-# Copyright (C) 2016-2018 all contributors <meta@public-inbox.org>
+# Copyright (C) 2016-2019 all contributors <meta@public-inbox.org>
# License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
# read-only utilities for Email::MIME
# Like Email::MIME::walk_parts, but this is:
# * non-recursive
# * passes depth and indices to the iterator callback
-sub msg_iter ($$) {
- my ($mime, $cb) = @_;
+sub msg_iter ($$;$$) {
+ my ($mime, $cb, $cb_arg, $do_undef) = @_;
my @parts = $mime->subparts;
if (@parts) {
+ $mime = $_[0] = undef if $do_undef; # saves some memory
my $i = 0;
@parts = map { [ $_, 1, ++$i ] } @parts;
while (my $p = shift @parts) {
@sub = map { [ $_, $depth, @idx, ++$i ] } @sub;
@parts = (@sub, @parts);
} else {
- $cb->($p);
+ $cb->($p, $cb_arg);
}
}
} else {
- $cb->([$mime, 0, 0]);
+ $cb->([$mime, 0, 0], $cb_arg);
}
}
sub msg_part_text ($$) {
my ($part, $ct) = @_;
+ # TODO: we may offer a separate sub for people who need to index
+ # HTML-only mail, but the majority of HTML mail is multipart/alternative
+ # with a text part which we don't have to waste cycles decoding
+ return if $ct =~ m!\btext/x?html\b!;
+
my $s = eval { $part->body_str };
my $err = $@;
# times when it should not have been:
# <87llgalspt.fsf@free.fr>
# <200308111450.h7BEoOu20077@mail.osdl.org>
- if ($ct =~ m!\btext/plain\b!i || $ct =~ m!\bmultipart/mixed\b!i) {
- # Try to assume UTF-8 because Alpine seems to
- # do wacky things and set charset=X-UNKNOWN
- $part->charset_set('UTF-8');
- $s = eval { $part->body_str };
+ if ($err && ($ct =~ m!\btext/\b!i ||
+ $ct =~ m!\bmultipart/mixed\b!i)) {
+ my $cte = $part->header_raw('Content-Transfer-Encoding');
+ if (defined($cte) && $cte =~ /\b7bit\b/i) {
+ $s = $part->body;
+ $err = undef if $s =~ /\A[[:ascii:]]+\z/s;
+ } else {
+ # Try to assume UTF-8 because Alpine seems to
+ # do wacky things and set charset=X-UNKNOWN
+ $part->charset_set('UTF-8');
+ $s = eval { $part->body_str };
+ }
# If forcing charset=UTF-8 failed,
# caller will warn further down...