Opened 6 years ago
Last modified 6 years ago
#57766 closed defect
py-graph-tool uses incorrect (default) python version 2.7, propagates to dependencies — at Version 2
Reported by: | ssllmit (ss) | Owned by: | |
---|---|---|---|
Priority: | Normal | Milestone: | |
Component: | ports | Version: | 2.5.4 |
Keywords: | Cc: | mamoll (Mark Moll) | |
Port: | py-graph-tool |
Description (last modified by ssllmit (ss))
py-graph-tool
uses the incorrect python version, default to v. 2.7 in its port file, which results in incorrect dependencies during the build.
First, clean everything and set python
to use v. 3.6:
sudo port clean --all py-graph-tool, sudo port select --set python python36 hash -r
Confirm:
$ python --version Python 3.6.7
But the correct python version dependency doesn't propagate:
$ port list depof:py-graph-tool py27-graph-tool @2.27 python/py-graph-tool
This results in boost
being compiled and installed with an incorrect python version:
$ sudo port -pN install py-graph-tool ---> Computing dependencies for py-graph-tool ---> Dependencies to be installed: py27-graph-tool boost cgal ---> Fetching archive for boost ---> Attempting to fetch boost-1.66.0_3+no_single+no_static+python27.darwin_17.x86_64.tbz2 from https://packages.macports.org/boost
Even an install with an explicit python version (v. 3.6) gets the wrong python version (default v. 2.7):
sudo port -pN install py36-graph-tool ---> Computing dependencies for py36-graph-tool ---> Dependencies to be installed: boost cgal ---> Fetching archive for boost ---> Attempting to fetch boost-1.66.0_3+no_single+no_static+python27.darwin_17.x86_64.tbz2 from https://packages.macports.org/boost
This issue appears to go beyond py-graph-tool
because boost
also shows the same dependency issue:
port list depof:boost bzip2 @1.0.6 archivers/bzip2 expat @2.2.6 textproc/expat icu @58.2 devel/icu libiconv @1.15 textproc/libiconv python27 @2.7.15 lang/python27 zlib @1.2.11 archivers/zlib
Admittedly, this may be a problem with my configuration, but I did my due diligence and cannot identify any local problems.
FWIW, I'm trying to track down a segmentation fault with gt.graph_draw
that I hope is related to this build problem…
gt.graph_draw(g) Segmentation fault: 11
Change History (2)
comment:1 Changed 6 years ago by ssllmit (ss)
Description: | modified (diff) |
---|
comment:2 Changed 6 years ago by ssllmit (ss)
Description: | modified (diff) |
---|