Fix for MacOS/X which uses STUPID BROKEN UNICODE COMPOSE CHARACTERS !

(rant off :-). Inspired by work from Benjamin Riefenstahl <Benjamin.Riefenstahl@epost.de>.
Also add MacOSX/Darwin configure fixes.
Jerry - can we put this in 3.0 release ? :-).
Jeremy.

2 files changed, 63 insertions, 27 deletions

diff --git a/source/lib/charcnv.c b/source/lib/charcnv.c
index e8ae40dbe29..dafc88fb77a 100644
--- a/source/lib/charcnv.c
+++ b/source/lib/charcnv.c
@@ -176,6 +176,14 @@ static size_t convert_string_internal(charset_t from, charset_t to,
 
 	descriptor = conv_handles[from][to];
 
+	if (srclen == (size_t)-1) {
+		if (from == CH_UCS2) {
+			srclen = (strlen_w((const smb_ucs2_t *)src)+1) * 2;
+		} else {
+			srclen = strlen((const char *)src)+1;
+		}
+	}
+
 	if (descriptor == (smb_iconv_t)-1 || descriptor == (smb_iconv_t)0) {
 		if (!conv_silent)
 			DEBUG(0,("convert_string_internal: Conversion not supported.\n"));
@@ -248,31 +256,40 @@ size_t convert_string(charset_t from, charset_t to,
 		      void const *src, size_t srclen, 
 		      void *dest, size_t destlen)
 {
+	/*
+	 * NB. We deliberately don't do a strlen here is srclen == -1.
+	 * This is very expensive over millions of calls and is taken
+	 * care of in the slow path in convert_string_internal. JRA.
+	 */
+
 	if (srclen == 0)
 		return 0;
 
 	if (from != CH_UCS2 && to != CH_UCS2) {
 		const unsigned char *p = (const unsigned char *)src;
 		unsigned char *q = (unsigned char *)dest;
+		size_t slen = srclen;
+		size_t dlen = destlen;
 		unsigned char lastp;
 		size_t retval = 0;
 
 		/* If all characters are ascii, fast path here. */
-		while (srclen && destlen) {
+		while (slen && dlen) {
 			if ((lastp = *p) <= 0x7f) {
 				*q++ = *p++;
-				if (srclen != (size_t)-1) {
-					srclen--;
+				if (slen != (size_t)-1) {
+					slen--;
 				}
-				destlen--;
+				dlen--;
 				retval++;
 				if (!lastp)
 					break;
 			} else {
-				if (srclen == (size_t)-1) {
-					srclen = strlen(p)+1;
-				}
-				return retval + convert_string_internal(from, to, p, srclen, q, destlen);
+#ifdef BROKEN_UNICODE_COMPOSE_CHARACTERS
+				goto general_case;
+#else
+				return retval + convert_string_internal(from, to, p, slen, q, dlen);
+#endif
 			}
 		}
 		return retval;
@@ -280,25 +297,28 @@ size_t convert_string(charset_t from, charset_t to,
 		const unsigned char *p = (const unsigned char *)src;
 		unsigned char *q = (unsigned char *)dest;
 		size_t retval = 0;
+		size_t slen = srclen;
+		size_t dlen = destlen;
 		unsigned char lastp;
 
 		/* If all characters are ascii, fast path here. */
-		while ((srclen >= 2) && destlen) {
+		while ((slen >= 2) && dlen) {
 			if (((lastp = *p) <= 0x7f) && (p[1] == 0)) {
 				*q++ = *p;
-				if (srclen != (size_t)-1) {
-					srclen -= 2;
+				if (slen != (size_t)-1) {
+					slen -= 2;
 				}
 				p += 2;
-				destlen--;
+				dlen--;
 				retval++;
 				if (!lastp)
 					break;
 			} else {
-				if (srclen == (size_t)-1) {
-					srclen = (strlen_w((const smb_ucs2_t *)p)+1) * 2;
-				}
-				return retval + convert_string_internal(from, to, p, srclen, q, destlen);
+#ifdef BROKEN_UNICODE_COMPOSE_CHARACTERS
+				goto general_case;
+#else
+				return retval + convert_string_internal(from, to, p, slen, q, dlen);
+#endif
 			}
 		}
 		return retval;
@@ -306,29 +326,36 @@ size_t convert_string(charset_t from, charset_t to,
 		const unsigned char *p = (const unsigned char *)src;
 		unsigned char *q = (unsigned char *)dest;
 		size_t retval = 0;
+		size_t slen = srclen;
+		size_t dlen = destlen;
 		unsigned char lastp;
 
 		/* If all characters are ascii, fast path here. */
-		while (srclen && (destlen >= 2)) {
+		while (slen && (dlen >= 2)) {
 			if ((lastp = *p) <= 0x7F) {
 				*q++ = *p++;
 				*q++ = '\0';
-				if (srclen != (size_t)-1) {
-					srclen--;
+				if (slen != (size_t)-1) {
+					slen--;
 				}
-				destlen -= 2;
+				dlen -= 2;
 				retval += 2;
 				if (!lastp)
 					break;
 			} else {
-				if (srclen == (size_t)-1) {
-					srclen = strlen(p)+1;
-				}
-				return retval + convert_string_internal(from, to, p, srclen, q, destlen);
+#ifdef BROKEN_UNICODE_COMPOSE_CHARACTERS
+				goto general_case;
+#else
+				return retval + convert_string_internal(from, to, p, slen, q, dlen);
+#endif
 			}
 		}
 		return retval;
 	}
+
+#ifdef BROKEN_UNICODE_COMPOSE_CHARACTERS
+  general_case:
+#endif
 	return convert_string_internal(from, to, src, srclen, dest, destlen);
 }
 
diff --git a/source/lib/util_str.c b/source/lib/util_str.c
index c065bfe9db6..15ac1639a9a 100644
--- a/source/lib/util_str.c
+++ b/source/lib/util_str.c
@@ -382,6 +382,10 @@ void string_replace(pstring s,char oldc,char newc)
 		return;
 
 	/* Slow (mb) path. */
+#ifdef BROKEN_UNICODE_COMPOSE_CHARACTERS
+	/* With compose characters we must restart from the beginning. JRA. */
+	p = s;
+#endif
 	push_ucs2(NULL, tmpbuf, p, sizeof(tmpbuf), STR_TERMINATE);
 	string_replace_w(tmpbuf, UCS2_CHAR(oldc), UCS2_CHAR(newc));
 	pull_ucs2(NULL, p, tmpbuf, -1, sizeof(tmpbuf), STR_TERMINATE);
@@ -1175,26 +1179,31 @@ char *string_truncate(char *s, unsigned int length)
  We convert via ucs2 for now.
 **/
 
-char *strchr_m(const char *s, char c)
+char *strchr_m(const char *src, char c)
 {
 	wpstring ws;
 	pstring s2;
 	smb_ucs2_t *p;
+	const char *s;
 
 	/* this is quite a common operation, so we want it to be
 	   fast. We optimise for the ascii case, knowing that all our
 	   supported multi-byte character sets are ascii-compatible
 	   (ie. they match for the first 128 chars) */
 
-	while (*s && !(((unsigned char)s[0]) & 0x80)) {
+	for (s = src; *s && !(((unsigned char)s[0]) & 0x80); s++) {
 		if (*s == c)
 			return s;
-		s++;
 	}
 
 	if (!*s)
 		return NULL;
 
+#ifdef BROKEN_UNICODE_COMPOSE_CHARACTERS
+	/* With compose characters we must restart from the beginning. JRA. */
+	s = src;
+#endif
+
 	push_ucs2(NULL, ws, s, sizeof(ws), STR_TERMINATE);
 	p = strchr_w(ws, UCS2_CHAR(c));
 	if (!p)