From 3f6e8ffe64b57b0eaba6812208e94500422ca40c Mon Sep 17 00:00:00 2001
From: Jeffrey Yasskin <jyasskin@gmail.com>
Date: Sun, 25 Jul 2010 00:36:03 -0700
Subject: Implement _str.len() to return the number of bytes, rename it to
 byte_len(), and add a test.

---
 src/lib/_str.rs | 15 ++++++++++-----
 1 file changed, 10 insertions(+), 5 deletions(-)

(limited to 'src/lib')

diff --git a/src/lib/_str.rs b/src/lib/_str.rs
index 7d1a2dbd..a607c7d5 100644
--- a/src/lib/_str.rs
+++ b/src/lib/_str.rs
@@ -3,7 +3,7 @@ import rustrt.sbuf;
 native "rust" mod rustrt {
   type sbuf;
   fn str_buf(str s) -> sbuf;
-  fn str_len(str s) -> uint;
+  fn str_byte_len(str s) -> uint;
   fn str_alloc(uint n_bytes) -> str;
   fn refcount[T](str s) -> uint;
 }
@@ -13,7 +13,7 @@ fn is_utf8(vec[u8] v) -> bool {
 }
 
 fn is_ascii(str s) -> bool {
-  let uint i = len(s);
+  let uint i = byte_len(s);
   while (i > 0u) {
     i -= 1u;
     if ((s.(i) & 0x80u8) != 0u8) {
@@ -27,8 +27,13 @@ fn alloc(uint n_bytes) -> str {
   ret rustrt.str_alloc(n_bytes);
 }
 
-fn len(str s) -> uint {
-  ret rustrt.str_len(s);
+// Returns the number of bytes (a.k.a. UTF-8 code units) in s.
+// Contrast with a function that would return the number of code
+// points (char's), combining character sequences, words, etc.  See
+// http://icu-project.org/apiref/icu4c/classBreakIterator.html for a
+// way to implement those.
+fn byte_len(str s) -> uint {
+  ret rustrt.str_byte_len(s);
 }
 
 fn buf(str s) -> sbuf {
@@ -39,5 +44,5 @@ fn bytes(&str s) -> vec[u8] {
   fn ith(str s, uint i) -> u8 {
     ret s.(i);
   }
-  ret _vec.init_fn[u8](bind ith(s, _), _str.len(s));
+  ret _vec.init_fn[u8](bind ith(s, _), _str.byte_len(s));
 }
-- 
cgit v1.2.3