------- You are receiving this mail because: -------
You are on the CC list for the bug.
http://bugs.exim.org/show_bug.cgi?id=1120
Summary: Compactifty code
Product: PCRE
Version: 8.12
Platform: x86
OS/Version: All
Status: NEW
Severity: wishlist
Priority: low
Component: Code
AssignedTo: ph10@???
ReportedBy: blueboar_2@???
CC: pcre-dev@???
In compile2() there is such thing:
switch (options & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE))
{
case 0:
case PCRE_BSR_ANYCRLF:
case PCRE_BSR_UNICODE:
break;
default: errorcode = ERR56; goto PCRE_EARLY_ERROR_RETURN;
}
Why so big?
Can't it be simply
if options & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE) =
PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE then ... error
or
options | ANYCRLF | UNICODE = options then ERROR!!!
Why so much switch and cases?
(BTW compiler does not optimize this, i've checked in Debian udeb package - it
checks all possibilities. Even worse, it first checks CRLF, then UNICODE, and
only then that both are unset - most common situation. Why he does it - i don't
know)
PS: I know, it doesn't make PCRE significally faster, but it is more beautiful,
mode compact, and more faster, even several nanoseconds.
--
Configure bugmail:
http://bugs.exim.org/userprefs.cgi?tab=email
