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");
}