Mercurial > hg > nginx-vendor-current

diff src/core/ngx_string.c @ 390:0b6053502c55 NGINX_0_7_7

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
nginx 0.7.7 *) Change: now the EAGAIN error returned by connect() is not considered as temporary error. *) Change: now the $ssl_client_cert variable value is a certificate with TAB character intended before each line except first one; an unchanged certificate is available in the $ssl_client_raw_cert variable. *) Feature: the "ask" parameter in the "ssl_verify_client" directive. *) Feature: byte-range processing improvements. Thanks to Maxim Dounin. *) Feature: the "directio" directive. *) Feature: MacOSX 1.5 sendfile() support. *) Bugfix: now in MacOSX and Cygwin locations are tested in case insensitive mode; however, the compare is provided by single-byte locales only. *) Bugfix: mail proxy SSL connections hanged, if select, poll, or /dev/poll methods were used. *) Bugfix: UTF-8 encoding usage in the ngx_http_autoindex_module.
author Igor Sysoev <http://sysoev.ru>
date Wed, 30 Jul 2008 00:00:00 +0400
parents 984bb0b1399b
children 34fb3a573548
line wrap: on
line diff
--- a/src/core/ngx_string.c
+++ b/src/core/ngx_string.c
@@ -952,16 +952,16 @@ ngx_decode_base64(ngx_str_t *dst, ngx_st
 
 
 /*
- * ngx_utf_decode() decodes two and more bytes UTF sequences only
+ * ngx_utf8_decode() decodes two and more bytes UTF sequences only
  * the return values:
  *    0x80 - 0x10ffff         valid character
- *    0x10ffff - 0xfffffffd   invalid sequence
+ *    0x110000 - 0xfffffffd   invalid sequence
  *    0xfffffffe              incomplete sequence
  *    0xffffffff              error
  */
 
 uint32_t
-ngx_utf_decode(u_char **p, size_t n)
+ngx_utf8_decode(u_char **p, size_t n)
 {
     size_t    len;
     uint32_t  u, i, valid;
@@ -1018,31 +1018,26 @@ ngx_utf_decode(u_char **p, size_t n)
 
 
 size_t
-ngx_utf_length(u_char *p, size_t n)
+ngx_utf8_length(u_char *p, size_t n)
 {
-    u_char      c;
-    size_t      len;
-    ngx_uint_t  i;
+    u_char  c, *last;
+    size_t  len;
 
-    for (len = 0, i = 0; i < n; len++, i++) {
+    last = p + n;
 
-        c = p[i];
+    for (len = 0; p < last; len++) {
+
+        c = *p;
 
         if (c < 0x80) {
+            p++;
             continue;
         }
 
-        if (c >= 0xc0) {
-            for (c <<= 1; c & 0x80; c <<= 1) {
-                i++;
-            }
-
-            continue;
+        if (ngx_utf8_decode(&p, n) > 0x10ffff) {
+            /* invalid UTF-8 */
+            return n;
         }
-
-        /* invalid utf */
-
-        return n;
     }
 
     return len;
@@ -1050,36 +1045,45 @@ ngx_utf_length(u_char *p, size_t n)
 
 
 u_char *
-ngx_utf_cpystrn(u_char *dst, u_char *src, size_t n)
+ngx_utf8_cpystrn(u_char *dst, u_char *src, size_t n, size_t len)
 {
-    u_char  c;
+    u_char  c, *next;
 
     if (n == 0) {
         return dst;
     }
 
-    for ( /* void */ ; --n; dst++, src++) {
+    while (--n) {
 
         c = *src;
         *dst = c;
 
         if (c < 0x80) {
-            if (*dst != '\0') {
+
+            if (c != '\0') {
+                dst++;
+                src++;
+                len--;
+
                 continue;
             }
 
             return dst;
         }
 
-        if (c >= 0xc0) {
-            for (c <<= 1; c & 0x80; c <<= 1) {
-               *++dst = *++src;
-            }
+        next = src;
 
-            continue;
+        if (ngx_utf8_decode(&next, len) > 0x10ffff) {
+            /* invalid UTF-8 */
+            break;
         }
 
-        /* invalid utf */
+        len--;
+
+        while (src < next) {
+            *++dst = *++src;
+            len--;
+        }
     }
 
     *dst = '\0';