/*
 * 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       = nullptr;
	space_start		 = '{';
	space_end		 = '}';
	option_delimiter = ',';
	input_as_utf8    = true;
}



void SpaceParser::use_utf8(bool utf)
{
	input_as_utf8 = utf;
}


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



SpaceParser::Status SpaceParser::parse_json_file(const char * file_name, Space & out_space, bool clear_space)
{
	reading_from_file = true;
	parsing_space = false;
	root_space = &out_space;

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



SpaceParser::Status SpaceParser::parse_json_file(const std::string & file_name, Space & out_space, bool clear_space)
{
	return parse_json_file(file_name.c_str(), out_space, clear_space);
}




SpaceParser::Status SpaceParser::parse_json_file(const wchar_t * file_name, Space & out_space, bool clear_space)
{
	std::string file_name_utf8;

	wide_to_utf8(file_name, file_name_utf8);
	return parse_json_file(file_name_utf8.c_str(), out_space, clear_space);
}



SpaceParser::Status SpaceParser::parse_json_file(const std::wstring & file_name, Space & out_space, bool clear_space)
{
	return parse_json_file(file_name.c_str(), out_space, clear_space);
}





SpaceParser::Status SpaceParser::parse_space_file(const char * file_name, Space & out_space, bool clear_space)
{
	reading_from_file = true;
	parsing_space = true;
	root_space = &out_space;

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

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

return status;
}



SpaceParser::Status SpaceParser::parse_space_file(const std::string & file_name, Space & out_space, bool clear_space)
{
	return parse_space_file(file_name.c_str(), out_space, clear_space);
}




SpaceParser::Status SpaceParser::parse_space_file(const wchar_t * file_name, Space & out_space, bool clear_space)
{
	std::string file_name_utf8;

	wide_to_utf8(file_name, file_name_utf8);
	return parse_space_file(file_name_utf8.c_str(), out_space, clear_space);
}



SpaceParser::Status SpaceParser::parse_space_file(const std::wstring & file_name, Space & out_space, bool clear_space)
{
	return parse_space_file(file_name.c_str(), out_space, clear_space);
}



SpaceParser::Status SpaceParser::parse_json(const char * str, Space & out_space, bool clear_space)
{
	reading_from_file         = false;
	reading_from_wchar_string = false;
	pchar_ascii               = str;
	pchar_unicode             = 0;
	parsing_space             = false;
	root_space                = &out_space;

	parse_root_space(clear_space);

	return status;
}


SpaceParser::Status SpaceParser::parse_json(const std::string & str, Space & out_space, bool clear_space)
{
	return parse_json(str.c_str(), out_space, clear_space);
}


SpaceParser::Status SpaceParser::parse_json(const wchar_t * str, Space & out_space, bool clear_space)
{
	reading_from_file         = false;
	reading_from_wchar_string = true;
	pchar_unicode             = str;
	pchar_ascii               = 0;
	parsing_space             = false;
	root_space                = &out_space;

	parse_root_space(clear_space);

	return status;
}


SpaceParser::Status SpaceParser::parse_json(const std::wstring & str, Space & out_space, bool clear_space)
{
	return parse_json(str.c_str(), out_space, clear_space);
}





SpaceParser::Status SpaceParser::parse_space(const char * str, Space & out_space, bool clear_space)
{
	reading_from_file         = false;
	reading_from_wchar_string = false;
	pchar_ascii               = str;
	pchar_unicode             = 0;
	parsing_space             = true;
	root_space                = &out_space;

	parse_root_space(clear_space);

	return status;
}


SpaceParser::Status SpaceParser::parse_space(const std::string & str, Space & out_space, bool clear_space)
{
	return parse_space(str.c_str(), out_space, clear_space);
}


SpaceParser::Status SpaceParser::parse_space(const wchar_t * str, Space & out_space, bool clear_space)
{
	reading_from_file         = false;
	reading_from_wchar_string = true;
	pchar_unicode             = str;
	pchar_ascii               = 0;
	parsing_space             = true;
	root_space                = &out_space;

	parse_root_space(clear_space);

	return status;
}


SpaceParser::Status SpaceParser::parse_space(const std::wstring & str, Space & out_space, bool clear_space)
{
	return parse_space(str.c_str(), out_space, clear_space);
}





void SpaceParser::parse_root_space(bool clear_root_space)
{
	line    = 1;
	status  = ok;

	if( clear_root_space )
	{
		root_space->set_empty_object();
	}

	read_char(); // put first character to lastc

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

	skip_white();

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

	token.clear();
}


void SpaceParser::parse(Space * space, bool is_object_value, bool is_table_value)
{
	skip_white();

	if( lastc == space_start )
	{
		parse_space(space);
	}
	else
	if( lastc == table_start )
	{
		parse_table(space);
	}
	else
	if( lastc == '"' ) // IMPROVEME define a variable
	{
		parse_text_value(space);
	}
	else
	{
		read_string_value(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() )
		{
			parse_integer_value(space);
		}
		else
		if( is_floating_point_token() )
		{
			parse_floating_point_value(space);
		}
		else
		{
			if( parsing_space )
			{
				space->set(token);
			}
			else
			{
				status = syntax_error;
			}
		}
	}
}





void SpaceParser::parse_space(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 )
	{
		read_char(); // inserting a next character after the space_start char to lastc
	}

	if( !space->is_object() )
		space->set_empty_object();

	parse_key_value_pairs(space);

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





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

	if( parsing_space )
		read_multiline_token_quoted(*str);
	else
		read_token_quoted(*str);
}


void SpaceParser::parse_integer_value(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::parse_floating_point_value(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::parse_table(Space * space)
{
	read_char(); // inserting a next character after the table_start char to lastc
	space->set_empty_table();
	parse_values_list(space);

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




void SpaceParser::parse_key_value_pairs(Space * space)
{
	bool is_first = true;
	skip_white();

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

			if( lastc == option_delimiter )
			{
				read_char(); // 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 option_delimiter
					skip_white();

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

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

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

				if( lastc == separator )
				{
					read_char(); // 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
				{
					status = syntax_error;
				}
			}
		}

		is_first = false;
		skip_white();
	}
}




void SpaceParser::parse_values_list(Space * space)
{
	bool is_first = true;
	skip_white();

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

			if( lastc == option_delimiter ) // may add a new delimiter for tables? default the same as for objects...
			{
				read_char(); // 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
					skip_white();

					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;
		skip_white();
	}
}



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::is_white(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::skip_line()
{
	while( lastc != -1 && (char_was_escaped || lastc != '\n') )
		read_char();
}



void SpaceParser::skip_white()
{
	if( parsing_space )
	{
		while( is_white(lastc) || (!char_was_escaped && lastc == '#') )
		{
			if( lastc == '#' )
				skip_line();
			else
				read_char();
		}
	}
	else
	{
		while( is_white(lastc) )
		{
			read_char();
		}
	}
}


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

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

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



void SpaceParser::read_token_until_delimiter(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);
		read_char();
	}

	trim_last_white(token);
}


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

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


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


void SpaceParser::read_space_field_token(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);
		read_char();
	}

	trim_last_white(token);
}


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

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

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


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

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

	if( !char_was_escaped && lastc == '"' )
	{
		read_char(); // 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::read_key()
{
	skip_white();

	if( parsing_space )
	{
		if( lastc == '"' )
		{
			read_multiline_token_quoted(token);
		}
		else
		{
			read_space_field_token(token);
		}
	}
	else
	{
		if( lastc == '"' )
		{
			read_token_quoted(token);
		}
		else
		{
			status = syntax_error;
		}
	}
}





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


int SpaceParser::hex_to_int(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;
}


/*
 * format: \uHHHH where H is a hex digit 0-F
 */
bool SpaceParser::read_unicode_four_digit_format(bool has_first_byte, int first_byte)
{
int c;
int value = 0;

	for(int i=0 ; i<4 ; ++i)
	{
		if( i == 0 && has_first_byte )
		{
			c = first_byte;
		}
		else
		{
			c = read_char_no_escape();
		}

		if( !is_hex_digit(c) )
		{
			return false;
		}

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

	lastc = static_cast<wchar_t>(value);
	return true;
}



/*
 * format: \uHHHH and optionally following by \uHHHH
 *
 */
void SpaceParser::read_unicode_json_format(bool has_first_byte, int first_byte)
{
	bool ok = read_unicode_four_digit_format(has_first_byte, first_byte);

	if( ok && pt::is_first_surrogate_char(lastc) )
	{
		int c1 = lastc;
		int c = read_char_no_escape();

		ok = ok && (c == '\\');

		if( ok )
		{
			c = read_char_no_escape();
			ok = ok && (c == 'u');
			ok = ok && read_unicode_four_digit_format(false, 0);

			if( ok && pt::is_second_surrogate_char(lastc) )
			{
				int c2 = lastc;
				ok = ok && pt::surrogate_pair_to_int(c1, c2, lastc);
			}
		}
	}

	if( !ok || !pt::utf8_check_range(lastc) )
	{
		lastc = 0xFFFD; // U+FFFD "replacement character";
	}
}


/*
 * format: \u{H...} where H is a hex digit 0-F, minimum digits: 1, maximum digits: 6
 */
void SpaceParser::read_unicode_floating_format()
{
int c;
int value = 0;
int i;

	// max 6 hex digits + '}'
	for(i=0 ; i<7 ; ++i)
	{
		c = read_char_no_escape();

		if( !is_hex_digit(c) )
		{
			break;
		}

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

	if( i > 0 && c == '}' && pt::utf8_check_range(value) )
	{
		lastc = static_cast<wchar_t>(value);
	}
	else
	{
		lastc = 0xFFFD; // U+FFFD "replacement character";
	}
}





void SpaceParser::read_unicode_code_point()
{
	if( parsing_space )
	{
		int c = read_char_no_escape();

		if( c == '{' )
		{
			read_unicode_floating_format();
		}
		else
		{
			read_unicode_json_format(true, c);
		}
	}
	else
	{
		read_unicode_json_format(false, 0);
	}
}


int SpaceParser::read_char()
{
	char_was_escaped = false;
	read_char_no_escape();

	if( lastc == '\\' )
	{
		char_was_escaped = true;
		read_char_no_escape();

		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':	read_unicode_code_point(); break;
		// "in other cases we return the last character, so two \\ returns one \ "
		}
	}

return lastc;
}







} // namespace