From: Alexander Moisseev <moiseev@mezonplus.ru>
Date: Mon, 13 Apr 2026 15:39:24 +0000 (+0300)
Subject: [Fix] Skip ICU conversion for x-binaryenc charset in all detection paths
X-Git-Url: http://git.ipfire.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=0ef6ab9b33273a698fa836cddcbe0f22c80c9ae6;p=thirdparty%2Frspamd.git

[Fix] Skip ICU conversion for x-binaryenc charset in all detection paths

x-binaryenc is a synthetic name returned by CED (Google Compact
Encoding Detection) to signal binary content, not a real text
encoding.  ICU has no converter for it and always fails with
U_FILE_ACCESS_ERROR, producing a misleading warning in the logs.

Add an early exit in rspamd_mime_text_part_maybe_convert() for all
three detection paths: announced-charset missing, announced-charset
unknown, and rspamd_mime_charset_utf_check content-heuristic. When
the detected charset is x-binaryenc, mark the part as raw binary
immediately, consistent with what the existing fallback already does.

Extract set_part_binary() helper and RSPAMD_BINARYENC_CHARSET constant
to eliminate the resulting code duplication.

Fixes the spurious "cannot open converter for x-binaryenc" warning
seen when processing messages with binary MIME parts.
---

diff --git a/src/libmime/mime_encoding.c b/src/libmime/mime_encoding.c
index 995706d10c..b24d9c036e 100644
--- a/src/libmime/mime_encoding.c
+++ b/src/libmime/mime_encoding.c
@@ -31,6 +31,7 @@
 #include <math.h>
 
 #define UTF8_CHARSET "UTF-8"
+#define RSPAMD_BINARYENC_CHARSET "x-binaryenc"
 
 #define RSPAMD_CHARSET_FLAG_UTF (1 << 0)
 #define RSPAMD_CHARSET_FLAG_ASCII (1 << 1)
@@ -725,6 +726,18 @@ rspamd_mime_charset_utf_check(rspamd_ftok_t *charset,
 	return FALSE;
 }
 
+static void
+set_part_binary(struct rspamd_task *task,
+				struct rspamd_mime_text_part *text_part,
+				GByteArray *part_content,
+				const char *charset)
+{
+	msg_debug_task("text part contains binary data (detected charset: %s), skip conversion",
+				   charset);
+	SET_PART_RAW(text_part);
+	text_part->utf_raw_content = part_content;
+}
+
 void rspamd_mime_text_part_maybe_convert(struct rspamd_task *task,
 										 struct rspamd_mime_text_part *text_part)
 {
@@ -771,6 +784,10 @@ void rspamd_mime_text_part_maybe_convert(struct rspamd_task *task,
 																	  text_part->parsed.len);
 
 			if (charset != NULL) {
+				if (g_ascii_strcasecmp(charset, RSPAMD_BINARYENC_CHARSET) == 0) {
+					set_part_binary(task, text_part, part_content, charset);
+					return;
+				}
 				msg_info_task("detected charset %s", charset);
 			}
 
@@ -794,6 +811,10 @@ void rspamd_mime_text_part_maybe_convert(struct rspamd_task *task,
 			if (need_charset_heuristic) {
 				charset = rspamd_mime_charset_find_by_content_maybe_split(part_content->data,
 																		  part_content->len);
+				if (charset != NULL && g_ascii_strcasecmp(charset, RSPAMD_BINARYENC_CHARSET) == 0) {
+					set_part_binary(task, text_part, part_content, charset);
+					return;
+				}
 				msg_info_task("detected charset: %s", charset);
 				checked = TRUE;
 				text_part->real_charset = charset;
@@ -840,6 +861,11 @@ void rspamd_mime_text_part_maybe_convert(struct rspamd_task *task,
 		else {
 			charset = charset_tok.begin;
 
+			if (g_ascii_strcasecmp(charset, RSPAMD_BINARYENC_CHARSET) == 0) {
+				set_part_binary(task, text_part, part_content, charset);
+				return;
+			}
+
 			if (!rspamd_mime_text_part_utf8_convert(task, text_part,
 													part_content, charset, &err)) {
 				msg_warn_task("<%s>: cannot convert from %s to utf8: %s",