database/mariadb-columnstore-engine
1
0
Fork 0
mirror of https://github.com/mariadb-corporation/mariadb-columnstore-engine.git synced 2025-07-29 08:21:15 +03:00
Code Activity

MCOL-4580 extent elimination for dictionary-based text/varchar types

Browse Source 
The idea is relatively simple - encode prefixes of collated strings as
integers and use them to compute extents' ranges. Then we can eliminate
extents with strings.

The actual patch does have all the code there but miss one important
step: we do not keep collation index, we keep charset index. Because of
this, some of the tests in the bugfix suite fail and thus main
functionality is turned off.

The reason of this patch to be put into PR at all is that it contains
changes that made CHAR/VARCHAR columns unsigned. This change is needed in
vectorization work.
This commit is contained in:
Serguey Zefirov
2022-02-04 11:55:09 +00:00
parent a66a8dfabf
commit 53b9a2a0f9
54 changed files with 698 additions and 227 deletions
Download Patch File Download Diff File Expand all files Collapse all files

22
writeengine/shared/we_brm.h
View File

@ -24,6 +24,7 @@
#pragma once
#include <iostream>
#include <memory>
#include <vector>
#include <boost/thread.hpp>
#include <boost/thread/tss.hpp>
@ -49,17 +50,19 @@ namespace WriteEngine
// forward reference
class DbFileOp;
/** @brief Extended CPInfo - with type handler for all type-related information */
/** @brief Extended CPInfo - with all type-related information and associated range data */
struct ExtCPInfo
{
execplan::CalpontSystemCatalog::ColDataType fColType;
int fColWidth;
BRM::CPInfo fCPInfo;
std::shared_ptr<std::vector<int64_t>> fStringsPrefixes;
ExtCPInfo(execplan::CalpontSystemCatalog::ColDataType colType, int colWidth)
: fColType(colType), fColWidth(colWidth)
{
fCPInfo.isBinaryColumn = (unsigned int)colWidth > datatypes::MAXLEGACYWIDTH;
}
void toInvalid()
{
auto mm = datatypes::MinMaxInfo::invalidRange(fColType);
@ -68,7 +71,22 @@ struct ExtCPInfo
fCPInfo.bigMax = mm.int128Max;
fCPInfo.bigMin = mm.int128Min;
}
void addStringPrefix(int64_t strPrefix)
{
if (!fStringsPrefixes)
{
fStringsPrefixes.reset(new std::vector<int64_t>());
}
fStringsPrefixes->push_back(strPrefix);
}
bool hasStringsPrefixes() const
{
return fStringsPrefixes.get() != nullptr;
}
int64_t* stringsPrefixes() const
{
return hasStringsPrefixes() ? fStringsPrefixes->data() : nullptr;
}
bool isInvalid()
{
datatypes::MinMaxInfo mm;

2
writeengine/shared/we_type.h
View File

@ -344,6 +344,7 @@ struct DctnryStruct /** @brief Dctnry Interface Struct*/
uint16_t fColSegment; /** @brief Segment for column file */
uint16_t fColDbRoot; /** @brief DBRoot for column file */
int fCompressionType; /** @brief Compression tpye for column file */
int fCharsetNumber; /** @brief Charset number to account for collation when computing string prefixes */
DctnryStruct()
: dctnryOid(0)
, columnOid(0)
@ -353,6 +354,7 @@ struct DctnryStruct /** @brief Dctnry Interface Struct*/
, fColSegment(0)
, fColDbRoot(0)
, fCompressionType(idbdatafile::IDBPolicy::useHdfs() ? 2 : 0)
, fCharsetNumber(8)
{
}
};