pikotools/src/space/spaceparser.cpp

/*
 * This file is a part of PikoTools
 * and is distributed under the (new) BSD licence.
 * Author: Tomasz Sowa <t.sowa@ttmath.org>
 */

/* 
 * Copyright (c) 2012-2021, Tomasz Sowa
 * All rights reserved.
 * 
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions are met:
 * 
 *  * Redistributions of source code must retain the above copyright notice,
 *    this list of conditions and the following disclaimer.
 *    
 *  * Redistributions in binary form must reproduce the above copyright
 *    notice, this list of conditions and the following disclaimer in the
 *    documentation and/or other materials provided with the distribution.
 *    
 *  * Neither the name Tomasz Sowa nor the names of contributors to this
 *    project may be used to endorse or promote products derived
 *    from this software without specific prior written permission.
 *
 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
 * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
 * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
 * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
 * THE POSSIBILITY OF SUCH DAMAGE.
 */

#include <cstdlib>
#include <wchar.h>
#include "spaceparser.h"
#include "utf8/utf8.h"
#include "convert/strtoint.h"


namespace pt
{




SpaceParser::SpaceParser()
{
	root_space = 0;
	SetDefault();
}


void SpaceParser::SetSpace(Space * pspace)
{
	root_space = pspace;
}


void SpaceParser::SetSpace(Space & pspace)
{
	root_space = &pspace;
}


void SpaceParser::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        = ':';
	space_start		 = '{';
	space_end		 = '}';
	table_start      = '[';
	table_end        = ']';
	option_delimiter = ',';
	skip_empty       = false;
	use_escape_char  = true;
	input_as_utf8    = true;
}



void SpaceParser::SkipEmpty(bool skip)
{
	skip_empty = skip;
}


void SpaceParser::UseEscapeChar(bool escape)
{
	use_escape_char = escape;
}


void SpaceParser::UTF8(bool utf)
{
	input_as_utf8 = utf;
}


int SpaceParser::get_last_parsed_line()
{
	return line;
}



SpaceParser::Status SpaceParser::ParseJSONFile(const char * file_name)
{
	reading_from_file = true;
	parsing_space = false;

	file.clear();
	file.open(file_name, std::ios_base::binary | std::ios_base::in);
	
	if( file )
	{
		ParseRootSpace();
		file.close();
	}
	else
	{
		status = cant_open_file;
	}
	
return status;
}



SpaceParser::Status SpaceParser::ParseJSONFile(const std::string & file_name)
{
	return ParseJSONFile(file_name.c_str());
}




SpaceParser::Status SpaceParser::ParseJSONFile(const wchar_t * file_name)
{
	std::string file_name_utf8;

	WideToUTF8(file_name, file_name_utf8);
	return ParseJSONFile(file_name_utf8.c_str());
}



SpaceParser::Status SpaceParser::ParseJSONFile(const std::wstring & file_name)
{
	return ParseJSONFile(file_name.c_str());
}





SpaceParser::Status SpaceParser::ParseSpaceFile(const char * file_name)
{
	reading_from_file = true;
	parsing_space = true;

	file.clear();
	file.open(file_name, std::ios_base::binary | std::ios_base::in);

	if( file )
	{
		ParseRootSpace();
		file.close();
	}
	else
	{
		status = cant_open_file;
	}

return status;
}



SpaceParser::Status SpaceParser::ParseSpaceFile(const std::string & file_name)
{
	return ParseSpaceFile(file_name.c_str());
}




SpaceParser::Status SpaceParser::ParseSpaceFile(const wchar_t * file_name)
{
	std::string file_name_utf8;

	WideToUTF8(file_name, file_name_utf8);
	return ParseSpaceFile(file_name_utf8.c_str());
}



SpaceParser::Status SpaceParser::ParseSpaceFile(const std::wstring & file_name)
{
	return ParseSpaceFile(file_name.c_str());
}



SpaceParser::Status SpaceParser::ParseJSON(const char * str)
{
	reading_from_file         = false;
	reading_from_wchar_string = false;
	pchar_ascii               = str;
	pchar_unicode             = 0;
	parsing_space             = false;

	ParseRootSpace();

return status;
}


SpaceParser::Status SpaceParser::ParseJSON(const std::string & str)
{
	return ParseJSON(str.c_str());
}


SpaceParser::Status SpaceParser::ParseJSON(const wchar_t * str)
{
	reading_from_file         = false;
	reading_from_wchar_string = true;
	pchar_unicode             = str;
	pchar_ascii               = 0;
	parsing_space             = false;

	ParseRootSpace();

return status;
}


SpaceParser::Status SpaceParser::ParseJSON(const std::wstring & str)
{
	return ParseJSON(str.c_str());
}





SpaceParser::Status SpaceParser::ParseSpace(const char * str)
{
	reading_from_file         = false;
	reading_from_wchar_string = false;
	pchar_ascii               = str;
	pchar_unicode             = 0;
	parsing_space             = true;

	ParseRootSpace();

return status;
}


SpaceParser::Status SpaceParser::ParseSpace(const std::string & str)
{
	return ParseSpace(str.c_str());
}


SpaceParser::Status SpaceParser::ParseSpace(const wchar_t * str)
{
	reading_from_file         = false;
	reading_from_wchar_string = true;
	pchar_unicode             = str;
	pchar_ascii               = 0;
	parsing_space             = true;

	ParseRootSpace();

return status;
}


SpaceParser::Status SpaceParser::ParseSpace(const std::wstring & str)
{
	return ParseSpace(str.c_str());
}





void SpaceParser::ParseRootSpace()
{
	line    = 1;
	status  = ok;

	if( !root_space )
	{
		status = no_space;
		return;
	}

	ReadChar(); // put first character to lastc

	if( parsing_space )
	{
		separator = '=';
		table_start = '(';
		table_end = ')';
		ParseSpace(root_space);
	}
	else
	{
		separator = ':';
		table_start = '[';
		table_end = ']';
		Parse(root_space, false, false);
	}

	SkipWhite();

	if( lastc != -1 )
		status = syntax_error;

	token.clear();
}


void SpaceParser::Parse(Space * space, bool is_object_value, bool is_table_value)
{
	SkipWhite();

	if( lastc == space_start )
	{
		ParseSpace(space);
	}
	else
	if( lastc == table_start )
	{
		ParseTable(space);
	}
	else
	if( lastc == '"' ) // IMPROVEME define a variable
	{
		ParseTextValue(space);
	}
	else
	{
		ReadStringValue(token, is_object_value, is_table_value);

		if( token == L"null" )
		{
			space->set_null();
		}
		else
		if( token == L"true" )
		{
			space->set(true);
		}
		else
		if( token == L"false" )
		{
			space->set(false);
		}
		else
		if( is_integer_token() )
		{
			ParseIntegerValue(space);
		}
		else
		if( is_floating_point_token() )
		{
			ParseFloatingPointValue(space);
		}
		else
		{
			if( parsing_space )
			{
				space->set(token);
			}
			else
			{
				status = syntax_error;
			}
		}
	}
}





void SpaceParser::ParseSpace(Space * space)
{
	/*
	 * in Space format in global namespace the space start character is not required
	 */
	bool need_space_start_character = !parsing_space || space != root_space;

	if( need_space_start_character )
	{
		ReadChar(); // inserting a next character after the space_start char to lastc
	}

	space->set_empty_object();
	ParseKeyValuePairs(space);

	if( need_space_start_character )
	{
		if( lastc == space_end )
		{
			ReadChar();
		}
		else
		{
			status = syntax_error;
		}
	}
}





void SpaceParser::ParseTextValue(Space * space)
{
	space->set_empty_wstring();
	std::wstring * str = space->get_wstr();

	if( parsing_space )
		ReadMultilineTokenQuoted(*str);
	else
		ReadTokenQuoted(*str);
}


void SpaceParser::ParseIntegerValue(Space * space)
{
	const wchar_t * after_str = nullptr;
	bool was_overflow = false;
	int base = 10;

	if( parsing_space )
	{
		// in Space format when the integer value begins with a zero it means
		// this is an octal number
		if( !token.empty() && token[0] == '0' )
			base = 8;
	}

	long long val = Toll(token.c_str(), base, &after_str, &was_overflow, false);

	if( was_overflow )
	{
		status = syntax_error;
	}
	else
	if( size_t(after_str - token.c_str()) != token.size() )
	{
		status = syntax_error;
	}
	else
	{
		space->set(val);
	}
}


void SpaceParser::ParseFloatingPointValue(Space * space)
{
	wchar_t * after_str = nullptr;
	double val = wcstod(token.c_str(), &after_str);

	if( errno == ERANGE )
	{
		status = syntax_error;
	}
	else
	if( size_t(after_str - token.c_str()) != token.size() )
	{
		status = syntax_error;
	}
	else
	{
		space->set(val);
	}
}




void SpaceParser::ParseTable(Space * space)
{
	space->set_empty_table();


	ReadChar(); // inserting a next character after the table_start char to lastc
	space->set_empty_table();
	ParseValuesList(space);

	if( lastc == table_end )
	{
		ReadChar();
	}
	else
	{
		status = syntax_error;
	}
}




void SpaceParser::ParseKeyValuePairs(Space * space)
{
	bool is_first = true;
	SkipWhite();

	while( status == ok && lastc != space_end && lastc != -1 )
	{
		if( !is_first )
		{
			SkipWhite();

			if( lastc == option_delimiter )
			{
				ReadChar(); // inserting a next character after the option_delimiter to lastc

				if( parsing_space )
				{
					// in space format a space_end character is allowed to be after the last table item
					SkipWhite();

					if( lastc == space_end )
						break;
				}
			}
			else
			if( !parsing_space )
			{
				// in json format the option_delimiter is required
				status = syntax_error;
			}
		}

		if( status == ok )
		{
			ReadKey();

			if( status == ok )
			{
				SkipWhite();

				if( lastc == separator )
				{
					ReadChar(); // inserting a next character after the separator to lastc

					Space & new_space = space->add(token.c_str(), new Space());
					Parse(&new_space, true, false);
				}
				else
				if( parsing_space && lastc == space_start )
				{
					Space & new_space = space->add_child_space(token.c_str());
					ParseSpace(&new_space);
				}
				else
				{
					status = syntax_error;
				}
			}
		}

		is_first = false;
		SkipWhite();
	}
}




void SpaceParser::ParseValuesList(Space * space)
{
	bool is_first = true;
	SkipWhite();

	while( status == ok && lastc != table_end && lastc != -1 )
	{
		if( !is_first )
		{
			SkipWhite();

			if( lastc == option_delimiter ) // may add a new delimiter for tables? default the same as for objects...
			{
				ReadChar(); // inserting a next character after the delimiter

				if( parsing_space )
				{
					// in space format a table_end character is allowed to be after the last table item
					SkipWhite();

					if( lastc == table_end )
						break;
				}
			}
			else
			if( !parsing_space )
			{
				// in json format the option_delimiter is required
				status = syntax_error;
			}
		}

		if( status == ok )
		{
			Space * new_space = &space->add(new Space());
			Parse(new_space, false, true);
		}

		is_first = false;
		SkipWhite();
	}
}



bool SpaceParser::is_integer_token()
{
	if( token.empty() )
		return false;

	size_t i = 0;

	if( token[i] == '-' )
	{
		i += 1;

		if( token.size() == 1 )
			return false;
	}

	for( ; i < token.size() ; ++i)
	{
		if( token[i] < '0' || token[i] > '9' )
		{
			return false;
		}
	}

	return true;
}


bool SpaceParser::is_floating_point_token()
{
	bool was_dot = false;
	bool was_exponential = false;
	bool was_plus_minus_sign = false;

	if( token.empty() )
		return false;

	size_t i = 0;

	if( token[i] == '-' )
	{
		i += 1;

		if( token.size() == 1 )
			return false;
	}

	for( ; i < token.size() ; ++i)
	{
		if( token[i] == '.' )
		{
			if( was_dot || was_exponential )
				return false;

			was_dot = true;
		}
		else
		if( token[i] == 'e' || token[i]=='E' )
		{
			if( was_exponential )
				return false;

			was_exponential = true;

			// the exponential character cannot be the last character
			if( i + 1 == token.size() )
				return false;
		}
		else
		if( token[i] == '+' || token[i] == '-' )
		{
			if( was_plus_minus_sign || !was_exponential )
				return false;

			// the plus or minus should be after the exponential character
			if( i > 0 && (token[i-1] != 'e' && token[i-1] != 'E') )
				return false;

			was_plus_minus_sign = true;
		}
		else
		if( token[i] < '0' || token[i] > '9' )
		{
			return false;
		}
	}

	return true;
}



bool SpaceParser::IsWhite(int c)
{
	// 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 || c==10 )
		return true;
		
return false;
}


bool SpaceParser::is_alfa_numeric_char(int c)
{
	return  (c >= 'a' && c <= 'z') ||
			(c >= 'A' && c <= 'Z') ||
			(c >= '0' && c <= '9') ||
			c == '.' || c=='-' || c=='+';
}


void SpaceParser::SkipLine()
{
	while( lastc != -1 && (char_was_escaped || lastc != '\n') )
		ReadChar();
}



void SpaceParser::SkipWhite()
{
	if( parsing_space )
	{
		while( IsWhite(lastc) || (!char_was_escaped && lastc == '#') )
		{
			if( lastc == '#' )
				SkipLine();
			else
				ReadChar();
		}
	}
	else
	{
		while( IsWhite(lastc) )
		{
			ReadChar();
		}
	}
}


void SpaceParser::TrimLastWhite(std::wstring & s)
{
	std::wstring::size_type i;

	for(i=s.size() ; i>0 && IsWhite(s[i-1]) ; --i)
	{
	}

	if( i < s.size() )
	{
		s.erase(i, std::wstring::npos);
	}
}


/*
void SpaceParser::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 consists of 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 ; i<s.size() && IsWhite(s[i]) ; ++i);
	
	// deleting white characters at the beginning
	if( i != 0 )
		s.erase(0, i);
}
*/


void SpaceParser::ReadTokenUntilDelimiter(std::wstring & token, int delimiter1, int delimiter2)
{
	token.clear();

	while( lastc != -1 && (char_was_escaped || (lastc != '\n' && lastc != '#' && lastc != delimiter1 && lastc != delimiter2)) )
	{
		token += static_cast<wchar_t>(lastc);
		ReadChar();
	}

	TrimLastWhite(token);
}


void SpaceParser::ReadAlfaNumericToken(std::wstring & token)
{
	token.clear();

	while( is_alfa_numeric_char(lastc) )
	{
		token += static_cast<wchar_t>(lastc);
		ReadChar();
	}
}


void SpaceParser::ReadStringValue(std::wstring & token, bool is_object_value, bool is_table_value)
{
	if( parsing_space )
	{
		if( is_object_value )
		{
			ReadTokenUntilDelimiter(token, space_end, -1);
		}
		else
		if( is_table_value )
		{
			ReadTokenUntilDelimiter(token, table_end, option_delimiter);
		}
		else
		{
			ReadTokenUntilDelimiter(token, -1, -1);
		}
	}
	else
	{
		ReadAlfaNumericToken(token);
	}
}


// rename to something like ReadSpaceFieldToken???
void SpaceParser::ReadSpaceFieldToken(std::wstring & token)
{
	token.clear();

	while( lastc != -1 && (char_was_escaped || (lastc != separator && lastc != 10 && lastc != space_start && lastc != '#' )) )
	{
		token += static_cast<wchar_t>(lastc);
		ReadChar();
	}

	TrimLastWhite(token);
}


// IMPROVEME in JSON we should not allow non-escaped a new line character
void SpaceParser::ReadTokenQuoted(std::wstring & token)
{
	token.clear();
	ReadChar(); // skipping the first quotation mark

	while( lastc != -1 && (char_was_escaped || (lastc != '"' && lastc != 10)) )
	{
		token += static_cast<wchar_t>(lastc);
		ReadChar();
	}

	if( !char_was_escaped && lastc == '"' )
	{
		ReadChar(); // skipping the last quotation mark
	}
	else
	{
		status = syntax_error;
	}
}


void SpaceParser::ReadMultilineTokenQuoted(std::wstring & token)
{
	token.clear();
	ReadChar(); // skipping the first quotation mark

	while( lastc != -1 && (char_was_escaped || lastc != '"') )
	{
		token += static_cast<wchar_t>(lastc);
		ReadChar();
	}

	if( !char_was_escaped && lastc == '"' )
	{
		ReadChar(); // skipping the last quotation mark
	}
	else
	{
		status = syntax_error;
	}
}


/*
 * this method is used to read the field name (key) in an object
 * or to read the space child name (used in Space format)
 */
void SpaceParser::ReadKey()
{
	SkipWhite();

	if( parsing_space )
	{
		if( lastc == '"' )
		{
			ReadMultilineTokenQuoted(token);
		}
		else
		{
			ReadSpaceFieldToken(token);
		}
	}
	else
	{
		if( lastc == '"' )
		{
			ReadTokenQuoted(token);
		}
		else
		{
			status = syntax_error;
		}
	}
}




int SpaceParser::ReadUTF8Char()
{
int c;
bool correct;

	lastc = -1;

	do
	{
		UTF8ToInt(file, c, correct);

		if( !file )
			return lastc;
	}
	while( !correct );

	lastc = c;

	if( lastc == '\n' )
		++line;
	
return lastc;
}



int SpaceParser::ReadASCIIChar()
{
	lastc = file.get();

	if( lastc == '\n' )
		++line;
	
return lastc;
}




int SpaceParser::ReadCharFromWcharString()
{
	if( *pchar_unicode == 0 )
		lastc = -1;
	else
		lastc = *(pchar_unicode++);

	if( lastc == '\n' )
		++line;

return lastc;
}


int SpaceParser::ReadCharFromUTF8String()
{
int c;
bool correct;

	lastc = -1;

	do
	{
		size_t len = UTF8ToInt(pchar_ascii, c, correct);
		pchar_ascii += len;
	}
	while( *pchar_ascii && !correct );

	if( correct )
		lastc = c;

	if( lastc == '\n' )
		++line;
	
return lastc;
	
}


int SpaceParser::ReadCharFromAsciiString()
{
	if( *pchar_ascii == 0 )
		lastc = -1;
	else
		lastc = *(pchar_ascii++);

	if( lastc == '\n' )
		++line;

return lastc;
}


int SpaceParser::ReadCharNoEscape()
{
	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();
		}
	}
}

bool SpaceParser::IsHexDigit(wchar_t c)
{
	return ((c>='0' && c<='9') ||
			(c>='a' && c<='f') ||
			(c>='A' && c<='F') );
}


int SpaceParser::HexToInt(wchar_t c)
{
	if( c>='0' && c<='9' )
		return c - '0';

	if( c>='a' && c<='f' )
		return c - 'a' + 10;

	if( c>='A' && c<='F' )
		return c - 'A' + 10;

return 0;
}


void SpaceParser::ReadUnicodeCodePoint()
{
wchar_t c;
int value = 0;

	for(int i=0 ; i<4 ; ++i)
	{
		c = ReadCharNoEscape();

		if( !IsHexDigit(c) )
		{
			status = syntax_error;
			return;
		}

		value = (value << 4) | HexToInt(c);
	}

	lastc = (wchar_t)value;
}


int SpaceParser::ReadChar()
{
	char_was_escaped = false;
	ReadCharNoEscape();

	if( use_escape_char && lastc == '\\' )
	{
		char_was_escaped = true;
		ReadCharNoEscape();

		switch(lastc)
		{
		case '0':	lastc = 0;		break;
		case 't':	lastc = '\t';	break;
		case 'r':	lastc = '\r';	break;
		case 'n':	lastc = '\n';	break;
		case 'b':	lastc = 0x08;	break;
		case 'f':	lastc = 0x0c;	break;
		case 'u':	ReadUnicodeCodePoint(); break;
		// "in other cases we return the last character, so two \\ returns one \ "
		}
	}

return lastc;
}



} // namespace