return -1;
}
+static void U(char *&local, unsigned point) {
+ if (false) {
+ } else if (point < 0x000080) {
+ *local++ = point;
+ } else if (point < 0x000800) {
+ *local++ = 0xc0 | point >> 0x06 & 0x1f;
+ goto one;
+ } else if (point < 0x010000) {
+ *local++ = 0xe0 | point >> 0x0c & 0x0f;
+ goto two;
+ } else if (point < 0x110000) {
+ *local++ = 0xf0 | point >> 0x12 & 0x07;
+ *local++ = 0x80 | point >> 0x0c & 0x3f;
+ two:
+ *local++ = 0x80 | point >> 0x06 & 0x3f;
+ one:
+ *local++ = 0x80 | point >> 0x00 & 0x3f;
+ } else _assert(false);
+}
+
+static void U(char *&local, const char *text, yy_size_t &i) {
+ unsigned point;
+
+ char next(text[++i]);
+ if (next != '{') {
+ point = H(text[i + 0]) << 12 | H(text[i + 1]) << 8 | H(text[i + 2]) << 4 | H(text[i + 3]);
+ i += 3;
+ } else {
+ point = 0;
+ for (;;) {
+ next = text[++i];
+ if (next == '}')
+ break;
+ point = (point << 4) | H(next);
+ }
+ }
+
+ U(local, point);
+}
+
#define YY_INPUT(data, value, size) { \
if (yyextra->data_.eof()) \
value = YY_NULL; \
%option full
%option ecs
-%option align
-Escape \\[\\'"bfnrtv]|\\[0-7]|\\[4-7][0-7]|\\[0-3][0-7][0-7]?|\\x[0-9a-fA-F]{2}|\\u[0-9a-fA-F]{4}|\\\n
+U1 [\x00-\x7f]
+U0 [\x80-\xbf]
+U2 [\xc2-\xdf]
+U3 [\xe0-\xef]
+U4 [\xf0-\xf4]
-IdentifierStart [a-zA-Z$_]
-IdentifierPart [a-zA-Z$_0-9]
+HexDigit [0-9a-fA-F]
+LineTerminatorSequence \r?\n|\r|\xe2\x80[\xa8\xa9]
+WhiteSpace [\x09\x0b\x0c\x20]|\xc2\xa0|\xef\xbb\xbf
+UnicodeEscape \\u({HexDigit}{4}|\{{HexDigit}+\})
+
+OctalEscape \\[1-7]|\\[4-7][0-7]|\\[0-3][0-7][0-7]?
+StringEscape \\['"\\bfnrtv]|\\0|{OctalEscape}|\\x{HexDigit}{2}|{UnicodeEscape}
+StringExtra {StringEscape}|\\{LineTerminatorSequence}
+SingleString ([^'\\\n]|{StringExtra})*
+DoubleString ([^"\\\n]|{StringExtra})*
+StringPrefix '{SingleString}|\"{DoubleString}
+
+@include UnicodeIDStart.l
+@include UnicodeIDContinue.l
+
+IdentifierMore [$_]
+
+UnicodeStart {IdentifierMore}|{UnicodeIDStart}
+UnicodePart {IdentifierMore}|\xe2\x80[\x8c\x8d]|{UnicodeIDContinue}
+UnicodeFail {U2}|{U3}|{U3}{U0}|{U4}|{U4}{U0}|{U4}{U0}{U0}
+UnicodeScrap {UnicodePart}*{UnicodeFail}?
+
+IdentifierStart {UnicodeStart}|{UnicodeEscape}
+IdentifierPart {UnicodePart}|{UnicodeEscape}
+IdentifierFail {UnicodeFail}|\\(u({HexDigit}{0,3}|\{{HexDigit}*))?
+IdentifierScrap {IdentifierPart}*{IdentifierFail}?
NonTerminator [^\n]
BackslashSequence \\{NonTerminator}
RegularExpressionFirstChar [^\n*\\/]|{BackslashSequence}
RegularExpressionChar [^\n\\/]|{BackslashSequence}
-RegularExpressionFlags {IdentifierPart}*
+RegularExpressionFlags {UnicodePart}*
RegularExpressionChars {RegularExpressionChar}*
RegularExpressionBody {RegularExpressionFirstChar}{RegularExpressionChars}
%%
<RegExp>\/{RegularExpressionBody}\/{RegularExpressionFlags} L C I(literal, RegEx(Y), tk::RegularExpressionLiteral, hi::Constant);
+<RegExp>\/{RegularExpressionBody}\/{RegularExpressionFlags}{UnicodeFail} L E("invalid flags")
<RegExp>\/{RegularExpressionBody}?\\? L E("unterminated regex")
#![^\n]* L M
"=" L C F(tk::Equal, hi::Operator);
"==" L C F(tk::EqualEqual, hi::Operator);
"===" L C F(tk::EqualEqualEqual, hi::Operator);
-"=>" L C F(tk::EqualRight, hi::Operator);
+"=>" L C F(yylval->newline_ ? tk::EqualRight_ : tk::EqualRight, hi::Operator);
"!" L C F(tk::Exclamation, hi::Operator);
"!=" L C F(tk::ExclamationEqual, hi::Operator);
"!==" L C F(tk::ExclamationEqualEqual, hi::Operator);
"@import" L C F(tk::AtImport, hi::Special);
"@selector" L C F(tk::AtSelector, hi::Meta);
+"@null" L C F(tk::AtNull, hi::Constant);
+"@YES" L C F(tk::AtYes, hi::Constant);
+"@NO" L C F(tk::AtNo, hi::Constant);
+"@true" L C F(tk::AtTrue, hi::Constant);
+"@false" L C F(tk::AtFalse, hi::Constant);
+
"NULL" L C I(identifier, Identifier("NULL"), tk::Identifier_, hi::Constant);
"nil" L C I(identifier, Identifier("nil"), tk::Identifier_, hi::Constant);
"YES" L C I(identifier, Identifier("YES"), tk::Yes, hi::Constant);
"volatile" L C I(identifier, Identifier("volatile"), tk::Volatile, hi::Meta);
"let" L C I(identifier, Identifier("let"), tk::Let, hi::Meta);
-"yield" L C I(identifier, Identifier("yield"), tk::Yield, hi::Control);
+"yield" L R I(identifier, Identifier("yield"), tk::Yield, hi::Control);
"each" L C I(identifier, Identifier("each"), tk::Each, hi::Control);
"of" L C I(identifier, Identifier("of"), tk::Of, hi::Operator);
"xml" L C I(identifier, Identifier("xml"), tk::XML, hi::Meta);
@end
-{IdentifierStart}{IdentifierPart}* L C I(identifier, Identifier(Y), tk::Identifier_, hi::Identifier);
+{UnicodeStart}{UnicodePart}* L C I(identifier, Identifier(Y), tk::Identifier_, hi::Identifier);
+
+{IdentifierStart}{IdentifierPart}* L C {
+ char *value(A char[yyleng + 1]);
+ char *local(value);
+
+ for (yy_size_t i(0), e(yyleng); i != e; ++i) {
+ char next(yytext[i]);
+ if (next != '\\')
+ *local++ = next;
+ else
+ U(local, yytext, ++i);
+ }
+
+ *local = '\0';
+ I(identifier, Identifier(value), tk::Identifier_, hi::Identifier);
+}
+
+({IdentifierStart}{IdentifierPart}*)?{IdentifierFail} L E("invalid identifier")
0[0-7]+ L C I(number, Number(strtoull(yytext + 1, NULL, 8)), tk::NumericLiteral, hi::Constant);
0[0-9]+ L C I(number, Number(strtoull(yytext + 1, NULL, 10)), tk::NumericLiteral, hi::Constant);
0[bB][0-1]+ L C I(number, Number(strtoull(yytext + 2, NULL, 2)), tk::NumericLiteral, hi::Constant);
(\.[0-9]+|(0|[1-9][0-9]*)(\.[0-9]*)?)([eE][+-]?[0-9]+)? L C I(number, Number(strtod(yytext, NULL)), tk::NumericLiteral, hi::Constant);
-(\.[0-9]+|(0|[1-9][0-9]*)(\.[0-9]*)?)[eE][+-]?{IdentifierPart}* L E("invalid exponent")
-(\.?[0-9]|(0|[1-9][0-9]*)\.){IdentifierPart}* L E("invalid number")
+(\.[0-9]+|(0|[1-9][0-9]*)(\.[0-9]*)?)[eE][+-]?{IdentifierScrap} L E("invalid exponent")
+(\.?[0-9]|(0|[1-9][0-9]*)\.){IdentifierScrap} L E("invalid number")
-\"([^"\\\n]|{Escape})*\"|'([^'\\\n]|{Escape})*' L C {
+'{SingleString}'|\"{DoubleString}\" L C {
char *value(A char[yyleng]);
char *local(value);
char next(yytext[i]);
if (yytext[i] == '\\')
- switch (next = yytext[++i]) {
- case '\n': continue;
+ // XXX: support more line continuation characters
+ if (false) line: {
+ yylloc->end.lines(1);
+ yylloc->end.columns(yyleng - i);
+ } else switch (next = yytext[++i]) {
+ case '\n': goto line;
+
case '\\': next = '\\'; break;
case '\'': next = '\''; break;
case '"': next = '"'; break;
break;
case 'x':
- next = H(yytext[i + 1]) << 4 | H(yytext[i + 2]);
+ U(local, H(yytext[i + 1]) << 4 | H(yytext[i + 2]));
i += 2;
- break;
+ continue;
+
+ case 'u':
+ U(local, yytext, i);
+ continue;
}
*local++ = next;
I(string, String(value, local - value), tk::StringLiteral, hi::Constant);
}
-(\"([^"\\\n]|{Escape})*|'([^'\\\n]|{Escape})*)(\\(x.{0,2}|u.{0,4})?)? L E("invalid escape")
+{StringPrefix}\\(x.{0,2}|u([^{].{0,3}|\{[^}]*)?|{UnicodeFail})? L E("invalid escape")
+{StringPrefix} L E("invalid string")
-\r?\n|\r|\xe2\x80[\xa8\xa9] yylloc->step(); yylloc->end.lines(); N
+{LineTerminatorSequence} yylloc->step(); yylloc->end.lines(); N
-[ \t] L
+{WhiteSpace} L
<<EOF>> if (yyextra->auto_) { yyextra->auto_ = false; F(tk::AutoComplete, hi::Nothing); } L yyterminate();
-@{IdentifierPart}+|\xe2.|. L E("unknown token")
+@({UnicodeStart}{UnicodeScrap}|{UnicodeFail}) L E("invalid keyword")
+
+. L E("invalid character")
%%
projects / cycript.git / blobdiff