database/mariadb-columnstore-engine
1
0
Fork 0
mirror of https://github.com/mariadb-corporation/mariadb-columnstore-engine.git synced 2025-04-23 07:05:36 +03:00
Code
mariadb-columnstore-engine/utils/funcexp/funcexp.cpp

558 lines
19 KiB
C++
Raw Permalink Blame History

/* Copyright (C) 2014 InfiniDB, Inc.
Copyright (C) 2019 MariaDB Corporation
This program is free software; you can redistribute it and/or
modify it under the terms of the GNU General Public License
as published by the Free Software Foundation; version 2 of
the License.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program; if not, write to the Free Software
Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
MA 02110-1301, USA. */
/****************************************************************************
* $Id: funcexp.cpp 3954 2013-07-08 16:30:15Z bpaul $
*
*
****************************************************************************/
#include "functor_json.h"
#include <boost/thread/mutex.hpp>
#include "funcexp.h"
#include "functor_all.h"
#include "functor_bool.h"
#include "functor_dtm.h"
#include "functor_int.h"
#include "functor_real.h"
#include "functor_str.h"
#include "functor_export.h"
#include "calpontsystemcatalog.h"
using namespace execplan;
#include "joblisttypes.h"
using namespace joblist;
#ifndef SKIP_UDF
#include "../udfsdk/udfsdk.h"
#endif
#include "mcs_decimal.h"
namespace funcexp
{
/* static */
FuncExp* FuncExp::fInstance = 0;
/* static */
boost::mutex FuncExp::fInstanceMutex;
FuncExp* FuncExp::instance()
{
boost::mutex::scoped_lock lk(fInstanceMutex);
if (!fInstance)
fInstance = new FuncExp();
return fInstance;
}
FuncExp::FuncExp()
{
fFuncMap["<<"] = new Func_leftshift();
fFuncMap[">>"] = new Func_rightshift();
fFuncMap["|"] = new Func_bitor();
fFuncMap["^"] = new Func_bitxor();
fFuncMap["&"] = new Func_bitand();
fFuncMap["abs"] = new Func_abs();
fFuncMap["acos"] = new Func_acos();
fFuncMap["add_time"] = new Func_add_time();
fFuncMap["addtime"] = new Func_add_time();
fFuncMap["subtime"] = new Func_add_time();
fFuncMap["asin"] = new Func_asin();
fFuncMap["ascii"] = new Func_ascii();
fFuncMap["atan"] = new Func_atan();
fFuncMap["atan2"] = new Func_atan();
fFuncMap["between"] = new Func_between();
fFuncMap["bit_count"] = new Func_bit_count();
fFuncMap["case_searched"] = new Func_searched_case();
fFuncMap["case_simple"] = new Func_simple_case();
fFuncMap["cast_as_signed"] = new Func_cast_signed(); // dlh
fFuncMap["cast_as_unsigned"] = new Func_cast_unsigned(); // dch
fFuncMap["cast_as_char"] = new Func_cast_char(); // dlh
fFuncMap["cast_as_date"] = new Func_cast_date(); // dlh
fFuncMap["cast_as_datetime"] = new Func_cast_datetime(); // dlh
fFuncMap["decimal_typecast"] = new Func_cast_decimal(); // dlh
fFuncMap["double_typecast"] = new Func_cast_double();
fFuncMap["ceil"] = new Func_ceil(); // dlh
fFuncMap["ceiling"] = new Func_ceil(); // dlh
fFuncMap["char"] = new Func_char(); // dlh
fFuncMap["char_length"] = new Func_char_length(); // dlh
fFuncMap["character_length"] = new Func_char_length(); // dlh
fFuncMap["coalesce"] = new Func_coalesce();
fFuncMap["concat_operator_oracle"] = new Func_concat_oracle();
fFuncMap["concat"] = new Func_concat();
fFuncMap["concat_ws"] = new Func_concat_ws();
fFuncMap["conv"] = new Func_conv();
fFuncMap["cos"] = new Func_cos();
fFuncMap["cot"] = new Func_cot();
fFuncMap["convert"] = new Func_cast_char(); // dlh
fFuncMap["convert_tz"] = new Func_convert_tz(); // BT
fFuncMap["crc32"] = new Func_crc32();
fFuncMap["date_add_interval"] = new Func_date_add(); // dlh
fFuncMap["date_format"] = new Func_date_format();
fFuncMap["day"] = new Func_day(); // dlh
fFuncMap["dayname"] = new Func_dayname();
fFuncMap["dayofmonth"] = new Func_day(); // dlh
fFuncMap["dayofweek"] = new Func_dayofweek(); // dlh
fFuncMap["dayofyear"] = new Func_dayofyear(); // dlh
fFuncMap["decode"] = new Func_decode(); // BT
fFuncMap["decode_oracle"] = new Func_decode_oracle(); // BT
fFuncMap["degrees"] = new Func_degrees();
fFuncMap["DIV"] = new Func_div(); // MySQL use upper case for this function name
fFuncMap["elt"] = new Func_elt();
fFuncMap["encode"] = new Func_encode(); // BT
fFuncMap["exp"] = new Func_exp();
fFuncMap["extract"] = new Func_extract(); // dlh
fFuncMap["find_in_set"] = new Func_find_in_set();
fFuncMap["floor"] = new Func_floor(); // dlh
fFuncMap["format"] = new Func_format(); // dlh
fFuncMap["from_days"] = new Func_from_days();
fFuncMap["from_unixtime"] = new Func_from_unixtime();
fFuncMap["get_format"] = new Func_get_format(); // dlh
fFuncMap["greatest"] = new Func_greatest(); // dlh
fFuncMap["hex"] = new Func_hex();
fFuncMap["hour"] = new Func_hour(); // dlh
fFuncMap["idbpartition"] = new Func_idbpartition(); // pseudo column
fFuncMap["if"] = new Func_if();
fFuncMap["ifnull"] = new Func_ifnull();
fFuncMap["in"] = new Func_in();
fFuncMap[" IN "] = new Func_in();
fFuncMap["inet_aton"] = new Func_inet_aton();
fFuncMap["inet_ntoa"] = new Func_inet_ntoa();
fFuncMap["insert"] = new Func_insert();
fFuncMap["instr"] = new Func_instr();
fFuncMap["isnull"] = new Func_isnull(false);
fFuncMap["isnotnull"] = new Func_isnull(true);
fFuncMap["istrue"] = new Func_IsTrue();
fFuncMap["isnottrue"] = new Func_IsNotTrue();
fFuncMap["isfalse"] = new Func_IsFalse();
fFuncMap["isnotfalse"] = new Func_IsNotFalse();
fFuncMap["json_array"] = new Func_json_array();
fFuncMap["json_array_append"] = new Func_json_array_append();
fFuncMap["json_array_insert"] = new Func_json_array_insert();
fFuncMap["json_contains"] = new Func_json_contains();
fFuncMap["json_contains_path"] = new Func_json_contains_path();
fFuncMap["json_compact"] = new Func_json_format(Func_json_format::COMPACT);
fFuncMap["json_depth"] = new Func_json_depth();
fFuncMap["json_equals"] = new Func_json_equals();
fFuncMap["json_exists"] = new Func_json_exists();
fFuncMap["json_extract"] = new Func_json_extract();
fFuncMap["json_format"] = new Func_json_format();
fFuncMap["json_insert"] = new Func_json_insert();
fFuncMap["json_keys"] = new Func_json_keys();
fFuncMap["json_length"] = new Func_json_length();
fFuncMap["json_loose"] = new Func_json_format(Func_json_format::LOOSE);
fFuncMap["json_merge"] = new Func_json_merge();
fFuncMap["json_merge_patch"] = new Func_json_merge_patch();
fFuncMap["json_merge_preserve"] = new Func_json_merge();
fFuncMap["json_normalize"] = new Func_json_normalize();
fFuncMap["json_object"] = new Func_json_object();
#ifdef MYSQL_GE_1009
fFuncMap["json_overlaps"] = new Func_json_overlaps();
#endif
fFuncMap["json_query"] = new Func_json_query();
fFuncMap["json_quote"] = new Func_json_quote();
fFuncMap["json_remove"] = new Func_json_remove();
fFuncMap["json_replace"] = new Func_json_insert(Func_json_insert::REPLACE);
fFuncMap["json_search"] = new Func_json_search();
fFuncMap["json_set"] = new Func_json_insert(Func_json_insert::SET);
fFuncMap["json_type"] = new Func_json_type();
fFuncMap["json_unquote"] = new Func_json_unquote();
fFuncMap["json_valid"] = new Func_json_valid();
fFuncMap["json_value"] = new Func_json_value();
fFuncMap["last_day"] = new Func_last_day();
fFuncMap["lcase"] = new Func_lcase(); // dlh
fFuncMap["least"] = new Func_least(); // dlh
fFuncMap["left"] = new Func_left(); // dlh
fFuncMap["length"] = new Func_length();
fFuncMap["octet_length"] = new Func_length(); // MariaDB 10.3
fFuncMap["ln"] = new Func_log();
fFuncMap["locate"] = new Func_instr();
fFuncMap["log"] = new Func_log();
fFuncMap["log2"] = new Func_log2();
fFuncMap["log10"] = new Func_log10();
fFuncMap["lower"] = new Func_lcase(); // dlh
fFuncMap["lpad"] = new Func_lpad(); // dlh
fFuncMap["ltrim"] = new Func_ltrim(); // dlh
fFuncMap["ltrim_oracle"] = new Func_ltrim_oracle(); // dlh
fFuncMap["makedate"] = new Func_makedate();
fFuncMap["maketime"] = new Func_maketime();
fFuncMap["microsecond"] = new Func_microsecond();
fFuncMap["minute"] = new Func_minute(); // dlh
fFuncMap["mod"] = new Func_mod(); // dlh
fFuncMap["MOD"] = new Func_mod(); // MariaDB 10.3
fFuncMap["%"] = new Func_mod(); // dlh
fFuncMap["md5"] = new Func_md5();
fFuncMap["mid"] = new Func_substr();
fFuncMap["month"] = new Func_month(); // dlh
fFuncMap["monthname"] = new Func_monthname();
fFuncMap["notin"] = new Func_notin();
fFuncMap["not IN "] = new Func_notin();
fFuncMap["notbetween"] = new Func_notbetween();
fFuncMap["nullif"] = new Func_nullif();
fFuncMap["period_add"] = new Func_period_add(); // dlh
fFuncMap["period_diff"] = new Func_period_diff(); // dlh
fFuncMap["position"] = new Func_instr(); // dlh
fFuncMap["pow"] = new Func_pow();
fFuncMap["power"] = new Func_pow();
fFuncMap["quote"] = new Func_quote();
fFuncMap["quarter"] = new Func_quarter();
fFuncMap["radians"] = new Func_radians(); // dlh
fFuncMap["rand"] = new Func_rand();
fFuncMap["regexp"] = new Func_regexp();
fFuncMap["regexp_instr"] = new Func_regexp_instr();
fFuncMap["regexp_replace"] = new Func_regexp_replace();
fFuncMap["regexp_substr"] = new Func_regexp_substr(); // dlh
fFuncMap["repeat"] = new Func_repeat(); // dlh
fFuncMap["replace"] = new Func_replace(); // dlh
fFuncMap["replace_oracle"] = new Func_replace_oracle(); // dlh
fFuncMap["reverse"] = new Func_reverse(); // dlh
fFuncMap["right"] = new Func_right(); // dlh
fFuncMap["round"] = new Func_round();
fFuncMap["rpad"] = new Func_rpad(); // dlh
fFuncMap["rtrim"] = new Func_rtrim(); // dlh
fFuncMap["rtrim_oracle"] = new Func_rtrim_oracle(); // dlh
fFuncMap["second"] = new Func_second(); // dlh
fFuncMap["sec_to_time"] = new Func_sec_to_time();
fFuncMap["sha"] = new Func_sha();
fFuncMap["sha1"] = new Func_sha();
fFuncMap["sign"] = new Func_sign();
fFuncMap["sin"] = new Func_sin();
fFuncMap["space"] = new Func_space();
fFuncMap["sqrt"] = new Func_sqrt();
fFuncMap["str_to_date"] = new Func_str_to_date();
fFuncMap["strcmp"] = new Func_strcmp();
fFuncMap["substr"] = new Func_substr();
fFuncMap["substring"] = new Func_substr(); // dlh
fFuncMap["substring_index"] = new Func_substring_index(); // dlh
fFuncMap["sysdate"] = new Func_sysdate(); // dlhFuncExp
fFuncMap["cast_as_time"] = new Func_time(); // dlh
fFuncMap["tan"] = new Func_tan();
fFuncMap["timediff"] = new Func_timediff(); // dlh
fFuncMap["timestampdiff"] = new Func_timestampdiff();
fFuncMap["time_format"] = new Func_time_format(); // dlh
fFuncMap["time_to_sec"] = new Func_time_to_sec(); // dlh
fFuncMap["to_days"] = new Func_to_days(); // dlh
fFuncMap["trim"] = new Func_trim(); // dlh
fFuncMap["trim_oracle"] = new Func_trim_oracle(); // dlh
fFuncMap["truncate"] = new Func_truncate(); // dlh
fFuncMap["ucase"] = new Func_ucase(); // dlh
fFuncMap["unhex"] = new Func_unhex();
fFuncMap["unix_timestamp"] = new Func_unix_timestamp();
fFuncMap["upper"] = new Func_ucase(); // dlh
fFuncMap["week"] = new Func_week(); // dlh
fFuncMap["weekday"] = new Func_weekday();
fFuncMap["weekofyear"] = new Func_week(); // dlh
fFuncMap["year"] = new Func_year(); // dlh
fFuncMap["yearweek"] = new Func_yearweek(); // dlh
#ifndef SKIP_UDF
udfsdk::UDFSDK sdk;
FuncMap sdkfm = sdk.UDFMap();
FuncMap::const_iterator iter = sdkfm.begin();
FuncMap::iterator funcMapIter;
for (; iter != sdkfm.end(); ++iter)
{
// add sdkfm to fFuncMap
funcMapIter = fFuncMap.find(iter->first);
if (funcMapIter != fFuncMap.end())
{
// silently ignore the udf function? log it?
}
else
{
fFuncMap[iter->first] = iter->second;
}
}
#endif
}
Func* FuncExp::getFunctor(std::string& funcName)
{
FuncMap::iterator iter = fFuncMap.find(funcName);
if (iter == fFuncMap.end())
return NULL;
else
return (*iter).second;
}
void FuncExp::evaluate(rowgroup::Row& row, std::vector<execplan::SRCP>& expression)
{
bool isNull;
for (uint32_t i = 0; i < expression.size(); i++)
{
isNull = false;
switch (expression[i]->resultType().colDataType)
{
case CalpontSystemCatalog::DATE:
{
int64_t val = expression[i]->getIntVal(row, isNull);
// @bug6061, workaround date_add always return datetime for both date and datetime
if (val & 0xFFFFFFFF00000000)
val = (((val >> 32) & 0xFFFFFFC0) | 0x3E);
if (isNull)
row.setUintField<4>(DATENULL, expression[i]->outputIndex());
else
row.setUintField<4>(val, expression[i]->outputIndex());
break;
}
case CalpontSystemCatalog::DATETIME:
{
int64_t val = expression[i]->getDatetimeIntVal(row, isNull);
if (isNull)
row.setUintField<8>(DATETIMENULL, expression[i]->outputIndex());
else
row.setUintField<8>(val, expression[i]->outputIndex());
break;
}
case CalpontSystemCatalog::TIMESTAMP:
{
int64_t val = expression[i]->getTimestampIntVal(row, isNull);
if (isNull)
row.setUintField<8>(TIMESTAMPNULL, expression[i]->outputIndex());
else
row.setUintField<8>(val, expression[i]->outputIndex());
break;
}
case CalpontSystemCatalog::TIME:
{
int64_t val = expression[i]->getTimeIntVal(row, isNull);
if (isNull)
row.setIntField<8>(TIMENULL, expression[i]->outputIndex());
else
row.setIntField<8>(val, expression[i]->outputIndex());
break;
}
case CalpontSystemCatalog::CHAR:
case CalpontSystemCatalog::VARCHAR:
// TODO: might not be right thing for BLOB
case CalpontSystemCatalog::BLOB:
case CalpontSystemCatalog::TEXT:
{
const utils::NullString& val = expression[i]->getStrVal(row, isNull);
// XXX: TODO: we may as well set the string field directly.
if (isNull)
{
utils::NullString nullstr;
row.setStringField(nullstr, expression[i]->outputIndex());
}
else
row.setStringField(val, expression[i]->outputIndex());
break;
}
case CalpontSystemCatalog::BIGINT:
{
int64_t val = expression[i]->getIntVal(row, isNull);
if (isNull)
row.setIntField<8>(BIGINTNULL, expression[i]->outputIndex());
else
row.setIntField<8>(val, expression[i]->outputIndex());
break;
}
case CalpontSystemCatalog::UBIGINT:
{
uint64_t val = expression[i]->getUintVal(row, isNull);
if (isNull)
row.setUintField<8>(UBIGINTNULL, expression[i]->outputIndex());
else
row.setUintField<8>(val, expression[i]->outputIndex());
break;
}
case CalpontSystemCatalog::INT:
case CalpontSystemCatalog::MEDINT:
{
int64_t val = expression[i]->getIntVal(row, isNull);
if (isNull)
row.setIntField<4>(INTNULL, expression[i]->outputIndex());
else
row.setIntField<4>(val, expression[i]->outputIndex());
break;
}
case CalpontSystemCatalog::UINT:
case CalpontSystemCatalog::UMEDINT:
{
uint64_t val = expression[i]->getUintVal(row, isNull);
if (isNull)
row.setUintField<4>(UINTNULL, expression[i]->outputIndex());
else
row.setUintField<4>(val, expression[i]->outputIndex());
break;
}
case CalpontSystemCatalog::SMALLINT:
{
int64_t val = expression[i]->getIntVal(row, isNull);
if (isNull)
row.setIntField<2>(SMALLINTNULL, expression[i]->outputIndex());
else
row.setIntField<2>(val, expression[i]->outputIndex());
break;
}
case CalpontSystemCatalog::USMALLINT:
{
uint64_t val = expression[i]->getUintVal(row, isNull);
if (isNull)
row.setUintField<2>(USMALLINTNULL, expression[i]->outputIndex());
else
row.setUintField<2>(val, expression[i]->outputIndex());
break;
}
case CalpontSystemCatalog::TINYINT:
{
int64_t val = expression[i]->getIntVal(row, isNull);
if (isNull)
row.setIntField<1>(TINYINTNULL, expression[i]->outputIndex());
else
row.setIntField<1>(val, expression[i]->outputIndex());
break;
}
case CalpontSystemCatalog::UTINYINT:
{
uint64_t val = expression[i]->getUintVal(row, isNull);
if (isNull)
row.setUintField<1>(UTINYINTNULL, expression[i]->outputIndex());
else
row.setUintField<1>(val, expression[i]->outputIndex());
break;
}
// In this case, we're trying to load a double output column with float data. This is the
// case when you do sum(floatcol), e.g.
case CalpontSystemCatalog::DOUBLE:
case CalpontSystemCatalog::UDOUBLE:
{
double val = expression[i]->getDoubleVal(row, isNull);
if (isNull)
row.setIntField<8>(DOUBLENULL, expression[i]->outputIndex());
else
row.setDoubleField(val, expression[i]->outputIndex());
break;
}
case CalpontSystemCatalog::FLOAT:
case CalpontSystemCatalog::UFLOAT:
{
float val = expression[i]->getFloatVal(row, isNull);
if (isNull)
row.setIntField<4>(FLOATNULL, expression[i]->outputIndex());
else
row.setFloatField(val, expression[i]->outputIndex());
break;
}
case CalpontSystemCatalog::LONGDOUBLE:
{
long double val = expression[i]->getLongDoubleVal(row, isNull);
if (isNull)
row.setLongDoubleField(LONGDOUBLENULL, expression[i]->outputIndex());
else
row.setLongDoubleField(val, expression[i]->outputIndex());
break;
}
case CalpontSystemCatalog::DECIMAL:
case CalpontSystemCatalog::UDECIMAL:
{
IDB_Decimal val = expression[i]->getDecimalVal(row, isNull);
if (expression[i]->resultType().colWidth == datatypes::MAXDECIMALWIDTH)
{
if (isNull)
{
row.setBinaryField_offset(const_cast<int128_t*>(&datatypes::Decimal128Null),
expression[i]->resultType().colWidth,
row.getOffset(expression[i]->outputIndex()));
}
else
{
row.setBinaryField_offset(&val.s128Value, expression[i]->resultType().colWidth,
row.getOffset(expression[i]->outputIndex()));
}
}
else
{
if (isNull)
row.setIntField<8>(BIGINTNULL, expression[i]->outputIndex());
else
row.setIntField<8>(val.value, expression[i]->outputIndex());
}
break;
}
default: // treat as int64
{
throw std::runtime_error("funcexp::evaluate(): non support datatype to set field.");
}
}
}
}
} // namespace funcexp
View Git Blame Copy Permalink