"Make" error with Boost lib
liujianqiao opened this issue · 6 comments
Hi, Robert,
I'm very interested in your project and trying to install your tool on my home dir. I do follow the steps in your doc, but get some problems with the "boost" (I do have several versions of boost installed in the machine):
-bash-4.1$ make
Making all in aux/sais-lite-2.4.1/
make[1]: Entering directory /home/min/a/liu1274/TTBS/apps/LCP/LCPKasai-master/aux/sais-lite-2.4.1' gcc -DPACKAGE_NAME=\"LCPKasai\" -DPACKAGE_TARNAME=\"lcpkasai\" -DPACKAGE_VERSION=\"1.0\" -DPACKAGE_STRING=\"LCPKasai\ 1.0\" -DPACKAGE_BUGREPORT=\"\" -DPACKAGE_URL=\"\" -DPACKAGE=\"lcpkasai\" -DVERSION=\"1.0\" -DHAVE_BOOST=/\*\*/ -DHAVE_BOOST_SYSTEM=/\*\*/ -DHAVE_BOOST_FILESYSTEM=/\*\*/ -DHAVE_BOOST_PROGRAM_OPTIONS=/\*\*/ -I. -g -O2 -MT sais.o -MD -MP -MF .deps/sais.Tpo -c -o sais.o sais.c mv -f .deps/sais.Tpo .deps/sais.Po rm -f libsais.a ar cru libsais.a sais.o ranlib libsais.a make[1]: Leaving directory/home/min/a/liu1274/TTBS/apps/LCP/LCPKasai-master/aux/sais-lite-2.4.1'
Making all in src/apps/
make[1]: Entering directory /home/min/a/liu1274/TTBS/apps/LCP/LCPKasai-master/src/apps' g++ -DPACKAGE_NAME=\"LCPKasai\" -DPACKAGE_TARNAME=\"lcpkasai\" -DPACKAGE_VERSION=\"1.0\" -DPACKAGE_STRING=\"LCPKasai\ 1.0\" -DPACKAGE_BUGREPORT=\"\" -DPACKAGE_URL=\"\" -DPACKAGE=\"lcpkasai\" -DVERSION=\"1.0\" -DHAVE_BOOST=/\*\*/ -DHAVE_BOOST_SYSTEM=/\*\*/ -DHAVE_BOOST_FILESYSTEM=/\*\*/ -DHAVE_BOOST_PROGRAM_OPTIONS=/\*\*/ -I. -I../../src/include -I../../aux/sais-lite-2.4.1 -I/usr/include -std=c++0x -g -O2 -MT LCPKasai-LCPKasai.o -MD -MP -MF .deps/LCPKasai-LCPKasai.Tpo -c -o LCPKasai-LCPKasai.otest -f 'LCPKasai.cpp' || echo './'LCPKasai.cpp mv -f .deps/LCPKasai-LCPKasai.Tpo .deps/LCPKasai-LCPKasai.Po g++ -std=c++0x -g -O2 -o LCPKasai LCPKasai-LCPKasai.o -lboost_program_options ../../aux/sais-lite-2.4.1/libsais.a LCPKasai-LCPKasai.o: In functionboost::program_options::variables_map SetOptions<int, char**>(int&, char*_&)':
/home/min/a/liu1274/TTBS/apps/LCP/LCPKasai-master/src/apps/LCPKasai.cpp:70: undefined reference to boost::program_options::options_description::options_description(std::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, unsigned int, unsigned int)' collect2: ld returned 1 exit status make[1]: *_\* [LCPKasai] Error 1 make[1]: Leaving directory /home/min/a/liu1274/TTBS/apps/LCP/LCPKasai-master/src/apps'
make: *** [all-recursive] Error 1
Thanks very much!
Hi,
I apologize for the late replay. The problem you have looks like it is connected to the boost library. It is as if some linking is missing. When you ran ./configure was the boost lib properly set ??
Hi, Robert,
I'm very sorry for the late reply to your previous email due to my travel.
I eventually figured out the way to solve the Boost problem by adding
"-L" after the configure command. For some reason, the
"--with-boost" or "--with-boost-libdir=LIB_DIR" flags don't work when I
have multiple Boost versions and none of them are installed under default
path (/usr/bin or /usr/lib). Thanks very much for your attention!
Best Regards,
Jianqiao
2016-09-18 6:02 GMT-07:00 Robert Bakaric notifications@github.com:
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
#1 (comment), or mute
the thread
https://github.com/notifications/unsubscribe-auth/AGNRlxqFlJvWclr0wiFC2wQ8MatWsewgks5qrTZQgaJpZM4J8cR6
.
:) Well that is odd, but I had my share of complaints regarding the usage of boost in my projects, so maybe in the future I'll just stick to the STL. Also one more comment, I don't know why are you interested in LCP computation. If it is just some sort of a class project than I guess this implementation should suffice. however if not let me know and I'll put much faster non-Kasai version.
Cheers
Robert
The STL version would be much easier to use. Actually I’m a PhD student of Purdue University, and doing research with tree traversal. I’m writing a benchmark suite paper about various tree traversals. It looks like the LCP computation uses a suffix array to simulate the “bottom – up” traversal of a suffix tree. The LCP’s bottom-up traversal is very interesting to me. I’m looking into your code and will definitely cite your work if you allow me to include your code into my paper. I’m not sure about the “non-Kasai” version. Is it still a bottom-up traversal of a tree?
Best regards,
Jianqiao
From: Robert Bakaric notifications@github.com
Reply-To: RobertBakaric/LCPKasai reply@reply.github.com
Date: Sunday, September 18, 2016 at 10:41 AM
To: RobertBakaric/LCPKasai LCPKasai@noreply.github.com
Cc: liujianqiao liujianqiao.0001@gmail.com, Author author@noreply.github.com
Subject: Re: [RobertBakaric/LCPKasai] "Make" error with Boost lib (#1)
:) Well that is odd, but I had my share of complaints regarding the usage of boost in my projects, so maybe in the future I'll just stick to the STL. Also one more comment, I don't know why are you interested in LCP computation. If it is just some sort of a class project than I guess this implementation should suffice. however if not let me know and I'll put much faster non-Kasai version.
Cheers
Robert
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub, or mute the thread.
Any citations are more than welcomed. As far as the tree traversal is concerned the short answer is no. But it exploits L2 and L3 cash memory allowing the LCP computations to be preformed in the amount of time equal to that of the suffix array pass x 2. Also to spare you potentially some search time, the suffix array implementation done by Yuta Mori (you can find it in my repository) is the fastest one available today (checked it myself) so I strongly recommend it for any project based on the SAs.
cheers
Robert
Thanks very much for your information! I’ll let you know if I get any issues or progress about the LCP. Thanks, again! J
Jianqiao
From: Robert Bakaric notifications@github.com
Reply-To: RobertBakaric/LCPKasai reply@reply.github.com
Date: Sunday, September 18, 2016 at 11:02 AM
To: RobertBakaric/LCPKasai LCPKasai@noreply.github.com
Cc: liujianqiao liujianqiao.0001@gmail.com, Author author@noreply.github.com
Subject: Re: [RobertBakaric/LCPKasai] "Make" error with Boost lib (#1)
Any citations are more than welcomed. As far as the tree traversal is concerned the short answer is no. But it exploits L2 and L3 cash memory allowing the LCP computations to be preformed in the amount of time equal to that of the suffix array pass x 2. Also to spare you potentially some search time, the suffix array implementation done by Yuta Mori (you can find it in my repository) is the fastest one available today (checked it myself) so I strongly recommend it for any project based on the SAs.
cheers
Robert
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub, or mute the thread.