-
Notifications
You must be signed in to change notification settings - Fork 29
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
libReg.a fails to link to Boost in Ubuntu 20.04 #1037
Comments
However, I think the problem is with
|
We do install It is arguably a problem with the Boost version? I don't think so because I'm building with Ubuntu 18.04 should have 1.65, but in some messages during the build I could see 1.36 (STIR's CMake output during config) and 1.48, which buffles me. |
great
We require 1.36, which is probably what you see. Maybe STIR requires 1.48. not sure. Of course, I think we likely need more recent boost... Not sure what's going on. I guess you'' find out if we really use the |
I wonder now if this is because the Boost version was more recent than the CMake version (as discussed in SyneRBI/SIRF-SuperBuild#667). Which CMake and Boost version was this using? |
@KrisThielemans is this still relevant? |
yes. In fact, GHA is on I first thought we could merge db4479d via a PR to simplify our files, but I now think that this code could cover the case where CMake is older than Boost (in which case Of course, I have no idea what happened when you had this problem, so there might be some problem lurking somewhere, but finding out is going to be tough. Summary : I'm fine with closing this issue. |
* implemented basic-functionality listmode data class in C++ and Python * added objective function type for lismode reconstruction * introduced abstract base class PETScanData for all raw data objects * [ci skip] interfaced PETScanData into Python * Minimal backbone for LM recons * New test for LM data (test6) * fix * Working test6 * Generalize the paths in test6 * Minor renaming * simplify link library to Boost closes #1037 * remove dangling endif * [ci skip] moved data folder TBPET out of SIRF repo * removed unused stuff from cstir_test6.cpp * restored backward compatibility of run_test6.cpp * implemented C interface for list mode reconstruction objective function * started interfacing reconstruction from listmode into Python * implemented minimal Python interface for reconstruction from listmode data * resolved some issues raised by Kris * small amendments in STIR.py * removed cstir_test6 from ctest tests * minor fixes to listmode classes * removed some unneeded data_sptr() methods * replaced cache_path quick fix with a more proper handling * [ci skip] corrected copyrights in cSTIR/tests/test6.cpp * make listmode recon compatible with STIR 5.1.0 * rename PETScanData to ScanData(Python) or STIRScanData (C++) This is consistent with our renaming of AcquisitionData * No longer derive ListmodeData from DataContainer * update listmode test (WIP) - update to recent SIRF - use sample data in STIR - still quite slow as it's precomputing the sensitivity * Derive DataContainer from ContainerBase This allows us to check the type of a container using standard C++ RTTI. * speed-up listmode recon test and use SIRF example data we had some listmode example data in SIRFdata already, so use that. reduce segment etc such that the test doesn't take too long * attended to Codacy issues * attended to further Codacy issues * fix ListmodeData hierarchy and add get_info to STIR containers - derive ListmodeData from ContainerBase - add ListmodeData to C wrappers - add get_info() to STIRImageData and ListmodeData - expand STIRAcquisitionData::get_info() to add exam-info string - minimal documentation This now works fine through Python, but demos still need work. * renamed ListmodeData to STIRListmodeData for consistency * add set_acquisition_model to listmode obj-fun * minor doc corrections * listmode recon improvements - add ListmodeData::acquisition_data_template() - let ListmodeToSinograms::set_input() cope with either filename or ListmodeData object * minor fix in demo * fix storing filename of listmode file * add SIRF_DATA_PATH to STIR C++ test * fixed bugs and issues found in lmrecontest2 * [ci skip] got rid of gpu stuff in reconstruct_from_listmode.py * [ci skip] removed osem_lm_reconstruction.py, superseded by listmode_reconstruction.py * [ci skip] updated CHANGES.md --------- Co-authored-by: Nikos Efthimiou <[email protected]> Co-authored-by: Edoardo Pasca <[email protected]> Co-authored-by: Nikos Efthimiou <[email protected]> Co-authored-by: Kris Thielemans <[email protected]> Co-authored-by: Kris Thielemans <[email protected]>
In SyneRBI/SIRF-SuperBuild#649 it seems there is a linker error for
libReg.a
Apparently boost linking is taken care of by
SIRF/src/Registration/cReg/CMakeLists.txt
Lines 76 to 87 in 6c0b1fe
The current
if
statement has been merged in long ago, before we used CMake 3.5.Given that now we require CMake 3.9 I believe the
if
is not useful, probably harmful. I suggest to remove them in the 3 places where this is done.The text was updated successfully, but these errors were encountered: