- parser.c parserInternals.c encoding.c: Since Notepad on Win2k
outputs a BOM in UTF8, an errata has been issued to avoid the
problem, that was the most reasonable solution... Add support
for a leading UTF8 BOM in entities.
Daniel
diff --git a/encoding.c b/encoding.c
index f86adf3..df760f7 100644
--- a/encoding.c
+++ b/encoding.c
@@ -1131,6 +1131,15 @@
(in[2] == 0x78) && (in[3] == 0x6D))
return(XML_CHAR_ENCODING_UTF8);
}
+ if (len >= 3) {
+ /*
+ * Errata on XML-1.0 June 20 2001
+ * We now allow an UTF8 encoded BOM
+ */
+ if ((in[0] == 0xEF) && (in[1] == 0xBB) &&
+ (in[2] == 0xBF))
+ return(XML_CHAR_ENCODING_UTF8);
+ }
if (len >= 2) {
if ((in[0] == 0xFE) && (in[1] == 0xFF))
return(XML_CHAR_ENCODING_UTF16BE);