matz	2008-02-15 17:14:40 +0900 (Fri, 15 Feb 2008)

  New Revision: 15486

  Modified files:
    trunk/ChangeLog
    trunk/encoding.c
    trunk/string.c

  Log:
    * string.c (str_strlen): use search_nonascii() for performance.
    
    * string.c (str_nth): ditto.

  http://svn.ruby-lang.org/cgi-bin/viewvc.cgi/trunk/string.c?r1=15486&r2=15485&diff_format=u
  http://svn.ruby-lang.org/cgi-bin/viewvc.cgi/trunk/ChangeLog?r1=15486&r2=15485&diff_format=u
  http://svn.ruby-lang.org/cgi-bin/viewvc.cgi/trunk/encoding.c?r1=15486&r2=15485&diff_format=u

Index: encoding.c
===================================================================
--- encoding.c	(revision 15485)
+++ encoding.c	(revision 15486)
@@ -645,8 +645,8 @@
     rb_encoding *enc = rb_enc_compatible(str1, str2);
     if (!enc)
 	rb_raise(rb_eArgError, "character encodings differ: %s and %s",
-            rb_enc_name(rb_enc_get(str1)),
-            rb_enc_name(rb_enc_get(str2)));
+		 rb_enc_name(rb_enc_get(str1)),
+		 rb_enc_name(rb_enc_get(str2)));
     return enc;
 }
 
Index: ChangeLog
===================================================================
--- ChangeLog	(revision 15485)
+++ ChangeLog	(revision 15486)
@@ -1,3 +1,9 @@
+Fri Feb 15 17:12:41 2008  Yukihiro Matsumoto  <matz@r...>
+
+	* string.c (str_strlen): use search_nonascii() for performance.
+
+	* string.c (str_nth): ditto.
+
 Fri Feb 15 16:22:49 2008  Yukihiro Matsumoto  <matz@r...>
 
 	* io.c (open_key_args): allow specifying both :mode and :encoding.
Index: string.c
===================================================================
--- string.c	(revision 15485)
+++ string.c	(revision 15486)
@@ -591,11 +591,21 @@
 static long
 str_strlen(VALUE str, rb_encoding *enc)
 {
-    long len;
+    long len = 0;
+    const char *p, *e;
 
     if (single_byte_optimizable(str)) return RSTRING_LEN(str);
     if (!enc) enc = STR_ENC_GET(str);
-    len = rb_enc_strlen(RSTRING_PTR(str), RSTRING_END(str), enc);
+    p = RSTRING_PTR(str);
+    e = RSTRING_END(str);
+    if (rb_enc_asciicompat(enc)) {
+	const char *p2 = search_nonascii(p, e);
+
+	if (!p2) return RSTRING_LEN(str);
+	len = p2 - p;
+	p = p2;
+    }
+    len += rb_enc_strlen(p, e, enc);
     if (len < 0) {
 	rb_raise(rb_eArgError, "invalid mbstring sequence");
     }
@@ -886,8 +896,17 @@
 {
     if (singlebyte)
 	p += nth;
-    else
+    else {
+	if (rb_enc_asciicompat(enc)) {
+	    const char *p2 = search_nonascii(p, e);
+
+	    if (!p2 || p + nth < p2)
+		return (char*)p + nth;
+	    nth -= p2 - p;
+	    p = p2;
+	}
 	p = rb_enc_nth(p, e, nth, enc);
+    }
     if (!p) return 0;
     if (p > e) p = e;
     return (char *)p;
@@ -2805,14 +2824,12 @@
         enc = rb_enc_compatible(str, repl);
         if (!enc) {
             rb_encoding *str_enc = STR_ENC_GET(str);
-            if (coderange_scan(RSTRING_PTR(str), BEG(0), str_enc) !=
-                    ENC_CODERANGE_7BIT ||
+            if (coderange_scan(RSTRING_PTR(str), BEG(0), str_enc) != ENC_CODERANGE_7BIT ||
                 coderange_scan(RSTRING_PTR(str)+END(0),
-                    RSTRING_LEN(str)-END(0), str_enc) !=
-                    ENC_CODERANGE_7BIT) {
+			       RSTRING_LEN(str)-END(0), str_enc) != ENC_CODERANGE_7BIT) {
                 rb_raise(rb_eArgError, "character encodings differ: %s and %s",
-                    rb_enc_name(str_enc),
-                    rb_enc_name(STR_ENC_GET(repl)));
+			 rb_enc_name(str_enc),
+			 rb_enc_name(STR_ENC_GET(repl)));
             }
             enc = STR_ENC_GET(repl);
         }

--
ML: ruby-changes@q...
Info: http://www.atdot.net/~ko1/quickml/