send-email: align RFC 2047 decoding more closely with the spec
More specifically: * Add "\" to the list of characters not allowed in a token (see RFC 2047 errata). * Share regexes between unquote_rfc2047 and is_rfc2047_quoted. Besides removing duplication, this also makes unquote_rfc2047 more stringent. * Allow both "q" and "Q" to identify the encoding. * Allow lowercase hexadecimal digits in the "Q" encoding. And, more on the cosmetic side: * Change the "encoded-text" regex to exclude rather than include characters, for clarity and consistency with "token". Signed-off-by: Роман Донченко <dpb@corrigendum.ru> Acked-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Роман Донченко committed
Dec 14, 2014 at 18:59 UTC
11f70a7e29ff7167e89aad186d763a4b017e48d9
1 file changed
+19
-11
git-send-email.perl
+19
-11
@@ -143,6 +143,11 @@ my $have_mail_address = eval { require Mail::Address; 1 };
143
my $smtp;
144
my $auth;
145
146
+# Regexes for RFC 2047 productions.
147
+my $re_token = qr/[^][()<>@,;:\\"\/?.= \000-\037\177-\377]+/;
148
+my $re_encoded_text = qr/[^? \000-\037\177-\377]+/;
149
+my $re_encoded_word = qr/=\?($re_token)\?($re_token)\?($re_encoded_text)\?=/;
150
+
151
# Variables we fill in automatically, or via prompting:
152
my (@to,$no_to,@initial_to,@cc,$no_cc,@initial_cc,@bcclist,$no_bcc,@xh,
153
$initial_reply_to,$initial_subject,@files,
@@ -906,15 +911,20 @@ $time = time - scalar $#files;
911
912
sub unquote_rfc2047 {
913
local ($_) = @_;
909
- my $encoding;
910
- s{=\?([^?]+)\?q\?(.*?)\?=}{
911
- $encoding = $1;
912
- my $e = $2;
913
- $e =~ s/_/ /g;
914
- $e =~ s/=([0-9A-F]{2})/chr(hex($1))/eg;
915
- $e;
914
+ my $charset;
915
+ s{$re_encoded_word}{
916
+ $charset = $1;
917
+ my $encoding = $2;
918
+ my $text = $3;
919
+ if ($encoding eq 'q' || $encoding eq 'Q') {
920
+ $text =~ s/_/ /g;
921
+ $text =~ s/=([0-9A-F]{2})/chr(hex($1))/egi;
922
+ $text;
923
+ } else {
924
+ $&; # other encodings not supported yet
925
+ }
926
}eg;
917
- return wantarray ? ($_, $encoding) : $_;
927
+ return wantarray ? ($_, $charset) : $_;
928
}
929
930
sub quote_rfc2047 {
@@ -927,10 +937,8 @@ sub quote_rfc2047 {
937
938
sub is_rfc2047_quoted {
939
my $s = shift;
930
- my $token = qr/[^][()<>@,;:"\/?.= \000-\037\177-\377]+/;
931
- my $encoded_text = qr/[!->@-~]+/;
940
length($s) <= 75 &&
933
- $s =~ m/^(?:"[[:ascii:]]*"|=\?$token\?$token\?$encoded_text\?=)$/o;
941
+ $s =~ m/^(?:"[[:ascii:]]*"|$re_encoded_word)$/o;
942
}
943
944
sub subject_needs_rfc2047_quoting {