Don't put UTF8 last in the list of encodings

Sometimes, BIG5 will be selected for english text if quotes make it
not representable in ASCII. See [1] for the original bug report.

This default list is arguably European-centric but at least it documents a good
amount of the alternative encodings.

[1]: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=659920

Reported-by: James Cloos <cloos@jhcloos.com>
Signed-off-by: Etienne Millon <me@emillon.org>

diff --git a/rss2email/config.py b/rss2email/config.py
index d8dc0731ecd9ee6d9e32a711f0de9287ac0b4740..afe2725560f96706eeb617c67935faf2843d9cf9 100644 (file)
--- a/rss2email/config.py
+++ b/rss2email/config.py
@@ -89,9 +89,8 @@ CONFIG['DEFAULT'] = _collections.OrderedDict((
         ('trust-guid', str(True)),
         # To most correctly encode emails with international
         # characters, we iterate through the list below and use the
-        # first character set that works Eventually (and
-        # theoretically) UTF-8 is our catch-all failsafe.
-        ('encodings', 'US-ASCII, BIG5, ISO-2022-JP, ISO-8859-1, UTF-8'),
+        # first character set that works.
+        ('encodings', 'US-ASCII, ISO-8859-1, UTF-8, BIG5, ISO-2022-JP'),
         ## HTML conversion
         # True: Send text/html messages when possible.
         # False: Convert HTML to plain text.