/* * This file is a part of Winix * and is not publicly distributed * * Copyright (c) 2008-2011, Tomasz Sowa * All rights reserved. * */ #include #include #include "confparser.h" #include "utf8.h" /* * * * Space * * */ Space::Space() { parent = 0; } Space::~Space() { Clear(); } Space::Space(const Space & s) { operator=(s); } Space & Space::operator=(const Space & s) { Clear(); name = s.name; table_single = s.table_single; table = s.table; parent = s.parent; for(size_t i=0 ; iparent = this; spaces.push_back(pspace); } return *this; } void Space::Clear() { name.clear(); table_single.clear(); table.clear(); for(size_t i=0 ; isecond; } else { Table::iterator t = table.find(name); if( t == table.end() || t->second.empty() ) { return 0; } else { return &t->second[0]; } } } std::wstring & Space::Text(const wchar_t * name) { tmp_name = name; return Text(tmp_name, L""); } std::wstring & Space::Text(const wchar_t * name, const wchar_t * def) { tmp_name = name; return Text(tmp_name, def); } std::wstring & Space::Text(const std::wstring & name, const wchar_t * def) { std::wstring * value = GetValue(name); if( value ) { return *value; } else { tmp_value_text = def; return tmp_value_text; } } std::string & Space::AText(const wchar_t * name) { tmp_name = name; return AText(tmp_name, ""); } std::string & Space::AText(const wchar_t * name, const char * def) { tmp_name = name; return AText(tmp_name, def); } std::string & Space::AText(const std::wstring & name, const char * def) { std::wstring * value = GetValue(name); if( value ) { Ezc::WideToUTF8(*value, tmp_value_text_ascii); return tmp_value_text_ascii; } else { tmp_value_text_ascii = def; return tmp_value_text_ascii; } } int Space::Int(const wchar_t * name) { tmp_name = name; return Int(tmp_name, 0); } int Space::Int(const wchar_t * name, int def) { tmp_name = name; return Int(tmp_name, def); } int Space::ToInt(const std::wstring & value) { long res = (value[0] == '0')? wcstol(value.c_str() + 1, 0, 8) : wcstol(value.c_str(), 0, 10); return static_cast(res); } int Space::Int(const std::wstring & name, int def) { std::wstring * value = GetValue(name); if( value ) return ToInt(*value); return def; } size_t Space::Size(const wchar_t * name) { tmp_name = name; return Size(tmp_name, 0); } size_t Space::Size(const wchar_t * name, size_t def) { tmp_name = name; return Size(tmp_name, def); } size_t Space::ToSize(const std::wstring & value) { unsigned long res = (value[0] == '0')? wcstoul(value.c_str() + 1, 0, 8) : wcstoul(value.c_str(), 0, 10); return static_cast(res); } size_t Space::Size(const std::wstring & name, size_t def) { std::wstring * value = GetValue(name); if( value ) return ToSize(*value); return def; } bool Space::Bool(const wchar_t * name) { tmp_name = name; return Bool(tmp_name, false); } bool Space::Bool(const wchar_t * name, bool def) { tmp_name = name; return Bool(tmp_name, def); } bool Space::ToBool(const std::wstring & value) { return ( EqualNoCase(value.c_str(), L"true") || EqualNoCase(value.c_str(), L"yes") || EqualNoCase(value.c_str(), L"1") ); } bool Space::Bool(const std::wstring & name, bool def) { std::wstring * value = GetValue(name); if( value ) return ToBool(*value); return def; } // in lists we don't use default values bool Space::ListText(const wchar_t * name, std::vector & list) { tmp_name = name; return ListText(tmp_name, list); } bool Space::ListText(const std::wstring & name, std::vector & list) { list.clear(); TableSingle::iterator i = table_single.find(name); if( i != table_single.end() ) { list.push_back(i->second); return true; } else { Table::iterator t = table.find(name); if( t != table.end() ) { list = t->second; return true; } } return false; } wchar_t Space::ToSmall(wchar_t c) { if( c>='A' && c<='Z' ) c = c - 'A' + 'a'; return c; } bool Space::EqualNoCase(const wchar_t * str1, const wchar_t * str2) { while( *str1 && *str2 && ToSmall(*str1) == ToSmall(*str2) ) { ++str1; ++str2; } if( *str1 == 0 && *str2 == 0 ) return true; return false; } /* * * * ConfParser * * */ ConfParser::ConfParser() { root_space = 0; SetDefault(); } void ConfParser::SetSpace(Space * pspace) { root_space = pspace; } void ConfParser::SetSpace(Space & pspace) { root_space = &pspace; } void ConfParser::SetDefault() { // you can change this separators to what you want // you shoud not use only white characters here (as expected by IsWhite() method) // and new line characters ('\n') separator = '='; commentary = '#'; list_start = '('; list_end = ')'; list_delimiter = ','; split_single = false; skip_empty = false; use_escape_char = true; input_as_utf8 = false; } void ConfParser::SplitSingle(bool split) { split_single = split; } void ConfParser::SkipEmpty(bool skip) { skip_empty = skip; } void ConfParser::UseEscapeChar(bool escape) { use_escape_char = escape; } void ConfParser::UTF8(bool utf) { input_as_utf8 = utf; } ConfParser::Status ConfParser::Parse(const char * file_name) { reading_from_file = true; file.clear(); file.open( file_name ); if( file ) { Parse(); file.close(); } else { status = cant_open_file; } return status; } ConfParser::Status ConfParser::Parse(const std::string & file_name) { return Parse(file_name.c_str()); } ConfParser::Status ConfParser::Parse(const wchar_t * file_name) { Ezc::WideToUTF8(file_name, afile_name); return Parse(afile_name.c_str()); } ConfParser::Status ConfParser::Parse(const std::wstring & file_name) { return Parse(file_name.c_str()); } ConfParser::Status ConfParser::ParseString(const char * str) { reading_from_file = false; reading_from_wchar_string = false; pchar_ascii = str; pchar_unicode = 0; Parse(); return status; } ConfParser::Status ConfParser::ParseString(const std::string & str) { return ParseString(str.c_str()); } ConfParser::Status ConfParser::ParseString(const wchar_t * str) { reading_from_file = false; reading_from_wchar_string = true; pchar_unicode = str; pchar_ascii = 0; Parse(); return status; } ConfParser::Status ConfParser::ParseString(const std::wstring & str) { return ParseString(str.c_str()); } void ConfParser::Parse() { if( !root_space ) { status = no_space; return; } line = 1; status = ok; space = root_space; ReadChar(); SkipWhiteLines(); ParseLoop(); if( status == ok && space != root_space ) { // last closing a space characters ')' are missing status = syntax_error; } } void ConfParser::ParseLoop() { while( status == ok && lastc != -1 ) { if( lastc == list_end ) { SpaceEnds(); } else { ReadVariable(); if( lastc == list_start ) { SpaceStarts(); } else if( lastc == separator && !variable.empty() ) { ReadAddValue(); } else { status = syntax_error; } } if( status == ok ) SkipWhiteLines(); } } void ConfParser::SpaceEnds() { if( space == root_space ) { // there cannot be a loose list end character in the global space status = syntax_error; } else { space = space->parent; ReadChar(); } } void ConfParser::SpaceStarts() { Space * new_space = new Space(); space->spaces.push_back(new_space); new_space->parent = space; new_space->name = variable; space = new_space; ReadChar(); } void ConfParser::ReadAddValue() { ReadChar(); // skipping separator '=' if( ReadValue() ) { AddOption(); } else { status = syntax_error; } } bool ConfParser::IsVariableChar(int c) { if( (c>='a' && c<='z') || (c>='A' && c<='Z') || (c>='0' && c<='9') || c=='.' || c==',' || c=='_' ) return true; return false; } bool ConfParser::IsWhite(int c) { // dont use '\n' here // 13 (\r) is at the end of a line in a dos file \r\n // 160 is an unbreakable space if( c==' ' || c=='\t' || c==13 || c==160 ) return true; return false; } void ConfParser::SkipWhite() { while( IsWhite(lastc) || lastc == commentary ) { if( lastc == commentary ) SkipLine(); else ReadChar(); } } void ConfParser::SkipWhiteLines() { while( IsWhite(lastc) || lastc == commentary || lastc=='\n' ) { if( lastc == commentary ) SkipLine(); else ReadChar(); } } void ConfParser::SkipLine() { while( lastc != -1 && lastc != '\n' ) ReadChar(); } void ConfParser::Trim(std::wstring & s) { std::wstring::size_type i; if( s.empty() ) return; // looking for white characters at the end for(i=s.size()-1 ; i>0 && IsWhite(s[i]) ; --i); if( i==0 && IsWhite(s[i]) ) { // the whole string has white characters s.clear(); return; } // deleting white characters at the end if( i != s.size() - 1 ) s.erase(i+1, std::wstring::npos); // looking for white characters at the beginning for(i=0 ; itable_single[variable] = value[0]; DeleteFromTable(variable); } else { space->table[variable] = value; DeleteFromTableSingle(variable); } } void ConfParser::DeleteFromTable(const std::wstring & var) { Space::Table::iterator i = space->table.find(var); if( i != space->table.end() ) space->table.erase(i); } void ConfParser::DeleteFromTableSingle(const std::wstring & var) { Space::TableSingle::iterator i = space->table_single.find(var); if( i != space->table_single.end() ) space->table_single.erase(i); } void ConfParser::ReadVariable() { variable.clear(); SkipWhite(); while( IsVariableChar(lastc) ) { variable += lastc; ReadChar(); } SkipWhite(); } bool ConfParser::ReadValue() { value.clear(); SkipWhite(); if( lastc == list_start ) return ReadValueList(); else return ReadValueNoList(); } bool ConfParser::ReadValueList() { ReadChar(); // skipping first list character '(' SkipWhiteLines(); // lists can be split into several lines while( lastc != -1 && lastc != list_end ) { if( !ReadValueNoList(true) ) return false; if( lastc == list_delimiter ) ReadChar(); SkipWhiteLines(); } if( lastc != list_end ) return false; ReadChar(); // skipping last list character ')' SkipWhite(); return true; } bool ConfParser::ReadValueNoList(bool use_list_delimiter) { bool res; value_item.clear(); if( lastc == '"' ) { res = ReadValueQuoted(); // quoted value if( res ) value.push_back(value_item); } else { res = ReadValueSimple(use_list_delimiter); if( res && !value_item.empty() ) value.push_back(value_item); } return res; } bool ConfParser::ReadValueQuoted() { ReadChar(); // skipping the first quote // !! dodac obsluge innych escapowanych znakow w szczegolnosci \0 (serializator Space juz tak zapisuje) while( lastc != '"' && lastc != -1 ) { if( use_escape_char && lastc == '\\' ) ReadChar(); value_item += lastc; ReadChar(); } if( lastc != '"' ) return false; ReadChar(); // skipping the last quote SkipWhite(); return true; } bool ConfParser::ReadValueSimple(bool use_list_delimiter) { int list_delimiter1 = -1; int list_delimiter2 = -1; if( use_list_delimiter ) list_delimiter1 = list_delimiter; if( use_list_delimiter || space != root_space ) list_delimiter2 = list_end; while( lastc!=-1 && lastc!='\n' && lastc!=commentary && lastc!=list_delimiter1 && lastc!=list_delimiter2 ) { value_item += lastc; ReadChar(); } Trim(value_item); SkipWhite(); return true; } int ConfParser::ReadUTF8Char() { int c; bool correct; lastc = -1; do { Ezc::UTF8ToInt(file, c, correct); if( !file ) return lastc; } while( !correct ); lastc = c; if( lastc == '\n' ) ++line; return lastc; } int ConfParser::ReadASCIIChar() { lastc = file.get(); if( lastc == '\n' ) ++line; return lastc; } int ConfParser::ReadCharFromWcharString() { if( *pchar_unicode == 0 ) lastc = -1; else lastc = *(pchar_unicode++); if( lastc == '\n' ) ++line; return lastc; } int ConfParser::ReadCharFromUTF8String() { int c; bool correct; lastc = -1; do { size_t len = Ezc::UTF8ToInt(pchar_ascii, c, correct); pchar_ascii += len; if( *pchar_ascii == 0 ) return lastc; } while( !correct ); lastc = c; if( lastc == '\n' ) ++line; return lastc; } int ConfParser::ReadCharFromAsciiString() { if( *pchar_ascii == 0 ) lastc = -1; else lastc = *(pchar_ascii++); if( lastc == '\n' ) ++line; return lastc; } int ConfParser::ReadChar() { if( reading_from_file ) { if( input_as_utf8 ) return ReadUTF8Char(); else return ReadASCIIChar(); } else { if( reading_from_wchar_string ) { return ReadCharFromWcharString(); } else { if( input_as_utf8 ) return ReadCharFromUTF8String(); else return ReadCharFromAsciiString(); } } }