Files
doris/be/src/exec/odbc_scan_node.cpp
HappenLee e1d7233e9c [feature](vectorization) Support Vectorized Exec Engine In Doris (#7785)
# Proposed changes

Issue Number: close #6238

    Co-authored-by: HappenLee <happenlee@hotmail.com>
    Co-authored-by: stdpain <34912776+stdpain@users.noreply.github.com>
    Co-authored-by: Zhengguo Yang <yangzhgg@gmail.com>
    Co-authored-by: wangbo <506340561@qq.com>
    Co-authored-by: emmymiao87 <522274284@qq.com>
    Co-authored-by: Pxl <952130278@qq.com>
    Co-authored-by: zhangstar333 <87313068+zhangstar333@users.noreply.github.com>
    Co-authored-by: thinker <zchw100@qq.com>
    Co-authored-by: Zeno Yang <1521564989@qq.com>
    Co-authored-by: Wang Shuo <wangshuo128@gmail.com>
    Co-authored-by: zhoubintao <35688959+zbtzbtzbt@users.noreply.github.com>
    Co-authored-by: Gabriel <gabrielleebuaa@gmail.com>
    Co-authored-by: xinghuayu007 <1450306854@qq.com>
    Co-authored-by: weizuo93 <weizuo@apache.org>
    Co-authored-by: yiguolei <guoleiyi@tencent.com>
    Co-authored-by: anneji-dev <85534151+anneji-dev@users.noreply.github.com>
    Co-authored-by: awakeljw <993007281@qq.com>
    Co-authored-by: taberylyang <95272637+taberylyang@users.noreply.github.com>
    Co-authored-by: Cui Kaifeng <48012748+azurenake@users.noreply.github.com>


## Problem Summary:

### 1. Some code from clickhouse

**ClickHouse is an excellent implementation of the vectorized execution engine database,
so here we have referenced and learned a lot from its excellent implementation in terms of
data structure and function implementation.
We are based on ClickHouse v19.16.2.2 and would like to thank the ClickHouse community and developers.**

The following comment has been added to the code from Clickhouse, eg:
// This file is copied from
// https://github.com/ClickHouse/ClickHouse/blob/master/src/Interpreters/AggregationCommon.h
// and modified by Doris

### 2. Support exec node and query:
* vaggregation_node
* vanalytic_eval_node
* vassert_num_rows_node
* vblocking_join_node
* vcross_join_node
* vempty_set_node
* ves_http_scan_node
* vexcept_node
* vexchange_node
* vintersect_node
* vmysql_scan_node
* vodbc_scan_node
* volap_scan_node
* vrepeat_node
* vschema_scan_node
* vselect_node
* vset_operation_node
* vsort_node
* vunion_node
* vhash_join_node

You can run exec engine of SSB/TPCH and 70% TPCDS stand query test set.

### 3. Data Model

Vec Exec Engine Support **Dup/Agg/Unq** table, Support Block Reader Vectorized.
Segment Vec is working in process.

### 4. How to use

1. Set the environment variable `set enable_vectorized_engine = true; `(required)
2. Set the environment variable `set batch_size = 4096; ` (recommended)

### 5. Some diff from origin exec engine

https://github.com/doris-vectorized/doris-vectorized/issues/294

## Checklist(Required)

1. Does it affect the original behavior: (No)
2. Has unit tests been added: (Yes)
3. Has document been added or modified: (No)
4. Does it need to update dependencies: (No)
5. Are there any changes that cannot be rolled back: (Yes)
2022-01-18 10:07:15 +08:00

258 lines
8.8 KiB
C++

// Licensed to the Apache Software Foundation (ASF) under one
// or more contributor license agreements. See the NOTICE file
// distributed with this work for additional information
// regarding copyright ownership. The ASF licenses this file
// to you under the Apache License, Version 2.0 (the
// "License"); you may not use this file except in compliance
// with the License. You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing,
// software distributed under the License is distributed on an
// "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
// KIND, either express or implied. See the License for the
// specific language governing permissions and limitations
// under the License.
#include "odbc_scan_node.h"
#include <sstream>
#include "exec/text_converter.hpp"
#include "gen_cpp/PlanNodes_types.h"
#include "runtime/row_batch.h"
#include "runtime/runtime_state.h"
#include "runtime/string_value.h"
#include "runtime/tuple_row.h"
#include "util/runtime_profile.h"
namespace doris {
OdbcScanNode::OdbcScanNode(ObjectPool* pool, const TPlanNode& tnode, const DescriptorTbl& descs,
std::string scan_node_type)
: ScanNode(pool, tnode, descs),
_is_init(false),
_scan_node_type(scan_node_type),
_table_name(tnode.odbc_scan_node.table_name),
_connect_string(std::move(tnode.odbc_scan_node.connect_string)),
_query_string(std::move(tnode.odbc_scan_node.query_string)),
_tuple_id(tnode.odbc_scan_node.tuple_id),
_tuple_desc(nullptr),
_slot_num(0) {}
OdbcScanNode::~OdbcScanNode() {}
Status OdbcScanNode::prepare(RuntimeState* state) {
VLOG_CRITICAL << _scan_node_type << "::Prepare";
if (_is_init) {
return Status::OK();
}
if (nullptr == state) {
return Status::InternalError("input pointer is null.");
}
RETURN_IF_ERROR(ScanNode::prepare(state));
// get tuple desc
_tuple_desc = state->desc_tbl().get_tuple_descriptor(_tuple_id);
if (nullptr == _tuple_desc) {
return Status::InternalError("Failed to get tuple descriptor.");
}
_slot_num = _tuple_desc->slots().size();
_odbc_param.connect_string = std::move(_connect_string);
_odbc_param.query_string = std::move(_query_string);
_odbc_param.tuple_desc = _tuple_desc;
_odbc_scanner.reset(new (std::nothrow) ODBCConnector(_odbc_param));
if (_odbc_scanner.get() == nullptr) {
return Status::InternalError("new a odbc scanner failed.");
}
_tuple_pool.reset(new (std::nothrow) MemPool(mem_tracker().get()));
if (_tuple_pool.get() == nullptr) {
return Status::InternalError("new a mem pool failed.");
}
_text_converter.reset(new (std::nothrow) TextConverter('\\'));
if (_text_converter.get() == nullptr) {
return Status::InternalError("new a text convertor failed.");
}
_is_init = true;
return Status::OK();
}
Status OdbcScanNode::open(RuntimeState* state) {
RETURN_IF_ERROR(ExecNode::open(state));
VLOG_CRITICAL << _scan_node_type << "::Open";
if (nullptr == state) {
return Status::InternalError("input pointer is null.");
}
if (!_is_init) {
return Status::InternalError("used before initialize.");
}
RETURN_IF_ERROR(exec_debug_action(TExecNodePhase::OPEN));
RETURN_IF_CANCELLED(state);
SCOPED_TIMER(_runtime_profile->total_time_counter());
RETURN_IF_ERROR(_odbc_scanner->open());
RETURN_IF_ERROR(_odbc_scanner->query());
// check materialize slot num
return Status::OK();
}
Status OdbcScanNode::write_text_slot(char* value, int value_length, SlotDescriptor* slot,
RuntimeState* state) {
if (!_text_converter->write_slot(slot, _tuple, value, value_length, true, false,
_tuple_pool.get())) {
std::stringstream ss;
ss << "Fail to convert odbc value:'" << value << "' to " << slot->type() << " on column:`"
<< slot->col_name() + "`";
return Status::InternalError(ss.str());
}
return Status::OK();
}
Status OdbcScanNode::get_next(RuntimeState* state, RowBatch* row_batch, bool* eos) {
VLOG_CRITICAL << _scan_node_type << "::GetNext";
if (nullptr == state || nullptr == row_batch || nullptr == eos) {
return Status::InternalError("input is nullptr pointer");
}
if (!_is_init) {
return Status::InternalError("used before initialize.");
}
RETURN_IF_ERROR(exec_debug_action(TExecNodePhase::GETNEXT));
RETURN_IF_CANCELLED(state);
SCOPED_TIMER(_runtime_profile->total_time_counter());
if (reached_limit()) {
*eos = true;
return Status::OK();
}
// create new tuple buffer for row_batch
int tuple_buffer_size = row_batch->capacity() * _tuple_desc->byte_size();
void* tuple_buffer = _tuple_pool->allocate(tuple_buffer_size);
if (nullptr == tuple_buffer) {
return Status::InternalError("Allocate memory failed.");
}
_tuple = reinterpret_cast<Tuple*>(tuple_buffer);
// Indicates whether there are more rows to process. Set in _odbc_scanner.next().
bool odbc_eos = false;
while (true) {
RETURN_IF_CANCELLED(state);
if (reached_limit() || row_batch->is_full()) {
// hang on to last allocated chunk in pool, we'll keep writing into it in the
// next get_next() call
row_batch->tuple_data_pool()->acquire_data(_tuple_pool.get(), !reached_limit());
*eos = reached_limit();
return Status::OK();
}
RETURN_IF_ERROR(_odbc_scanner->get_next_row(&odbc_eos));
if (odbc_eos) {
row_batch->tuple_data_pool()->acquire_data(_tuple_pool.get(), false);
*eos = true;
return Status::OK();
}
int row_idx = row_batch->add_row();
TupleRow* row = row_batch->get_row(row_idx);
// scan node is the first tuple of tuple row
row->set_tuple(0, _tuple);
memset(_tuple, 0, _tuple_desc->num_null_bytes());
int j = 0;
for (int i = 0; i < _slot_num; ++i) {
auto slot_desc = _tuple_desc->slots()[i];
// because the fe planner filter the non_materialize column
if (!slot_desc->is_materialized()) {
continue;
}
const auto& column_data = _odbc_scanner->get_column_data(j);
if (column_data.strlen_or_ind == SQL_NULL_DATA) {
if (slot_desc->is_nullable()) {
_tuple->set_null(slot_desc->null_indicator_offset());
} else {
std::stringstream ss;
ss << "nonnull column contains nullptr. table=" << _table_name
<< ", column=" << slot_desc->col_name();
return Status::InternalError(ss.str());
}
} else if (column_data.strlen_or_ind > column_data.buffer_length) {
std::stringstream ss;
ss << "nonnull column contains nullptr. table=" << _table_name
<< ", column=" << slot_desc->col_name();
return Status::InternalError(ss.str());
} else {
RETURN_IF_ERROR(write_text_slot(static_cast<char*>(column_data.target_value_ptr),
column_data.strlen_or_ind, slot_desc, state));
}
j++;
}
ExprContext* const* ctxs = &_conjunct_ctxs[0];
int num_ctxs = _conjunct_ctxs.size();
// ODBC scanner can not filter conjunct with function, need check conjunct again.
if (ExecNode::eval_conjuncts(ctxs, num_ctxs, row)) {
row_batch->commit_last_row();
++_num_rows_returned;
COUNTER_SET(_rows_returned_counter, _num_rows_returned);
char* new_tuple = reinterpret_cast<char*>(_tuple);
new_tuple += _tuple_desc->byte_size();
_tuple = reinterpret_cast<Tuple*>(new_tuple);
}
}
}
Status OdbcScanNode::close(RuntimeState* state) {
if (is_closed()) {
return Status::OK();
}
RETURN_IF_ERROR(exec_debug_action(TExecNodePhase::CLOSE));
SCOPED_TIMER(_runtime_profile->total_time_counter());
_tuple_pool.reset();
return ExecNode::close(state);
}
void OdbcScanNode::debug_string(int indentation_level, std::stringstream* out) const {
*out << string(indentation_level * 2, ' ');
*out << _scan_node_type << "(tupleid=" << _tuple_id << " table=" << _table_name;
*out << ")" << std::endl;
for (int i = 0; i < _children.size(); ++i) {
_children[i]->debug_string(indentation_level + 1, out);
}
}
Status OdbcScanNode::set_scan_ranges(const std::vector<TScanRangeParams>& scan_ranges) {
return Status::OK();
}
} // namespace doris