UPDATE: py-numpy 0.9.6 and py-scipy04 0.4.8

[erickt@…]

Version bump for these two projects. Markus, could you upload these while I wait for commit approval?

Thanks again!

-e

Diff for py-scipy04:

--- ../../../rsync.rsync.darwinports.org_dpupdate_dports/python/py-scipy04/Portfile     2006-04-14 01:03:31.000000000 -0700
+++ Portfile    2006-04-19 17:01:37.000000000 -0700
@@ -1,10 +1,11 @@
-# $Id: Portfile,v 1.1 2006/04/14 08:03:31 blb Exp $
+# $Id: Portfile,v 1.1 2005/01/06 15:07:03 mww Exp $
 
 PortSystem 1.0
 PortGroup  python24 1.0
 
 name         py-scipy04
-version      0.4.6
+version      0.4.8
+revision     1
 categories   python
 platforms    darwin
 description  An opensource library of scientific tools for Python
@@ -14,7 +15,7 @@
 master_sites  sourceforge:scipy
 distname      scipy-${version}
 
-checksums md5 cdf236f1d93fca286d27a7c10b6fe446
+checksums md5 854db037c56a0a359f877e8f852925ea
 
 depends_lib    lib:gnuplot:gnuplot \
                port:py-numpy

diff for py-numpy:

--- ../../../rsync.rsync.darwinports.org_dpupdate_dports/python/py-numpy/Portfile       2006-03-07 10:30:43.000000000 -0800
+++ Portfile    2006-04-19 18:47:40.000000000 -0700
@@ -1,10 +1,9 @@
-# $Id: Portfile,v 1.1 2006/03/07 18:30:43 mww Exp $
-
 PortSystem 1.0
 PortGroup  python24 1.0
 
 name        py-numpy
-version     0.9.5
+version     0.9.6
+revision    1
 categories  python
 platforms   darwin
 maintainers erickt@dslextreme.com
@@ -15,10 +14,10 @@
 master_sites  sourceforge:numpy
 distname      numpy-${version}
 
-checksums md5 47fe186078c5e1f68b37e0ffc4994178
+checksums md5 eac9ae05bd7511cecf1ffd12da0e1993
 
-depends_build  port:gcc41
-depends_lib-append     lib:fftw:fftw-3 
+depends_build       port:gcc41
+depends_lib-append lib:fftw:fftw-3 
 
 patchfiles patch-gnu.py \
            patch-system_info.py

[erickt@…]

Also, it looks like the py-numpy patch never got updated to change it to use gcc 4.1. This patch should fix it.

--- ../../../rsync.rsync.darwinports.org_dpupdate_dports/python/py-numpy/files/patch-gnu.py     2006-03-07 10:30:43.000000000 -0800
+++ files/patch-gnu.py  2006-04-19 19:04:20.000000000 -0700
@@ -16,7 +16,7 @@
      # Debian: GNU Fortran 95 (GCC 4.0.3 20051023 (prerelease) (Debian 4.0.2-3))
  
 -    for fc_exe in map(find_executable,['gfortran','f95']):
-+    for fc_exe in map(find_executable,['gfortran-dp-4.0','f95']):
++    for fc_exe in map(find_executable,['gfortran-dp-4.1','f95']):
          if os.path.isfile(fc_exe):
              break
      executables = {

[erickt@…]

oh and that last patch was for py-numpy

[markd@…]

In light of this pending update, it looks like #4233 is obsolete. But if py-scipy is broken, shouldn't py-scipy and py-scipy04 be merged?

[markd@…]

Well numpy is at 0.9.8 and scipy 0.4.9 before this was committed. It would be great if the patches could be updated first. But could py-scuipy and py-scipy04 be merged?

[erickt@…]

I was under the impression that scipy 0.4 broke a lot of stuff written for 0.3.2. I didn't want to overwrite the scipy package in order to prevent breaking other peoples code. I guess this should be bubbled up to include other senior level darwinports folks to see what they think.

-e

[markd@…]

Erick, what is the status of this ticket? I see you committed an update to py-scipy04 after the ticket was created, but not py-numpy nor the rest of the changes the patches in this ticket specify that I can see. What should be done with the ports at this point and also this ticket?

[markd@…]

Ports are now updated to latest.