rtpsw commented on code in PR #13880:
URL: https://github.com/apache/arrow/pull/13880#discussion_r954622419
##########
cpp/src/arrow/compute/exec/asof_join_node.cc:
##########
@@ -17,33 +17,69 @@
#include <condition_variable>
#include <mutex>
-#include <set>
#include <thread>
#include <unordered_map>
+#include <unordered_set>
+#include "arrow/array/builder_binary.h"
#include "arrow/array/builder_primitive.h"
#include "arrow/compute/exec/exec_plan.h"
+#include "arrow/compute/exec/key_hash.h"
#include "arrow/compute/exec/options.h"
#include "arrow/compute/exec/schema_util.h"
#include "arrow/compute/exec/util.h"
+#include "arrow/compute/light_array.h"
#include "arrow/record_batch.h"
#include "arrow/result.h"
#include "arrow/status.h"
+#include "arrow/type_traits.h"
#include "arrow/util/checked_cast.h"
#include "arrow/util/future.h"
#include "arrow/util/make_unique.h"
#include "arrow/util/optional.h"
+#include "arrow/util/string_view.h"
namespace arrow {
namespace compute {
-// Remove this when multiple keys and/or types is supported
-typedef int32_t KeyType;
+template <typename T, typename V = typename T::value_type>
+inline typename T::const_iterator std_find(const T& container, const V& val) {
+ return std::find(container.begin(), container.end(), val);
+}
+
+template <typename T, typename V = typename T::value_type>
+inline bool std_has(const T& container, const V& val) {
+ return container.end() != std_find(container, val);
+}
+
+typedef uint64_t ByType;
+typedef uint64_t OnType;
+typedef uint64_t HashType;
Review Comment:
Do you mean `HashType` and `ByType`? The current code only works if `ByType`
is wider than `HashType` due to both being returned in `GetLatestKey`. It's
possible that performance would be improved if `HashType` is defined as
`uint32_t` and `Hashing32` is used instead of `Hashing64`; I didn't try this as
yet. It's also possible a 128-bit or 256-bit wide `ByType` would provide
improve performance for decimal types of this width that would be added in the
future; I haven't tried this either.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
