github-actions[bot] commented on code in PR #25493:
URL: https://github.com/apache/doris/pull/25493#discussion_r1366430993
##########
be/src/vec/common/hash_table/hash_map.h:
##########
@@ -193,10 +196,156 @@ class HashMapTable : public HashTable<Key, Cell, Hash,
Grower, Allocator> {
bool has_null_key_data() const { return false; }
};
+template <typename Key, typename Cell, typename Hash = DefaultHash<Key>,
+ typename Grower = HashTableGrower<>, typename Allocator =
HashTableAllocator>
+class JoinHashMapTable : public HashMapTable<Key, Cell, Hash, Grower,
Allocator> {
+public:
+ using Self = JoinHashMapTable;
+ using Base = HashMapTable<Key, Cell, Hash, Grower, Allocator>;
+
+ using key_type = Key;
+ using value_type = typename Cell::value_type;
+ using mapped_type = typename Cell::Mapped;
+
+ using LookupResult = typename Base::LookupResult;
+
+ using HashMapTable<Key, Cell, Hash, Grower, Allocator>::HashMapTable;
+
+ static uint32_t calc_bucket_size(size_t num_elem) {
+ size_t expect_bucket_size = static_cast<size_t>(num_elem) + (num_elem
- 1) / 7;
+ return phmap::priv::NormalizeCapacity(expect_bucket_size) + 1;
+ }
+
+ void reserve(int num_elem) {
+ bucket_size = calc_bucket_size(num_elem + 1);
+ first.resize(bucket_size, 0);
+ next.resize(num_elem);
+ }
+
+ void build(const Key* __restrict keys, const size_t* __restrict
hash_values, size_t num_elem,
+ int batch_size) {
+ _batch_size = batch_size;
+ bucket_size = calc_bucket_size(num_elem);
+ first.resize(bucket_size, 0);
+ next.resize(num_elem);
+
+ build_keys = keys;
+ for (size_t i = 1; i < num_elem; i++) {
+ next[i] = first[hash_values[i]];
+ first[hash_values[i]] = i;
+ }
+ }
+
+ template <int JoinOpType>
+ auto find_batch(const Key* __restrict keys, const size_t* __restrict
hash_values, int probe_idx,
+ int probe_rows, std::vector<uint32_t>& probe_idxs,
+ std::vector<uint32_t>& build_idxs) {
+ if constexpr (JoinOpType == doris::TJoinOp::INNER_JOIN ||
+ JoinOpType == doris::TJoinOp::LEFT_OUTER_JOIN) {
+ return _find_batch_inner_outer_join<JoinOpType>(keys, hash_values,
probe_idx,
+ probe_rows,
probe_idxs, build_idxs);
+ }
+ if constexpr (JoinOpType == doris::TJoinOp::LEFT_ANTI_JOIN ||
+ JoinOpType == doris::TJoinOp::LEFT_SEMI_JOIN) {
+ return _find_batch_left_semi_anti<JoinOpType>(keys, hash_values,
probe_idx, probe_rows,
+ probe_idxs);
+ }
+ return std::pair {0, 0};
+ }
+
+ size_t get_bucket_mask() { return bucket_size - 1; }
+
+private:
+ template <int JoinOpType>
+ auto _find_batch_left_semi_anti(const Key* __restrict keys,
+ const size_t* __restrict hash_values, int
probe_idx,
+ int probe_rows, std::vector<uint32_t>&
probe_idxs) {
+ int matched_cnt = 0;
+ while (LIKELY(probe_idx < probe_rows && matched_cnt < _batch_size)) {
Review Comment:
warning: boolean expression can be simplified by DeMorgan's theorem
[readability-simplify-boolean-expr]
```cpp
while (LIKELY(probe_idx < probe_rows && matched_cnt < _batch_size)) {
^
```
<details>
<summary>Additional context</summary>
**be/src/common/compiler_util.h:34:** expanded from macro 'LIKELY'
```cpp
#define LIKELY(expr) __builtin_expect(!!(expr), 1)
^
```
</details>
##########
be/src/vec/common/hash_table/hash_map.h:
##########
@@ -193,10 +196,156 @@ class HashMapTable : public HashTable<Key, Cell, Hash,
Grower, Allocator> {
bool has_null_key_data() const { return false; }
};
+template <typename Key, typename Cell, typename Hash = DefaultHash<Key>,
+ typename Grower = HashTableGrower<>, typename Allocator =
HashTableAllocator>
+class JoinHashMapTable : public HashMapTable<Key, Cell, Hash, Grower,
Allocator> {
+public:
+ using Self = JoinHashMapTable;
+ using Base = HashMapTable<Key, Cell, Hash, Grower, Allocator>;
+
+ using key_type = Key;
+ using value_type = typename Cell::value_type;
+ using mapped_type = typename Cell::Mapped;
+
+ using LookupResult = typename Base::LookupResult;
+
+ using HashMapTable<Key, Cell, Hash, Grower, Allocator>::HashMapTable;
+
+ static uint32_t calc_bucket_size(size_t num_elem) {
+ size_t expect_bucket_size = static_cast<size_t>(num_elem) + (num_elem
- 1) / 7;
+ return phmap::priv::NormalizeCapacity(expect_bucket_size) + 1;
+ }
+
+ void reserve(int num_elem) {
+ bucket_size = calc_bucket_size(num_elem + 1);
+ first.resize(bucket_size, 0);
+ next.resize(num_elem);
+ }
+
+ void build(const Key* __restrict keys, const size_t* __restrict
hash_values, size_t num_elem,
+ int batch_size) {
+ _batch_size = batch_size;
+ bucket_size = calc_bucket_size(num_elem);
+ first.resize(bucket_size, 0);
+ next.resize(num_elem);
+
+ build_keys = keys;
+ for (size_t i = 1; i < num_elem; i++) {
+ next[i] = first[hash_values[i]];
+ first[hash_values[i]] = i;
+ }
+ }
+
+ template <int JoinOpType>
+ auto find_batch(const Key* __restrict keys, const size_t* __restrict
hash_values, int probe_idx,
+ int probe_rows, std::vector<uint32_t>& probe_idxs,
+ std::vector<uint32_t>& build_idxs) {
+ if constexpr (JoinOpType == doris::TJoinOp::INNER_JOIN ||
+ JoinOpType == doris::TJoinOp::LEFT_OUTER_JOIN) {
+ return _find_batch_inner_outer_join<JoinOpType>(keys, hash_values,
probe_idx,
+ probe_rows,
probe_idxs, build_idxs);
+ }
+ if constexpr (JoinOpType == doris::TJoinOp::LEFT_ANTI_JOIN ||
+ JoinOpType == doris::TJoinOp::LEFT_SEMI_JOIN) {
+ return _find_batch_left_semi_anti<JoinOpType>(keys, hash_values,
probe_idx, probe_rows,
+ probe_idxs);
+ }
+ return std::pair {0, 0};
+ }
+
+ size_t get_bucket_mask() { return bucket_size - 1; }
+
+private:
+ template <int JoinOpType>
+ auto _find_batch_left_semi_anti(const Key* __restrict keys,
+ const size_t* __restrict hash_values, int
probe_idx,
+ int probe_rows, std::vector<uint32_t>&
probe_idxs) {
+ int matched_cnt = 0;
+ while (LIKELY(probe_idx < probe_rows && matched_cnt < _batch_size)) {
+ uint32_t build_idx = first[hash_values[probe_idx]];
+ while (build_idx) {
+ if (keys[probe_idx] == build_keys[build_idx]) {
+ break;
+ }
+ build_idx = next[build_idx];
+ }
+ const bool matched =
+ JoinOpType == doris::TJoinOp::LEFT_SEMI_JOIN ? build_idx
!= 0 : build_idx == 0;
+ matched_cnt += matched;
+ probe_idxs[matched_cnt - matched] = probe_idx++;
+ }
+ return std::pair {probe_idx, matched_cnt};
+ }
+
+ template <int JoinOpType>
+ auto _find_batch_inner_outer_join(const Key* __restrict keys,
+ const size_t* __restrict hash_values,
int probe_idx,
+ int probe_rows, std::vector<uint32_t>&
probe_idxs,
+ std::vector<uint32_t>& build_idxs) {
+ int matched_cnt = 0;
+ uint32_t build_idx = 0;
+
+ auto do_the_probe = [&]() {
+ while (build_idx && LIKELY(matched_cnt < _batch_size)) {
+ if (keys[probe_idx] == build_keys[build_idx]) {
+ probe_idxs[matched_cnt] = probe_idx;
+ build_idxs[matched_cnt] = build_idx;
+ matched_cnt++;
+ }
+ build_idx = next[build_idx];
+ }
+
+ if constexpr (JoinOpType != doris::TJoinOp::INNER_JOIN) {
+ // `(!matched_cnt || probe_idxs[matched_cnt - 1] !=
probe_idx)` means not match one build side
+ if (!build_idx && (!matched_cnt || probe_idxs[matched_cnt - 1]
!= probe_idx)) {
+ probe_idxs[matched_cnt] = probe_idx;
+ build_idxs[matched_cnt] = build_idx;
+ matched_cnt++;
+ }
+ }
+
+ probe_idx++;
+ };
+
+ // some row over the batch_size, need dispose first
+ if (probe_idx == current_probe_idx) {
+ current_probe_idx = -1;
+ build_idx = current_build_idx;
+ current_build_idx = 0;
+ do_the_probe();
+ }
+ while (LIKELY(probe_idx < probe_rows && matched_cnt < _batch_size)) {
Review Comment:
warning: boolean expression can be simplified by DeMorgan's theorem
[readability-simplify-boolean-expr]
```cpp
while (LIKELY(probe_idx < probe_rows && matched_cnt < _batch_size)) {
^
```
<details>
<summary>Additional context</summary>
**be/src/common/compiler_util.h:34:** expanded from macro 'LIKELY'
```cpp
#define LIKELY(expr) __builtin_expect(!!(expr), 1)
^
```
</details>
--
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: commits-unsubscr...@doris.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscr...@doris.apache.org
For additional commands, e-mail: commits-h...@doris.apache.org
