From cf7e585bb45970fe0b5a8a6aa6653cd50583d052 Mon Sep 17 00:00:00 2001
From: Rory Byrne <rory@jinsky.com>
Date: Sat, 16 Jan 2010 14:47:43 +0000
Subject: Fix re:replace/4 to handle unicode charlist Replacement argument

A bug in re:replace/4 causes a badarg exception to be thrown when the
Replacement argument is a charlist containing non-ascii codepoints.

The problem is that the code incorrectly assumes that the Replacement
text is iodata() and calls iolist_to_binary/1 on it. This patch fixes
it to obey the 'unicode' option and handle charlist() Replacement
arguments correctly.
---
 lib/stdlib/src/re.erl        | 17 ++++++++++++++---
 lib/stdlib/test/re_SUITE.erl |  1 +
 2 files changed, 15 insertions(+), 3 deletions(-)

diff --git a/lib/stdlib/src/re.erl b/lib/stdlib/src/re.erl
index f934fdcba1..889d273f6f 100644
--- a/lib/stdlib/src/re.erl
+++ b/lib/stdlib/src/re.erl
@@ -229,7 +229,19 @@ replace(Subject,RE,Replacement,Options) ->
 			iolist_to_binary(Subject)
 		end
 	end,
-    case do_replace(FlatSubject,Subject,RE,Replacement,NewOpt) of
+    FlatReplacement =
+	case is_binary(Replacement) of
+	    true ->
+		Replacement;
+	    false ->
+		case Unicode of
+		    true ->
+			unicode:characters_to_binary(Replacement,unicode);
+		    false ->
+			iolist_to_binary(Replacement)
+		end
+	end,
+    case do_replace(FlatSubject,Subject,RE,FlatReplacement,NewOpt) of
 	{error,_Err} ->
 	    throw(badre);
 	IoList ->
@@ -329,8 +341,7 @@ process_split_params([H|T],C,U,L,S,G) ->
     {[H|NT],NC,NU,NL,NS,NG}.
 
 apply_mlist(Subject,Replacement,Mlist) ->
-    do_mlist(Subject,Subject,0,precomp_repl(iolist_to_binary(Replacement)),
-	     Mlist).
+    do_mlist(Subject,Subject,0,precomp_repl(Replacement), Mlist).
 
 
 precomp_repl(<<>>) ->
diff --git a/lib/stdlib/test/re_SUITE.erl b/lib/stdlib/test/re_SUITE.erl
index f8667bfcee..e0e0670676 100644
--- a/lib/stdlib/test/re_SUITE.erl
+++ b/lib/stdlib/test/re_SUITE.erl
@@ -275,6 +275,7 @@ replace_input_types(Config) when is_list(Config) ->
     Dog = ?t:timetrap(?t:minutes(3)),
     ?line <<"abcd">> = re:replace("abcd","Z","X",[{return,binary},unicode]),
     ?line <<"abcd">> = re:replace("abcd","\x{400}","X",[{return,binary},unicode]),
+    ?line <<"a",208,128,"cd">> = re:replace(<<"abcd">>,"b","\x{400}",[{return,binary},unicode]),
     ?t:timetrap_cancel(Dog),
     ok.
 
-- 
cgit v1.2.3