Added support for multiline and comments.

* automake.y : Updated the grammar to handle multiline statements and two
*              type of comments.
* Lexer.pm, Tree.pm, parser.pl : Updated to support new features.
* input.txt : Updated to test combination of multiline statements and
*             comments.

8 files changed, 287 insertions, 89 deletions

diff --git a/lib/Automake/Parser/Converter.pl b/lib/Automake/Parser/Converter.pl
index ca6a315f0..ab2184486 100644
--- a/lib/Automake/Parser/Converter.pl
+++ b/lib/Automake/Parser/Converter.pl
@@ -17,15 +17,15 @@ while( <> )
 {
 	#Finding label word in the current line as every node and edge description 
 	#contains label property. The value of label is extracted.
-	if(m/label=\"(.*)\"/)
+	if( m/label=\"(.*)\"/ )
 	{
 		my $token = $1;
 		#Every edge is denoted as state_number -> state_number . The current 
 		#line is searched for this and to and from state number are extracted.
-		if(m/(\d+) -> (\d+)/)
+		if( m/(\d+) -> (\d+)/ )
 		{
 			# "$end" token is replaced with end.
-			if($token eq "\$end")
+			if( $token eq "\$end" )
 			{
 				$table[ $1 ]{ end } = $2;
 			}
@@ -37,7 +37,7 @@ while( <> )
 		#The line describing the node contains State word in its description 
 		#followed by state number. The state number is extracted and its value 
 		#is stored.
-		elsif(m/State (\d+)\\n/)
+		elsif( m/State (\d+)\\n/ )
 		{
 			$labels[ $1 ] = $token;
 		}
@@ -46,7 +46,7 @@ while( <> )
 	#state_number -> state_number R production_number.
 	#production_number denotes the production by which reduction is to happen. 
 	#It is extracted from the label value of the specified state.
-	elsif(m/(\d+) -> "\d+R(\d+)"/)
+	elsif( m/(\d+) -> "\d+R(\d+)"/ )
 	{
 		my $state_number = $1;
 		my $production_number = $2;
@@ -57,7 +57,7 @@ while( <> )
 		#with value equal to an array having number of words on right side and 
 		#function with name of the value of left side.
 		$labels[$state_number] =~ m/$production_number (.+): (.+)\.\\l/;
-		if($1 eq "\$accept")
+		if( $1 eq "\$accept" )
 		{
 			$table[ $state_number ] = {};
 			$acceptstate = $state_number;
@@ -78,7 +78,7 @@ for my $href ( @table )
 	my @hashval;
 	for my $key ( keys %$href )
 	{
-		if($key eq "reduce")
+		if( $key eq "reduce" )
 		{
 			push @hashval , sprintf( "$key => [%s]",join(", ",@{ $href -> { $key } }));
 		}
diff --git a/lib/Automake/Parser/Lexer.pm b/lib/Automake/Parser/Lexer.pm
index 8eff3151c..7446e8114 100644
--- a/lib/Automake/Parser/Lexer.pm
+++ b/lib/Automake/Parser/Lexer.pm
@@ -5,44 +5,162 @@ use Exporter;
 our @ISA = qw(Exporter);
 our @EXPORT = qw(lex);
 
-# lex(string)
-# Takes as input a string of line. Divides it into tokens as specified 
-# by Regex and outputs an array of Tokens. Every Token is an array having 
-# two values: token name and its value. If its an operator, it has only 
-# one value.
-sub lex($)
+# lex(string,multiline)
+# Takes as input a string of line and multiline variable deciding whether 
+# current line is related to the previous line. Divides it into tokens as 
+# specified by Regex and outputs an array of Tokens. Every Token is an 
+# array having two values: token name and its value. If its an operator, 
+# it has only one value.
+sub lex($$)
 {
+	my ( $curr_line , $multiline ) = @_;
 	my @tokens;
 	my $rhs = 0;
+	$_ = $curr_line;
 	while( $_ )
 	{
-		if( $rhs && s/^(.+)//o )
+		if( $multiline )
 		{
-			push @tokens, ["rhs",$1];
-			$rhs=0;
+			my @vals = split;
+			if( $multiline eq 'automake_comment' )
+			{
+				if( $vals[ -1 ] ne '\\' )
+				{
+					$multiline = undef;
+					push @tokens, [ "newline" ];
+				}
+				$_ = undef;
+				last;
+			}
+			elsif( $multiline eq 'comment' )
+			{
+			    my $comment;
+				foreach my $val ( @vals )
+				{
+					if($val =~ m/^##/)
+					{
+						if($vals[ -1 ] eq '\\')
+						{
+							$multiline = 'automake_comment';
+						}
+						$_ = undef;
+						last;
+					}
+					elsif( $val eq '\\' )
+					{
+						last;
+					}
+					else
+					{
+						$comment .= " ".$val;
+					}
+				}
+				if( $comment )
+				{
+					push @tokens, [ "comment" , $comment ];
+				}
+				if($vals[ -1 ] ne '\\')
+				{
+					$multiline = undef;
+					push @tokens, [ "newline" ];
+				}
+				$_ = undef;
+			}
+			else
+			{
+				$multiline = undef;
+				$rhs = 1;
+			}
 		}
-		elsif(s/^(PROGRAMS|LIBRARIES|LTLIBRARIES|LISP|PYTHON|JAVA|SCRIPTS|DATA|HEADERS|MASN|TEXINFOS)//o)
+		elsif( $rhs  )
+		{
+		    my @vals = split;
+			my $comment;
+			foreach my $val ( @vals )
+			{
+				if( $val =~ m/^##/ )
+				{
+					if($vals[ -1 ] eq '\\' )
+					{
+						$multiline = 'automake_comment';
+					}
+					$_ = undef;
+					last;
+				}
+				elsif( $val =~ m/^#(.*)/ )
+				{
+					if($vals[ -1 ] eq '\\' )
+					{
+						$multiline = 'comment';
+					}
+					$comment .= " ".$1;
+				}
+				elsif( $val =~ m/\\/ )
+				{
+					if( !$multiline )
+					{
+						$multiline = 'rhsval';
+					}
+				}
+				elsif( $comment )
+				{
+					$comment .= " ".$val;
+				}
+				else
+				{
+					push @tokens, [ "rhsval" , $val];
+				}
+			}
+			if( $comment )
+			{
+				push @tokens, [ "comment" , $comment];
+			}
+			if( !$multiline )
+			{
+				push @tokens, [ "newline" ];
+			}
+			$_ = undef;
+		}
+		elsif( s/^##.*\n$//o )
+		{
+		}
+		elsif( s/^#(.*)\n$//o )
+		{
+			my $val = $1;
+			if($val =~ m/(.*?)\\/o)
+			{
+				push @tokens, [ "comment" , substr( $1 , 0 , -1 )];
+				$multiline = 'comment';
+			}
+			else
+			{
+				push @tokens, [ "comment" , $1];
+				push @tokens, [ "newline" ];
+			}
+		}
+		elsif( s/^(PROGRAMS|LIBRARIES|LTLIBRARIES|LISP|PYTHON|JAVA|SCRIPTS|DATA|HEADERS|MASN|TEXINFOS)//o)
 		{
 			push @tokens, [$1];
 		}
-		elsif(s/^([a-zA-Z0-9]+)//o)
+		elsif( s/^([a-zA-Z0-9]+)//o )
 		{
 			push @tokens, ["value",$1];
 		}
-		elsif(s/^(=)//o)
+		elsif( s/^(=)//o )
 		{
 			push @tokens, [$1];
 			$rhs = 1;
 		}
-		elsif(s/^(:|_)//o)
+		elsif( s/^(:|_)//o )
 		{
 			push @tokens, [$1];
 		}
-		elsif(s/^\n//o)
+		elsif( s/^\n//o )
 		{
 			push @tokens, ["newline"];
+			$multiline = undef;
 		}
-		elsif(s/^(\r| )//o)
+		elsif( s/^(\r|\s*)//o )
 		{
 		}
 		else
@@ -50,6 +168,6 @@ sub lex($)
 			die "Incorrect input $_";
 		}
 	}
-	return @tokens;
+	return ( \@tokens , $multiline );
 }
 
diff --git a/lib/Automake/Parser/Makefile b/lib/Automake/Parser/Makefile
index 3d5a269e2..9e1c41322 100644
--- a/lib/Automake/Parser/Makefile
+++ b/lib/Automake/Parser/Makefile
@@ -1,9 +1,9 @@
 
 all: ast.png
 
-ast.png: Lexer.pm Tree.pm ParserTable.pm parser.pl
+ast.png: Lexer.pm Tree.pm ParserTable.pm parser.pl input.txt
 	perl -I. parser.pl
-	unflatten -f -l 5 -c 6 -o ast1.gv ast.gv
+	unflatten -f -l 10 -c 10 -o ast1.gv ast.gv
 	dot -Tpng ast1.gv > ast.png
 	rm ast1.gv
 
@@ -12,7 +12,7 @@ build: automake.dot ParserTable.pm
 automake.dot: automake.y
 	bison --graph automake.y
 	rm automake.tab.c
-	unflatten -f -l 5 -c 6 -o automake1.dot automake.dot
+	unflatten -f -l 16 -c 9 -o automake1.dot automake.dot
 	dot -Tpng automake1.dot > automake.png
 	rm automake1.dot
 
diff --git a/lib/Automake/Parser/ParserTable.pm b/lib/Automake/Parser/ParserTable.pm
index 12aaaa414..d215a0a7d 100644
--- a/lib/Automake/Parser/ParserTable.pm
+++ b/lib/Automake/Parser/ParserTable.pm
@@ -6,23 +6,26 @@ use Tree;
 our @ISA=qw(Exporter);
 our @Export=qw(@table $accept);
 
-our $accept=9;
+our $accept=11;
 
 our @table=(
-		{stmt => 4, input => 2, lhs => 5, stmts => 3, value => 1, optionlist => 6},
-		{'_' => 8, ':' => 7},
-		{end => 9},
-		{value => 1, optionlist => 6, stmt => 10, lhs => 5, reduce => [1,  \&input]},
-		{newline => 11},
-		{'=' => 12},
-		{DATA => 21, PYTHON => 18, LIBRARIES => 15, TEXINFOS => 24, PROGRAMS => 14, MASN => 23, value => 13, JAVA => 19, HEADERS => 22, SCRIPTS => 20, LTLIBRARIES => 16, primaries => 25, LISP => 17},
-		{rhs => 26},
+		{commentlist => 7, optionlist => 8, stmts => 4, comment => 2, value => 1, stmt => 5, lhs => 6, input => 3},
+		{'_' => 10, ':' => 9},
+		{reduce => [1,  \&commentlist]},
+		{end => 11},
+		{lhs => 6, reduce => [1,  \&input], comment => 2, commentlist => 7, value => 1, optionlist => 8, stmt => 12},
+		{newline => 13},
+		{'=' => 14},
+		{reduce => [1,  \&stmt], comment => 15},
+		{PYTHON => 21, HEADERS => 25, JAVA => 22, LTLIBRARIES => 19, PROGRAMS => 17, primaries => 28, MASN => 26, value => 16, SCRIPTS => 23, LISP => 20, DATA => 24, LIBRARIES => 18, TEXINFOS => 27},
+		{rhs => 30, rhsval => 29},
 		{reduce => [2,  \&optionlist]},
 		{},
-		{newline => 27},
+		{newline => 31},
 		{reduce => [2,  \&stmts]},
-		{rhs => 28},
-		{'_' => 29, reduce => [1,  \&primaries]},
+		{rhsval => 29, rhs => 32},
+		{reduce => [2,  \&commentlist]},
+		{reduce => [1,  \&primaries], '_' => 33},
 		{reduce => [1,  \&primaries]},
 		{reduce => [1,  \&primaries]},
 		{reduce => [1,  \&primaries]},
@@ -35,8 +38,11 @@ our @table=(
 		{reduce => [1,  \&primaries]},
 		{reduce => [1,  \&primaries]},
 		{reduce => [2,  \&lhs]},
-		{reduce => [3,  \&stmt]},
+		{reduce => [1,  \&rhs]},
+		{reduce => [3,  \&stmt], rhsval => 34},
 		{reduce => [3,  \&stmts]},
-		{reduce => [3,  \&stmt]},
-		{reduce => [3,  \&optionlist]}
+		{commentlist => 35, reduce => [3,  \&stmt], comment => 2, rhsval => 34},
+		{reduce => [3,  \&optionlist]},
+		{reduce => [2,  \&rhs]},
+		{reduce => [4,  \&stmt], comment => 15}
 );
\ No newline at end of file
diff --git a/lib/Automake/Parser/Tree.pm b/lib/Automake/Parser/Tree.pm
index cad08fcde..92aa7673a 100644
--- a/lib/Automake/Parser/Tree.pm
+++ b/lib/Automake/Parser/Tree.pm
@@ -3,33 +3,44 @@ package Tree;
 use Exporter;
 
 our @ISA = qw(Exporter);
-our @EXPORT = qw(input stmt stmts lhs primaries optionlist traverse printgraph);
+our @EXPORT = qw(input stmt stmts lhs rhs primaries optionlist commentlist traverse printgraph);
 
 # Grammar Rule : (1) input => stmts
 # Create a node having child as stmts.
 sub input($)
 {
-	my ($val) = @_;
-	my %node = (name => input, childs => [$val]);
+	my ( $val ) = @_;
+	my %node = (name => input, childs => [ $val ]);
 	return \%node;
 }
 
 # Grammar Rule : (1) stmt => lhs '=' rhs
-# Create a node for Automake rule. It has lhs and rhs as childs.
-#				 (2) stmt => value ':' rhs
-# Create a node for Make rule. It has value and rhs as childs.
-sub stmt($$$)
+# Create a node for Automake rule having lhs and rhs as its childs.
+#				 (2) stmt => lhs '=' rhs commentlist
+# Create a node for Automake rule having lhs, rhs and comments as its child.
+#				 (3) stmt => value ':' rhs
+# Create a node for Make rule having lhs and rhs as its childs.
+#				 (4) stmt => commentlist
+# Create a node for comments.
+sub stmt($;$$;$)
 {
-	my ($lhs, $sym, $rhs) = @_;
+	my ( $val1, $val2, $val3, $val4 ) = @_;
 	my %node;
-	if($sym -> [0] eq '=')
+	if( !$val2 )
 	{
-		my %rhsnode = (name => rhs, val => $rhs -> [1]);
-		%node = (name => stmt, childs => [$lhs, \%rhsnode],type => automake);
+		%node = (name => stmt, childs => [ $val1 ], type => comment);
+	}
+	elsif( $val2 -> [0] eq '=' )
+	{
+		%node = (name => stmt, childs => [ $val1,$val3 ],type => automake);
+		if( $val4 )
+		{
+			push @{ $node{ childs }}, $val4;
+		}
 	}
 	else
 	{
-		%node = (name => stmt, lhs => $lhs, rhs =>$rhs->[1],type => make);
+		%node = (name => stmt, childs => [ $val1,$val3 ],type => make);
 	} 
 	return \%node;
 }
@@ -41,17 +52,17 @@ sub stmt($$$)
 # the childs array of the stmts(First Argument).
 sub stmts($$;$)
 {
-	my ($val1,$val2,$val3) = @_;
+	my ( $val1, $val2, $val3) = @_;
 	if($val3 == undef)
 	{
-		my %node = (name => stmts, childs => [$val1]);
-		my %nodeval = (name => stmts, childs => [\%node]);
+		my %node = (name => stmts, childs => [ $val1 ]);
+		my %nodeval = (name => stmts, childs => [ \%node ]);
 		return \%nodeval;
 	}
 	else
 	{
-		my %node = (name => stmts,childs => [$val2]);
-		push @{$val1->{childs}},\%node;
+		my %node = (name => stmts,childs => [ $val2 ]);
+		push @{ $val1 -> { childs }}, \%node;
 		return $val1;
 	}
 }
@@ -61,11 +72,49 @@ sub stmts($$;$)
 # option list and primary.
 sub lhs($$)
 {
-	my ($val1, $val2) = @_;
-	my %node = (name => lhs, childs => [$val1, $val2]);
+	my ( $val1, $val2 ) = @_;
+	my %node = (name => lhs, childs => [ $val1, $val2 ]);
 	return \%node;
 }
 
+# Grammar Rule : (1) rhs => rhsval
+# Creates a node having rhsval as its value.
+#				(2) rhs => rhs rhsval
+# Inserts rhsval into the array pointed by value key in rhs.		
+sub rhs($;$)
+{
+	my ( $val1, $val2 ) = @_;
+	if($val2 == undef)
+	{
+		my %node = ( name => rhs, value => [$val1 -> [1]]);
+		return \%node;
+	}
+	else
+	{
+		push @{ $val1 -> { value }}, $val2 -> [1];
+		return $val1;
+	}
+}
+
+# Grammar Rule : (1) commentlist => comment
+# Creates a node having comment as its value.
+#				(2) commentlist => commentlist comment
+# Inserts comment into the array pointed by value key in commentlist.
+sub commentlist($;$)
+{
+	my ( $val1, $val2 ) = @_;
+	if($val2 == undef)
+	{
+		my %node = ( name => commentlist, value => [ $val1 -> [1]]);
+		return \%node;
+	}
+	else
+	{
+		push @{ $val1 -> {value}} , $val2 -> [1];
+		return $val1;
+	}
+}
+
 # Grammar Rule : (1) primaries : PROGRAMS
 #				 (2) primaries : LIBRARIES
 #				 (3) primaries : LTLIBRARIES
@@ -81,7 +130,7 @@ sub lhs($$)
 # Creates a node corresponding to the given primary.
 sub primaries($)
 {
-	my ($val) = @_;
+	my ( $val ) = @_;
 	my %node;
 	if( $val -> [0] eq 'value')
 	{
@@ -89,7 +138,7 @@ sub primaries($)
 	}
 	else
 	{
-		%node = ( name => primaries, val => $val1);
+		%node = ( name => primaries, val => $val);
 	}
 	return \%node;
 }
@@ -100,7 +149,7 @@ sub primaries($)
 # Add the data value to val key in the node pointed by optionlist(First Argument).
 sub optionlist($$;$)
 {
-	my ($val1, $val2, $val3) = @_;
+	my ( $val1, $val2, $val3 ) = @_;
 	if($val3 == undef)
 	{
 		my %node = (name => optionlist, val => [$val1 -> [1]]);
@@ -118,30 +167,34 @@ sub optionlist($$;$)
 sub printgraph($)
 {
 	my $FH;
-	open($FH, '>', 'ast.gv') or die $!;
+	open( $FH, '>', 'ast.gv' ) or die $!;
 	print $FH "graph graphname {\n";
-	my ($ref) = @_;
+	my ( $ref ) = @_;
 	print $FH "0 [label=\"Root\"];";
-	traverse($ref, $FH, 0, 1);
+	traverse( $ref, $FH, 0);
 	print $FH "}\n";
 	close $FH;
 }
-# traverse(Hash, File Handle, Parent Id, Node Id)
+
+#Stores the next id to be alloted to new node.
+my $id=0;
+
+# traverse(Hash, File Handle, Parent Id)
 # Traverses the tree recursively. Prints the information about the current 
-# node to file. Call all its child with Parent Id equal to current Node Id 
-# and Node Id equal to (Parent Id*2+i) where i is the ith Child.
-sub traverse($$$$)
+# node to file. Call all its child with Parent Id equal to current Node Id.
+sub traverse($$$)
 {
-	my ($ref,$FH,$parent,$id)=@_;
+	my ( $ref,$FH,$parent ) = @_;
+	$id++;
+	my $curr_id = $id;
 	my %node = %$ref;
-	#print $level," ",$pos," ",$node{name}," ";
 	print $FH "$parent--$id;\n";
 	my $label = "";
 	@keys = sort grep {!/^childs/} keys %node;
-	foreach $key (@keys)
+	foreach $key ( @keys )
 	{
 		$label .= $key."=>";
-		if(ref($node{$key}) eq 'ARRAY')
+		if(ref( $node{ $key }) eq 'ARRAY')
 		{
 			$label .= join(" ",@{$node{$key}})."\n";
 		}
@@ -150,15 +203,14 @@ sub traverse($$$$)
 			$label .= $node{$key}." ";
 		}
 	}
-	print $FH "$id [label=\"$label\"];";
+	print $FH "$curr_id [label=\"$label\"];";
 	if( $node{childs} )
 	{
 		my $val1 = $node{childs};
-		my $i = 1;
 		foreach $child (@$val1)
 		{
-			traverse($child,$FH,$id,2*$id+$i);
-			$i++;
+			traverse($child,$FH,$curr_id);
 		}
 	}
-}
\ No newline at end of file
+}
+1;
\ No newline at end of file
diff --git a/lib/Automake/Parser/automake.y b/lib/Automake/Parser/automake.y
index a75adce1c..45fdc1598 100644
--- a/lib/Automake/Parser/automake.y
+++ b/lib/Automake/Parser/automake.y
@@ -1,12 +1,24 @@
-%token value rhs PROGRAMS LIBRARIES LTLIBRARIES LISP PYTHON JAVA SCRIPTS DATA HEADERS MASN TEXINFOS newline
+%token value rhsval comment PROGRAMS LIBRARIES LTLIBRARIES LISP PYTHON JAVA SCRIPTS DATA HEADERS MASN TEXINFOS newline
 %%
 
-input : stmts ;
+input : stmts
+;
 stmts : stmt newline
 		| stmts stmt newline
-stmt  : lhs '=' rhs  
-		| value ':' rhs 
+;
+stmt  : lhs '=' rhs
+		| lhs '=' rhs commentlist
+		| value ':' rhs
+		| commentlist
+;		
 lhs   : optionlist primaries
+;
+rhs   : rhsval
+		| rhs rhsval
+;
+commentlist: comment
+			 | commentlist comment
+;
 primaries : PROGRAMS 
 			| LIBRARIES
 			| LTLIBRARIES
@@ -19,6 +31,8 @@ primaries : PROGRAMS
 			| MASN
 			| TEXINFOS
 			| value
+;
 optionlist : value '_'
 			| optionlist value '_'
+;
 %%
\ No newline at end of file
diff --git a/lib/Automake/Parser/input.txt b/lib/Automake/Parser/input.txt
index c65f5d013..fe3ca91ea 100644
--- a/lib/Automake/Parser/input.txt
+++ b/lib/Automake/Parser/input.txt
@@ -1,6 +1,12 @@
-dist_bin_PROGRAMS = server client
-server_SOURCES = server.c db.c
-client_SOURCES = client.c dep.c
+## Process this file with automake to produce Makefile.in
+dist_bin_PROGRAMS = server \
+client
+server_SOURCES = server.c db.c   ## Server Files \
+Database Files  
+#Comment Testing Here \
+Same here
+client_SOURCES = client.c dep.c #Multiline comment \
+Client dependencies
 noinst_LIBRARIES = libfoo.a
 noinst_LTLIBRARIES = foolib.a
 files_JAVA = a.java b.java
diff --git a/lib/Automake/Parser/parser.pl b/lib/Automake/Parser/parser.pl
index 00555d220..4a3386d14 100644
--- a/lib/Automake/Parser/parser.pl
+++ b/lib/Automake/Parser/parser.pl
@@ -13,9 +13,12 @@ open ( $data, "<input.txt" );
 #Stores the list of tokens generated by lexer.
 my @tokens; 
 
+my $multiline = 0;
+my $curr_tokens;
 while ( <$data> )
 {
-	push @tokens, lex($_);
+	( $curr_tokens, $multiline) = lex($_ , $multiline);
+	push @tokens,@$curr_tokens;
 }
 if( $debug )
 {
@@ -25,8 +28,7 @@ if( $debug )
 		print join(" ", @{$token}), "\n";
 	}
 }
-
-push @tokens, ["end"];
+push @tokens, [ "end" ];
 my @stack = (0);
 print "Parser Output\n" if $debug;