Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

More efficient constructor for SerializationEnum #57887

Merged
merged 2 commits into from Dec 19, 2023
Merged

More efficient constructor for SerializationEnum #57887

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
6e6eb5c
fast get default enum serialization
canhld94 Dec 15, 2023
b07cde5
use const reference instead for pointer to values
canhld94 Dec 18, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Jump to
Jump to file
Failed to load files.
Diff view
Diff view
2 changes: 1 addition & 1 deletion src/DataTypes/DataTypeEnum.cpp
View file
Open in desktop
Expand Up @@ -170,7 +170,7 @@ bool DataTypeEnum<Type>::contains(const IDataType & rhs) const
template <typename Type>
SerializationPtr DataTypeEnum<Type>::doGetDefaultSerialization() const
{
return std::make_shared<SerializationEnum<Type>>(this->getValues());
return std::make_shared<SerializationEnum<Type>>(std::static_pointer_cast<const DataTypeEnum<Type>>(shared_from_this()));
}


Expand Down
24 changes: 12 additions & 12 deletions src/DataTypes/Serializations/SerializationEnum.cpp
View file
Open in desktop
Expand Up @@ -11,13 +11,13 @@ namespace DB
template <typename Type>
void SerializationEnum<Type>::serializeText(const IColumn & column, size_t row_num, WriteBuffer & ostr, const FormatSettings &) const
{
writeString(this->getNameForValue(assert_cast<const ColumnType &>(column).getData()[row_num]), ostr);
writeString(ref_enum_values.getNameForValue(assert_cast<const ColumnType &>(column).getData()[row_num]), ostr);
}

template <typename Type>
void SerializationEnum<Type>::serializeTextEscaped(const IColumn & column, size_t row_num, WriteBuffer & ostr, const FormatSettings &) const
{
writeEscapedString(this->getNameForValue(assert_cast<const ColumnType &>(column).getData()[row_num]).toView(), ostr);
writeEscapedString(ref_enum_values.getNameForValue(assert_cast<const ColumnType &>(column).getData()[row_num]).toView(), ostr);
}

template <typename Type>
Expand All @@ -30,22 +30,22 @@ void SerializationEnum<Type>::deserializeTextEscaped(IColumn & column, ReadBuffe
/// NOTE It would be nice to do without creating a temporary object - at least extract std::string out.
std::string field_name;
readEscapedString(field_name, istr);
assert_cast<ColumnType &>(column).getData().push_back(this->getValue(StringRef(field_name), true));
assert_cast<ColumnType &>(column).getData().push_back(ref_enum_values.getValue(StringRef(field_name), true));
}
}

template <typename Type>
void SerializationEnum<Type>::serializeTextQuoted(const IColumn & column, size_t row_num, WriteBuffer & ostr, const FormatSettings &) const
{
writeQuotedString(this->getNameForValue(assert_cast<const ColumnType &>(column).getData()[row_num]), ostr);
writeQuotedString(ref_enum_values.getNameForValue(assert_cast<const ColumnType &>(column).getData()[row_num]), ostr);
}

template <typename Type>
void SerializationEnum<Type>::deserializeTextQuoted(IColumn & column, ReadBuffer & istr, const FormatSettings &) const
{
std::string field_name;
readQuotedStringWithSQLStyle(field_name, istr);
assert_cast<ColumnType &>(column).getData().push_back(this->getValue(StringRef(field_name)));
assert_cast<ColumnType &>(column).getData().push_back(ref_enum_values.getValue(StringRef(field_name)));
}

template <typename Type>
Expand All @@ -61,20 +61,20 @@ void SerializationEnum<Type>::deserializeWholeText(IColumn & column, ReadBuffer
{
std::string field_name;
readStringUntilEOF(field_name, istr);
assert_cast<ColumnType &>(column).getData().push_back(this->getValue(StringRef(field_name), true));
assert_cast<ColumnType &>(column).getData().push_back(ref_enum_values.getValue(StringRef(field_name), true));
}
}

template <typename Type>
void SerializationEnum<Type>::serializeTextJSON(const IColumn & column, size_t row_num, WriteBuffer & ostr, const FormatSettings & settings) const
{
writeJSONString(this->getNameForValue(assert_cast<const ColumnType &>(column).getData()[row_num]).toView(), ostr, settings);
writeJSONString(ref_enum_values.getNameForValue(assert_cast<const ColumnType &>(column).getData()[row_num]).toView(), ostr, settings);
}

template <typename Type>
void SerializationEnum<Type>::serializeTextXML(const IColumn & column, size_t row_num, WriteBuffer & ostr, const FormatSettings &) const
{
writeXMLStringForTextElement(this->getNameForValue(assert_cast<const ColumnType &>(column).getData()[row_num]).toView(), ostr);
writeXMLStringForTextElement(ref_enum_values.getNameForValue(assert_cast<const ColumnType &>(column).getData()[row_num]).toView(), ostr);
}

template <typename Type>
Expand All @@ -86,14 +86,14 @@ void SerializationEnum<Type>::deserializeTextJSON(IColumn & column, ReadBuffer &
{
std::string field_name;
readJSONString(field_name, istr);
assert_cast<ColumnType &>(column).getData().push_back(this->getValue(StringRef(field_name)));
assert_cast<ColumnType &>(column).getData().push_back(ref_enum_values.getValue(StringRef(field_name)));
}
}

template <typename Type>
void SerializationEnum<Type>::serializeTextCSV(const IColumn & column, size_t row_num, WriteBuffer & ostr, const FormatSettings &) const
{
writeCSVString(this->getNameForValue(assert_cast<const ColumnType &>(column).getData()[row_num]), ostr);
writeCSVString(ref_enum_values.getNameForValue(assert_cast<const ColumnType &>(column).getData()[row_num]), ostr);
}

template <typename Type>
Expand All @@ -105,7 +105,7 @@ void SerializationEnum<Type>::deserializeTextCSV(IColumn & column, ReadBuffer &
{
std::string field_name;
readCSVString(field_name, istr, settings.csv);
assert_cast<ColumnType &>(column).getData().push_back(this->getValue(StringRef(field_name), true));
assert_cast<ColumnType &>(column).getData().push_back(ref_enum_values.getValue(StringRef(field_name), true));
}
}

Expand All @@ -114,7 +114,7 @@ void SerializationEnum<Type>::serializeTextMarkdown(
const IColumn & column, size_t row_num, WriteBuffer & ostr, const FormatSettings & settings) const
{
if (settings.markdown.escape_special_characters)
writeMarkdownEscapedString(this->getNameForValue(assert_cast<const ColumnType &>(column).getData()[row_num]).toView(), ostr);
writeMarkdownEscapedString(ref_enum_values.getNameForValue(assert_cast<const ColumnType &>(column).getData()[row_num]).toView(), ostr);
else
serializeTextEscaped(column, row_num, ostr, settings);
}
Expand Down
27 changes: 23 additions & 4 deletions src/DataTypes/Serializations/SerializationEnum.h
View file
Open in desktop
@@ -1,20 +1,35 @@
#pragma once

#include <memory>
#include <DataTypes/Serializations/SerializationNumber.h>
#include <DataTypes/EnumValues.h>
#include <DataTypes/DataTypeEnum.h>

namespace DB
{

template <typename Type>
class SerializationEnum : public SerializationNumber<Type>, public EnumValues<Type>
class SerializationEnum : public SerializationNumber<Type>
{
public:
using typename SerializationNumber<Type>::FieldType;
using typename SerializationNumber<Type>::ColumnType;
using typename EnumValues<Type>::Values;
using Values = EnumValues<Type>::Values;

explicit SerializationEnum(const Values & values_) : EnumValues<Type>(values_) {}
// SerializationEnum can be constructed in two ways:
/// - Make a copy of the Enum name-to-type mapping.
/// - Only store a reference to an existing mapping. This is faster if the Enum has a lot of different values or if SerializationEnum is
/// constructed very frequently. Make sure that the pointed-to mapping has a longer lifespan than SerializationEnum!

explicit SerializationEnum(const Values & values_)
: own_enum_values(values_), ref_enum_values(own_enum_values.value())
{
}

explicit SerializationEnum(const std::shared_ptr<const DataTypeEnum<Type>> & enum_type)
: own_enum_type(enum_type), ref_enum_values(*enum_type)
{
}

void serializeText(const IColumn & column, size_t row_num, WriteBuffer & ostr, const FormatSettings &) const override;
void serializeTextEscaped(const IColumn & column, size_t row_num, WriteBuffer & ostr, const FormatSettings &) const override;
Expand All @@ -35,8 +50,12 @@ class SerializationEnum : public SerializationNumber<Type>, public EnumValues<Ty
{
FieldType x;
readText(x, istr);
return this->findByValue(x)->first;
return ref_enum_values.findByValue(x)->first;
}

std::optional<EnumValues<Type>> own_enum_values;
std::shared_ptr<const DataTypeEnum<Type>> own_enum_type;
const EnumValues<Type> & ref_enum_values;
};

}