Allow scanner to backtrack (for UTF-8 whitespace).

[cycript.git] / Scanner.lpp.in

diff --git a/Scanner.lpp.in b/Scanner.lpp.in
index 677d7e91d3dd784e7037ef2321fd44f30c88b04c..96599f4da781a093d3007cfa108447c6f2b92860 100644 (file)
--- a/Scanner.lpp.in
+++ b/Scanner.lpp.in
@@ -64,6 +64,7 @@ typedef cy::parser::token tk;
 
 #define I(type, Type, value, highlight) do { \
     yylval->semantic_.type ## _ = A CY ## Type; \
+    yylval->semantic_.type ## _->location_ = *yylloc; \
     F(value, highlight); \
 } while (false)
 
@@ -83,13 +84,13 @@ typedef cy::parser::token tk;
             nl = reinterpret_cast<const char *>(memchr(nl + 1, '\n', left)); \
         } while (nl != NULL); \
         yylloc->step(); \
-        yylloc->end.lines(lines); \
-        yylloc->end.columns(left); \
+        yylloc->end.Lines(lines); \
+        yylloc->end.Columns(left); \
         more \
     } else L \
 }
 
-#define R yylloc->end.columns(yyleng);
+#define R yylloc->end.Columns(yyleng);
 #define L yylloc->step(); R
 
 #define H(value, highlight) do { \
@@ -216,6 +217,7 @@ U2 [\xc2-\xdf]
 U3 [\xe0-\xef]
 U4 [\xf0-\xf4]
 UN [\xc0-\xc1\xf5-\xff]
+UE {U1}|{U2}|{U3}|{U4}|{UN}
 
 HexDigit [0-9a-fA-F]
 LineTerminatorSequence \r?\n|\r|\xe2\x80[\xa8\xa9]
@@ -237,12 +239,12 @@ IdentifierMore [$_]
 
 UnicodeStart {IdentifierMore}|{UnicodeIDStart}
 UnicodePart {IdentifierMore}|\xe2\x80[\x8c\x8d]|{UnicodeIDContinue}
-UnicodeFail {U2}|{U3}|{U3}{U0}|{U4}|{U4}{U0}|{U4}{U0}{U0}|{UN}|{U0}
-UnicodeScrap {UnicodePart}*{UnicodeFail}?
+UnicodeScrap {U2}|{U3}{U0}{0,1}|{U4}{U0}{0,2}|{UN}|{U0}
+UnicodeError ({U2}|{U3}{U0}{0,1}|{U4}{U0}{0,2}){UE}|{UN}|{U0}
 
 IdentifierStart {UnicodeStart}|{UnicodeEscape}
 IdentifierPart {UnicodePart}|{UnicodeEscape}
-IdentifierFail {UnicodeFail}|\\(u({HexDigit}{0,3}|\{{HexDigit}*))?
+IdentifierFail {UnicodeError}|\\(u({HexDigit}{0,3}|\{{HexDigit}*))?
 IdentifierScrap {IdentifierPart}*{IdentifierFail}?
 
 RegularExpressionBackslashSequence \\{NoneTerminatorCharacter}
@@ -277,20 +279,22 @@ XMLName {XMLNameStart}{XMLNamePart}*
     /* RegEx {{{ */
 <RegularExpression>{
     \/{UnicodePart}* R CYLexBufferUnits(yytext, yyleng); CYLexBufferEnd(literal, RegEx, tk::RegularExpressionLiteral, hi::Constant);
-    \/{UnicodePart}*{UnicodeFail} R E("invalid flags")
+    \/{UnicodePart}*{UnicodeError} R E("invalid character");
 
     {RegExCharacter}+ R CYLexBufferUnits(yytext, yyleng);
-    {RegExCharacter}*{UnicodeFail} R E("invalid character");
 
     {RegularExpressionBackslashSequence} R CYLexBufferUnits(yytext, yyleng);
-    \\{UnicodeFail}? R E("invalid escape")
+    \\ R E("invalid escape")
+
+    (\\|{RegExCharacter}+)?{LineTerminatorSequence} R E("invalid newline");
+    (\\|{RegExCharacter}+)?{UnicodeScrap} R E("invalid character");
 
     "["{RegularExpressionClassChars}"]" R CYLexBufferUnits(yytext, yyleng);
     "["{RegularExpressionClassChars}\\? R E("invalid class");
-    "["{RegularExpressionClassChars}\\?{UnicodeFail} R E("invalid character");
+
     "["{RegularExpressionClassChars}\\?{LineTerminatorSequence} R E("invalid newline");
+    "["{RegularExpressionClassChars}\\?{UnicodeScrap} R E("invalid character");
 
-    (\\|{RegExCharacter}+)?{LineTerminatorSequence} R E("invalid newline");
     <<EOF>> R E("unterminated regex")
 }
     /* }}} */
@@ -302,9 +306,12 @@ XMLName {XMLNameStart}{XMLNamePart}*
 
 <MultiLine>{
     \**\*\/ R yy_pop_state(yyscanner); M N
-    \**{LineTerminatorSequence} yylloc->end.lines(); yyextra->last_ = true;
+    \**{LineTerminatorSequence} yylloc->end.Lines(); yyextra->last_ = true;
     \**{CommentCharacter}|\/ R
-    \**({UnicodeFail}|\*) R E("invalid comment");
+
+    \**{UnicodeScrap} R E("invalid character");
+    \**\* R E("invalid comment");
+
     <<EOF>> R E("invalid comment")
 }
     /* }}} */
@@ -424,7 +431,7 @@ XMLName {XMLNameStart}{XMLNamePart}*
 "@YES"            L F(tk::At_YES_, hi::Constant);
 @end
 
-@({UnicodeStart}{UnicodeScrap}|{UnicodeFail}) L E("invalid keyword")
+@({UnicodeStart}{UnicodePart}*{UnicodeError}?|{UnicodeError}) L E("invalid keyword")
     /* }}} */
     /* Highlight {{{ */
 "undefined"       L F(tk::_undefined_, hi::Operator);
@@ -460,6 +467,7 @@ XMLName {XMLNameStart}{XMLNamePart}*
 "enum"            L /*FFF*/ F(tk::_enum_, hi::Meta);
 "export"          L /*FFK*/ F(tk::_export_, hi::Meta);
 "extends"         L /*FFK*/ F(tk::_extends_, hi::Meta);
+"eval"            L /*III*/ F(tk::_eval_, hi::Special);
 "false"           L /*LLL*/ F(tk::_false_, hi::Constant);
 "final"           L /*FII*/ F(tk::_final_, hi::Meta);
 "finally"         L /*KKK*/ F(tk::_finally_, hi::Control);
@@ -473,6 +481,7 @@ XMLName {XMLNameStart}{XMLNamePart}*
 "implements"      L /*FSS*/ F(tk::_implements_, hi::Meta);
 "import"          L /*FFK*/ F(tk::_import_, hi::Meta);
 "in"              L /*KKK*/ F(yyextra->in_.top() ? tk::_in__ : tk::_in_, hi::Operator);
+"Infinity"        L /*III*/ F(tk::_Infinity_, hi::Constant);
 "instanceof"      L /*KKK*/ F(tk::_instanceof_, hi::Operator);
 "int"             L /*FII*/ F(tk::_int_, hi::Type);
 "interface"       L /*FSS*/ F(tk::_interface_, hi::Meta);
@@ -566,16 +575,16 @@ XMLName {XMLNameStart}{XMLNamePart}*
 <LegacySingleString,StrictSingleString>{
     \' R CYLexBufferEnd(string, String, tk::StringLiteral, hi::Constant);
     {SingleCharacter}+ R CYLexBufferUnits(yytext, yyleng);
-    {SingleCharacter}*{UnicodeFail} R E("invalid character");
-    {LineTerminatorSequence} R E("invalid newline");
+    {SingleCharacter}*{LineTerminatorSequence} R E("invalid newline");
+    {SingleCharacter}*{UnicodeScrap} R E("invalid character");
 }
 
 \" L CYLexBufferStart(LegacyDoubleString);
 <LegacyDoubleString,StrictDoubleString>{
     \" R CYLexBufferEnd(string, String, tk::StringLiteral, hi::Constant);
     {DoubleCharacter}+ R CYLexBufferUnits(yytext, yyleng);
-    {DoubleCharacter}*{UnicodeFail} R E("invalid character");
-    {LineTerminatorSequence} R E("invalid newline");
+    {DoubleCharacter}*{LineTerminatorSequence} R E("invalid newline");
+    {DoubleCharacter}*{UnicodeScrap} R E("invalid character");
 }
     /* }}} */
     /* Template {{{ */
@@ -589,8 +598,8 @@ XMLName {XMLNameStart}{XMLNamePart}*
     "$" R CYLexBufferUnit('$');
 
     {PlateCharacter}+ R CYLexBufferUnits(yytext, yyleng);
-    {PlateCharacter}*{UnicodeFail} R E("invalid character");
-    {LineTerminatorSequence} R E("invalid newline");
+    {PlateCharacter}*{LineTerminatorSequence} R E("invalid newline");
+    {PlateCharacter}*{UnicodeScrap} R E("invalid character");
 }
     /* }}} */
     /* Escapes {{{ */
@@ -624,21 +633,21 @@ XMLName {XMLNameStart}{XMLNamePart}*
         CYLexBufferPoint(point);
     }
 
-    \\{LineTerminatorSequence} yylloc->end.lines();
-    \\(.|{NotLineTerminator}) R CYLexBufferUnits(yytext + 1, yyleng - 1);
+    \\{LineTerminatorSequence} yylloc->end.Lines();
+    \\{NoneTerminatorCharacter} R CYLexBufferUnits(yytext + 1, yyleng - 1);
+    \\{UnicodeScrap} R E("invalid character");
 
-    \\(x{HexDigit}{0,1}|u({HexDigit}{0,3}|\{{HexDigit}*)|{UnicodeFail})? R E("invalid escape");
+    \\(x{HexDigit}{0,1}|u({HexDigit}{0,3}|\{{HexDigit}*))? R E("invalid escape");
     <<EOF>> R E("invalid string");
 }
     /* }}} */
 
-{LineTerminatorSequence} yylloc->step(); yylloc->end.lines(); yyextra->last_ = true; N
+{LineTerminatorSequence} yylloc->step(); yylloc->end.Lines(); yyextra->last_ = true; N
 {WhiteSpace} L
+{U1}|{UnicodeScrap} L E("invalid character");
 
 <<EOF>> if (yyextra->auto_) { yyextra->auto_ = false; F(tk::AutoComplete, hi::Nothing); } L yyterminate();
 
-. L E("invalid character")
-
 %%
 
 #undef yyextra