New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Compilation in Windows (MSVC2015) #115
Conversation
Next issue to be solved (ideas?):Fails on project Now it is
which generates the following command
and it fails when executed in CMD as |
Ah, yeah, the libsvm modules... (ugh). Let me see if I can convert those to use |
OK, can you try merging (or rebasing) in the latest commits from develop? I think 63401b7 should help here by making the libsvm modules be CMake projects themselves, instead of just hackily calling |
…ssion This error raises in ~/include/util/identifiers.h(139) with T=meta::sequence::tag_t which is of type std::string.
In MSVC2015 functions std::tolower, std::isdigit, std::isupper,... require two arguments, the second one the locale being used.
I've rebased and push changes following your comments in bb1b547 It compiles (libsvm modules too) ;D Pending:
|
Here it is the output with all the warnings: https://gist.github.com/jgsogo/b81bbd7210b0215d91a5 |
@skystrife Shall I address some of those warnings? I've seen that you have several commits in branch |
Looks like the build is failing because of the submodule commit hash for porter2_stemmer. Can you update it to the tip of master, which is 52c6b5c00c7e38a47013c55f0cc3cda7d5a48dca? |
As far as the warnings go, I'd like to fix them for sure but I wouldn't block merging this on them. Were you able to run the unit tests on the MSVC build? |
Unit tests fail because Failed
Passed
I have no time now to investigate further (although I'll try to) |
Where is |
|
Nevertheless there is an assert (string iterator not dereferencable) in
at this point Maybe it is interesting to have some tests associated with this set of functions... |
This is a bug upstream. (Upstream just so happens to be me, so I'll look into fixing it.) cpptoml is well tested (with sanitizers, too). I think this is innocuous (the iterator points to the null terminator, which is valid memory) but that MSVC is technically correct. Let me fix it. |
OK, try updating the cpptoml submodule to the tip of master and see if that helps. If there are more issues with the config file parsing, go ahead and file issues upstream for them and I can fix them there. |
Hi! I'm here again, sorry for dissappering, it's been a month full of changes, but now I'm quiet again and willing to contribute. I've updated cpptoml and previous issue is fixed, but now I come accross an I see a lot of changes (great!), should I merge/rebase |
Closing for inactivity. Please feel free to re-open if you have more time to dedicate to this. If someone else wants to take this on, I'd be more than happy to have things building with MSVC. In theory we could then add a job to AppVeyor so that we can keep the build for MSVC working even if no individual contributor is using it on a daily basis. |
This PR gathers commits in order to compile MeTA in MSVC2015. It is a work in progress, so this message may be edited if neccesary.
It also depends in modifications/PR of other libraries: