static int jsY_lexx(js_State *J)
{
J->newline = 0;
while (1) {
J->lexline = J->line; /* save location of beginning of token */
while (jsY_iswhite(J->lexchar))
jsY_next(J);
if (jsY_accept(J, '\n')) {
J->newline = 1;
if (isnlthcontext(J->lasttoken))
return ';';
continue;
}
if (jsY_accept(J, '/')) {
if (jsY_accept(J, '/')) {
lexlinecomment(J);
continue;
} else if (jsY_accept(J, '*')) {
if (lexcomment(J))
jsY_error(J, "multi-line comment not terminated");
continue;
} else if (isregexpcontext(J->lasttoken)) {
return lexregexp(J);
} else if (jsY_accept(J, '=')) {
return TK_DIV_ASS;
} else {
return '/';
}
}
if (J->lexchar >= '0' && J->lexchar <= '9') {
return lexnumber(J);
}
switch (J->lexchar) {
case '(': jsY_next(J); return '(';
case ')': jsY_next(J); return ')';
case ',': jsY_next(J); return ',';
case ':': jsY_next(J); return ':';
case ';': jsY_next(J); return ';';
case '?': jsY_next(J); return '?';
case '[': jsY_next(J); return '[';
case ']': jsY_next(J); return ']';
case '{': jsY_next(J); return '{';
case '}': jsY_next(J); return '}';
case '~': jsY_next(J); return '~';
case '\'':
case '"':
return lexstring(J);
case '.':
return lexnumber(J);
case '<':
jsY_next(J);
if (jsY_accept(J, '<')) {
if (jsY_accept(J, '='))
return TK_SHL_ASS;
return TK_SHL;
}
if (jsY_accept(J, '='))
return TK_LE;
return '<';
case '>':
jsY_next(J);
if (jsY_accept(J, '>')) {
if (jsY_accept(J, '>')) {
if (jsY_accept(J, '='))
return TK_USHR_ASS;
return TK_USHR;
}
if (jsY_accept(J, '='))
return TK_SHR_ASS;
return TK_SHR;
}
if (jsY_accept(J, '='))
return TK_GE;
return '>';
case '=':
jsY_next(J);
if (jsY_accept(J, '=')) {
if (jsY_accept(J, '='))
return TK_STRICTEQ;
return TK_EQ;
}
return '=';
case '!':
jsY_next(J);
if (jsY_accept(J, '=')) {
if (jsY_accept(J, '='))
return TK_STRICTNE;
return TK_NE;
}
return '!';
case '+':
jsY_next(J);
if (jsY_accept(J, '+'))
return TK_INC;
if (jsY_accept(J, '='))
return TK_ADD_ASS;
return '+';
case '-':
jsY_next(J);
if (jsY_accept(J, '-'))
return TK_DEC;
if (jsY_accept(J, '='))
return TK_SUB_ASS;
return '-';
case '*':
jsY_next(J);
if (jsY_accept(J, '='))
return TK_MUL_ASS;
return '*';
case '%':
jsY_next(J);
if (jsY_accept(J, '='))
return TK_MOD_ASS;
return '%';
case '&':
jsY_next(J);
if (jsY_accept(J, '&'))
return TK_AND;
if (jsY_accept(J, '='))
return TK_AND_ASS;
return '&';
case '|':
jsY_next(J);
if (jsY_accept(J, '|'))
return TK_OR;
if (jsY_accept(J, '='))
return TK_OR_ASS;
return '|';
case '^':
jsY_next(J);
if (jsY_accept(J, '='))
return TK_XOR_ASS;
return '^';
case 0:
return 0; /* EOF */
}
/* Handle \uXXXX escapes in identifiers */
jsY_unescape(J);
if (jsY_isidentifierstart(J->lexchar)) {
textinit(J);
textpush(J, J->lexchar);
jsY_next(J);
jsY_unescape(J);
while (jsY_isidentifierpart(J->lexchar)) {
textpush(J, J->lexchar);
jsY_next(J);
jsY_unescape(J);
}
textend(J);
return jsY_findkeyword(J, J->lexbuf.text);
}
if (J->lexchar >= 0x20 && J->lexchar <= 0x7E)
jsY_error(J, "unexpected character: '%c'", J->lexchar);
jsY_error(J, "unexpected character: \\u%04X", J->lexchar);
}
}
int
pdf_lex(fz_stream *f, unsigned char *buf, int n, int *sl)
{
int c;
while (1)
{
c = fz_peekbyte(f);
if (c == EOF)
return PDF_TEOF;
else if (iswhite(c))
lexwhite(f);
else if (c == '%')
lexcomment(f);
else if (c == '/')
{
fz_readbyte(f);
lexname(f, buf, n);
*sl = strlen(buf);
return PDF_TNAME;
}
else if (c == '(')
{
fz_readbyte(f);
*sl = lexstring(f, buf, n);
return PDF_TSTRING;
}
else if (c == '<')
{
fz_readbyte(f);
c = fz_peekbyte(f);
if (c == '<')
{
fz_readbyte(f);
return PDF_TODICT;
}
else
{
*sl = lexhexstring(f, buf, n);
return PDF_TSTRING;
}
}
else if (c == '>')
{
fz_readbyte(f);
c = fz_readbyte(f);
if (c == '>')
return PDF_TCDICT;
return PDF_TERROR;
}
else if (c == '[')
{
fz_readbyte(f);
return PDF_TOARRAY;
}
else if (c == ']')
{
fz_readbyte(f);
return PDF_TCARRAY;
}
else if (c == '{')
{
fz_readbyte(f);
return PDF_TOBRACE;
}
else if (c == '}')
{
fz_readbyte(f);
return PDF_TCBRACE;
}
else if (isnumber(c))
{
lexnumber(f, buf, n);
*sl = strlen(buf);
if (strchr(buf, '.'))
return PDF_TREAL;
return PDF_TINT;
}
else if (isregular(c))
{
lexname(f, buf, n);
*sl = strlen(buf);
return tokenfromkeyword(buf);
}
else
return PDF_TERROR;
}
}
fz_error
pdf_lex(int *tok, fz_stream *f, char *buf, int n, int *sl)
{
while (1)
{
int c = fz_readbyte(f);
switch (c)
{
case EOF:
*tok = PDF_TEOF;
return fz_okay;
case ISWHITE:
lexwhite(f);
break;
case '%':
lexcomment(f);
break;
case '/':
lexname(f, buf, n);
*sl = strlen(buf);
*tok = PDF_TNAME;
return fz_okay;
case '(':
*sl = lexstring(f, buf, n);
*tok = PDF_TSTRING;
return fz_okay;
case ')':
*tok = PDF_TERROR;
goto cleanuperror;
case '<':
c = fz_readbyte(f);
if (c == '<')
{
*tok = PDF_TODICT;
}
else
{
fz_unreadbyte(f);
*sl = lexhexstring(f, buf, n);
*tok = PDF_TSTRING;
}
return fz_okay;
case '>':
c = fz_readbyte(f);
if (c == '>')
{
*tok = PDF_TCDICT;
return fz_okay;
}
*tok = PDF_TERROR;
goto cleanuperror;
case '[':
*tok = PDF_TOARRAY;
return fz_okay;
case ']':
*tok = PDF_TCARRAY;
return fz_okay;
case '{':
*tok = PDF_TOBRACE;
return fz_okay;
case '}':
*tok = PDF_TCBRACE;
return fz_okay;
case ISNUMBER:
fz_unreadbyte(f);
*sl = lexnumber(f, buf, n, tok);
return fz_okay;
default: /* isregular: !isdelim && !iswhite && c != EOF */
fz_unreadbyte(f);
lexname(f, buf, n);
*sl = strlen(buf);
*tok = pdf_tokenfromkeyword(buf);
return fz_okay;
}
}
cleanuperror:
*tok = PDF_TERROR;
return fz_throw("lexical error");
}
fz_error
pdf_lex(pdf_token_e *tok, fz_stream *f, char *buf, int n, int *sl)
{
fz_error error;
int c;
while (1)
{
c = fz_peekbyte(f);
if (c == EOF)
{
*tok = PDF_TEOF;
goto cleanupokay;
}
else if (iswhite(c))
lexwhite(f);
else if (c == '%')
lexcomment(f);
else if (c == '/')
{
fz_readbyte(f);
lexname(f, buf, n);
*sl = strlen(buf);
*tok = PDF_TNAME;
goto cleanupokay;
}
else if (c == '(')
{
fz_readbyte(f);
*sl = lexstring(f, buf, n);
*tok = PDF_TSTRING;
goto cleanupokay;
}
else if (c == '<')
{
fz_readbyte(f);
c = fz_peekbyte(f);
if (c == '<')
{
fz_readbyte(f);
*tok = PDF_TODICT;
goto cleanupokay;
}
else
{
*sl = lexhexstring(f, buf, n);
*tok = PDF_TSTRING;
goto cleanupokay;
}
}
else if (c == '>')
{
fz_readbyte(f);
c = fz_readbyte(f);
if (c == '>')
{
*tok = PDF_TCDICT;
goto cleanupokay;
}
*tok = PDF_TERROR;
goto cleanuperror;
}
else if (c == '[')
{
fz_readbyte(f);
*tok = PDF_TOARRAY;
goto cleanupokay;
}
else if (c == ']')
{
fz_readbyte(f);
*tok = PDF_TCARRAY;
goto cleanupokay;
}
else if (c == '{')
{
fz_readbyte(f);
*tok = PDF_TOBRACE;
goto cleanupokay;
}
else if (c == '}')
{
fz_readbyte(f);
*tok = PDF_TCBRACE;
goto cleanupokay;
}
else if (isnumber(c))
{
lexnumber(f, buf, n);
*sl = strlen(buf);
if (strchr(buf, '.'))
{
*tok = PDF_TREAL;
goto cleanupokay;
}
*tok = PDF_TINT;
goto cleanupokay;
}
else if (isregular(c))
{
lexname(f, buf, n);
*sl = strlen(buf);
*tok = pdf_tokenfromkeyword(buf);
goto cleanupokay;
}
else
{
*tok = PDF_TERROR;
goto cleanuperror;
}
}
cleanupokay:
error = fz_readerror(f);
if (error)
{
*tok = PDF_TERROR;
return fz_rethrow(error, "cannot read token");
}
return fz_okay;
cleanuperror:
error = fz_readerror(f);
if (error)
{
*tok = PDF_TERROR;
return fz_rethrow(error, "cannot read token");
}
*tok = PDF_TERROR;
return fz_throw("lexical error");
}
