Summary: | media-libs/vigra-1.8.0 - make[2]: *** No rule to make target `/usr/lib64/libboost_python-mt.so', needed by `vigranumpy/src/core/colors.so'. Stop. | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Denis <deniss.samcuks> |
Component: | [OLD] Printing | Assignee: | Printing Team <printing> |
Status: | RESOLVED OBSOLETE | ||
Severity: | normal | CC: | gentoo_bugs.nu_q5v |
Priority: | Normal | ||
Version: | unspecified | ||
Hardware: | AMD64 | ||
OS: | Linux | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- | |
Attachments: | media-libs:vigra-1.8.0:20140304-203030.log |
Description
Denis
2014-03-04 20:32:44 UTC
emerge -pqv '=media-libs/vigra-1.8.0::gentoo' [ebuild N ] media-libs/vigra-1.8.0 USE="fftw hdf5 jpeg png python tiff -doc -openexr {-test}" Created attachment 371738 [details]
media-libs:vigra-1.8.0:20140304-203030.log
Other versions of vigra have the same problem. #ACCEPT_KEYWORDS="~amd64" emerge -1 =media-libs/vigra-1.8.0-r1 and # ACCEPT_KEYWORDS="~amd64" emerge -1 =media-libs/vigra-1.9.0 result with the same error: make[2]: *** No rule to make target `/usr/lib64/libboost_python-mt.so', needed by `vigranumpy/src/core/colors.so'. Stop. make[2]: *** Waiting for unfinished jobs.... I've managed to fix the problem. I upgraded dev-libs/boost to masked version 1.53.0-r1 and then deleted the broken link (/usr/lib64/libboost_python-mt.so -> libboost_python-mt-1_46.so). After that I was able to compile vigra. I guess that actually removing the broken link did the trick, because downgrading boost back to 1.52.0-r6 didn't affect vigra installation. Since I have two version of python installed, /usr/lib64/libboost_python-mt.so does not seem to be correct, because each version uses its own library. /usr/lib64/libboost_python-2.7-mt.so -> libboost_python-2.7.so /usr/lib64/libboost_python-2.7.so -> libboost_python-2.7.so.1.53.0 /usr/lib64/libboost_python-2.7.so.1.53.0 /usr/lib64/libboost_python-3.2-mt.so -> libboost_python-3.2.so /usr/lib64/libboost_python-3.2.so -> libboost_python-3.2.so.1.53.0 /usr/lib64/libboost_python-3.2.so.1.53.0 I had the same problem on x86; removing dangling links /usr/lib/libboost_python.so /usr/lib/libboost_python-mt.so solved it. They were not claimed by any package, so I wonder how they ended up there. I just had the same problem when compiling sci-chemistry/avogadro-1.1.1: "No rule to make target '/usr/lib64/libboost_python-mt.so'" Removing the dangling links mentioned above (/usr/lib/libboost_python.so, /usr/lib/libboost_python-mt.so) solved the problem. On my system the links were pointing to boost versions 1.46 and had a stat modification date in September 2012 within days from an upgrade to boost-1.49. I think this bug can be closed as obsolete. (In reply to Jérôme Borme from comment #6) > I just had the same problem when compiling sci-chemistry/avogadro-1.1.1: "No > rule to make target '/usr/lib64/libboost_python-mt.so'" > > Removing the dangling links mentioned above (/usr/lib/libboost_python.so, > /usr/lib/libboost_python-mt.so) solved the problem. > > On my system the links were pointing to boost versions 1.46 and had a stat > modification date in September 2012 within days from an upgrade to > boost-1.49. I think this bug can be closed as obsolete. OK thanks. |