Revision: 1425
http://vcs.pcre.org/viewvc?view=rev&revision=1425
Author: ph10
Date: 2013-12-31 17:44:40 +0000 (Tue, 31 Dec 2013)
Log Message:
-----------
Get rid of confusing RAWUCHAR set of macros.
Modified Paths:
--------------
code/trunk/ChangeLog
code/trunk/pcre_dfa_exec.c
code/trunk/pcre_exec.c
code/trunk/pcre_internal.h
code/trunk/pcre_string_utils.c
Modified: code/trunk/ChangeLog
===================================================================
--- code/trunk/ChangeLog 2013-12-31 11:22:31 UTC (rev 1424)
+++ code/trunk/ChangeLog 2013-12-31 17:44:40 UTC (rev 1425)
@@ -30,6 +30,11 @@
above a certain treshold (e.g: 256). The only limitation is that the value
must be bigger than the treshold as well. This function is useful, when
the characters above the treshold are handled in the same way.
+
+7. The macros RAWUCHAR and RAWUCHARTEST were identical (and the latter was not
+ testing anything, contrary to its name and comment) and were not really
+ fulfilling any useful function, so I have replaced their use by plain code.
+ Similarly for RAWUCHARINC and RAWUCHARINCTEST.
Version 8.34 15-December-2013
Modified: code/trunk/pcre_dfa_exec.c
===================================================================
--- code/trunk/pcre_dfa_exec.c 2013-12-31 11:22:31 UTC (rev 1424)
+++ code/trunk/pcre_dfa_exec.c 2013-12-31 17:44:40 UTC (rev 1425)
@@ -1473,7 +1473,7 @@
goto ANYNL01;
case CHAR_CR:
- if (ptr + 1 < end_subject && RAWUCHARTEST(ptr + 1) == CHAR_LF) ncount = 1;
+ if (ptr + 1 < end_subject && ptr[1] == CHAR_LF) ncount = 1;
/* Fall through */
ANYNL01:
@@ -1742,7 +1742,7 @@
goto ANYNL02;
case CHAR_CR:
- if (ptr + 1 < end_subject && RAWUCHARTEST(ptr + 1) == CHAR_LF) ncount = 1;
+ if (ptr + 1 < end_subject && ptr[1] == CHAR_LF) ncount = 1;
/* Fall through */
ANYNL02:
@@ -2012,7 +2012,7 @@
goto ANYNL03;
case CHAR_CR:
- if (ptr + 1 < end_subject && RAWUCHARTEST(ptr + 1) == CHAR_LF) ncount = 1;
+ if (ptr + 1 < end_subject && ptr[1] == CHAR_LF) ncount = 1;
/* Fall through */
ANYNL03:
@@ -2210,7 +2210,7 @@
if ((md->moptions & PCRE_PARTIAL_HARD) != 0)
reset_could_continue = TRUE;
}
- else if (RAWUCHARTEST(ptr + 1) == CHAR_LF)
+ else if (ptr[1] == CHAR_LF)
{
ADD_NEW_DATA(-(state_offset + 1), 0, 1);
}
@@ -3474,12 +3474,12 @@
{
pcre_uchar csc;
while (current_subject < end_subject &&
- (csc = RAWUCHARTEST(current_subject)) != first_char && csc != first_char2)
+ (csc = *current_subject) != first_char && csc != first_char2)
current_subject++;
}
else
while (current_subject < end_subject &&
- RAWUCHARTEST(current_subject) != first_char)
+ *current_subject != first_char)
current_subject++;
}
@@ -3509,10 +3509,9 @@
ANYCRLF, and we are now at a LF, advance the match position by one
more character. */
- if (RAWUCHARTEST(current_subject - 1) == CHAR_CR &&
+ if (current_subject[-1] == CHAR_CR &&
(md->nltype == NLTYPE_ANY || md->nltype == NLTYPE_ANYCRLF) &&
- current_subject < end_subject &&
- RAWUCHARTEST(current_subject) == CHAR_NL)
+ current_subject < end_subject && *current_subject == CHAR_NL)
current_subject++;
}
}
@@ -3523,7 +3522,7 @@
{
while (current_subject < end_subject)
{
- register pcre_uint32 c = RAWUCHARTEST(current_subject);
+ register pcre_uint32 c = *current_subject;
#ifndef COMPILE_PCRE8
if (c > 255) c = 255;
#endif
@@ -3589,7 +3588,7 @@
{
while (p < end_subject)
{
- register pcre_uint32 pp = RAWUCHARINCTEST(p);
+ register pcre_uint32 pp = *p++;
if (pp == req_char || pp == req_char2) { p--; break; }
}
}
@@ -3597,7 +3596,7 @@
{
while (p < end_subject)
{
- if (RAWUCHARINCTEST(p) == req_char) { p--; break; }
+ if (*p++ == req_char) { p--; break; }
}
}
@@ -3665,9 +3664,9 @@
not contain any explicit matches for \r or \n, and the newline option is CRLF
or ANY or ANYCRLF, advance the match position by one more character. */
- if (RAWUCHARTEST(current_subject - 1) == CHAR_CR &&
+ if (current_subject[-1] == CHAR_CR &&
current_subject < end_subject &&
- RAWUCHARTEST(current_subject) == CHAR_NL &&
+ *current_subject == CHAR_NL &&
(re->flags & PCRE_HASCRORLF) == 0 &&
(md->nltype == NLTYPE_ANY ||
md->nltype == NLTYPE_ANYCRLF ||
Modified: code/trunk/pcre_exec.c
===================================================================
--- code/trunk/pcre_exec.c 2013-12-31 11:22:31 UTC (rev 1424)
+++ code/trunk/pcre_exec.c 2013-12-31 17:44:40 UTC (rev 1425)
@@ -134,7 +134,7 @@
BOOL utf = md->utf;
if (is_subject && length > md->end_subject - p) length = md->end_subject - p;
while (length-- > 0)
- if (isprint(c = RAWUCHARINCTEST(p))) printf("%c", (char)c); else printf("\\x{%02x}", c);
+ if (isprint(c = *p++)) printf("%c", (char)c); else printf("\\x{%02x}", c);
}
#endif
@@ -237,8 +237,8 @@
{
pcre_uint32 cc, cp;
if (eptr >= md->end_subject) return -2; /* Partial match */
- cc = RAWUCHARTEST(eptr);
- cp = RAWUCHARTEST(p);
+ cc = *eptr;
+ cp = *p;
if (TABLE_GET(cp, md->lcc, cp) != TABLE_GET(cc, md->lcc, cc)) return -1;
p++;
eptr++;
@@ -254,7 +254,7 @@
while (length-- > 0)
{
if (eptr >= md->end_subject) return -2; /* Partial match */
- if (RAWUCHARINCTEST(p) != RAWUCHARINCTEST(eptr)) return -1;
+ if (*p++ != *eptr++) return -1;
}
}
@@ -2103,7 +2103,7 @@
eptr + 1 >= md->end_subject &&
NLBLOCK->nltype == NLTYPE_FIXED &&
NLBLOCK->nllen == 2 &&
- RAWUCHARTEST(eptr) == NLBLOCK->nl[0])
+ *eptr == NLBLOCK->nl[0])
{
md->hitend = TRUE;
if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
@@ -2147,7 +2147,7 @@
eptr + 1 >= md->end_subject &&
NLBLOCK->nltype == NLTYPE_FIXED &&
NLBLOCK->nllen == 2 &&
- RAWUCHARTEST(eptr) == NLBLOCK->nl[0])
+ *eptr == NLBLOCK->nl[0])
{
md->hitend = TRUE;
if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
@@ -2290,7 +2290,7 @@
eptr + 1 >= md->end_subject &&
NLBLOCK->nltype == NLTYPE_FIXED &&
NLBLOCK->nllen == 2 &&
- RAWUCHARTEST(eptr) == NLBLOCK->nl[0])
+ *eptr == NLBLOCK->nl[0])
{
md->hitend = TRUE;
if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
@@ -2444,7 +2444,7 @@
{
SCHECK_PARTIAL();
}
- else if (RAWUCHARTEST(eptr) == CHAR_LF) eptr++;
+ else if (*eptr == CHAR_LF) eptr++;
break;
case CHAR_LF:
@@ -3218,7 +3218,7 @@
CHECK_PARTIAL(); /* Not SCHECK_PARTIAL() */
RRETURN(MATCH_NOMATCH);
}
- while (length-- > 0) if (*ecode++ != RAWUCHARINC(eptr)) RRETURN(MATCH_NOMATCH);
+ while (length-- > 0) if (*ecode++ != *eptr++) RRETURN(MATCH_NOMATCH);
}
else
#endif
@@ -3258,7 +3258,7 @@
if (fc < 128)
{
- pcre_uint32 cc = RAWUCHAR(eptr);
+ pcre_uint32 cc = *eptr;
if (md->lcc[fc] != TABLE_GET(cc, md->lcc, cc)) RRETURN(MATCH_NOMATCH);
ecode++;
eptr++;
@@ -3527,7 +3527,7 @@
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
- cc = RAWUCHARTEST(eptr);
+ cc = *eptr;
if (fc != cc && foc != cc) RRETURN(MATCH_NOMATCH);
eptr++;
}
@@ -3545,7 +3545,7 @@
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
- cc = RAWUCHARTEST(eptr);
+ cc = *eptr;
if (fc != cc && foc != cc) RRETURN(MATCH_NOMATCH);
eptr++;
}
@@ -3562,7 +3562,7 @@
SCHECK_PARTIAL();
break;
}
- cc = RAWUCHARTEST(eptr);
+ cc = *eptr;
if (fc != cc && foc != cc) break;
eptr++;
}
@@ -3589,7 +3589,7 @@
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
- if (fc != RAWUCHARINCTEST(eptr)) RRETURN(MATCH_NOMATCH);
+ if (fc != *eptr++) RRETURN(MATCH_NOMATCH);
}
if (min == max) continue;
@@ -3606,7 +3606,7 @@
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
- if (fc != RAWUCHARINCTEST(eptr)) RRETURN(MATCH_NOMATCH);
+ if (fc != *eptr++) RRETURN(MATCH_NOMATCH);
}
/* Control never gets here */
}
@@ -3620,7 +3620,7 @@
SCHECK_PARTIAL();
break;
}
- if (fc != RAWUCHARTEST(eptr)) break;
+ if (fc != *eptr) break;
eptr++;
}
if (possessive) continue; /* No backtracking */
@@ -4375,7 +4375,7 @@
eptr + 1 >= md->end_subject &&
NLBLOCK->nltype == NLTYPE_FIXED &&
NLBLOCK->nllen == 2 &&
- RAWUCHAR(eptr) == NLBLOCK->nl[0])
+ *eptr == NLBLOCK->nl[0])
{
md->hitend = TRUE;
if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
@@ -4417,7 +4417,7 @@
default: RRETURN(MATCH_NOMATCH);
case CHAR_CR:
- if (eptr < md->end_subject && RAWUCHAR(eptr) == CHAR_LF) eptr++;
+ if (eptr < md->end_subject && *eptr == CHAR_LF) eptr++;
break;
case CHAR_LF:
@@ -4527,7 +4527,7 @@
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
- cc = RAWUCHAR(eptr);
+ cc = *eptr;
if (cc >= 128 || (md->ctypes[cc] & ctype_digit) == 0)
RRETURN(MATCH_NOMATCH);
eptr++;
@@ -4544,7 +4544,7 @@
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
- cc = RAWUCHAR(eptr);
+ cc = *eptr;
if (cc < 128 && (md->ctypes[cc] & ctype_space) != 0)
RRETURN(MATCH_NOMATCH);
eptr++;
@@ -4561,7 +4561,7 @@
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
- cc = RAWUCHAR(eptr);
+ cc = *eptr;
if (cc >= 128 || (md->ctypes[cc] & ctype_space) == 0)
RRETURN(MATCH_NOMATCH);
eptr++;
@@ -4578,7 +4578,7 @@
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
- cc = RAWUCHAR(eptr);
+ cc = *eptr;
if (cc < 128 && (md->ctypes[cc] & ctype_word) != 0)
RRETURN(MATCH_NOMATCH);
eptr++;
@@ -4595,7 +4595,7 @@
SCHECK_PARTIAL();
RRETURN(MATCH_NOMATCH);
}
- cc = RAWUCHAR(eptr);
+ cc = *eptr;
if (cc >= 128 || (md->ctypes[cc] & ctype_word) == 0)
RRETURN(MATCH_NOMATCH);
eptr++;
@@ -5156,7 +5156,7 @@
{
default: RRETURN(MATCH_NOMATCH);
case CHAR_CR:
- if (eptr < md->end_subject && RAWUCHAR(eptr) == CHAR_LF) eptr++;
+ if (eptr < md->end_subject && *eptr == CHAR_LF) eptr++;
break;
case CHAR_LF:
@@ -5695,7 +5695,7 @@
eptr + 1 >= md->end_subject &&
NLBLOCK->nltype == NLTYPE_FIXED &&
NLBLOCK->nllen == 2 &&
- RAWUCHAR(eptr) == NLBLOCK->nl[0])
+ *eptr == NLBLOCK->nl[0])
{
md->hitend = TRUE;
if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
@@ -5721,7 +5721,7 @@
eptr + 1 >= md->end_subject &&
NLBLOCK->nltype == NLTYPE_FIXED &&
NLBLOCK->nllen == 2 &&
- RAWUCHAR(eptr) == NLBLOCK->nl[0])
+ *eptr == NLBLOCK->nl[0])
{
md->hitend = TRUE;
if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);
@@ -5778,7 +5778,7 @@
if (c == CHAR_CR)
{
if (++eptr >= md->end_subject) break;
- if (RAWUCHAR(eptr) == CHAR_LF) eptr++;
+ if (*eptr == CHAR_LF) eptr++;
}
else
{
@@ -5941,8 +5941,8 @@
if (rrc != MATCH_NOMATCH) RRETURN(rrc);
eptr--;
BACKCHAR(eptr);
- if (ctype == OP_ANYNL && eptr > pp && RAWUCHAR(eptr) == CHAR_NL &&
- RAWUCHAR(eptr - 1) == CHAR_CR) eptr--;
+ if (ctype == OP_ANYNL && eptr > pp && *eptr == CHAR_NL &&
+ eptr[-1] == CHAR_CR) eptr--;
}
}
else
@@ -6789,10 +6789,10 @@
if (first_char != first_char2)
while (start_match < end_subject &&
- (smc = RAWUCHARTEST(start_match)) != first_char && smc != first_char2)
+ (smc = *start_match) != first_char && smc != first_char2)
start_match++;
else
- while (start_match < end_subject && RAWUCHARTEST(start_match) != first_char)
+ while (start_match < end_subject && *start_match != first_char)
start_match++;
}
@@ -6824,7 +6824,7 @@
if (start_match[-1] == CHAR_CR &&
(md->nltype == NLTYPE_ANY || md->nltype == NLTYPE_ANYCRLF) &&
start_match < end_subject &&
- RAWUCHARTEST(start_match) == CHAR_NL)
+ *start_match == CHAR_NL)
start_match++;
}
}
@@ -6835,7 +6835,7 @@
{
while (start_match < end_subject)
{
- register pcre_uint32 c = RAWUCHARTEST(start_match);
+ register pcre_uint32 c = *start_match;
#ifndef COMPILE_PCRE8
if (c > 255) c = 255;
#endif
@@ -6893,7 +6893,7 @@
{
while (p < end_subject)
{
- register pcre_uint32 pp = RAWUCHARINCTEST(p);
+ register pcre_uint32 pp = *p++;
if (pp == req_char || pp == req_char2) { p--; break; }
}
}
@@ -6901,7 +6901,7 @@
{
while (p < end_subject)
{
- if (RAWUCHARINCTEST(p) == req_char) { p--; break; }
+ if (*p++ == req_char) { p--; break; }
}
}
Modified: code/trunk/pcre_internal.h
===================================================================
--- code/trunk/pcre_internal.h 2013-12-31 11:22:31 UTC (rev 1424)
+++ code/trunk/pcre_internal.h 2013-12-31 17:44:40 UTC (rev 1425)
@@ -316,8 +316,7 @@
&(NLBLOCK->nllen), utf)) \
: \
((p) <= NLBLOCK->PSEND - NLBLOCK->nllen && \
- RAWUCHARTEST(p) == NLBLOCK->nl[0] && \
- (NLBLOCK->nllen == 1 || RAWUCHARTEST(p+1) == NLBLOCK->nl[1]) \
+ *p == NLBLOCK->nl[0] && (NLBLOCK->nllen == 1 || p[1] == NLBLOCK->nl[1]) \
) \
)
@@ -330,8 +329,8 @@
&(NLBLOCK->nllen), utf)) \
: \
((p) >= NLBLOCK->PSSTART + NLBLOCK->nllen && \
- RAWUCHARTEST(p - NLBLOCK->nllen) == NLBLOCK->nl[0] && \
- (NLBLOCK->nllen == 1 || RAWUCHARTEST(p - NLBLOCK->nllen + 1) == NLBLOCK->nl[1]) \
+ *(p - NLBLOCK->nllen) == NLBLOCK->nl[0] && \
+ (NLBLOCK->nllen == 1 || *(p - NLBLOCK->nllen + 1) == NLBLOCK->nl[1]) \
) \
)
@@ -600,10 +599,6 @@
#define GETCHARINC(c, eptr) c = *eptr++;
#define GETCHARINCTEST(c, eptr) c = *eptr++;
#define GETCHARLEN(c, eptr, len) c = *eptr;
-#define RAWUCHAR(eptr) (*(eptr))
-#define RAWUCHARINC(eptr) (*(eptr)++)
-#define RAWUCHARTEST(eptr) (*(eptr))
-#define RAWUCHARINCTEST(eptr) (*(eptr)++)
/* #define GETCHARLENTEST(c, eptr, len) */
/* #define BACKCHAR(eptr) */
/* #define FORWARDCHAR(eptr) */
@@ -776,30 +771,6 @@
c = *eptr; \
if (utf && c >= 0xc0) GETUTF8LEN(c, eptr, len);
-/* Returns the next uchar, not advancing the pointer. This is called when
-we know we are in UTF mode. */
-
-#define RAWUCHAR(eptr) \
- (*(eptr))
-
-/* Returns the next uchar, advancing the pointer. This is called when
-we know we are in UTF mode. */
-
-#define RAWUCHARINC(eptr) \
- (*((eptr)++))
-
-/* Returns the next uchar, testing for UTF mode, and not advancing the
-pointer. */
-
-#define RAWUCHARTEST(eptr) \
- (*(eptr))
-
-/* Returns the next uchar, testing for UTF mode, advancing the
-pointer. */
-
-#define RAWUCHARINCTEST(eptr) \
- (*((eptr)++))
-
/* If the pointer is not at the start of a character, move it back until
it is. This is called only in UTF-8 mode - we don't put a test within the macro
because almost all calls are already within a block of UTF-8 only code. */
@@ -895,30 +866,6 @@
c = *eptr; \
if (utf && (c & 0xfc00) == 0xd800) GETUTF16LEN(c, eptr, len);
-/* Returns the next uchar, not advancing the pointer. This is called when
-we know we are in UTF mode. */
-
-#define RAWUCHAR(eptr) \
- (*(eptr))
-
-/* Returns the next uchar, advancing the pointer. This is called when
-we know we are in UTF mode. */
-
-#define RAWUCHARINC(eptr) \
- (*((eptr)++))
-
-/* Returns the next uchar, testing for UTF mode, and not advancing the
-pointer. */
-
-#define RAWUCHARTEST(eptr) \
- (*(eptr))
-
-/* Returns the next uchar, testing for UTF mode, advancing the
-pointer. */
-
-#define RAWUCHARINCTEST(eptr) \
- (*((eptr)++))
-
/* If the pointer is not at the start of a character, move it back until
it is. This is called only in UTF-16 mode - we don't put a test within the
macro because almost all calls are already within a block of UTF-16 only
@@ -980,30 +927,6 @@
#define GETCHARLENTEST(c, eptr, len) \
GETCHARTEST(c, eptr)
-/* Returns the next uchar, not advancing the pointer. This is called when
-we know we are in UTF mode. */
-
-#define RAWUCHAR(eptr) \
- (*(eptr))
-
-/* Returns the next uchar, advancing the pointer. This is called when
-we know we are in UTF mode. */
-
-#define RAWUCHARINC(eptr) \
- (*((eptr)++))
-
-/* Returns the next uchar, testing for UTF mode, and not advancing the
-pointer. */
-
-#define RAWUCHARTEST(eptr) \
- (*(eptr))
-
-/* Returns the next uchar, testing for UTF mode, advancing the
-pointer. */
-
-#define RAWUCHARINCTEST(eptr) \
- (*((eptr)++))
-
/* If the pointer is not at the start of a character, move it back until
it is. This is called only in UTF-32 mode - we don't put a test within the
macro because almost all calls are already within a block of UTF-32 only
Modified: code/trunk/pcre_string_utils.c
===================================================================
--- code/trunk/pcre_string_utils.c 2013-12-31 11:22:31 UTC (rev 1424)
+++ code/trunk/pcre_string_utils.c 2013-12-31 17:44:40 UTC (rev 1425)
@@ -91,8 +91,8 @@
while (*str1 != '\0' || *str2 != '\0')
{
- c1 = RAWUCHARINC(str1);
- c2 = RAWUCHARINC(str2);
+ c1 = *str1++;
+ c2 = *str2++;
if (c1 != c2)
return ((c1 > c2) << 1) - 1;
}
@@ -131,7 +131,7 @@
while (*str1 != '\0' || *ustr2 != '\0')
{
- c1 = RAWUCHARINC(str1);
+ c1 = *str1++;
c2 = (pcre_uchar)*ustr2++;
if (c1 != c2)
return ((c1 > c2) << 1) - 1;
