[ET Trac] [Einstein Toolkit] #1910: ExternalLibraries/HDF5 does not set HDF5_INC_DIRS
Einstein Toolkit
trac-noreply at einsteintoolkit.org
Mon Jun 13 08:41:10 CDT 2016
#1910: ExternalLibraries/HDF5 does not set HDF5_INC_DIRS
----------------------------+-----------------------------------------------
Reporter: barry.wardell | Owner: eschnett
Type: defect | Status: assigned
Priority: unset | Milestone:
Component: Other | Version: development version
Resolution: | Keywords:
----------------------------+-----------------------------------------------
Comment (by knarf):
Replying to [comment:16 barry.wardell]:
> On my machine, hdf5 is installed in ```/usr/local```. I have traced the
evaluation through build.sh and bash_utils.sh. The library is found by
[https://bitbucket.org/cactuscode/cactus/src/01d8937b1fdd8a31060aeb4e387bc97bf5eeefe2/lib/make/bash_utils.sh?at=master&fileviewer
=file-view-default#bash_utils.sh-192 line 192 of bash_utils.sh] using
find_standardlib. This correctly calls set_make_vars, but then
[https://bitbucket.org/cactuscode/cactus/src/01d8937b1fdd8a31060aeb4e387bc97bf5eeefe2/lib/make/bash_utils.sh?at=master&fileviewer
=file-view-default#bash_utils.sh-63 strip-incdirs.sh and strip-libdirs.sh]
commands remove the ```/usr/local/include``` and ```/usr/local/lib``` that
are required (note that these are not part of the standard paths in the OS
X build system).
Ah, this then would be the problem. Cactus assumes paths like
/usr/local/... to be correctly set up and usable without specific paths
and thus removes those (has to). I cannot think of an easy and generic
check whether this is the case. A check whether the environment is OSX
would be far easier, but an actual OSX developer would need to implement
and test that.
--
Ticket URL: <https://trac.einsteintoolkit.org/ticket/1910#comment:17>
Einstein Toolkit <http://einsteintoolkit.org>
The Einstein Toolkit
More information about the Trac
mailing list