naruse	2009-08-21 15:16:53 +0900 (Fri, 21 Aug 2009)

  New Revision: 24605

  http://svn.ruby-lang.org/cgi-bin/viewvc.cgi?view=rev&revision=24605

  Log:
    Change modestr syntax for BOM to "BOM|UTF-*".
    * io .c (rb_io_fmode_modestr): change modestr syntax for BOM
      to "BOM|UTF-*".
    
    * io.c (parse_mode_enc): ditto.

  Modified files:
    trunk/ChangeLog
    trunk/io.c
    trunk/test/ruby/test_io_m17n.rb

Index: ChangeLog
===================================================================
--- ChangeLog	(revision 24604)
+++ ChangeLog	(revision 24605)
@@ -1,3 +1,10 @@
+Thu Aug 20 01:24:55 2009  NARUSE, Yui  <naruse@r...>
+
+	* io.c (rb_io_fmode_modestr): change modestr syntax for BOM
+	  to "BOM|UTF-*". [ruby-dev:39106]
+
+	* io.c (parse_mode_enc): ditto.
+
 Fri Aug 21 15:01:35 2009  NARUSE, Yui  <naruse@r...>
 
 	* ext/readline/readline.c (readline_readline): use rb_prep_terminal
Index: io.c
===================================================================
--- io.c	(revision 24604)
+++ io.c	(revision 24605)
@@ -3938,17 +3938,11 @@
 static int
 io_encname_bom_p(const char *name, long len)
 {
-    if (len) {
-	if (len > 4 && STRNCASECMP(name + len - 4, "-bom", 4) == 0)
-	    return 1;
-    }
-    else {
+    if (!len) {
 	const char *p = strchr(name, ':');
-	if (!p) p = name + strlen(name);
-	if (p - name > 4 && STRNCASECMP(p - 4, "-bom", 4) == 0)
-	    return 1;
+	len = p ? p - name : strlen(name);
     }
-    return 0;
+    return len > 8 && STRNCASECMP(name, "bom|utf-", 8) == 0;
 }
 
 int
@@ -4147,7 +4141,7 @@
 	    idx = -1;
 	else {
 	    if (io_encname_bom_p(estr, len))
-		len -= 4;
+		estr += 4;
 	    memcpy(encname, estr, len);
 	    encname[len] = '\0';
 	    estr = encname;
@@ -4157,7 +4151,7 @@
     else {
 	long len = strlen(estr);
 	if (io_encname_bom_p(estr, len)) {
-	    len -= 4;
+	    estr += 4;
 	    memcpy(encname, estr, len);
 	    encname[len] = '\0';
 	    estr = encname;
@@ -5481,9 +5475,9 @@
  *  read string will be tagged by the encoding in reading,
  *  and output string will be converted
  *  to the specified encoding in writing.
- *  If ext_enc ends with '-bom', check whether the input has a BOM. If
+ *  If ext_enc starts with 'BOM|', check whether the input has a BOM. If
  *  there is a BOM, strip it and set external encoding as
- *  what the BOM tells. If there is no BOM, use ext_enc without '-bom'.
+ *  what the BOM tells. If there is no BOM, use ext_enc without 'BOM|'.
  *  If two encoding names,
  *  ext_enc and int_enc (external encoding and internal encoding),
  *  are specified, the read string is converted from ext_enc
Index: test/ruby/test_io_m17n.rb
===================================================================
--- test/ruby/test_io_m17n.rb	(revision 24604)
+++ test/ruby/test_io_m17n.rb	(revision 24605)
@@ -1699,7 +1699,7 @@
         path = '%s-bom.txt' % name
         content = text.encode(name)
         generate_file(path, content)
-        result = File.read(path, mode: 'rb:utf-7-bom')
+        result = File.read(path, mode: 'rb:BOM|UTF-8')
         assert_equal(content[1].force_encoding("ascii-8bit"),
                      result.force_encoding("ascii-8bit"))
       end

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