X-Git-Url: https://wimlib.net/git/?a=blobdiff_plain;f=src%2Fencoding.c;h=5ff1df2312d906cb1f560a7aa9a756e12efaaa21;hb=ccaeb7b1b39754f82a084ff7d9307962d8142339;hp=b54e787bd4cc10a5b671146dbad7661b5e9101bc;hpb=3071e89c11d1be71cf45b694432e5908e0c4ded9;p=wimlib

diff --git a/src/encoding.c b/src/encoding.c
index b54e787b..5ff1df23 100644
--- a/src/encoding.c
+++ b/src/encoding.c
@@ -23,7 +23,13 @@
 #  include "config.h"
 #endif
 
+#include <errno.h>
+#include <iconv.h>
+#include <pthread.h>
+#include <string.h>
+
 #include "wimlib.h"
+#include "wimlib/alloca.h"
 #include "wimlib/assert.h"
 #include "wimlib/encoding.h"
 #include "wimlib/endianness.h"
@@ -31,15 +37,6 @@
 #include "wimlib/list.h"
 #include "wimlib/util.h"
 
-#include <errno.h>
-#include <iconv.h>
-#include <pthread.h>
-#include <stdlib.h>
-#include <string.h>
-
-#ifdef HAVE_ALLOCA_H
-#  include <alloca.h>
-#endif
 
 bool wimlib_mbs_is_utf8 = !TCHAR_IS_UTF16LE;
 
@@ -139,7 +136,7 @@ varname1##_to_##varname2##_nbytes(const chartype1 *in, size_t in_nbytes,\
 	bool buf_onheap;						\
 	bufsize = (worst_case_len_expr) * sizeof(chartype2);		\
 	/* Worst case length */						\
-	if (bufsize <= STACK_MAX) { 					\
+	if (bufsize <= STACK_MAX) {					\
 		buf = alloca(bufsize);					\
 		buf_onheap = false;					\
 	} else {							\
@@ -561,9 +558,35 @@ cmp_utf16le_strings(const utf16lechar *s1, size_t n1,
 	return (n1 < n2) ? -1 : 1;
 }
 
-/* Duplicate a UTF16-LE string which may not be null-terminated.  */
+/* Like cmp_utf16le_strings(), but assumes the strings are null terminated.  */
+int
+cmp_utf16le_strings_z(const utf16lechar *s1, const utf16lechar *s2,
+		      bool ignore_case)
+{
+	if (ignore_case) {
+		for (;;) {
+			u16 c1 = upcase[le16_to_cpu(*s1)];
+			u16 c2 = upcase[le16_to_cpu(*s2)];
+			if (c1 != c2)
+				return (c1 < c2) ? -1 : 1;
+			if (c1 == 0)
+				return 0;
+			s1++, s2++;
+		}
+	} else {
+		while (*s1 && *s1 == *s2)
+			s1++, s2++;
+		if (*s1 == *s2)
+			return 0;
+		return (le16_to_cpu(*s1) < le16_to_cpu(*s2)) ? -1 : 1;
+	}
+}
+
+/* Duplicate a UTF-16LE string.  The input string might not be null terminated
+ * and might be misaligned, but the returned string is guaranteed to be null
+ * terminated and properly aligned.  */
 utf16lechar *
-utf16le_dupz(const utf16lechar *ustr, size_t usize)
+utf16le_dupz(const void *ustr, size_t usize)
 {
 	utf16lechar *dup = MALLOC(usize + sizeof(utf16lechar));
 	if (dup) {
@@ -572,3 +595,32 @@ utf16le_dupz(const utf16lechar *ustr, size_t usize)
 	}
 	return dup;
 }
+
+/* Duplicate a null-terminated UTF-16LE string.  */
+utf16lechar *
+utf16le_dup(const utf16lechar *ustr)
+{
+	const utf16lechar *p = ustr;
+	while (*p++)
+		;
+	return memdup(ustr, (const u8 *)p - (const u8 *)ustr);
+}
+
+/* Return the length, in bytes, of a UTF-null terminated UTF-16 string,
+ * excluding the null terminator.  */
+size_t
+utf16le_len_bytes(const utf16lechar *s)
+{
+	const utf16lechar *p = s;
+	while (*p)
+		p++;
+	return (p - s) * sizeof(utf16lechar);
+}
+
+/* Return the length, in UTF-16 coding units, of a UTF-null terminated UTF-16
+ * string, excluding the null terminator.  */
+size_t
+utf16le_len_chars(const utf16lechar *s)
+{
+	return utf16le_len_bytes(s) / sizeof(utf16lechar);
+}