dovecot-2.0: message_header_decode*() should ignore LWSP between...

[dovecot at dovecot.org]

details:   http://hg.dovecot.org/dovecot-2.0/rev/9444d7964766
changeset: 9934:9444d7964766
user:      Timo Sirainen <tss at iki.fi>
date:      Sun Sep 13 19:01:01 2009 -0400
description:
message_header_decode*() should ignore LWSP between two encoded-words.

diffstat:

3 files changed, 77 insertions(+), 1 deletion(-)
src/lib-mail/Makefile.am                  |    5 ++
src/lib-mail/message-header-decode.c      |   15 +++++++
src/lib-mail/test-message-header-decode.c |   58 +++++++++++++++++++++++++++++

diffs (116 lines):

diff -r 27904121ae72 -r 9444d7964766 src/lib-mail/Makefile.am
--- a/src/lib-mail/Makefile.am	Sun Sep 13 18:37:05 2009 -0400
+++ b/src/lib-mail/Makefile.am	Sun Sep 13 19:01:01 2009 -0400
@@ -58,6 +58,7 @@ test_programs = \
 	test-message-address \
 	test-message-date \
 	test-message-decoder \
+	test-message-header-decode \
 	test-message-header-parser \
 	test-message-id \
 	test-message-parser \
@@ -94,6 +95,10 @@ test_message_decoder_LDADD = message-dec
 test_message_decoder_LDADD = message-decoder.lo rfc822-parser.lo rfc2231-parser.lo $(test_libs)
 test_message_decoder_DEPENDENCIES = message-decoder.lo rfc822-parser.lo rfc2231-parser.lo $(test_libs)
 
+test_message_header_decode_SOURCES = test-message-header-decode.c
+test_message_header_decode_LDADD = message-header-decode.lo quoted-printable.lo $(test_libs)
+test_message_header_decode_DEPENDENCIES = message-header-decode.lo quoted-printable.lo $(test_libs)
+
 test_message_header_parser_SOURCES = test-message-header-parser.c
 test_message_header_parser_LDADD = message-header-parser.lo $(test_libs)
 test_message_header_parser_DEPENDENCIES = message-header-parser.lo $(test_libs)
diff -r 27904121ae72 -r 9444d7964766 src/lib-mail/message-header-decode.c
--- a/src/lib-mail/message-header-decode.c	Sun Sep 13 18:37:05 2009 -0400
+++ b/src/lib-mail/message-header-decode.c	Sun Sep 13 19:01:01 2009 -0400
@@ -56,6 +56,18 @@ message_header_decode_encoded(const unsi
 	return start_pos[2] + 2;
 }
 
+static bool is_only_lwsp(const unsigned char *data, unsigned int size)
+{
+	unsigned int i;
+
+	for (i = 0; i < size; i++) {
+		if (!(data[i] == ' ' || data[i] == '\t' ||
+		      data[i] == '\r' || data[i] == '\n'))
+			return FALSE;
+	}
+	return TRUE;
+}
+
 void message_header_decode(const unsigned char *data, size_t size,
 			   message_header_decode_callback_t *callback,
 			   void *context)
@@ -73,7 +85,8 @@ void message_header_decode(const unsigne
 		}
 
 		/* encoded string beginning */
-		if (pos != start_pos) {
+		if (pos != start_pos &&
+		    !is_only_lwsp(data+start_pos, pos-start_pos)) {
 			/* send the unencoded data so far */
 			if (!callback(data + start_pos, pos - start_pos,
 				      NULL, context)) {
diff -r 27904121ae72 -r 9444d7964766 src/lib-mail/test-message-header-decode.c
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/src/lib-mail/test-message-header-decode.c	Sun Sep 13 19:01:01 2009 -0400
@@ -0,0 +1,58 @@
+/* Copyright (c) 2009 Dovecot authors, see the included COPYING file */
+
+#include "lib.h"
+#include "buffer.h"
+#include "str.h"
+#include "charset-utf8.h"
+#include "message-header-decode.h"
+#include "test-common.h"
+
+bool charset_is_utf8(const char *charset ATTR_UNUSED) { return TRUE; }
+
+int charset_to_utf8_begin(const char *charset ATTR_UNUSED,
+			  enum charset_flags flags ATTR_UNUSED,
+			  struct charset_translation **t_r ATTR_UNUSED) { return 0; }
+void charset_to_utf8_end(struct charset_translation **t ATTR_UNUSED) {}
+
+enum charset_result
+charset_to_utf8(struct charset_translation *t ATTR_UNUSED,
+		const unsigned char *src, size_t *src_size, buffer_t *dest)
+{
+	buffer_append(dest, src, *src_size);
+	return CHARSET_RET_OK;
+}
+
+static void test_message_header_decode(void)
+{
+	static const char *data[] = {
+		"a =?utf-8?q?=c3=a4?= b", "a ä b",
+		"a =?utf-8?q?=c3=a4?= b", "a ä b",
+		"a =?utf-8?q?=c3=a4?=\t\t\r\n =?utf-8?q?=c3=a4?= b", "a ää b",
+		"a =?utf-8?q?=c3=a4?=  x  =?utf-8?q?=c3=a4?= b", "a ä  x  ä b",
+		"a =?utf-8?b?w6TDpCDDpA==?= b", "a ää ä b",
+		"=?utf-8?b?w6Qgw6Q=?=", "ä ä",
+	};
+	string_t *dest;
+	unsigned int i;
+
+	test_begin("message header decode");
+
+	dest = t_str_new(256);
+	for (i = 0; i < N_ELEMENTS(data); i += 2) {
+		str_truncate(dest, 0);
+		test_assert(message_header_decode_utf8((const unsigned char *)data[i],
+						       strlen(data[i]),
+						       dest, FALSE));
+		test_assert(strcmp(str_c(dest), data[i+1]) == 0);
+	}
+	test_end();
+}
+
+int main(void)
+{
+	static void (*test_functions[])(void) = {
+		test_message_header_decode,
+		NULL
+	};
+	return test_run(test_functions);
+}