Since e0eeb904 changed b_slot::pathcode from size_type to uint16_t, MSVC 2015 complains:
1>cuckoohash_map.hh(1457): warning C4267: 'initializing': conversion from 'size_t' to 'uint16_t', possible loss of data
1> cuckoohash_map.hh(1456): note: while compiling class template member function 'cuckoohash_map<rs::String,rs::metrics::Metric,rs::string::Hash,rs::string::IsEqual,rs::Allocator<std::pair<const rs::String,rs::metrics::Metric>>,4>::b_slot::b_slot(const unsigned int64,const unsigned int64,const int8_t)'
This arithmetic should in fact be safe - we use static assertions to ensure that
pathcode will be big enough. To silence the compiler, this commit changes a
couple of other types to uint16_t to match pathcode.
manugoyal
commented
6 years ago
Since e0eeb904 changed b_slot::pathcode from size_type to uint16_t, MSVC 2015 complains: 1>cuckoohash_map.hh(1457): warning C4267: 'initializing': conversion from 'size_t' to 'uint16_t', possible loss of data 1> cuckoohash_map.hh(1456): note: while compiling class template member function 'cuckoohash_map<rs::String,rs::metrics::Metric,rs::string::Hash,rs::string::IsEqual,rs::Allocator<std::pair<const rs::String,rs::metrics::Metric>>,4>::b_slot::b_slot(const unsigned int64,const unsigned int64,const int8_t)'
This arithmetic should in fact be safe - we use static assertions to ensure that pathcode will be big enough. To silence the compiler, this commit changes a couple of other types to uint16_t to match pathcode.