Hi all,
I am trying to compile spdlib in a windows system and I am getting
more problems than expected. First of all, I had many issues with
dependencies compilation, and after that I had to modify the
CMakeList.txt file because it was UNIX-based written and It didn't
find the libraries and .h headers. Now everything seems to work well
and I will publish a little manual and the CMakeList.txt file when
have finish the compilation.
The compilation works smoothly for the libspdio and libspdalg
libraries, without errors nor warnings but it stops at:
Building CXX object src/CMakeFiles/libspd.dir/spd/SPDGenerateTiles.cpp.obj
with the next error (among others):
error: no match for 'operator+' (operand types are 'std::string {aka
std::basic_string<char>}' and 'const string_type {aka const
std::basic_string<wchar_t>}')
std::string outMess =
std::string("Directory \'") + outDir.native() + std::string("\' was
not present. Did you build the directory structure?");
I googled for a few time and it seems that the "native()" method of
the "path" class in the Boost library returns different variable types
depending on the OS. That is, it returns "wchar_t" while it is used
within Windows systems and "char" otherwise (see Boost documentation
[1]). I am currently using Boost v1.55 but I checked older Boost
versions have the same behavior.
I thought on casting the variable type the return value of native(),
but that would only solves the windows compilation and wouldn't be a
cross-platform solution.
Other solution would be to use 'boost::filesystem::path' all along the
code for "file names"-related strings but I do not know whether this
is possible and if so, how long would it take.
[1]
http://www.boost.org/doc/libs/1_55_0/libs/filesystem/doc/reference.html#path-Encoding-conversions
Any hint, solution, whatever would be much appreciated!
Roberto
--
Roberto
------------------------------------------------------------------------------
HPCC Systems Open Source Big Data Platform from LexisNexis Risk Solutions
Find What Matters Most in Your Big Data with HPCC Systems
Open Source. Fast. Scalable. Simple. Ideal for Dirty Data.
Leverages Graph Analysis for Fast Processing & Easy Data Exploration
http://p.sf.net/sfu/hpccsystems
_______________________________________________
spdlib-develop mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/spdlib-develop
