tomasz.sowa/winix
1
0
Fork 0
Code Issues 5 Pull Requests Releases Activity

improve AcceptBaseParser parsing algorithm

Browse Source 
remove AcceptParser - not needed now, AcceptBaseParser can prepare a table now
This commit is contained in:
Tomasz Sowa
2022-02-02 17:58:27 +01:00
parent d0d2cfb22c
commit 75daf37bbd
10 changed files with 230 additions and 207 deletions
Download Patch File Download Diff File Expand all files Collapse all files

52
winixd/utils/Makefile.dep
View File

@@ -1,3 +1,53 @@
# DO NOT DELETE
acceptbaseparser.o: acceptbaseparser.h ../../../tito/src/misc.h
acceptbaseparser.o: acceptbaseparser.h ../../../winix/winixd/core/header.h
acceptbaseparser.o: ../../../winix/winixd/core/log.h
acceptbaseparser.o: ../../../winix/winixd/core/logmanipulators.h
acceptbaseparser.o: ../../../pikotools/src/log/log.h
acceptbaseparser.o: ../../../pikotools/src/textstream/textstream.h
acceptbaseparser.o: ../../../pikotools/src/textstream/stream.h
acceptbaseparser.o: ../../../pikotools/src/space/space.h
acceptbaseparser.o: ../../../pikotools/src/textstream/types.h
acceptbaseparser.o: ../../../pikotools/src/convert/inttostr.h
acceptbaseparser.o: ../../../pikotools/src/utf8/utf8.h
acceptbaseparser.o: ../../../pikotools/src/textstream/stream.h
acceptbaseparser.o: ../../../pikotools/src/utf8/utf8_templates.h
acceptbaseparser.o: ../../../pikotools/src/utf8/utf8_private.h
acceptbaseparser.o: ../../../pikotools/src/date/date.h
acceptbaseparser.o: ../../../pikotools/src/membuffer/membuffer.h
acceptbaseparser.o: ../../../pikotools/src/textstream/types.h
acceptbaseparser.o: ../../../pikotools/src/log/filelog.h
acceptbaseparser.o: ../../../morm/src/morm.h ../../../morm/src/morm_types.h
acceptbaseparser.o: ../../../morm/src/model.h
acceptbaseparser.o: ../../../morm/src/modelconnector.h
acceptbaseparser.o: ../../../morm/src/clearer.h
acceptbaseparser.o: ../../../morm/src/dbconnector.h
acceptbaseparser.o: ../../../morm/src/queryresult.h ../../../morm/src/ft.h
acceptbaseparser.o: ../../../morm/src/flatconnector.h
acceptbaseparser.o: ../../../morm/src/dbexpression.h
acceptbaseparser.o: ../../../morm/src/baseexpression.h
acceptbaseparser.o: ../../../morm/src/modelenv.h
acceptbaseparser.o: ../../../morm/src/modeldata.h
acceptbaseparser.o: ../../../morm/src/cursorhelper.h
acceptbaseparser.o: ../../../morm/src/finderhelper.h
acceptbaseparser.o: ../../../morm/src/fieldvaluehelper.h
acceptbaseparser.o: ../../../morm/src/wrapper.h
acceptbaseparser.o: ../../../morm/src/spacewrapper.h
acceptbaseparser.o: ../../../morm/src/baseobjectwrapper.h
acceptbaseparser.o: ../../../morm/src/modelcontainerwrapper.h
acceptbaseparser.o: ../../../pikotools/src/convert/text.h
acceptbaseparser.o: ../../../morm/src/flatexpression.h
acceptbaseparser.o: ../../../morm/src/finder.h ../../../morm/src/cursor.h
acceptbaseparser.o: ../../../morm/src/jsonexpression.h
acceptbaseparser.o: ../../../morm/src/postgresqlexpression.h
acceptbaseparser.o: ../../../morm/src/jsonconnector.h
acceptbaseparser.o: ../../../morm/src/postgresqlconnector.h
acceptbaseparser.o: ../../../morm/src/postgresqlqueryresult.h
acceptbaseparser.o: ../../../pikotools/src/convert/convert.h
acceptbaseparser.o: ../../../pikotools/src/convert/inttostr.h
acceptbaseparser.o: ../../../pikotools/src/convert/patternreplacer.h
acceptbaseparser.o: ../../../pikotools/src/convert/strtoint.h
acceptbaseparser.o: ../../../pikotools/src/convert/text.h
acceptbaseparser.o: ../../../pikotools/src/convert/misc.h
acceptbaseparser.o: ../../../pikotools/src/utf8/utf8_stream.h
acceptbaseparser.o: ../../../pikotools/src/convert/double.h

156
winixd/utils/acceptbaseparser.cpp
View File

@@ -5,7 +5,7 @@
*/
/*
* Copyright (c) 2008-2014, Tomasz Sowa
* Copyright (c) 2008-2022, Tomasz Sowa
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
@@ -32,107 +32,167 @@
*
*/
#include <wchar.h>
#include "acceptbaseparser.h"
#include "core/misc.h"
#include "convert/convert.h"
namespace Winix
{
bool AcceptBaseParser::IsWhite(int c)
AcceptBaseParser::AcceptBaseParser()
{
if( c==' ' || c=='\t' )
return true;
}
return false;
AcceptBaseParser::~AcceptBaseParser()
{
}
void AcceptBaseParser::skip_white()
bool AcceptBaseParser::is_delimiter(wchar_t c, wchar_t delimiter)
{
while( IsWhite(*text) )
return delimiter != 0 && delimiter == c;
}
bool AcceptBaseParser::is_delimiter(wchar_t c, wchar_t delimiter1, wchar_t delimiter2, wchar_t delimiter3)
{
return is_delimiter(c, delimiter1) || is_delimiter(c, delimiter2) || is_delimiter(c, delimiter3);
}
void AcceptBaseParser::read_token(std::wstring & token, size_t max_len, wchar_t delimiter1, wchar_t delimiter2, wchar_t delimiter3)
{
token.clear();
text = pt::skip_white(text, false, false);
while( *text!=0 && !is_delimiter(*text, delimiter1, delimiter2, delimiter3) )
{
if( token.size() < max_len )
{
token += *text;
}
++text;
}
pt::trim_white(token);
}
void AcceptBaseParser::read_name()
{
read_token(name, MAX_NAME_LENGTH, ',', ';');
if( !name.empty() )
{
parsed_name(name);
}
}
void AcceptBaseParser::read_parameter()
{
param.clear();
skip_white();
param_value.clear();
read_token(param, MAX_PARAM_LENGTH, '=', ',', ';');
while( *text!=0 && *text!=',' && *text!=';' )
if( *text == '=' )
{
param += *text;
++text;
read_token(param_value, MAX_PARAM_VALUE_LENGTH, ';', ',');
}
TrimWhite(param);
if( !name.empty() && !param.empty() )
{
parsed_param(param, param_value);
}
if( param == L"q" && !param_value.empty() )
{
q = pt::to_double(param_value);
}
}
void AcceptBaseParser::ReadQ()
void AcceptBaseParser::read_loop(std::vector<HeaderValue> * header_values, size_t max_len)
{
q = 1.0;
skip_white();
while( *text != 0 )
{
q = 1.0;
read_name();
if( *text != ';' )
return;
while( *text == ';' )
{
++text;
read_parameter();
}
++text; // skipping a semicolon
if( !name.empty() && q > 0.0 )
{
if( q > 1.0 )
q = 1.0;
while( *text!=0 && *text!=',' && *text!='=' )
// skipping until ',' or '='
++text;
parsed_name_q(name, q);
if( *text==0 || *text==',' )
return;
if( header_values && (max_len == 0 || header_values->size() < max_len) )
{
header_values->resize(header_values->size() + 1);
header_values->back().value = name;
header_values->back().weight = q;
}
}
++text; // skipping '='
skip_white();
q = wcstod(text, (wchar_t**)&text);
if( *text != 0 )
++text;
}
}
void AcceptBaseParser::SkipParam()
void AcceptBaseParser::read(const wchar_t * str, std::vector<HeaderValue> * header_values, size_t max_len)
{
skip_white();
text = str;
init();
if( *text == ',' )
++text;
read_loop(header_values, max_len);
name.clear();
param.clear();
param_value.clear();
}
void AcceptBaseParser::parse(const wchar_t * str)
{
text = str;
init();
while( *text != 0 )
{
read_parameter();
ReadQ();
SkipParam();
Param(param, q);
}
read(str, nullptr, 0);
}
void AcceptBaseParser::parse(const std::wstring & str)
{
parse(str.c_str());
}
void AcceptBaseParser::parse(const wchar_t * str, std::vector<HeaderValue> & header_values, size_t max_len, bool clear_header_values)
{
if( clear_header_values )
header_values.clear();
AcceptBaseParser::read(str, &header_values, max_len);
std::sort(header_values.begin(), header_values.end(), [](HeaderValue & h1, HeaderValue & h2) -> bool {
return h1.weight > h2.weight;
});
}
void AcceptBaseParser::parse(const std::wstring & str, std::vector<HeaderValue> & header_values, size_t max_len, bool clear_header_values)
{
return parse(str.c_str(), header_values, max_len, clear_header_values);
}
} // namespace Winix

56
winixd/utils/acceptbaseparser.h
View File

@@ -5,7 +5,7 @@
*/
/*
* Copyright (c) 2008-2018, Tomasz Sowa
* Copyright (c) 2008-2022, Tomasz Sowa
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
@@ -32,44 +32,64 @@
*
*/
#ifndef headerfile_winix_core_acceptbaseparser
#define headerfile_winix_core_acceptbaseparser
#ifndef headerfile_winix_utils_acceptbaseparser
#define headerfile_winix_utils_acceptbaseparser
#include <string>
#include "core/winixbase.h"
#include "core/header.h"
namespace Winix
{
// sample (you must create your own class derived from this one):
// object.parse(L" text/html ; , ; q = 45, application / xhtml+xml ; q = 0.4 , application/xml ; q = 0.9 , */* ; q = 0.8 ");
class AcceptBaseParser : public WinixBase
class AcceptBaseParser
{
public:
static size_t constexpr MAX_NAME_LENGTH = 64;
static size_t constexpr MAX_PARAM_LENGTH = 32;
static size_t constexpr MAX_PARAM_VALUE_LENGTH = 32;
AcceptBaseParser();
virtual ~AcceptBaseParser();
/*
*
*
*
*
*/
// sample:
// object.parse(L" text/html , text/* ; q = 45, application / xhtml+xml ; q = 0.4 ; limit = 1 , application/xml ; charset = UTF-8 ; q = 0.9 , */* ; q = 0.8 ");
void parse(const wchar_t * str);
void parse(const std::wstring & str);
void parse(const wchar_t * str, std::vector<HeaderValue> & header_values, size_t max_len, bool clear_header_values = true);
void parse(const std::wstring & str, std::vector<HeaderValue> & header_values, size_t max_len, bool clear_header_values = true);
private:
virtual void init() {} ;
virtual void Param(const std::wstring & param, double q) = 0;
bool IsWhite(int c);
void skip_white();
bool is_delimiter(wchar_t c, wchar_t delimiter);
bool is_delimiter(wchar_t c, wchar_t delimiter1, wchar_t delimiter2, wchar_t delimiter3);
void read_token(std::wstring & token, size_t max_len, wchar_t delimiter1, wchar_t delimiter2 = 0, wchar_t delimiter3 = 0);
void read_name();
void read_parameter();
void ReadQ();
void SkipParam();
void read_loop(std::vector<HeaderValue> * header_values, size_t max_len);
void read(const wchar_t * str, std::vector<HeaderValue> * header_values, size_t max_len);
virtual void init() {} ;
virtual void parsed_name(const std::wstring & name) {};
virtual void parsed_param(const std::wstring & param, const std::wstring & param_value) {};
virtual void parsed_name_q(const std::wstring & name, double q) {};
const wchar_t * text;
std::wstring name;
std::wstring param;
std::wstring param_value;
double q;
};