Rewrote the Characer encoding section

1 files changed, 7 insertions, 9 deletions

diff --git a/InternetSecure.pm b/InternetSecure.pm
index c7f71b7..8590efb 100755
--- a/InternetSecure.pm
+++ b/InternetSecure.pm
@@ -617,15 +617,13 @@ be left undefined.
 
 =head2 Character encoding
 
-Since communication to/from InternetSecure is encoded with UTF-8, all Unicode
-characters are theoretically available when submitting information via
-B<submit>().  (Further restrictions may be imposed by InternetSecure itself.)
-
-When using non-ASCII characters, all data provided to B<submit>() should either
-be in the current native encoding (typically latin-1, unless it was modified
-via the C<encoding> pragma), or be decoded via the C<Encode> module.
-Conversely, all data returned after calling B<submit>() will be automatically
-decoded.
+When using non-ASCII characters, all data provided to B<contents>() should
+have been decoded beforehand via the C<Encode> module, unless your data is in
+ISO-8859-1 and you haven't meddled with the C<encoding> pragma.  (Please
+don't.)
+
+InternetSecure currently does not handle characters outside of ISO-8859-1, so
+these will be replaced with C<?> before being transmitted.
 
 
 =head1 EXPORT