fix utf-8 names in autoindex

3 files changed, 37 insertions, 32 deletions

diff --git a/src/core/ngx_string.c b/src/core/ngx_string.c
index c9193152c..64ba703b7 100644
--- a/src/core/ngx_string.c
+++ b/src/core/ngx_string.c
@@ -955,7 +955,7 @@ ngx_decode_base64(ngx_str_t *dst, ngx_str_t *src)
  * ngx_utf_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
  */
@@ -1020,29 +1020,24 @@ ngx_utf_decode(u_char **p, size_t n)
 size_t
 ngx_utf_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_utf_decode(&p, n) > 0x10ffff) {
+            /* invalid utf */
+            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_utf_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_utf_decode(&next, len) > 0x10ffff) {
+            /* invalid utf */
+            break;
         }
 
-        /* invalid utf */
+        len--;
+
+        while (src < next) {
+            *++dst = *++src;
+            len--;
+        }
     }
 
     *dst = '\0';
diff --git a/src/core/ngx_string.h b/src/core/ngx_string.h
index 3514e52f3..895925b6d 100644
--- a/src/core/ngx_string.h
+++ b/src/core/ngx_string.h
@@ -153,7 +153,7 @@ ngx_int_t ngx_decode_base64(ngx_str_t *dst, ngx_str_t *src);
 
 uint32_t ngx_utf_decode(u_char **p, size_t n);
 size_t ngx_utf_length(u_char *p, size_t n);
-u_char *ngx_utf_cpystrn(u_char *dst, u_char *src, size_t n);
+u_char *ngx_utf_cpystrn(u_char *dst, u_char *src, size_t n, size_t len);
 
 
 #define NGX_ESCAPE_URI         0
diff --git a/src/http/modules/ngx_http_autoindex_module.c b/src/http/modules/ngx_http_autoindex_module.c
index b5b6ef7f3..64b29a8b3 100644
--- a/src/http/modules/ngx_http_autoindex_module.c
+++ b/src/http/modules/ngx_http_autoindex_module.c
@@ -135,7 +135,7 @@ ngx_http_autoindex_handler(ngx_http_request_t *r)
 {
     u_char                         *last, *filename, scale;
     off_t                           length;
-    size_t                          len, copy, allocated, root;
+    size_t                          len, utf_len, allocated, root;
     ngx_tm_t                        tm;
     ngx_err_t                       err;
     ngx_buf_t                      *b;
@@ -412,15 +412,16 @@ ngx_http_autoindex_handler(ngx_http_request_t *r)
 
         len = entry[i].utf_len;
 
-        if (entry[i].name.len - len) {
+        if (entry[i].name.len != len) {
             if (len > NGX_HTTP_AUTOINDEX_NAME_LEN) {
-                copy = NGX_HTTP_AUTOINDEX_NAME_LEN - 3 + 1;
+                utf_len = NGX_HTTP_AUTOINDEX_NAME_LEN - 3 + 1;
 
             } else {
-                copy = NGX_HTTP_AUTOINDEX_NAME_LEN + 1;
+                utf_len = NGX_HTTP_AUTOINDEX_NAME_LEN + 1;
             }
 
-            b->last = ngx_utf_cpystrn(b->last, entry[i].name.data, copy);
+            b->last = ngx_utf_cpystrn(b->last, entry[i].name.data,
+                                      utf_len, entry[i].name.len + 1);
             last = b->last;
 
         } else {