use strict;
use warnings;
use PublicInbox::MIME;
-use Email::MIME::ContentType;
-$Email::MIME::ContentType::STRICT_PARAMS = 0; # user input is imperfect
use PublicInbox::Git;
use PublicInbox::Import;
use PublicInbox::MDA;
use PublicInbox::Spawn qw(spawn);
+use PublicInbox::InboxWritable;
use File::Temp qw//;
+use PublicInbox::Filter::Base;
+*REJECT = *PublicInbox::Filter::Base::REJECT;
sub new {
my ($class, $config) = @_;
$spamcheck = undef;
}
}
+
+ # need to make all inboxes writable for spam removal:
+ $config->each_inbox(sub { PublicInbox::InboxWritable->new($_[0]) });
+
foreach $k (keys %$config) {
$k =~ /\Apublicinbox\.([^\.]+)\.watch\z/ or next;
my $name = $1;
my ($self) = @_;
my $importers = $self->{importers};
foreach my $im (values %$importers) {
- $im->done if $im->{nchg};
- }
-
- my $opendirs = $self->{opendirs};
-
- # spamdir scanning means every importer remains open
- my $spamdir = $self->{spamdir};
- return if defined($spamdir) && $opendirs->{$spamdir};
-
- foreach my $im (values %$importers) {
- # not done if we're scanning
- next if $opendirs->{$im->{git}->{git_dir}};
$im->done;
}
}
# path must be marked as (S)een
$path =~ /:2,[A-R]*S[T-Za-z]*\z/ or return;
my $mime = _path_to_mime($path) or return;
- _force_mid($mime);
$self->{config}->each_inbox(sub {
my ($ibx) = @_;
eval {
my $im = _importer_for($self, $ibx);
$im->remove($mime, 'spam');
- if (my $scrub = _scrubber_for($ibx)) {
- my $scrubbed = $scrub->scrub($mime) or return;
- $scrubbed == 100 and return;
+ if (my $scrub = $ibx->filter) {
+ my $scrubbed = $scrub->scrub($mime, 1);
+ $scrubbed or return;
+ $scrubbed == REJECT() and return;
$im->remove($scrubbed, 'spam');
}
};
})
}
-# used to hash the relevant portions of a message when there are conflicts
-sub _hash_mime2 {
- my ($mime) = @_;
- require Digest::SHA;
- my $dig = Digest::SHA->new('SHA-1');
- $dig->add($mime->header_obj->header_raw('Subject'));
- $dig->add($mime->body_raw);
- $dig->hexdigest;
-}
-
-sub _force_mid {
- my ($mime) = @_;
- # probably a bad idea, but we inject a Message-Id if
- # one is missing, here..
- my $mid = $mime->header_obj->header_raw('Message-Id');
- if (!defined $mid || $mid =~ /\A\s*\z/) {
- $mid = '<' . _hash_mime2($mime) . '@generated>';
- $mime->header_set('Message-Id', $mid);
- }
-}
-
sub _try_path {
my ($self, $path) = @_;
- my @p = split(m!/+!, $path);
- return if $p[-1] !~ /\A[a-zA-Z0-9][\-\w:,=\.]+\z/;
- if ($p[-1] =~ /:2,([A-Z]+)\z/i) {
- my $flags = $1;
- return if $flags =~ /[DT]/; # no [D]rafts or [T]rashed mail
- }
- return unless -f $path;
+ return unless PublicInbox::InboxWritable::is_maildir_path($path);
if ($path !~ $self->{mdre}) {
warn "unrecognized path: $path\n";
return;
}
my $im = _importer_for($self, $inbox);
my $mime = _path_to_mime($path) or return;
- $mime->header_set($_) foreach @PublicInbox::MDA::BAD_HEADERS;
my $wm = $inbox->{-watchheader};
if ($wm) {
my $v = $mime->header_obj->header_raw($wm->[0]);
return unless ($v && $v =~ $wm->[1]);
}
- if (my $scrub = _scrubber_for($inbox)) {
+ if (my $scrub = $inbox->filter) {
my $ret = $scrub->scrub($mime) or return;
- $ret == 100 and return;
+ $ret == REJECT() and return;
$mime = $ret;
}
- _force_mid($mime);
$im->add($mime, $self->{spamcheck});
}
}
sub _importer_for {
- my ($self, $inbox) = @_;
- my $im = $inbox->{-import} ||= eval {
- my $git = $inbox->git;
- my $name = $inbox->{name};
- my $addr = $inbox->{-primary_address};
- PublicInbox::Import->new($git, $name, $addr, $inbox);
- };
-
+ my ($self, $ibx) = @_;
my $importers = $self->{importers};
+ my $im = $importers->{"$ibx"} ||= $ibx->importer(0);
if (scalar(keys(%$importers)) > 2) {
- delete $importers->{"$im"};
+ delete $importers->{"$ibx"};
_done_for_now($self);
}
- $importers->{"$im"} = $im;
-}
-
-sub _scrubber_for {
- my ($inbox) = @_;
- my $f = $inbox->{filter};
- if ($f && $f =~ /::/) {
- my @args = (-inbox => $inbox);
- # basic line splitting, only
- # Perhaps we can have proper quote splitting one day...
- ($f, @args) = split(/\s+/, $f) if $f =~ /\s+/;
-
- eval "require $f";
- if ($@) {
- warn $@;
- } else {
- # e.g: PublicInbox::Filter::Vger->new(@args)
- return $f->new(@args);
- }
- }
- undef;
+ $importers->{"$ibx"} = $im;
}
sub _spamcheck_cb {