Thanks to Sébastien and Keith Robinson for the suggestions.  I was able to 
compile with cmake, but still not with libz, for some reason.  I also couldn't 
figure out how to change the order of the tags as suggested by Keith.  I was 
certain I had been able to compile rc5 with libz however (as I have been 
feeding it gzipped files no problem), so I did a little bit of playing around 
this morning:

1. I recompiled 2.0-rc5 using "make HAVE_LIBZ=y".  Compiled no problem, "./Ray 
-version" showed "HAVE_LIBZ = y"
2. I recompiled 2.0.0-rc7 using "make HAVE_LIBZ=y".  Surprisingly, complied no 
problem, "./Ray -version" showed "HAVE_LIBZ = y".
3. Wondering how the heck I got that error message last week, I tried retracing 
my steps exactly from last week.  So I actually modified the Makefile from 
"HAVE_LIBZ = n" to "HAVE_LIBZ = y", then ran a simple "make" command. Boom.  
Error recreated.  

It looks like, for whatever reason, you have to feed options directly to the 
make command rather than modifying options in the Makefile.  Why?  I thought 
maybe it was the spaces flanking the "=" sign in the Makefile, but when I 
removed them (i.e. "HAVE_LIBZ=y" instead of "HAVE_LIBZ = y"), the error still 
recurred.  

I guess as long it's working, I don't really care.  I just wasn't expecting 
that modifying the option in the makefile versus feeding it to the make command 
could make such a difference.  

Thanks again for the help,
- Egon


On May 18, 2012, at 1:15 PM, Sébastien Boisvert wrote:

> Someone else is having problem linking the code os Mac OS
> 
> Le 2012-05-18 12:57, Egon Ozer a écrit :
>> 
>> I'm running into a little problem compiling 2.0.0-rc7 on OSX 10.6.8.  When I 
>> make with HAVE_LIBZ=y, I get the following error:
>> 
>> mpicxx -lz  code/TheRayGenomeAssembler.a RayPlatform/libRayPlatform.a -o Ray
>> Undefined symbols:
>>   "FastqGzLoader::getSize()", referenced from:
>>       Loader::load(std::basic_string<char, std::char_traits<char>, 
>> std::allocator<char> >, bool)in TheRayGenomeAssembler.a(Loader.o)
>>   "FastqGzLoader::open(std::basic_string<char, std::char_traits<char>, 
>> std::allocator<char> >, int)", referenced from:
>>       Loader::load(std::basic_string<char, std::char_traits<char>, 
>> std::allocator<char> >, bool)in TheRayGenomeAssembler.a(Loader.o)
>>       Loader::load(std::basic_string<char, std::char_traits<char>, 
>> std::allocator<char> >, bool)in TheRayGenomeAssembler.a(Loader.o)
>>   "FastqGzLoader::load(int, ArrayOfReads*, MyAllocator*, int)", referenced 
>> from:
>>       Loader::loadSequences()      in TheRayGenomeAssembler.a(Loader.o)
>> ld: symbol(s) not found
>> collect2: ld returned 1 exit status
>> make: *** [Ray] Error 1
>> 
>> When I run make with HAVE_LIBZ=n, compilation runs without a hitch.  I had 
>> no problems compiling 2.0.0-rc5 with libz.
> 
> 
> This is strange because the build system of Ray has not changed since 
> 2.0.0-rc5.
> 
> 
> Did you change something since 2.0.0-rc5 ?
> 
> 
>>   Just to make sure, I updated zlib with MacPorts to 1.2.7 (universal), but 
>> still no dice.   Any suggestions?
> 
> If you have cmake installed, you can build Ray with it instead. CMake is 
> cross-platform.
> 
> 
> 
> To build Ray with CMake:
> 
> 
> tar xjf Ray-v2.0.0-rc7.tar.bz2
> 
> cd Ray-v2.0.0-rc7
> 
> remove the line " RayPlatform/scheduling/SwitchMan_adapters.cpp" from 
> CMakeLists.txt
> 
> cd ..
> 
> mkdir Build-with-CMAKE
> 
> cd Build-with-CMAKE
> 
> cmake ../Ray-v2.0.0-rc7
> 
> make
> 
> ls Ray
> 
> 
> 
> Let me know if that works for you.
> 
>> Thanks,
>> - Egon
>> ------------------------------------------------------------------------------
>> Live Security Virtual Conference
>> Exclusive live event will cover all the ways today's security and 
>> threat landscape has changed and how IT managers can respond. Discussions 
>> will include endpoint security, mobile security and the latest in malware 
>> threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
>> _______________________________________________
>> Denovoassembler-users mailing list
>> Denovoassembler-users@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/denovoassembler-users
> 
> ------------------------------------------------------------------------------
> Live Security Virtual Conference
> Exclusive live event will cover all the ways today's security and 
> threat landscape has changed and how IT managers can respond. Discussions 
> will include endpoint security, mobile security and the latest in malware 
> threats. 
> http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/_______________________________________________
> Denovoassembler-users mailing list
> Denovoassembler-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/denovoassembler-users

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
Denovoassembler-users mailing list
Denovoassembler-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/denovoassembler-users

Reply via email to