#This script is designed to run with Understand - CodeCheck
use base qw(Understand::Codecheck);
use strict;


use constant ERR1 => 'Violation: invalid escape sequence \'\\%1\' in: %2';

sub register_tr_text {
	my $check = shift;
	$check->add_tr_text(ERR1);
}

sub name { '2-13-1 escape sequences are standardized' }

sub description { '2-13-1 (Required) Only escape sequences defined in the ISO C standard shall be used.' }

sub detailed_description { <<'END_DESC'
<p style="margin-bottom:0;padding-bottom:0;"><b>Rationale</b><br>
The use of an undefined escape sequence leads to undefined behaviour.
The defined escape sequences (ISO/IEC 14882:2003 [1] 2.13.2) are:
<pre style="margin-top:0;padding-top:0;">\n, \t, \v, \b, \r, \f, \a, \\, \?, \', \", \&lt;Octal Number&gt;, \x&lt;Hexadecimal Number&gt;
</pre>
END_DESC
}

sub test_language {
	my $language = shift;
	return $language eq 'C++';
}

sub test_entity { 1 }

sub test_global { 0 }

sub define_options { }


sub check {
	my $check = shift;
	my $file = shift;
	return unless $file->kind->check('file');
	
	my $lexer = $file->lexer(0);
	my @lexemes = $lexer->lexemes if $lexer;
	
	foreach my $lexeme (@lexemes){
		next unless $lexeme->token eq 'String';
		my $content = $lexeme->text;
		next unless $content =~ m! ^ ['"] !sx;
		while ($content =~ m! \G [^\\]* \\ (x[0-9a-fA-F]{1,2} | [0-7]{2,3} | .) !sxg) {
			my $escape = $1;
			
#			\n, \t, \v, \b, \r, \f, \a, \\, \?, \', \"
			next if $escape =~ m! ^ (?: n | t | v | b | r | f | a | \\ | \? | ' | " | 0 | [0-7]{2,3} | x[0-9a-fA-F]{1,2} ) $ !sx;
			
			$check->violation(0, $file, $lexeme->line_begin, $lexeme->column_begin, ERR1, $escape, $content);
		}
	}
	
	return;
}