Changeset b54d2f1 in mainline for kernel/generic/src/lib/string.c

Timestamp:

2009-03-31T22:11:11Z (16 years ago)

Author:

Jiri Svoboda <jirik.svoboda@…>

Branches:

lfn, master, serial, ticket/834-toolchain-update, topic/msim-upgrade, topic/simplify-dev-export

Children:

e1813cf

Parents:

32704cb

Message:

Semantics for 'index' parameter of utf8_encode/decode() should be more logical.

File:

• kernel/generic/src/lib/string.c (modified) (9 diffs)

kernel/generic/src/lib/string.c

@@ -60 +60 @@
  *
  * Decode a single UTF-8 character from a plain char NULL-terminated
- * string. Decoding starts at @index and this index is incremented
- * if the current UTF-8 string is encoded in more than a single byte.
+ * string. Decoding starts at @index and this index is moved to the
+ * beginning of the next character. In case of decoding error,
+ * index advances. However, index is never moved beyond (str+limit).
  *
  * @param str   Plain character NULL-terminated string.
@@ -79 +80 @@
         int cbytes;             /* Number of continuation bytes. */
 
-        if (*index > limit)
+        if (*index + 1 > limit)
                 return invalch;
 
-        b0 = (uint8_t) str[*index];
+        b0 = (uint8_t) str[(*index)++];
 
         /* Determine code length. */
@@ -115 +116 @@
         /* Decode continuation bytes. */
         while (cbytes > 0) {
-                b = (uint8_t) str[*index + 1];
-                ++(*index);
+                b = (uint8_t) str[(*index)++];
 
                 /* Must be 10xxxxxx. */
@@ -135 +135 @@
  * Encode a single UTF-32 character as UTF-8 and store it into
  * the given buffer at @index. Encoding starts at @index and
- * this index is incremented if the UTF-8 character takes
- * more than a single byte.
+ * this index is moved at the position where the next character
+ * can be written to.
  *
  * @param ch    Input UTF-32 character.
@@ -157 +157 @@
         int i;
 
-        if (*index > limit)
+        if (*index >= limit)
                 return false;
 
@@ -185 +185 @@
 
         /* Check for available space in buffer. */
-        if (*index + cbytes > limit)
+        if (*index + cbytes >= limit)
                 return false;
 
@@ -198 +198 @@
 
         /* Advance index. */
-        *index += cbytes;
+        *index += (1 + cbytes);
         
         return true;
@@ -220 +220 @@
         size_t size = 0;
         index_t index = 0;
-        
-        while ((utf8_decode(str, &index, UTF8_NO_LIMIT) != 0) && (size < count)) {
+        index_t iprev;
+        wchar_t ch;
+        
+        while (true) {
+                iprev = index;
+                if (size >= count)
+                        break;
+                ch = utf8_decode(str, &index, UTF8_NO_LIMIT);
+                if (ch == '\0') break;
+
                 size++;
-                index++;
-        }
-        
-        return index;
+        }
+        
+        return iprev;
 }
 
@@ -284 +291 @@
         while (utf8_decode(str, &index, UTF8_NO_LIMIT) != 0) {
                 size++;
-                index++;
         }