mirror of
https://github.com/ClickHouse/ClickHouse.git
synced 2024-11-19 14:11:58 +00:00
900 lines
26 KiB
C++
900 lines
26 KiB
C++
#pragma once
|
|
|
|
#include <cmath>
|
|
#include <cstring>
|
|
#include <limits>
|
|
#include <algorithm>
|
|
#include <iterator>
|
|
|
|
#include <type_traits>
|
|
|
|
#include <common/DateLUT.h>
|
|
#include <common/LocalDate.h>
|
|
#include <common/LocalDateTime.h>
|
|
|
|
#include <Core/Types.h>
|
|
#include <Core/UUID.h>
|
|
#include <common/StringRef.h>
|
|
#include <Common/Exception.h>
|
|
#include <Common/StringUtils/StringUtils.h>
|
|
#include <Common/Arena.h>
|
|
#include <Common/UInt128.h>
|
|
|
|
#include <Formats/FormatSettings.h>
|
|
|
|
#include <IO/ReadBuffer.h>
|
|
#include <IO/ReadBufferFromMemory.h>
|
|
#include <IO/VarInt.h>
|
|
|
|
#ifdef __clang__
|
|
#pragma clang diagnostic push
|
|
#pragma clang diagnostic ignored "-Wdouble-promotion"
|
|
#endif
|
|
|
|
#include <double-conversion/double-conversion.h>
|
|
|
|
#ifdef __clang__
|
|
#pragma clang diagnostic pop
|
|
#endif
|
|
|
|
|
|
#define DEFAULT_MAX_STRING_SIZE 0x00FFFFFFULL
|
|
|
|
|
|
namespace DB
|
|
{
|
|
|
|
namespace ErrorCodes
|
|
{
|
|
extern const int CANNOT_PARSE_DATE;
|
|
extern const int CANNOT_PARSE_DATETIME;
|
|
extern const int CANNOT_PARSE_UUID;
|
|
extern const int CANNOT_READ_ARRAY_FROM_TEXT;
|
|
extern const int CANNOT_PARSE_NUMBER;
|
|
extern const int ILLEGAL_TYPE_OF_ARGUMENT;
|
|
}
|
|
|
|
/// Helper functions for formatted input.
|
|
|
|
inline char parseEscapeSequence(char c)
|
|
{
|
|
switch(c)
|
|
{
|
|
case 'a':
|
|
return '\a';
|
|
case 'b':
|
|
return '\b';
|
|
case 'e':
|
|
return '\x1B'; /// \e escape sequence is non standard for C and C++ but supported by gcc and clang.
|
|
case 'f':
|
|
return '\f';
|
|
case 'n':
|
|
return '\n';
|
|
case 'r':
|
|
return '\r';
|
|
case 't':
|
|
return '\t';
|
|
case 'v':
|
|
return '\v';
|
|
case '0':
|
|
return '\0';
|
|
default:
|
|
return c;
|
|
}
|
|
}
|
|
|
|
|
|
/// These functions are located in VarInt.h
|
|
/// inline void throwReadAfterEOF()
|
|
|
|
|
|
inline void readChar(char & x, ReadBuffer & buf)
|
|
{
|
|
if (!buf.eof())
|
|
{
|
|
x = *buf.position();
|
|
++buf.position();
|
|
}
|
|
else
|
|
throwReadAfterEOF();
|
|
}
|
|
|
|
|
|
/// Read POD-type in native format
|
|
template <typename T>
|
|
inline void readPODBinary(T & x, ReadBuffer & buf)
|
|
{
|
|
buf.readStrict(reinterpret_cast<char *>(&x), sizeof(x));
|
|
}
|
|
|
|
template <typename T>
|
|
inline void readIntBinary(T & x, ReadBuffer & buf)
|
|
{
|
|
readPODBinary(x, buf);
|
|
}
|
|
|
|
template <typename T>
|
|
inline void readFloatBinary(T & x, ReadBuffer & buf)
|
|
{
|
|
readPODBinary(x, buf);
|
|
}
|
|
|
|
|
|
inline void readStringBinary(std::string & s, ReadBuffer & buf, size_t MAX_STRING_SIZE = DEFAULT_MAX_STRING_SIZE)
|
|
{
|
|
size_t size = 0;
|
|
readVarUInt(size, buf);
|
|
|
|
if (size > MAX_STRING_SIZE)
|
|
throw Poco::Exception("Too large string size.");
|
|
|
|
s.resize(size);
|
|
buf.readStrict(s.data(), size);
|
|
}
|
|
|
|
|
|
inline StringRef readStringBinaryInto(Arena & arena, ReadBuffer & buf)
|
|
{
|
|
size_t size = 0;
|
|
readVarUInt(size, buf);
|
|
|
|
char * data = arena.alloc(size);
|
|
buf.readStrict(data, size);
|
|
|
|
return StringRef(data, size);
|
|
}
|
|
|
|
|
|
template <typename T>
|
|
void readVectorBinary(std::vector<T> & v, ReadBuffer & buf, size_t MAX_VECTOR_SIZE = DEFAULT_MAX_STRING_SIZE)
|
|
{
|
|
size_t size = 0;
|
|
readVarUInt(size, buf);
|
|
|
|
if (size > MAX_VECTOR_SIZE)
|
|
throw Poco::Exception("Too large vector size.");
|
|
|
|
v.resize(size);
|
|
for (size_t i = 0; i < size; ++i)
|
|
readBinary(v[i], buf);
|
|
}
|
|
|
|
|
|
void assertString(const char * s, ReadBuffer & buf);
|
|
void assertEOF(ReadBuffer & buf);
|
|
void assertChar(char symbol, ReadBuffer & buf);
|
|
|
|
inline void assertString(const String & s, ReadBuffer & buf)
|
|
{
|
|
assertString(s.c_str(), buf);
|
|
}
|
|
|
|
bool checkString(const char * s, ReadBuffer & buf);
|
|
inline bool checkString(const String & s, ReadBuffer & buf)
|
|
{
|
|
return checkString(s.c_str(), buf);
|
|
}
|
|
|
|
inline bool checkChar(char c, ReadBuffer & buf)
|
|
{
|
|
if (buf.eof() || *buf.position() != c)
|
|
return false;
|
|
++buf.position();
|
|
return true;
|
|
}
|
|
|
|
bool checkStringCaseInsensitive(const char * s, ReadBuffer & buf);
|
|
inline bool checkStringCaseInsensitive(const String & s, ReadBuffer & buf)
|
|
{
|
|
return checkStringCaseInsensitive(s.c_str(), buf);
|
|
}
|
|
|
|
void assertStringCaseInsensitive(const char * s, ReadBuffer & buf);
|
|
inline void assertStringCaseInsensitive(const String & s, ReadBuffer & buf)
|
|
{
|
|
return assertStringCaseInsensitive(s.c_str(), buf);
|
|
}
|
|
|
|
/** Check that next character in buf matches first character of s.
|
|
* If true, then check all characters in s and throw exception if it doesn't match.
|
|
* If false, then return false, and leave position in buffer unchanged.
|
|
*/
|
|
bool checkStringByFirstCharacterAndAssertTheRest(const char * s, ReadBuffer & buf);
|
|
bool checkStringByFirstCharacterAndAssertTheRestCaseInsensitive(const char * s, ReadBuffer & buf);
|
|
|
|
inline bool checkStringByFirstCharacterAndAssertTheRest(const String & s, ReadBuffer & buf)
|
|
{
|
|
return checkStringByFirstCharacterAndAssertTheRest(s.c_str(), buf);
|
|
}
|
|
|
|
inline bool checkStringByFirstCharacterAndAssertTheRestCaseInsensitive(const String & s, ReadBuffer & buf)
|
|
{
|
|
return checkStringByFirstCharacterAndAssertTheRestCaseInsensitive(s.c_str(), buf);
|
|
}
|
|
|
|
|
|
inline void readBoolText(bool & x, ReadBuffer & buf)
|
|
{
|
|
char tmp = '0';
|
|
readChar(tmp, buf);
|
|
x = tmp != '0';
|
|
}
|
|
|
|
inline void readBoolTextWord(bool & x, ReadBuffer & buf)
|
|
{
|
|
if (buf.eof())
|
|
throwReadAfterEOF();
|
|
|
|
if (*buf.position() == 't')
|
|
{
|
|
assertString("true", buf);
|
|
x = true;
|
|
}
|
|
else
|
|
{
|
|
assertString("false", buf);
|
|
x = false;
|
|
}
|
|
}
|
|
|
|
template <typename T, typename ReturnType = void>
|
|
ReturnType readIntTextImpl(T & x, ReadBuffer & buf)
|
|
{
|
|
static constexpr bool throw_exception = std::is_same_v<ReturnType, void>;
|
|
|
|
bool negative = false;
|
|
x = 0;
|
|
if (buf.eof())
|
|
{
|
|
if constexpr (throw_exception)
|
|
throwReadAfterEOF();
|
|
else
|
|
return ReturnType(false);
|
|
}
|
|
|
|
while (!buf.eof())
|
|
{
|
|
switch (*buf.position())
|
|
{
|
|
case '+':
|
|
break;
|
|
case '-':
|
|
if (std::is_signed_v<T>)
|
|
negative = true;
|
|
else
|
|
{
|
|
if constexpr (throw_exception)
|
|
throw Exception("Unsigned type must not contain '-' symbol", ErrorCodes::CANNOT_PARSE_NUMBER);
|
|
else
|
|
return ReturnType(false);
|
|
}
|
|
break;
|
|
case '0': [[fallthrough]];
|
|
case '1': [[fallthrough]];
|
|
case '2': [[fallthrough]];
|
|
case '3': [[fallthrough]];
|
|
case '4': [[fallthrough]];
|
|
case '5': [[fallthrough]];
|
|
case '6': [[fallthrough]];
|
|
case '7': [[fallthrough]];
|
|
case '8': [[fallthrough]];
|
|
case '9':
|
|
x *= 10;
|
|
x += *buf.position() - '0';
|
|
break;
|
|
default:
|
|
if (negative)
|
|
x = -x;
|
|
return ReturnType(true);
|
|
}
|
|
++buf.position();
|
|
}
|
|
|
|
/// NOTE Signed integer overflow is undefined behaviour. Consider we have '128' that is parsed as Int8 and overflowed.
|
|
/// We are happy if it is overflowed to -128 and then 'x = -x' does nothing. But UBSan will warn.
|
|
if (negative)
|
|
x = -x;
|
|
|
|
return ReturnType(true);
|
|
}
|
|
|
|
template <typename T>
|
|
void readIntText(T & x, ReadBuffer & buf)
|
|
{
|
|
readIntTextImpl<T, void>(x, buf);
|
|
}
|
|
|
|
template <typename T>
|
|
bool tryReadIntText(T & x, ReadBuffer & buf)
|
|
{
|
|
return readIntTextImpl<T, bool>(x, buf);
|
|
}
|
|
|
|
/** More efficient variant (about 1.5 times on real dataset).
|
|
* Differs in following:
|
|
* - for numbers starting with zero, parsed only zero;
|
|
* - symbol '+' before number is not supported;
|
|
* - symbols :;<=>? are parsed as some numbers.
|
|
*/
|
|
template <typename T, bool throw_on_error = true>
|
|
void readIntTextUnsafe(T & x, ReadBuffer & buf)
|
|
{
|
|
bool negative = false;
|
|
x = 0;
|
|
|
|
auto on_error = []
|
|
{
|
|
if (throw_on_error)
|
|
throwReadAfterEOF();
|
|
};
|
|
|
|
if (unlikely(buf.eof()))
|
|
return on_error();
|
|
|
|
if (std::is_signed_v<T> && *buf.position() == '-')
|
|
{
|
|
++buf.position();
|
|
negative = true;
|
|
if (unlikely(buf.eof()))
|
|
return on_error();
|
|
}
|
|
|
|
if (*buf.position() == '0') /// There are many zeros in real datasets.
|
|
{
|
|
++buf.position();
|
|
return;
|
|
}
|
|
|
|
while (!buf.eof())
|
|
{
|
|
/// This check is suddenly faster than
|
|
/// unsigned char c = *buf.position() - '0';
|
|
/// if (c < 10)
|
|
/// for unknown reason on Xeon E5645.
|
|
|
|
if ((*buf.position() & 0xF0) == 0x30) /// It makes sense to have this condition inside loop.
|
|
{
|
|
x *= 10;
|
|
x += *buf.position() & 0x0F;
|
|
++buf.position();
|
|
}
|
|
else
|
|
break;
|
|
}
|
|
|
|
/// See note about undefined behaviour above.
|
|
if (std::is_signed_v<T> && negative)
|
|
x = -x;
|
|
}
|
|
|
|
template <typename T>
|
|
void tryReadIntTextUnsafe(T & x, ReadBuffer & buf)
|
|
{
|
|
return readIntTextUnsafe<T, false>(x, buf);
|
|
}
|
|
|
|
|
|
/// Look at readFloatText.h
|
|
template <typename T> void readFloatText(T & x, ReadBuffer & in);
|
|
template <typename T> bool tryReadFloatText(T & x, ReadBuffer & in);
|
|
|
|
|
|
/// simple: all until '\n' or '\t'
|
|
void readString(String & s, ReadBuffer & buf);
|
|
|
|
void readEscapedString(String & s, ReadBuffer & buf);
|
|
|
|
void readQuotedString(String & s, ReadBuffer & buf);
|
|
void readQuotedStringWithSQLStyle(String & s, ReadBuffer & buf);
|
|
|
|
void readDoubleQuotedString(String & s, ReadBuffer & buf);
|
|
void readDoubleQuotedStringWithSQLStyle(String & s, ReadBuffer & buf);
|
|
|
|
void readJSONString(String & s, ReadBuffer & buf);
|
|
|
|
void readBackQuotedString(String & s, ReadBuffer & buf);
|
|
void readBackQuotedStringWithSQLStyle(String & s, ReadBuffer & buf);
|
|
|
|
void readStringUntilEOF(String & s, ReadBuffer & buf);
|
|
|
|
|
|
/** Read string in CSV format.
|
|
* Parsing rules:
|
|
* - string could be placed in quotes; quotes could be single: ' if FormatSettings::CSV::allow_single_quotes is true
|
|
* or double: " if FormatSettings::CSV::allow_double_quotes is true;
|
|
* - or string could be unquoted - this is determined by first character;
|
|
* - if string is unquoted, then it is read until next delimiter,
|
|
* either until end of line (CR or LF),
|
|
* or until end of stream;
|
|
* but spaces and tabs at begin and end of unquoted string are consumed but ignored (note that this behaviour differs from RFC).
|
|
* - if string is in quotes, then it will be read until closing quote,
|
|
* but sequences of two consecutive quotes are parsed as single quote inside string;
|
|
*/
|
|
void readCSVString(String & s, ReadBuffer & buf, const FormatSettings::CSV & settings);
|
|
|
|
|
|
/// Read and append result to array of characters.
|
|
template <typename Vector>
|
|
void readStringInto(Vector & s, ReadBuffer & buf);
|
|
|
|
template <typename Vector>
|
|
void readEscapedStringInto(Vector & s, ReadBuffer & buf);
|
|
|
|
template <bool enable_sql_style_quoting, typename Vector>
|
|
void readQuotedStringInto(Vector & s, ReadBuffer & buf);
|
|
|
|
template <bool enable_sql_style_quoting, typename Vector>
|
|
void readDoubleQuotedStringInto(Vector & s, ReadBuffer & buf);
|
|
|
|
template <bool enable_sql_style_quoting, typename Vector>
|
|
void readBackQuotedStringInto(Vector & s, ReadBuffer & buf);
|
|
|
|
template <typename Vector>
|
|
void readStringUntilEOFInto(Vector & s, ReadBuffer & buf);
|
|
|
|
template <typename Vector>
|
|
void readCSVStringInto(Vector & s, ReadBuffer & buf, const FormatSettings::CSV & settings);
|
|
|
|
/// ReturnType is either bool or void. If bool, the function will return false instead of throwing an exception.
|
|
template <typename Vector, typename ReturnType = void>
|
|
ReturnType readJSONStringInto(Vector & s, ReadBuffer & buf);
|
|
|
|
template <typename Vector>
|
|
bool tryReadJSONStringInto(Vector & s, ReadBuffer & buf)
|
|
{
|
|
return readJSONStringInto<Vector, bool>(s, buf);
|
|
}
|
|
|
|
/// This could be used as template parameter for functions above, if you want to just skip data.
|
|
struct NullSink
|
|
{
|
|
void append(const char *, size_t) {}
|
|
void push_back(char) {}
|
|
};
|
|
|
|
void parseUUID(const UInt8 * src36, UInt8 * dst16);
|
|
void parseUUID(const UInt8 * src36, std::reverse_iterator<UInt8 *> dst16);
|
|
|
|
template <typename IteratorSrc, typename IteratorDst>
|
|
void formatHex(IteratorSrc src, IteratorDst dst, const size_t num_bytes);
|
|
|
|
|
|
template <typename ReturnType>
|
|
ReturnType readDateTextFallback(LocalDate & date, ReadBuffer & buf);
|
|
|
|
/// In YYYY-MM-DD format.
|
|
/// For convenience, Month and Day parts can have single digit instead of two digits.
|
|
/// Any separators other than '-' are supported.
|
|
template <typename ReturnType = void>
|
|
inline ReturnType readDateTextImpl(LocalDate & date, ReadBuffer & buf)
|
|
{
|
|
/// Optimistic path, when whole value is in buffer.
|
|
if (buf.position() + 10 <= buf.buffer().end())
|
|
{
|
|
UInt16 year = (buf.position()[0] - '0') * 1000 + (buf.position()[1] - '0') * 100 + (buf.position()[2] - '0') * 10 + (buf.position()[3] - '0');
|
|
buf.position() += 5;
|
|
|
|
UInt8 month = buf.position()[0] - '0';
|
|
if (isNumericASCII(buf.position()[1]))
|
|
{
|
|
month = month * 10 + buf.position()[1] - '0';
|
|
buf.position() += 3;
|
|
}
|
|
else
|
|
buf.position() += 2;
|
|
|
|
UInt8 day = buf.position()[0] - '0';
|
|
if (isNumericASCII(buf.position()[1]))
|
|
{
|
|
day = day * 10 + buf.position()[1] - '0';
|
|
buf.position() += 2;
|
|
}
|
|
else
|
|
buf.position() += 1;
|
|
|
|
date = LocalDate(year, month, day);
|
|
return ReturnType(true);
|
|
}
|
|
else
|
|
return readDateTextFallback<ReturnType>(date, buf);
|
|
}
|
|
|
|
template <typename ReturnType = void>
|
|
inline ReturnType readDateTextImpl(DayNum & date, ReadBuffer & buf)
|
|
{
|
|
static constexpr bool throw_exception = std::is_same_v<ReturnType, void>;
|
|
|
|
LocalDate local_date;
|
|
|
|
if constexpr (throw_exception)
|
|
readDateTextImpl<ReturnType>(local_date, buf);
|
|
else if (!readDateTextImpl<ReturnType>(local_date, buf))
|
|
return false;
|
|
|
|
date = DateLUT::instance().makeDayNum(local_date.year(), local_date.month(), local_date.day());
|
|
return ReturnType(true);
|
|
}
|
|
|
|
|
|
inline void readDateText(LocalDate & date, ReadBuffer & buf)
|
|
{
|
|
readDateTextImpl<void>(date, buf);
|
|
}
|
|
|
|
inline void readDateText(DayNum & date, ReadBuffer & buf)
|
|
{
|
|
readDateTextImpl<void>(date, buf);
|
|
}
|
|
|
|
inline bool tryReadDateText(LocalDate & date, ReadBuffer & buf)
|
|
{
|
|
return readDateTextImpl<bool>(date, buf);
|
|
}
|
|
|
|
inline bool tryReadDateText(DayNum & date, ReadBuffer & buf)
|
|
{
|
|
return readDateTextImpl<bool>(date, buf);
|
|
}
|
|
|
|
|
|
inline void readUUIDText(UUID & uuid, ReadBuffer & buf)
|
|
{
|
|
char s[36];
|
|
size_t size = buf.read(s, 36);
|
|
|
|
if (size != 36)
|
|
{
|
|
s[size] = 0;
|
|
throw Exception(std::string("Cannot parse uuid ") + s, ErrorCodes::CANNOT_PARSE_UUID);
|
|
}
|
|
|
|
parseUUID(reinterpret_cast<const UInt8 *>(s), std::reverse_iterator<UInt8 *>(reinterpret_cast<UInt8 *>(&uuid) + 16));
|
|
}
|
|
|
|
|
|
template <typename T>
|
|
inline T parse(const char * data, size_t size);
|
|
|
|
template <typename T>
|
|
inline T parseFromString(const String & str)
|
|
{
|
|
return parse<T>(str.data(), str.size());
|
|
}
|
|
|
|
UInt128 stringToUUID(const String & str);
|
|
|
|
|
|
template <typename ReturnType = void>
|
|
ReturnType readDateTimeTextFallback(time_t & datetime, ReadBuffer & buf, const DateLUTImpl & date_lut);
|
|
|
|
/** In YYYY-MM-DD hh:mm:ss format, according to specified time zone.
|
|
* As an exception, also supported parsing of unix timestamp in form of decimal number.
|
|
*/
|
|
template <typename ReturnType = void>
|
|
inline ReturnType readDateTimeTextImpl(time_t & datetime, ReadBuffer & buf, const DateLUTImpl & date_lut)
|
|
{
|
|
/** Read 10 characters, that could represent unix timestamp.
|
|
* Only unix timestamp of 5-10 characters is supported.
|
|
* Then look at 5th charater. If it is a number - treat whole as unix timestamp.
|
|
* If it is not a number - then parse datetime in YYYY-MM-DD hh:mm:ss format.
|
|
*/
|
|
|
|
/// Optimistic path, when whole value is in buffer.
|
|
const char * s = buf.position();
|
|
if (s + 19 <= buf.buffer().end())
|
|
{
|
|
if (s[4] < '0' || s[4] > '9')
|
|
{
|
|
UInt16 year = (s[0] - '0') * 1000 + (s[1] - '0') * 100 + (s[2] - '0') * 10 + (s[3] - '0');
|
|
UInt8 month = (s[5] - '0') * 10 + (s[6] - '0');
|
|
UInt8 day = (s[8] - '0') * 10 + (s[9] - '0');
|
|
|
|
UInt8 hour = (s[11] - '0') * 10 + (s[12] - '0');
|
|
UInt8 minute = (s[14] - '0') * 10 + (s[15] - '0');
|
|
UInt8 second = (s[17] - '0') * 10 + (s[18] - '0');
|
|
|
|
if (unlikely(year == 0))
|
|
datetime = 0;
|
|
else
|
|
datetime = date_lut.makeDateTime(year, month, day, hour, minute, second);
|
|
|
|
buf.position() += 19;
|
|
return ReturnType(true);
|
|
}
|
|
else
|
|
/// Why not readIntTextUnsafe? Because for needs of AdFox, parsing of unix timestamp with leading zeros is supported: 000...NNNN.
|
|
return readIntTextImpl<time_t, ReturnType>(datetime, buf);
|
|
}
|
|
else
|
|
return readDateTimeTextFallback<ReturnType>(datetime, buf, date_lut);
|
|
}
|
|
|
|
inline void readDateTimeText(time_t & datetime, ReadBuffer & buf, const DateLUTImpl & date_lut = DateLUT::instance())
|
|
{
|
|
readDateTimeTextImpl<void>(datetime, buf, date_lut);
|
|
}
|
|
|
|
inline bool tryReadDateTimeText(time_t & datetime, ReadBuffer & buf, const DateLUTImpl & date_lut = DateLUT::instance())
|
|
{
|
|
return readDateTimeTextImpl<bool>(datetime, buf, date_lut);
|
|
}
|
|
|
|
inline void readDateTimeText(LocalDateTime & datetime, ReadBuffer & buf)
|
|
{
|
|
char s[19];
|
|
size_t size = buf.read(s, 19);
|
|
if (19 != size)
|
|
{
|
|
s[size] = 0;
|
|
throw Exception(std::string("Cannot parse datetime ") + s, ErrorCodes::CANNOT_PARSE_DATETIME);
|
|
}
|
|
|
|
datetime.year((s[0] - '0') * 1000 + (s[1] - '0') * 100 + (s[2] - '0') * 10 + (s[3] - '0'));
|
|
datetime.month((s[5] - '0') * 10 + (s[6] - '0'));
|
|
datetime.day((s[8] - '0') * 10 + (s[9] - '0'));
|
|
|
|
datetime.hour((s[11] - '0') * 10 + (s[12] - '0'));
|
|
datetime.minute((s[14] - '0') * 10 + (s[15] - '0'));
|
|
datetime.second((s[17] - '0') * 10 + (s[18] - '0'));
|
|
}
|
|
|
|
|
|
/// Generic methods to read value in native binary format.
|
|
template <typename T>
|
|
inline std::enable_if_t<std::is_arithmetic_v<T>, void>
|
|
readBinary(T & x, ReadBuffer & buf) { readPODBinary(x, buf); }
|
|
|
|
inline void readBinary(String & x, ReadBuffer & buf) { readStringBinary(x, buf); }
|
|
inline void readBinary(Int128 & x, ReadBuffer & buf) { readPODBinary(x, buf); }
|
|
inline void readBinary(UInt128 & x, ReadBuffer & buf) { readPODBinary(x, buf); }
|
|
inline void readBinary(UInt256 & x, ReadBuffer & buf) { readPODBinary(x, buf); }
|
|
inline void readBinary(LocalDate & x, ReadBuffer & buf) { readPODBinary(x, buf); }
|
|
|
|
|
|
/// Generic methods to read value in text tab-separated format.
|
|
template <typename T>
|
|
inline std::enable_if_t<std::is_integral_v<T>, void>
|
|
readText(T & x, ReadBuffer & buf) { readIntText(x, buf); }
|
|
|
|
template <typename T>
|
|
inline std::enable_if_t<std::is_floating_point_v<T>, void>
|
|
readText(T & x, ReadBuffer & buf) { readFloatText(x, buf); }
|
|
|
|
inline void readText(bool & x, ReadBuffer & buf) { readBoolText(x, buf); }
|
|
inline void readText(String & x, ReadBuffer & buf) { readEscapedString(x, buf); }
|
|
inline void readText(LocalDate & x, ReadBuffer & buf) { readDateText(x, buf); }
|
|
inline void readText(LocalDateTime & x, ReadBuffer & buf) { readDateTimeText(x, buf); }
|
|
inline void readText(UUID & x, ReadBuffer & buf) { readUUIDText(x, buf); }
|
|
inline void readText(UInt128 &, ReadBuffer &)
|
|
{
|
|
/** Because UInt128 isn't a natural type, without arithmetic operator and only use as an intermediary type -for UUID-
|
|
* it should never arrive here. But because we used the DataTypeNumber class we should have at least a definition of it.
|
|
*/
|
|
throw Exception("UInt128 cannot be read as a text", ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT);
|
|
}
|
|
|
|
/// Generic methods to read value in text format,
|
|
/// possibly in single quotes (only for data types that use quotes in VALUES format of INSERT statement in SQL).
|
|
template <typename T>
|
|
inline std::enable_if_t<std::is_arithmetic_v<T>, void>
|
|
readQuoted(T & x, ReadBuffer & buf) { readText(x, buf); }
|
|
|
|
inline void readQuoted(String & x, ReadBuffer & buf) { readQuotedString(x, buf); }
|
|
|
|
inline void readQuoted(LocalDate & x, ReadBuffer & buf)
|
|
{
|
|
assertChar('\'', buf);
|
|
readDateText(x, buf);
|
|
assertChar('\'', buf);
|
|
}
|
|
|
|
inline void readQuoted(LocalDateTime & x, ReadBuffer & buf)
|
|
{
|
|
assertChar('\'', buf);
|
|
readDateTimeText(x, buf);
|
|
assertChar('\'', buf);
|
|
}
|
|
|
|
|
|
/// Same as above, but in double quotes.
|
|
template <typename T>
|
|
inline std::enable_if_t<std::is_arithmetic_v<T>, void>
|
|
readDoubleQuoted(T & x, ReadBuffer & buf) { readText(x, buf); }
|
|
|
|
inline void readDoubleQuoted(String & x, ReadBuffer & buf) { readDoubleQuotedString(x, buf); }
|
|
|
|
inline void readDoubleQuoted(LocalDate & x, ReadBuffer & buf)
|
|
{
|
|
assertChar('"', buf);
|
|
readDateText(x, buf);
|
|
assertChar('"', buf);
|
|
}
|
|
|
|
inline void readDoubleQuoted(LocalDateTime & x, ReadBuffer & buf)
|
|
{
|
|
assertChar('"', buf);
|
|
readDateTimeText(x, buf);
|
|
assertChar('"', buf);
|
|
}
|
|
|
|
|
|
/// CSV, for numbers, dates: quotes are optional, no special escaping rules.
|
|
template <typename T>
|
|
inline void readCSVSimple(T & x, ReadBuffer & buf)
|
|
{
|
|
if (buf.eof())
|
|
throwReadAfterEOF();
|
|
|
|
char maybe_quote = *buf.position();
|
|
|
|
if (maybe_quote == '\'' || maybe_quote == '\"')
|
|
++buf.position();
|
|
|
|
readText(x, buf);
|
|
|
|
if (maybe_quote == '\'' || maybe_quote == '\"')
|
|
assertChar(maybe_quote, buf);
|
|
}
|
|
|
|
template <typename T>
|
|
inline std::enable_if_t<std::is_arithmetic_v<T>, void>
|
|
readCSV(T & x, ReadBuffer & buf) { readCSVSimple(x, buf); }
|
|
|
|
inline void readCSV(String & x, ReadBuffer & buf, const FormatSettings::CSV & settings) { readCSVString(x, buf, settings); }
|
|
inline void readCSV(LocalDate & x, ReadBuffer & buf) { readCSVSimple(x, buf); }
|
|
inline void readCSV(LocalDateTime & x, ReadBuffer & buf) { readCSVSimple(x, buf); }
|
|
inline void readCSV(UUID & x, ReadBuffer & buf) { readCSVSimple(x, buf); }
|
|
inline void readCSV(UInt128 &, ReadBuffer &)
|
|
{
|
|
/** Because UInt128 isn't a natural type, without arithmetic operator and only use as an intermediary type -for UUID-
|
|
* it should never arrive here. But because we used the DataTypeNumber class we should have at least a definition of it.
|
|
*/
|
|
throw Exception("UInt128 cannot be read as a text", ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT);
|
|
}
|
|
|
|
template <typename T>
|
|
void readBinary(std::vector<T> & x, ReadBuffer & buf)
|
|
{
|
|
size_t size = 0;
|
|
readVarUInt(size, buf);
|
|
|
|
if (size > DEFAULT_MAX_STRING_SIZE)
|
|
throw Poco::Exception("Too large vector size.");
|
|
|
|
x.resize(size);
|
|
for (size_t i = 0; i < size; ++i)
|
|
readBinary(x[i], buf);
|
|
}
|
|
|
|
template <typename T>
|
|
void readQuoted(std::vector<T> & x, ReadBuffer & buf)
|
|
{
|
|
bool first = true;
|
|
assertChar('[', buf);
|
|
while (!buf.eof() && *buf.position() != ']')
|
|
{
|
|
if (!first)
|
|
{
|
|
if (*buf.position() == ',')
|
|
++buf.position();
|
|
else
|
|
throw Exception("Cannot read array from text", ErrorCodes::CANNOT_READ_ARRAY_FROM_TEXT);
|
|
}
|
|
|
|
first = false;
|
|
|
|
x.push_back(T());
|
|
readQuoted(x.back(), buf);
|
|
}
|
|
assertChar(']', buf);
|
|
}
|
|
|
|
template <typename T>
|
|
void readDoubleQuoted(std::vector<T> & x, ReadBuffer & buf)
|
|
{
|
|
bool first = true;
|
|
assertChar('[', buf);
|
|
while (!buf.eof() && *buf.position() != ']')
|
|
{
|
|
if (!first)
|
|
{
|
|
if (*buf.position() == ',')
|
|
++buf.position();
|
|
else
|
|
throw Exception("Cannot read array from text", ErrorCodes::CANNOT_READ_ARRAY_FROM_TEXT);
|
|
}
|
|
|
|
first = false;
|
|
|
|
x.push_back(T());
|
|
readDoubleQuoted(x.back(), buf);
|
|
}
|
|
assertChar(']', buf);
|
|
}
|
|
|
|
template <typename T>
|
|
void readText(std::vector<T> & x, ReadBuffer & buf)
|
|
{
|
|
readQuoted(x, buf);
|
|
}
|
|
|
|
|
|
/// Skip whitespace characters.
|
|
inline void skipWhitespaceIfAny(ReadBuffer & buf)
|
|
{
|
|
while (!buf.eof() && isWhitespaceASCII(*buf.position()))
|
|
++buf.position();
|
|
}
|
|
|
|
/// Skips json value.
|
|
void skipJSONField(ReadBuffer & buf, const StringRef & name_of_field);
|
|
|
|
|
|
/** Read serialized exception.
|
|
* During serialization/deserialization some information is lost
|
|
* (type is cut to base class, 'message' replaced by 'displayText', and stack trace is appended to 'message')
|
|
* Some additional message could be appended to exception (example: you could add information about from where it was received).
|
|
*/
|
|
void readException(Exception & e, ReadBuffer & buf, const String & additional_message = "");
|
|
void readAndThrowException(ReadBuffer & buf, const String & additional_message = "");
|
|
|
|
|
|
/** Helper function for implementation.
|
|
*/
|
|
template <typename T>
|
|
static inline const char * tryReadIntText(T & x, const char * pos, const char * end)
|
|
{
|
|
ReadBufferFromMemory in(pos, end - pos);
|
|
tryReadIntText(x, in);
|
|
return pos + in.count();
|
|
}
|
|
|
|
|
|
/// Convenient methods for reading something from string in text format.
|
|
template <typename T>
|
|
inline T parse(const char * data, size_t size)
|
|
{
|
|
T res;
|
|
ReadBufferFromMemory buf(data, size);
|
|
readText(res, buf);
|
|
return res;
|
|
}
|
|
|
|
template <typename T>
|
|
inline T parse(const char * data)
|
|
{
|
|
return parse<T>(data, strlen(data));
|
|
}
|
|
|
|
template <typename T>
|
|
inline T parse(const String & s)
|
|
{
|
|
return parse<T>(s.data(), s.size());
|
|
}
|
|
|
|
|
|
/** Skip UTF-8 BOM if it is under cursor.
|
|
* As BOM is usually located at start of stream, and buffer size is usually larger than three bytes,
|
|
* the function expects, that all three bytes of BOM is fully in buffer (otherwise it don't skip anything).
|
|
*/
|
|
inline void skipBOMIfExists(ReadBuffer & buf)
|
|
{
|
|
if (!buf.eof()
|
|
&& buf.position() + 3 < buf.buffer().end()
|
|
&& buf.position()[0] == '\xEF'
|
|
&& buf.position()[1] == '\xBB'
|
|
&& buf.position()[2] == '\xBF')
|
|
{
|
|
buf.position() += 3;
|
|
}
|
|
}
|
|
|
|
|
|
/// Skip to next character after next \n. If no \n in stream, skip to end.
|
|
void skipToNextLineOrEOF(ReadBuffer & buf);
|
|
|
|
/// Skip to next character after next unescaped \n. If no \n in stream, skip to end. Does not throw on invalid escape sequences.
|
|
void skipToUnescapedNextLineOrEOF(ReadBuffer & buf);
|
|
|
|
}
|