string_utf8.c in utf8-0.1.7

- old
+ new

@@ -91,11 +91,11 @@
 }
 
 /*
  * call-seq: valid?(max_codepoint=nil)
  *
- * Iterates over the string, yielding one UTF-8 codepoint at a time
+ * Iterates over the string, returning true/false if it's valid UTF-8
  *
  * max_codepoint - an optional Fixnum used to declare this string invalid
  *                 if a codepoint higher than that value is found
  *                 if nothing is passed, the UTF-8 maximum of 0x10FFFF is assumed
  *
@@ -335,46 +335,75 @@
  * Iterates over the string, replacing invalid UTF-8 characters with '?'
  *
  * Returns: a new String
  */
 static VALUE rb_cString_UTF8_clean(VALUE self) {
-  unsigned char *str;
-  unsigned char *out;
-  unsigned char replace;
-  size_t len;
+  unsigned char *inBuf, *inBufCur;
+  unsigned char *outBuf, *outBufCur;
+  size_t len, i;
   int8_t curCharLen;
-  size_t i;
   VALUE rb_out;
 
-  str = (unsigned char *)RSTRING_PTR(self);
+  inBuf = (unsigned char *)RSTRING_PTR(self);
+  inBufCur = inBuf;
   len = RSTRING_LEN(self);
-  replace = REPLACEMENT_CHAR;
-  out = xmalloc(len);
+  outBuf = malloc(len);
+  outBufCur = outBuf;
 
   for(i=0; i<len; i+=curCharLen) {
-    curCharLen = utf8CharLen(str+i, len);
+    curCharLen = utf8CharLen(inBufCur, len);
     if (curCharLen < 0) {
-      *(out+i) = replace;
+      if (inBufCur-inBuf > 0) {
+        memcpy(outBufCur, inBuf, inBufCur-inBuf);
+        outBufCur += inBufCur-inBuf;
+      }
+      *outBufCur++ = REPLACEMENT_CHAR;
+      inBuf += (inBufCur-inBuf)+1;
       curCharLen = 1;
-    } else {
-      memcpy(out+i, str+i, curCharLen);
     }
+
+    inBufCur += curCharLen;
   }
 
-  rb_out = rb_str_new((const char*)out, len);
+  if (inBufCur-inBuf > 0) {
+    memcpy(outBufCur, inBuf, inBufCur-inBuf);
+  }
+
+  rb_out = rb_str_new((const char*)outBuf, len);
   AS_UTF8(rb_out);
 
-  xfree(out);
+  free(outBuf);
 
   return rb_out;
 }
 
+/*
+ * call-seq: clean
+ *
+ * Iterates over the string, returning true/false if it's within the low ASCII range
+ *
+ * Returns: a Boolean - true if the string is within the low ASCII range, false if not
+ */
+static VALUE rb_cString_UTF8_ascii_only(VALUE self) {
+  unsigned char *str = (unsigned char *)RSTRING_PTR(self);
+  size_t len = RSTRING_LEN(self), i=0;
+
+  for(; i<len; i+=1) {
+    if (str[i] > 0x7f) {
+      return Qfalse;
+    }
+  }
+
+  return Qtrue;
+}
+
 void init_String_UTF8() {
   VALUE rb_cString_UTF8 = rb_define_class_under(rb_cString, "UTF8", rb_cString);
 
-  rb_define_method(rb_cString_UTF8, "length",    rb_cString_UTF8_length, 0);
+  rb_define_method(rb_cString_UTF8, "length", rb_cString_UTF8_length, 0);
   rb_define_method(rb_cString_UTF8, "each_char", rb_cString_UTF8_each_char, -1);
-  rb_define_method(rb_cString_UTF8, "[]",        rb_cString_UTF8_slice, -1);
+  rb_define_method(rb_cString_UTF8, "[]", rb_cString_UTF8_slice, -1);
   rb_define_method(rb_cString_UTF8, "each_codepoint", rb_cString_UTF8_each_codepoint, -1);
   rb_define_method(rb_cString_UTF8, "valid?", rb_cString_UTF8_valid, -1);
   rb_define_method(rb_cString_UTF8, "clean", rb_cString_UTF8_clean, 0);
+  rb_define_method(rb_cString_UTF8, "ascii_only?", rb_cString_UTF8_ascii_only, 0);
 }
\ No newline at end of file