/xemacs-21.4.19/dir Error deleting directory /usr/local/lib/xemacs-21.4.19/lisp: Directory not empty Error deleting directory /usr/local/lib/xemacs-21.4.19: Directory not empty --- xemacs-sumo-21.20060510 --- Error deleting directory /usr/local/lib/xemacs/xemacs-packages/info: Directory not empty Error deleting directory /usr/local/lib/xemacs/xemacs-packages: Directory not empty Error deleting directory /usr/local/lib/xemacs: Directory not empty This is the file in /usr/local/lib/xemacs/xemacs-packages/info -rw-r--r-- 1 root wheel 838 Aug 15 00:46 /usr/local/lib/xemacs/xemacs-packages/info/dir The /usr/local/lib/xemacs-21.4.19/lisp directory has an empty mule directory teletron1:/usr/local/lib/xemacs-21.4.19/lisp/mule$ ls teletron1:/usr/local/lib/xemacs-21.4.19/lisp/mule$ I am running a current snapshot from 2006/08/13 Tan
Re: NEW: py-numpy
On 6/25/06, Damien Miller [EMAIL PROTECTED] wrote: It looks like the regress failure is a harmless problem with the test, and not the library itself. I have updated the port (attached) with the equivalent regress test from numpy's SVN -current and it passes on sparc64. Ok? Good here on i386 and sparc64. -d
Re: NEW: py-numpy
On 6/15/06, Damien Miller [EMAIL PROTECTED] wrote: Hi, This is a port of numpy, the successor and replacement for the py-Numeric library. The port includes a fix for a segv from numpy -current and an OpenBSD-specific FP exception handling fix that has been fed back to the developers Please test. Okay here on i386. On sparc64 I get this error on regress. Found 0 tests for __main__ ...F == FAIL: check_types (numpy.core.tests.test_scalarmath.test_types) -- Traceback (most recent call last): File /usr/ports/mystuff/math/py-numpy/w-py-numpy-0.9.8/fake-sparc64/usr/local/lib/python2.4/site-packages/numpy/core/tests/test_scalarmath.py, line 63, in check_types assert val.dtype.num == typeconv[k,l] and \ AssertionError: error with (0,7) -- Ran 364 tests in 7.928s FAILED (failures=1) The package will install and simple examples will work on sparc64 though. I am running a snapshot from Jun 3 on sparc64. tan -d
Re: update: vim7
On 5/10/06, Chris Kuethe [EMAIL PROTECTED] wrote: rumor has it the attachment got eaten... Ok here on sparc64, but it leaves share/vim/vim70/lang behind on deinstall. -- GDB has a 'break' feature; why doesn't it have 'fix' too?
Re: NEW: lang/chicken
On 5/6/06, Deanna Phillips [EMAIL PROTECTED] wrote:
steven mestdagh [EMAIL PROTECTED] writes:
this doesn't build on my sparc64, no meaningful error, i might
be short on memory - can anyone else try this?
Hmm, NetBSD has this in their port's Makefile:
# Avoid an ICE in gcc2 on sparc64
CONFIGURE_ENV+=F77=${FALSE:Q}
Adding this to the Makefile allowed me to build this on sparc64. The
package process went fine, but I did not try any examples.
tan
--
deanna at sdf
Re: NEW: math/plplot 5.6.0
comments? regress works on i386 and sparc64 for me also. Did not do any tests with the app though. steven Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm
Re: NEW: math/plplot
On 4/19/06, steven mestdagh [EMAIL PROTECTED] wrote: hi, here is a port of the PLplot library. seems to work okay on i386/amd64. testing/feedback appreciated... ok on i386. on sparc64 I could not get make regress to finish. Had the thing going all weekend with 100% user time. I had to kill the regress. Besides regress on sparc64, most of the port testing guide steps went fine. make deinstall did give me this error, but I think I have a local problem. $ sudo make deinstall === Deinstalling for plplot-5.5.3 Can't remove plplot-5.5.3 without also removing: py-plplot-5.5.3 *** Error code 1 Stop in /usr/ports/mystuff/math/plplot (line 2738 of /usr/ports/infrastructure/mk/bsd.port.mk). Tan Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm
Re: NEW: math/lapack
On 3/24/06, steven mestdagh [EMAIL PROTECTED] wrote: steven mestdagh [2006-03-11, 10:35:11]: LAPACK is written in Fortran77 and provides routines for solving systems of simultaneous linear equations, least-squares solutions of linear systems of equations, eigenvalue problems, and singular value problems. please test/comment. First time port testing so pardon any mistakes that I may make. Seems to work fine on i386 running snapshot from March 2. All the steps from the port testing guide seem to work. I get this message from make install though -bash-3.1$ sudo make install === Installing lapack-3.0 from /usr/ports/packages/i386/all/lapack-3.0.tgz lapack-3.0: complete Unknown manpage type /usr/local/man/manl/csrot.l Unknown manpage type /usr/local/man/manl/zbcon.l Unknown manpage type /usr/local/man/manl/zdrot.l zbcon.l is an empty file. Also I can not seem to man any of the manpages that the package installed. Was able to compile and run a C program that used lapack libraries. thanks for testing. i'm not sure what to do about those man pages yet. okay, tweaked the man pages to go into section 3. new tgz attached. comments/ok ? Looks good on i386 and sparc64 now. Tan steven Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm
Re: NEW: math/lapack
On 3/10/06, steven mestdagh [EMAIL PROTECTED] wrote: well, it's not exactly brand new, but not in our tree yet. this library can be used by several math applications. LAPACK is written in Fortran77 and provides routines for solving systems of simultaneous linear equations, least-squares solutions of linear systems of equations, eigenvalue problems, and singular value problems. please test/comment. First time port testing so pardon any mistakes that I may make. Seems to work fine on i386 running snapshot from March 2. All the steps from the port testing guide seem to work. I get this message from make install though -bash-3.1$ sudo make install === Installing lapack-3.0 from /usr/ports/packages/i386/all/lapack-3.0.tgz lapack-3.0: complete Unknown manpage type /usr/local/man/manl/csrot.l Unknown manpage type /usr/local/man/manl/zbcon.l Unknown manpage type /usr/local/man/manl/zdrot.l zbcon.l is an empty file. Also I can not seem to man any of the manpages that the package installed. Was able to compile and run a C program that used lapack libraries. Tan steven Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm
