Commits

Wes McKinney authored 5d3dbd05ab4
ARROW-8660: [C++][Gandiva] Reduce usage of Boost in Gandiva codebase I noticed this while reading the Gandiva codebase as part of the C++ precompiled kernels revamp project. In general we've tried to reduce our use of Boost -- if we can eliminate Boost altogether from library builds in the future that would be ideal. I replaced usages of `boost::hash_combine` (vendored), `boost::iequals` (reimplemented), and `boost::algorithm::to_lower_copy` (reimplemented). There's a little bit of code scrubbing / IWYU fixes also. The only Boost remaining is boost/multiprecision (for int256_t) Closes #7077 from wesm/gandiva-less-boost Authored-by: Wes McKinney <wesm+git@apache.org> Signed-off-by: Wes McKinney <wesm+git@apache.org>