]> git.saurik.com Git - cycript.git/blobdiff - Cycript.l
Got jQuery parsing (empty finally block, function statements, carriage return) and...
[cycript.git] / Cycript.l
index 20a4ede060ed74026fdc04f929a149f4ab44b9b2..d62a14b09f3413553c69293d337db5d43a1482ba 100644 (file)
--- a/Cycript.l
+++ b/Cycript.l
@@ -9,6 +9,15 @@ typedef cy::parser::token tk;
 #define C T yyextra->state_ = CYClear;
 #define R T yyextra->state_ = CYRestricted;
 
+#define E(prefix) L C BEGIN(INITIAL); { \
+    char *value(reinterpret_cast<char *>(apr_palloc(yyextra->pool_, yyleng + sizeof(prefix)))); \
+    memcpy(value, prefix, sizeof(prefix) - 1); \
+    memcpy(value + sizeof(prefix) - 1, yytext, yyleng); \
+    value[yyleng + sizeof(prefix) - 1] = '\0'; \
+    yylval->literal_ = new(yyextra->pool_) CYRegEx(value); \
+    return tk::RegularExpressionLiteral; \
+}
+
 #define N \
     if (yyextra->state_ != CYNewLine) { \
         bool restricted(yyextra->state_ == CYRestricted); \
@@ -63,10 +72,50 @@ int H(char c) {
 Exponent [eE][+-]?[0-9]+
 Escape   \\[\\'"bfnrtv]|\\0|\\x[0-9a-fA-F]{2}|\\u[0-9a-fA-F]{4}|\\\n
 
+IdentifierStart [a-zA-Z$_]
+IdentifierPart  [a-zA-Z$_0-9]
+
+NonTerminator [^\n]
+BackslashSequence \\{NonTerminator}
+RegularExpressionFirstChar [^\n*\\/]|{BackslashSequence}
+RegularExpressionChar [^\n\\/]|{BackslashSequence}
+RegularExpressionFlags {IdentifierPart}*
+RegularExpressionChars {RegularExpressionChar}*
+RegularExpressionBody_ {RegularExpressionChars}
+RegularExpressionBody {RegularExpressionFirstChar}{RegularExpressionBody_}
+
+RegularExpressionEnd_ \/{RegularExpressionFlags}
+RegularExpressionRest_ {RegularExpressionBody_}{RegularExpressionEnd_}
+RegularExpressionStart_ {RegularExpressionBody}{RegularExpressionEnd_}
+
+%x res
+%x rer
+
 %%
 
-\/\/[^\n]* ;
-\/\*(\n|[^\*]|\*[^/])\*\/ if (memchr(yytext, '\n', yyleng) != NULL) N // XXX: supposedly I will be screwed on very very long multi-line comments and need to replace this with a manual lexer. http://websrv.cs.fsu.edu/~engelen/courses/COP5621/Pr2.pdf ; XXX: this rule doesn't work anyway, fucking A :(
+<res>{RegularExpressionStart_} E("/")
+<rer>{RegularExpressionRest_} E("/=")
+
+\/\/[^\n]* L
+
+\/\*(\n|[^\*]|\*[^/])*\*\/ {
+    // XXX: supposedly I will be screwed on very very long multi-line comments and need to replace this with a manual lexer. http://websrv.cs.fsu.edu/~engelen/courses/COP5621/Pr2.pdf
+
+    if (const char *nl = reinterpret_cast<const char *>(memchr(yytext, '\n', yyleng))) {
+        unsigned lines(0);
+        size_t left;
+
+        do {
+            ++lines;
+            left = yyleng - (nl - yytext) - 1;
+            nl = reinterpret_cast<const char *>(memchr(nl + 1, '\n', left));
+        } while (nl != NULL);
+
+        yylloc->end.lines(lines);
+        yylloc->end.columns(left);
+        yylloc->step();
+    } else L
+}
 
 "&"    L C return tk::Ampersand;
 "&&"   L C return tk::AmpersandAmpersand;
@@ -126,84 +175,84 @@ Escape   \\[\\'"bfnrtv]|\\0|\\x[0-9a-fA-F]{2}|\\u[0-9a-fA-F]{4}|\\\n
 "@end"         L C return tk::AtEnd;
 "@selector"    L C return tk::AtSelector;
 
-"false"        L C yylval->false_ = new CYFalse(); return tk::False;
-"null"         L C yylval->null_ = new CYNull(); return tk::Null;
-"true"         L C yylval->true_ = new CYTrue(); return tk::True;
-
-"break"        L R yylval->word_ = new CYWord("break"); return tk::Break;
-"case"         L C yylval->word_ = new CYWord("case"); return tk::Case;
-"catch"        L C yylval->word_ = new CYWord("catch"); return tk::Catch;
-"continue"     L R yylval->word_ = new CYWord("continue"); return tk::Continue;
-"default"      L C yylval->word_ = new CYWord("default"); return tk::Default;
-"delete"       L C yylval->word_ = new CYWord("delete"); return tk::Delete;
-"do"           L C yylval->word_ = new CYWord("do"); return tk::Do;
-"else"         L C yylval->word_ = new CYWord("else"); return tk::Else;
-"finally"      L C yylval->word_ = new CYWord("finally"); return tk::Finally;
-"for"          L C yylval->word_ = new CYWord("for"); return tk::For;
-"function"     L C yylval->word_ = new CYWord("function"); return tk::Function;
-"if"           L C yylval->word_ = new CYWord("if"); return tk::If;
-"in"           L C yylval->word_ = new CYWord("in"); return tk::In;
-"instanceof"   L C yylval->word_ = new CYWord("instanceof"); return tk::InstanceOf;
-"new"          L C yylval->word_ = new CYWord("new"); return tk::New;
-"return"       L R yylval->word_ = new CYWord("return"); return tk::Return;
-"switch"       L C yylval->word_ = new CYWord("switch"); return tk::Switch;
-"this"         L C yylval->this_ = new CYThis(); return tk::This;
-"throw"        L R yylval->word_ = new CYWord("throw"); return tk::Throw;
-"try"          L C yylval->word_ = new CYWord("try"); return tk::Try;
-"typeof"       L C yylval->word_ = new CYWord("typeof"); return tk::TypeOf;
-"var"          L C yylval->word_ = new CYWord("var"); return tk::Var;
-"void"         L C yylval->word_ = new CYWord("void"); return tk::Void;
-"while"        L C yylval->word_ = new CYWord("while"); return tk::While;
-"with"         L C yylval->word_ = new CYWord("with"); return tk::With;
-
-"debugger"     L C yylval->word_ = new CYWord("debugger"); return tk::Debugger;
-
-"const"        L C yylval->word_ = new CYWord("const"); return tk::Const;
-
-"class"        L C yylval->word_ = new CYWord("class"); return tk::Class;
-"enum"         L C yylval->word_ = new CYWord("enum"); return tk::Enum;
-"export"       L C yylval->word_ = new CYWord("export"); return tk::Export;
-"extends"      L C yylval->word_ = new CYWord("extends"); return tk::Extends;
-"import"       L C yylval->word_ = new CYWord("import"); return tk::Import;
-"super"        L C yylval->word_ = new CYWord("super"); return tk::Super;
-
-"implements"   L C yylval->identifier_ = new CYIdentifier("implements"); return tk::Implements;
-"interface"    L C yylval->identifier_ = new CYIdentifier("interface"); return tk::Interface;
-"package"      L C yylval->identifier_ = new CYIdentifier("package"); return tk::Package;
-"private"      L C yylval->identifier_ = new CYIdentifier("private"); return tk::Private;
-"protected"    L C yylval->identifier_ = new CYIdentifier("protected"); return tk::Protected;
-"public"       L C yylval->identifier_ = new CYIdentifier("public"); return tk::Public;
-"static"       L C yylval->identifier_ = new CYIdentifier("static"); return tk::Static;
-
-"abstract"     L C yylval->identifier_ = new CYIdentifier("abstract"); return tk::Abstract;
-"boolean"      L C yylval->identifier_ = new CYIdentifier("boolean"); return tk::Boolean;
-"byte"         L C yylval->identifier_ = new CYIdentifier("byte"); return tk::Byte;
-"char"         L C yylval->identifier_ = new CYIdentifier("char"); return tk::Char;
-"double"       L C yylval->identifier_ = new CYIdentifier("double"); return tk::Double;
-"final"        L C yylval->identifier_ = new CYIdentifier("final"); return tk::Final;
-"float"        L C yylval->identifier_ = new CYIdentifier("float"); return tk::Float;
-"goto"         L C yylval->identifier_ = new CYIdentifier("goto"); return tk::Goto;
-"int"          L C yylval->identifier_ = new CYIdentifier("int"); return tk::Int;
-"long"         L C yylval->identifier_ = new CYIdentifier("long"); return tk::Long;
-"native"       L C yylval->identifier_ = new CYIdentifier("native"); return tk::Native;
-"short"        L C yylval->identifier_ = new CYIdentifier("short"); return tk::Short;
-"synchronized" L C yylval->identifier_ = new CYIdentifier("synchronized"); return tk::Synchronized;
-"throws"       L C yylval->identifier_ = new CYIdentifier("throws"); return tk::Throws;
-"transient"    L C yylval->identifier_ = new CYIdentifier("transient"); return tk::Transient;
-"volatile"     L C yylval->identifier_ = new CYIdentifier("volatile"); return tk::Volatile;
-
-"let"          L C yylval->identifier_ = new CYIdentifier("let"); return tk::Let;
-"yield"        L C yylval->identifier_ = new CYIdentifier("yield"); return tk::Yield;
-
-"each"         L C yylval->identifier_ = new CYIdentifier("each"); return tk::Each;
-
-[a-zA-Z$_][a-zA-Z$_0-9]* yylval->identifier_ = new CYIdentifier(apr_pstrmemdup(yyextra->pool_, yytext, yyleng)); L C return tk::Identifier_;
-
-(\.[0-9]+|(0|[1-9][0-9]*)(\.[0-9]*)?){Exponent}? yylval->number_ = new CYNumber(strtod(yytext, NULL)); L C return tk::NumericLiteral;
-
-0[xX][0-9a-fA-F]+ L C yylval->number_ = new CYNumber(strtoull(yytext + 2, NULL, 16)); return tk::NumericLiteral;
-
-0[bB][0-1]+ L C yylval->number_ = new CYNumber(strtoull(yytext + 2, NULL, 2)); return tk::NumericLiteral;
+"false"        L C yylval->false_ = new(yyextra->pool_) CYFalse(); return tk::False;
+"null"         L C yylval->null_ = new(yyextra->pool_) CYNull(); return tk::Null;
+"true"         L C yylval->true_ = new(yyextra->pool_) CYTrue(); return tk::True;
+
+"break"        L R yylval->word_ = new(yyextra->pool_) CYWord("break"); return tk::Break;
+"case"         L C yylval->word_ = new(yyextra->pool_) CYWord("case"); return tk::Case;
+"catch"        L C yylval->word_ = new(yyextra->pool_) CYWord("catch"); return tk::Catch;
+"continue"     L R yylval->word_ = new(yyextra->pool_) CYWord("continue"); return tk::Continue;
+"default"      L C yylval->word_ = new(yyextra->pool_) CYWord("default"); return tk::Default;
+"delete"       L C yylval->word_ = new(yyextra->pool_) CYWord("delete"); return tk::Delete;
+"do"           L C yylval->word_ = new(yyextra->pool_) CYWord("do"); return tk::Do;
+"else"         L C yylval->word_ = new(yyextra->pool_) CYWord("else"); return tk::Else;
+"finally"      L C yylval->word_ = new(yyextra->pool_) CYWord("finally"); return tk::Finally;
+"for"          L C yylval->word_ = new(yyextra->pool_) CYWord("for"); return tk::For;
+"function"     L C yylval->word_ = new(yyextra->pool_) CYWord("function"); return tk::Function;
+"if"           L C yylval->word_ = new(yyextra->pool_) CYWord("if"); return tk::If;
+"in"           L C yylval->word_ = new(yyextra->pool_) CYWord("in"); return tk::In;
+"instanceof"   L C yylval->word_ = new(yyextra->pool_) CYWord("instanceof"); return tk::InstanceOf;
+"new"          L C yylval->word_ = new(yyextra->pool_) CYWord("new"); return tk::New;
+"return"       L R yylval->word_ = new(yyextra->pool_) CYWord("return"); return tk::Return;
+"switch"       L C yylval->word_ = new(yyextra->pool_) CYWord("switch"); return tk::Switch;
+"this"         L C yylval->this_ = new(yyextra->pool_) CYThis(); return tk::This;
+"throw"        L R yylval->word_ = new(yyextra->pool_) CYWord("throw"); return tk::Throw;
+"try"          L C yylval->word_ = new(yyextra->pool_) CYWord("try"); return tk::Try;
+"typeof"       L C yylval->word_ = new(yyextra->pool_) CYWord("typeof"); return tk::TypeOf;
+"var"          L C yylval->word_ = new(yyextra->pool_) CYWord("var"); return tk::Var;
+"void"         L C yylval->word_ = new(yyextra->pool_) CYWord("void"); return tk::Void;
+"while"        L C yylval->word_ = new(yyextra->pool_) CYWord("while"); return tk::While;
+"with"         L C yylval->word_ = new(yyextra->pool_) CYWord("with"); return tk::With;
+
+"debugger"     L C yylval->word_ = new(yyextra->pool_) CYWord("debugger"); return tk::Debugger;
+
+"const"        L C yylval->word_ = new(yyextra->pool_) CYWord("const"); return tk::Const;
+
+"class"        L C yylval->word_ = new(yyextra->pool_) CYWord("class"); return tk::Class;
+"enum"         L C yylval->word_ = new(yyextra->pool_) CYWord("enum"); return tk::Enum;
+"export"       L C yylval->word_ = new(yyextra->pool_) CYWord("export"); return tk::Export;
+"extends"      L C yylval->word_ = new(yyextra->pool_) CYWord("extends"); return tk::Extends;
+"import"       L C yylval->word_ = new(yyextra->pool_) CYWord("import"); return tk::Import;
+"super"        L C yylval->word_ = new(yyextra->pool_) CYWord("super"); return tk::Super;
+
+"implements"   L C yylval->identifier_ = new(yyextra->pool_) CYIdentifier("implements"); return tk::Implements;
+"interface"    L C yylval->identifier_ = new(yyextra->pool_) CYIdentifier("interface"); return tk::Interface;
+"package"      L C yylval->identifier_ = new(yyextra->pool_) CYIdentifier("package"); return tk::Package;
+"private"      L C yylval->identifier_ = new(yyextra->pool_) CYIdentifier("private"); return tk::Private;
+"protected"    L C yylval->identifier_ = new(yyextra->pool_) CYIdentifier("protected"); return tk::Protected;
+"public"       L C yylval->identifier_ = new(yyextra->pool_) CYIdentifier("public"); return tk::Public;
+"static"       L C yylval->identifier_ = new(yyextra->pool_) CYIdentifier("static"); return tk::Static;
+
+"abstract"     L C yylval->identifier_ = new(yyextra->pool_) CYIdentifier("abstract"); return tk::Abstract;
+"boolean"      L C yylval->identifier_ = new(yyextra->pool_) CYIdentifier("boolean"); return tk::Boolean;
+"byte"         L C yylval->identifier_ = new(yyextra->pool_) CYIdentifier("byte"); return tk::Byte;
+"char"         L C yylval->identifier_ = new(yyextra->pool_) CYIdentifier("char"); return tk::Char;
+"double"       L C yylval->identifier_ = new(yyextra->pool_) CYIdentifier("double"); return tk::Double;
+"final"        L C yylval->identifier_ = new(yyextra->pool_) CYIdentifier("final"); return tk::Final;
+"float"        L C yylval->identifier_ = new(yyextra->pool_) CYIdentifier("float"); return tk::Float;
+"goto"         L C yylval->identifier_ = new(yyextra->pool_) CYIdentifier("goto"); return tk::Goto;
+"int"          L C yylval->identifier_ = new(yyextra->pool_) CYIdentifier("int"); return tk::Int;
+"long"         L C yylval->identifier_ = new(yyextra->pool_) CYIdentifier("long"); return tk::Long;
+"native"       L C yylval->identifier_ = new(yyextra->pool_) CYIdentifier("native"); return tk::Native;
+"short"        L C yylval->identifier_ = new(yyextra->pool_) CYIdentifier("short"); return tk::Short;
+"synchronized" L C yylval->identifier_ = new(yyextra->pool_) CYIdentifier("synchronized"); return tk::Synchronized;
+"throws"       L C yylval->identifier_ = new(yyextra->pool_) CYIdentifier("throws"); return tk::Throws;
+"transient"    L C yylval->identifier_ = new(yyextra->pool_) CYIdentifier("transient"); return tk::Transient;
+"volatile"     L C yylval->identifier_ = new(yyextra->pool_) CYIdentifier("volatile"); return tk::Volatile;
+
+"let"          L C yylval->identifier_ = new(yyextra->pool_) CYIdentifier("let"); return tk::Let;
+"yield"        L C yylval->identifier_ = new(yyextra->pool_) CYIdentifier("yield"); return tk::Yield;
+
+"each"         L C yylval->identifier_ = new(yyextra->pool_) CYIdentifier("each"); return tk::Each;
+
+{IdentifierStart}{IdentifierPart}* yylval->identifier_ = new(yyextra->pool_) CYIdentifier(apr_pstrmemdup(yyextra->pool_, yytext, yyleng)); L C return tk::Identifier_;
+
+(\.[0-9]+|(0|[1-9][0-9]*)(\.[0-9]*)?){Exponent}? yylval->number_ = new(yyextra->pool_) CYNumber(strtod(yytext, NULL)); L C return tk::NumericLiteral;
+
+0[xX][0-9a-fA-F]+ L C yylval->number_ = new(yyextra->pool_) CYNumber(strtoull(yytext + 2, NULL, 16)); return tk::NumericLiteral;
+
+0[bB][0-1]+ L C yylval->number_ = new(yyextra->pool_) CYNumber(strtoull(yytext + 2, NULL, 2)); return tk::NumericLiteral;
 
 \"([^"\\\n]|{Escape})*\"|'([^'\\\n]|{Escape})*' L C {
     char *value(reinterpret_cast<char *>(apr_palloc(yyextra->pool_, yyleng)));
@@ -236,11 +285,11 @@ Escape   \\[\\'"bfnrtv]|\\0|\\x[0-9a-fA-F]{2}|\\u[0-9a-fA-F]{4}|\\\n
     }
 
     *local = '\0';
-    yylval->string_ = new CYString(value, local - value);
+    yylval->string_ = new(yyextra->pool_) CYString(value, local - value);
     return tk::StringLiteral;
 }
 
-\n yylloc->end.lines(); yylloc->step(); N
+\r?\n yylloc->end.lines(); yylloc->step(); N
 
 [ \t] L
 <<EOF>> L yyterminate();
@@ -263,3 +312,18 @@ void CYDriver::ScannerInit() {
 void CYDriver::ScannerDestroy() {
     cylex_destroy(scanner_);
 }
+
+void CYDriver::SetCondition(Condition condition) {
+    struct yyguts_t *yyg(reinterpret_cast<struct yyguts_t *>(scanner_));
+
+    switch (condition) {
+        case RegExStart:
+            BEGIN(res);
+            break;
+        case RegExRest:
+            BEGIN(rer);
+            break;
+        default:
+            _assert(false);
+    }
+}