Arseny Tolmachev
Arseny Tolmachev
At least MurmurHash internal hashing representation should use fixed 64 bit integers. But sized integer hygiene is... not good, yep.
Thanks for digging. I'm busy with a paper and upcoming PhD thesis right now, so I can't guarantee when I look into it, but I will try.
If your project is sorta largish, I strongly recommend you to use Juman++ via gRPC. You will be protected from crashes and other weird things that way. https://github.com/eiennohito/jumanpp-grpc
What kind of warnings? There could be signed-unsigned conversion/comparison warnings over the place. g++/clang builds on macos/linux are (mostly) warning-free with `-Wall`. One place is the usage of tmpfile in...
Ah, unsafe Windows CRT functions. I see. Need to setup a windows environment and see the warnings for myself as it seems.
On warnings: (I will update this list) ## Ignore: * Everything from jumanpp_rnn_legacy * C4141 (inline used more than once)
YES! I need to do this! Definitely! But right now I invite you to look into the main launcher https://github.com/ku-nlp/jumanpp/blob/master/src/jumandic/main/jumanpp.cc
Yes, if you can work on that it should be wonderful. The C++ itself code should be compilable with MSVC 2017 (I haven't tried), or definitely by Windows build of...
The other thing is, on Windows, filesystem access should use W versions of API (and convert paths from/to utf-8).
@DoumanAsh @kou what do you think. I've reused the issue to be more related to what actually should be done.