Commits
Kenta Murata authored and Antoine Pitrou committed 0fbaff6b9ba
ARROW-5736: [Format][C++] Support small bit-width indices in sparse tensor I'd like to enable all integer types to become the value type of sparse tensor index. Closes #5290 from mrkn/ARROW-5736 and squashes the following commits: b0be574c3 <Kenta Murata> Add check SparseCOOIndex::coords_->ndim() in the constructor ad725c24e <Kenta Murata> Remove needless include 84e672624 <Kenta Murata> Fix typo 76e5ca73d <Kenta Murata> Increase coverage rate 2226c9a28 <Kenta Murata> Fix compile errors on MSVC f1cf53faa <Kenta Murata> Insert an explicit down cast 56ffc1b08 <Kenta Murata> Add #include <limits> a3fd50242 <Kenta Murata> Refactoring f0222abfa <Kenta Murata> Suppor reading and writing f2a192030 <Kenta Murata> Use fixtures in TestSparseCSRMatrix 4079906b9 <Kenta Murata> Refactoring of TestSparseCOOTensor 980cf9668 <Kenta Murata> Refactoring becc425fc <Kenta Murata> Enable all integer types to become the value type of SparseCOOIndex 46acdb10b <Kenta Murata> Refactor ipc-read-write-test with a typed test case b0053decb <Kenta Murata> Add assertions for dim_names and dim_name b7b47c75a <Kenta Murata> Refactor sparse_tensor-test with a typed test case 45be5c4e3 <Kenta Murata> Support reading and writing row-major SparseCOOIndex 33b6d1ff0 <Kenta Murata> Make SparseCOOIndex support row-major index 5f8ba9f4d <Kenta Murata> Add indicesStrides in SparseTensorIndexCOO 5ec7e6554 <Kenta Murata> Modify the comment of SparseTensorIndexCOO 8a0f65b63 <Kenta Murata> Add type fields of sparse index types Authored-by: Kenta Murata <mrkn@mrkn.jp> Signed-off-by: Antoine Pitrou <antoine@python.org>