diff options
| author | Rory Byrne <[email protected]> | 2010-01-16 14:47:43 +0000 |
|---|---|---|
| committer | Björn Gustavsson <[email protected]> | 2010-01-27 12:27:35 +0100 |
| commit | cf7e585bb45970fe0b5a8a6aa6653cd50583d052 (patch) | |
| tree | 4ebe87470ab3203029a9a7b5e8830569bbaf32b1 | |
| parent | 1d2a481cfd016f204183b44d8f95798161b423e3 (diff) | |
| download | otp-cf7e585bb45970fe0b5a8a6aa6653cd50583d052.tar.gz otp-cf7e585bb45970fe0b5a8a6aa6653cd50583d052.tar.bz2 otp-cf7e585bb45970fe0b5a8a6aa6653cd50583d052.zip | |
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.
| -rw-r--r-- | lib/stdlib/src/re.erl | 17 | ||||
| -rw-r--r-- | 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. |