Strings

1 files changed, 61 insertions, 3 deletions

diff --git a/ast.cpp b/ast.cpp
index 68d4d4b..f1cbcc0 100644
--- a/ast.cpp
+++ b/ast.cpp
@@ -113,12 +113,40 @@ public:
 				nextat++;
 			}
 			return true;
+		} else if(source[index]=='"'){
+			for(nextat=index+1;nextat<size;nextat++){
+				if(source[nextat]=='\\'){
+					nextat++;
+					if(nextat==size){
+						throw ParseError("Unexpected end of file after '\\' in string");
+					}
+				} else if(source[nextat]=='"'){
+					break;
+				}
+			}
+			if(nextat==size){
+				throw ParseError("Unexpected end of file in string");
+			}
+			nextat++;
+			return true;
 		} else {
 			throw ParseError(string("Invalid token starting with '")+source[index]+"'");
 		}
 	}
 };
 
+static char parseHex(char c){
+	if(c>='0'&&c<='9'){
+		return c-'0';
+	} else if(c>='a'&&c<='f'){
+		return c-'a'+10;
+	} else if(c>='A'&&c<='F'){
+		return c-'A'+10;
+	} else {
+		return -1;
+	}
+}
+
 AST& AST::parse(AST::Tokeniser &tokeniser){
 	if(tokeniser.eof()){
 		throw ParseError("No tokens to parse");
@@ -128,15 +156,12 @@ AST& AST::parse(AST::Tokeniser &tokeniser){
 		throw logic_error("Unexpected empty token from tokeniser");
 	}
 	if(isdigit(token[0])||(token.size()>1&&token[0]=='-'&&isdigit(token[1]))){
-		// cerr<<"AST::parse: type = number"<<endl;
 		type=Type::number;
 		numval=strtol(token.data(),NULL,10);
 	} else if(iswordchar(token[0])){
-		// cerr<<"AST::parse: type = name"<<endl;
 		type=Type::name;
 		nameval=token;
 	} else if(token=="("){
-		// cerr<<"AST::parse: type = tuple"<<endl;
 		type=Type::tuple;
 		while(true){
 			if(!tokeniser.advance()){
@@ -165,6 +190,39 @@ AST& AST::parse(AST::Tokeniser &tokeniser){
 			throw ParseError("Body expected after argument of '\\'");
 		}
 		lambdaval.body=AST().parse(tokeniser);
+	} else if(token[0]=='"'){
+		if(token.back()!='"'){
+			throw logic_error("No string terminator at end of string token from tokeniser");
+		}
+		type=Type::string;
+		for(i64 i=1;i<(i64)token.size()-1;i++){
+			if(token[i]=='\\'){
+				switch(token[i+1]){
+					case 't': strval+='\t'; break;
+					case 'n': strval+='\n'; break;
+					case 'r': strval+='\r'; break;
+					case 'x':{
+						if(i+3>=(i64)token.size()-1){
+							throw ParseError("Not enough characters after hexadecimal"
+							                 " escape sequence");
+						}
+						char hex1=parseHex(token[i+2]),
+						     hex2=parseHex(token[i+3]);
+						if(hex1==-1||hex2==-1){
+							throw ParseError("Invalid characters in hexadecimal"
+							                 " escape sequence");
+						}
+						strval+=(hex1<<4)|hex2;
+						i+=2;
+						break;
+					}
+					default: strval+=token[i+1]; break;
+				}
+				i++;
+			} else {
+				strval+=token[i];
+			}
+		}
 	} else if(token==")"){
 		throw ParseError("Unexpected token '"+token+"'");
 	} else {