I will say, though, that I am running Arch Linux at the moment. Aside from keeping things on the newish side, Arch seems to have most packages optimized -O2. Maybe that has something to do with this?
This is caused by an apparent incompatibility between Boost Filesystem V2 and V3. For now you can comment that if statement -- the code isn't really used for anything right now -- I had to update it to get rid of file_exists().
error: could not revert 2a5ade8... Loading/saving: Use boost::filesystem and BOOST_TYPEOF_TPL instead of typeof().
hint: after resolving the conflicts, mark the corrected paths
hint: with 'git add <paths>' or 'git rm <paths>'
Alright, I'll try to install a more recent Boost in a virtual machine environment to see if I can find a solution.
EDIT: (EDIT#2: scratch that)
EDIT#2: You could also try the very latest revision from the repo; Apparently I was using a deprecated "path" constructor. It now compiles on F15 (which also has a newer Boost), but immediately segfaults on startup.
Last edited by AnonymousHero on Sun Apr 03, 2011 4:31 pm, edited 1 time in total.
That makes it compile... However, there are bunch of "no matching function to call for min/max" errors in casting_context.cc. Casting those as ints almost makes it compile... Then I get this when it tries to link the binary:
It could be that there are some stale build files still lying around. (EDIT: See (*))
Could you please provide a diff of the changes you did to std::max, std::min to get the code compiling? It's a bit difficult to spot all the cases just by looking at the code .
(*) EDIT: That is to say: I've sometimes observed "make clean" not cleaning "enough". What you can do to avoid this is issue is to have the build directory separate from the source directory. Something like this:
$ mkdir src
$ cd src
$ git clone wherever-you-are-cloning-from
$ cd ..
$ mkdir build
$ cd build
$ cmake ../src
Then you can just nuke the "build" directory completely if you want to be absolutely sure that everything is rebuilt without having to worry about any local changes to the source.
AnonymousHero wrote:
Could you please provide a diff of the changes you did to std::max, std::min to get the code compiling? It's a bit difficult to spot all the cases just by looking at the code .
I just did a search/replace on all instances of "min(" and "max(", changing them "min<int>(" and "max<int>(". Ham-fisted, stupid, and probably capable of causing all kinds of problems, but it appeared to work.