dovecot: If we encounter an unknown charset, get the ASCII (or U...

[dovecot at dovecot.org]

details:   http://hg.dovecot.org/dovecot/rev/aee731c09668
changeset: 6954:aee731c09668
user:      Timo Sirainen <tss at iki.fi>
date:      Sat Dec 08 15:59:48 2007 +0200
description:
If we encounter an unknown charset, get the ASCII (or UTF-8 actually)
characters instead of ignoring it completely.

diffstat:

1 file changed, 4 insertions(+), 1 deletion(-)
src/lib-mail/message-header-decode.c |    5 ++++-

diffs (15 lines):

diff -r edd296d164db -r aee731c09668 src/lib-mail/message-header-decode.c
--- a/src/lib-mail/message-header-decode.c	Sat Dec 08 15:55:09 2007 +0200
+++ b/src/lib-mail/message-header-decode.c	Sat Dec 08 15:59:48 2007 +0200
@@ -154,7 +154,10 @@ decode_utf8_callback(const unsigned char
 
 	flags = ctx->dtcase ? CHARSET_FLAG_DECOMP_TITLECASE : 0;
 	if (charset_to_utf8_begin(charset, flags, &t) < 0) {
-		/* let's just ignore this part */
+		/* data probably still contains some valid ASCII characters.
+		   append them. */
+		if (uni_utf8_get_valid_data(data, size, ctx->dest))
+			buffer_append(ctx->dest, data, size);
 		return TRUE;
 	}