Let's move this off the list for a bit until we get it resolved.

Can you please do: "VERBOSE=1 make testReaction" and send the output just
to me?

Thanks,
-greg


On Wed, Jun 21, 2017 at 10:24 AM, JP <jeanpaul.ebe...@inhibox.com> wrote:

> Looks like it is finding the correct version of boost (and it is finding
> boost serialize).
>
> /opt/rdkit/rdkit-Release_2017_03_2/build$ cmake -D
> RDK_BUILD_INCHI_SUPPORT=ON -D BOOST_ROOT=/opt/boost_1_61_0/ -D
> Boost_NO_SYSTEM_PATHS=ON ..
> -- Could NOT find InChI in system locations (missing:  INCHI_LIBRARY
> INCHI_INCLUDE_DIR)
> CUSTOM_INCHI_PATH = /opt/rdkit/rdkit-Release_2017_03_2/External/INCHI-API
> -- Found InChI software locally
> -- Boost version: 1.61.0
> -- Found the following Boost libraries:
> --   python
> -- Boost version: 1.61.0
> -- Found the following Boost libraries:
> --   thread
> --   system
> --   chrono
> --   date_time
> --   atomic
> -- Boost version: 1.61.0
> -- Found the following Boost libraries:
> --   serialization
> == Using strict rotor definition
> == Updating Filters.cpp from pains file
> == Done updating pains files
> -- Boost version: 1.61.0
> -- Found the following Boost libraries:
> --   regex
> -- Configuring done
> -- Generating done
> -- Build files have been written to: /opt/rdkit/rdkit-Release_2017_
> 03_2/build
>
> But the error prevails.
>
> If I set up rdkit to use just the system wide boost (1.58), not my
> specific install (1.61), rdkit builds successfully.  I do this by clearing
> BOOST_ROOT, LD_LIBRARY_PATH, and removing the boost-related cmake flags.
> So this is certainly an issue with wrong libraries being picked up.
>
>
>
>
> On 21 June 2017 at 07:29, Greg Landrum <greg.land...@gmail.com> wrote:
>
>> did you build boost serialize?
>>
>> On Mon, Jun 19, 2017 at 12:03 PM, JP <jeanpaul.ebe...@inhibox.com> wrote:
>>
>>> Hi Greg !
>>>
>>> Unfortunately that didn't help (I delete everything in my build
>>> directory, then):
>>>
>>> cmake -DRDK_BUILD_INCHI_SUPPORT=ON -DBOOST_ROOT=/opt/boost_1_61_0/
>>> -DBoost_NO_SYSTEM_PATHS=ON ..
>>>
>>> and make as usual.
>>>
>>> [ 62%] Linking CXX executable testReaction
>>> ../../../lib/libRDKitChemReactions.so.1.2017.03.2: undefined reference
>>> to `boost::archive::text_iarchive_impl<boost::archive::text_iar
>>> chive>::load_override(boost::archive::class_name_type&)'
>>> ../../../lib/libRDKitChemReactions.so.1.2017.03.2: undefined reference
>>> to 
>>> `boost::archive::archive_exception::archive_exception(boost::archive::archive_exception
>>> const&)'
>>> collect2: error: ld returned 1 exit status
>>> Code/GraphMol/ChemReactions/CMakeFiles/testReaction.dir/build.make:116:
>>> recipe for target 'Code/GraphMol/ChemReactions/testReaction' failed
>>> make[2]: *** [Code/GraphMol/ChemReactions/testReaction] Error 1
>>> CMakeFiles/Makefile2:4157: recipe for target
>>> 'Code/GraphMol/ChemReactions/CMakeFiles/testReaction.dir/all' failed
>>> make[1]: *** [Code/GraphMol/ChemReactions/C
>>> MakeFiles/testReaction.dir/all] Error 2
>>> Makefile:160: recipe for target 'all' failed
>>> make: *** [all] Error 2
>>>
>>> Perhaps this isn't related to the system vs user-install, boost after
>>> all?
>>>
>>>
>>> On 19 June 2017 at 10:38, Greg Landrum <greg.land...@gmail.com> wrote:
>>>
>>>> If you have a system boost install that you do not want to use, you
>>>> should be sure to add "-D Boost_NO_SYSTEM_PATHS=ON" to the cmake arguments.
>>>> This will (well, should) disable any usage of the system boost.
>>>>
>>>> -greg
>>>>
>>>>
>>>> On Mon, Jun 19, 2017 at 9:39 AM, JP <jeanpaul.ebe...@inhibox.com>
>>>> wrote:
>>>>
>>>>> HI Paul,
>>>>>
>>>>> Funny you should mention that.  I have boost 1.61 (installed manually
>>>>> in /opt) and system boost I installed via sudo apt-get install
>>>>>
>>>>> /opt/rdkit/rdkit-Release_2017_03_2/build$ dpkg -s libboost-dev | grep
>>>>> 'Version'
>>>>> Version: 1.58.0.1ubuntu1
>>>>>
>>>>> However I pass the BOOST path to cmake via:
>>>>>
>>>>> cmake -DRDK_BUILD_INCHI_SUPPORT=ON -DBOOST_ROOT=/opt/boost_1_61_0/  ..
>>>>>
>>>>> (I also have the $BOOST_ROOT env variable set, so I think that is
>>>>> redundant.  Whatever).  cmake output clearly shows it is finding/using
>>>>> boost 1.61
>>>>>
>>>>> Using make VERBOSE=1 I get:
>>>>>
>>>>> [ 62%] Linking CXX executable testReaction
>>>>> cd /opt/rdkit/rdkit-Release_2017_03_2/build/Code/GraphMol/ChemReactions
>>>>> && /usr/bin/cmake -E cmake_link_script 
>>>>> CMakeFiles/testReaction.dir/link.txt
>>>>> --verbose=1
>>>>> /usr/bin/c++    -mpopcnt -Wno-deprecated -Wno-unused-function
>>>>> -fno-strict-aliasing -fPIC -Wall -Wextra -O3 -DNDEBUG
>>>>> CMakeFiles/testReaction.dir/testReaction.cpp.o  -o testReaction
>>>>> -rdynamic ../../../lib/libRDKitChemReactions.so.1.2017.03.2
>>>>> ../../../lib/libRDKitChemTransforms.so.1.2017.03.2
>>>>> ../../../lib/libRDKitDescriptors.so.1.2017.03.2
>>>>> ../../../lib/libRDKitFingerprints.so.1.2017.03.2
>>>>> ../../../lib/libRDKitDepictor.so.1.2017.03.2
>>>>> ../../../lib/libRDKitFileParsers.so.1.2017.03.2 -lboost_serialization
>>>>> ../../../lib/libRDKitPartialCharges.so.1.2017.03.2
>>>>> ../../../lib/libRDKitMolTransforms.so.1.2017.03.2
>>>>> ../../../lib/libRDKitEigenSolvers.so.1.2017.03.2
>>>>> ../../../lib/libRDKitFilterCatalog.so.1.2017.03.2
>>>>> ../../../lib/libRDKitSubgraphs.so.1.2017.03.2
>>>>> ../../../lib/libRDKitSmilesParse.so.1.2017.03.2
>>>>> ../../../lib/libRDKitSubstructMatch.so.1.2017.03.2
>>>>> ../../../lib/libRDKitGraphMol.so.1.2017.03.2
>>>>> ../../../lib/libRDKitRDGeometryLib.so.1.2017.03.2
>>>>> ../../../lib/libRDKitDataStructs.so.1.2017.03.2 -lboost_serialization
>>>>> ../../../lib/libRDKitCatalogs.so.1.2017.03.2
>>>>> ../../../lib/libRDKitRDGeneral.so.1.2017.03.2 -lboost_thread
>>>>> -lboost_system -lpthread -Wl,-rpath,/opt/rdkit/rdkit-Re
>>>>> lease_2017_03_2/build/lib
>>>>> ../../../lib/libRDKitChemReactions.so.1.2017.03.2: undefined
>>>>> reference to `boost::archive::text_iarchive
>>>>> _impl<boost::archive::text_iarchive>::load_override(boost::a
>>>>> rchive::class_name_type&)'
>>>>> ../../../lib/libRDKitChemReactions.so.1.2017.03.2: undefined
>>>>> reference to `boost::archive::archive_excep
>>>>> tion::archive_exception(boost::archive::archive_exception const&)'
>>>>> collect2: error: ld returned 1 exit status
>>>>> Code/GraphMol/ChemReactions/CMakeFiles/testReaction.dir/build.make:116:
>>>>> recipe for target 'Code/GraphMol/ChemReactions/testReaction' failed
>>>>> make[2]: *** [Code/GraphMol/ChemReactions/testReaction] Error 1
>>>>> make[2]: Leaving directory '/opt/rdkit/rdkit-Release_2017_03_2/build'
>>>>> CMakeFiles/Makefile2:4157: recipe for target
>>>>> 'Code/GraphMol/ChemReactions/CMakeFiles/testReaction.dir/all' failed
>>>>> make[1]: *** [Code/GraphMol/ChemReactions/C
>>>>> MakeFiles/testReaction.dir/all] Error 2
>>>>> make[1]: Leaving directory '/opt/rdkit/rdkit-Release_2017_03_2/build'
>>>>> Makefile:160: recipe for target 'all' failed
>>>>> make: *** [all] Error 2
>>>>>
>>>>> /opt/boost_1_61_0/lib is also the first path in LD_LIBRARY_PATH.
>>>>>
>>>>> Thanks for your help and time.  I really appreciate it.
>>>>>
>>>>> Cheers
>>>>>
>>>>> On 16 June 2017 at 14:12, Paul Emsley <pems...@mrc-lmb.cam.ac.uk>
>>>>> wrote:
>>>>>
>>>>>> On 16/06/2017 12:08, JP wrote:
>>>>>>
>>>>>>> Hi Folks,
>>>>>>>
>>>>>>> Must have been eons ago last time I posted to this mailing list.
>>>>>>> The italians have a saying "chi non muore si rivede".
>>>>>>>
>>>>>>> I am trying to install the RDKit (release 2017_03_2) from source,
>>>>>>> without conda, and I thought this will be a breeze.  But I am getting an
>>>>>>> error.  I am pretty sure this is because of the boost version I am using
>>>>>>> (1.61).
>>>>>>>
>>>>>>> The error is:
>>>>>>>
>>>>>>>     [ 62%] Linking CXX executable testReaction
>>>>>>>     ../../../lib/libRDKitChemReactions.so.1.2017.03.2: undefined
>>>>>>> reference to
>>>>>>>     `boost::archive::text_iarchive_impl<boost::archive::text_iar
>>>>>>> chive>::load_override(boost::archive::class_name_type&)'
>>>>>>>     ../../../lib/libRDKitChemReactions.so.1.2017.03.2: undefined
>>>>>>> reference to
>>>>>>>     `boost::archive::archive_exception::archive_exception(boost:
>>>>>>> :archive::archive_exception
>>>>>>>     const&)'
>>>>>>>     collect2: error: ld returned 1 exit status
>>>>>>>     
>>>>>>> Code/GraphMol/ChemReactions/CMakeFiles/testReaction.dir/build.make:116:
>>>>>>> recipe for
>>>>>>>     target 'Code/GraphMol/ChemReactions/testReaction' failed
>>>>>>>     make[2]: *** [Code/GraphMol/ChemReactions/testReaction] Error 1
>>>>>>>     CMakeFiles/Makefile2:4157: recipe for target
>>>>>>>     'Code/GraphMol/ChemReactions/CMakeFiles/testReaction.dir/all'
>>>>>>> failed
>>>>>>>     make[1]: *** [Code/GraphMol/ChemReactions/C
>>>>>>> MakeFiles/testReaction.dir/all] Error 2
>>>>>>>     Makefile:160: recipe for target 'all' failed
>>>>>>>     make: *** [all] Error 2
>>>>>>>
>>>>>>
>>>>>> If you are using an outdated version of boost, this is not the most
>>>>>> elegant way for RDKit to tell you so.
>>>>>>
>>>>>> But I don't think that that's the case. I think that you are not
>>>>>> linking with the correct boost libraries. Hard to say at the moment.
>>>>>>
>>>>>> $ make  VERBOSE=1
>>>>>>
>>>>>>>
>>>>>>> As I side note, but this is just a deprecation warning, I do get a
>>>>>>> ton of these,
>>>>>>> /opt/boost_1_61_0/include/boost/type_traits/detail/template_arity_spec.hpp:13:84:
>>>>>>> note: #pragma message: NOTE: Use of this header 
>>>>>>> (template_arity_spec.hpp)
>>>>>>> is deprecated
>>>>>>>   # pragma message("NOTE: Use of this header
>>>>>>> (template_arity_spec.hpp) is deprecated")
>>>>>>>
>>>>>>
>>>>>> I see such things also - I comment out the pragmas in the header.
>>>>>>
>>>>>>
>>>>>> Paul.
>>>>>>
>>>>>> ------------------------------------------------------------
>>>>>> ------------------
>>>>>> Check out the vibrant tech community on one of the world's most
>>>>>> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
>>>>>> _______________________________________________
>>>>>> Rdkit-discuss mailing list
>>>>>> Rdkit-discuss@lists.sourceforge.net
>>>>>> https://lists.sourceforge.net/lists/listinfo/rdkit-discuss
>>>>>>
>>>>>
>>>>>
>>>>> ------------------------------------------------------------
>>>>> ------------------
>>>>> Check out the vibrant tech community on one of the world's most
>>>>> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
>>>>> _______________________________________________
>>>>> Rdkit-discuss mailing list
>>>>> Rdkit-discuss@lists.sourceforge.net
>>>>> https://lists.sourceforge.net/lists/listinfo/rdkit-discuss
>>>>>
>>>>>
>>>>
>>>
>>
>
------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Rdkit-discuss mailing list
Rdkit-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss

Reply via email to