-# Copyright (C) 2016-2019 all contributors <meta@public-inbox.org>
+# Copyright (C) 2016-2020 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 = $@;
($s, $err);
}
+# returns an array of quoted or unquoted sections
+sub split_quotes {
+ # Quiet "Complex regular subexpression recursion limit" warning
+ # in case an inconsiderate sender quotes 32K of text at once.
+ # The warning from Perl is harmless for us since our callers can
+ # tolerate less-than-ideal matches which work within Perl limits.
+ no warnings 'regexp';
+ split(/((?:^>[^\n]*\n)+)/sm, shift);
+}
+
1;