about summary refs log tree commit homepage
path: root/lib/PublicInbox/Filter
diff options
context:
space:
mode:
Diffstat (limited to 'lib/PublicInbox/Filter')
-rw-r--r--lib/PublicInbox/Filter/Base.pm2
-rw-r--r--lib/PublicInbox/Filter/Gmane.pm2
-rw-r--r--lib/PublicInbox/Filter/Mirror.pm2
-rw-r--r--lib/PublicInbox/Filter/RubyLang.pm63
-rw-r--r--lib/PublicInbox/Filter/SubjectTag.pm2
-rw-r--r--lib/PublicInbox/Filter/Vger.pm8
6 files changed, 47 insertions, 32 deletions
diff --git a/lib/PublicInbox/Filter/Base.pm b/lib/PublicInbox/Filter/Base.pm
index d54570fd..f6355e1b 100644
--- a/lib/PublicInbox/Filter/Base.pm
+++ b/lib/PublicInbox/Filter/Base.pm
@@ -1,4 +1,4 @@
-# Copyright (C) 2016-2020 all contributors <meta@public-inbox.org>
+# Copyright (C) 2016-2021 all contributors <meta@public-inbox.org>
 # License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
 #
 # base class for creating per-list or per-project filters
diff --git a/lib/PublicInbox/Filter/Gmane.pm b/lib/PublicInbox/Filter/Gmane.pm
index c326faca..a18b77d2 100644
--- a/lib/PublicInbox/Filter/Gmane.pm
+++ b/lib/PublicInbox/Filter/Gmane.pm
@@ -1,4 +1,4 @@
-# Copyright (C) 2018-2020 all contributors <meta@public-inbox.org>
+# Copyright (C) 2018-2021 all contributors <meta@public-inbox.org>
 # License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
 
 # Filter for importing some archives from gmane
diff --git a/lib/PublicInbox/Filter/Mirror.pm b/lib/PublicInbox/Filter/Mirror.pm
index 9f6dd342..fe915fc3 100644
--- a/lib/PublicInbox/Filter/Mirror.pm
+++ b/lib/PublicInbox/Filter/Mirror.pm
@@ -1,4 +1,4 @@
-# Copyright (C) 2016-2020 all contributors <meta@public-inbox.org>
+# Copyright (C) 2016-2021 all contributors <meta@public-inbox.org>
 # License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
 
 # Dumb filter for blindly accepting everything
diff --git a/lib/PublicInbox/Filter/RubyLang.pm b/lib/PublicInbox/Filter/RubyLang.pm
index a65a5971..57ebbe78 100644
--- a/lib/PublicInbox/Filter/RubyLang.pm
+++ b/lib/PublicInbox/Filter/RubyLang.pm
@@ -1,11 +1,10 @@
-# Copyright (C) 2017-2020 all contributors <meta@public-inbox.org>
+# Copyright (C) all contributors <meta@public-inbox.org>
 # License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
 
 # Filter for lists.ruby-lang.org trailers
 package PublicInbox::Filter::RubyLang;
-use base qw(PublicInbox::Filter::Base);
-use strict;
-use warnings;
+use v5.10.1;
+use parent qw(PublicInbox::Filter::Base);
 use PublicInbox::MID qw(mids);
 
 my $l1 = qr/Unsubscribe:\s
@@ -16,7 +15,7 @@ sub new {
         my ($class, %opts) = @_;
         my $altid = delete $opts{-altid};
         my $self = $class->SUPER::new(%opts);
-        my $ibx = $self->{-inbox};
+        my $ibx = $self->{ibx};
         # altid = serial:ruby-core:file=msgmap.sqlite3
         if (!$altid && $ibx && $ibx->{altid}) {
                 $altid ||= $ibx->{altid}->[0];
@@ -28,34 +27,50 @@ sub new {
         $self;
 }
 
+sub scrub_part ($) {
+        my ($part) = @_;
+        my $ct = $part->content_type;
+        if (!$ct || $ct =~ m{\btext/plain\b}i) {
+                my $s = eval { $part->body_str };
+                if (defined $s && $s =~ s/\n?$l1\n$l2\n\z//os) {
+                        $part->body_str_set($s);
+                        return 1;
+                }
+        }
+        0;
+}
+
 sub scrub {
         my ($self, $mime, $for_remove) = @_;
-        # no msg_iter here, that is only for read-only access
-        $mime->walk_parts(sub {
-                my ($part) = $_[0];
-                my $ct = $part->content_type;
-                if (!$ct || $ct =~ m{\btext/plain\b}i) {
-                        my $s = eval { $part->body_str };
-                        if (defined $s && $s =~ s/\n?$l1\n$l2\n\z//os) {
-                                $part->body_str_set($s);
-                        }
-                }
-        });
+        # no msg_iter here, msg_iter is only for read-only access
+        if (my @sub = $mime->subparts) {
+                my $changed = 0;
+                $changed |= scrub_part($_) for @sub;
+                $mime->parts_set(\@sub) if $changed;
+        } else {
+                scrub_part($mime);
+        }
         my $altid = $self->{-altid};
         if ($altid && !$for_remove) {
                 my $hdr = $mime->header_obj;
                 my $mids = mids($hdr);
                 return $self->REJECT('Message-ID missing') unless (@$mids);
-                my @v = $hdr->header_raw('X-Mail-Count');
                 my $n;
-                foreach (@v) {
-                        /\A\s*([0-9]+)\s*\z/ or next;
-                        $n = $1;
-                        last;
-                }
-                unless (defined $n) {
-                        return $self->REJECT('X-Mail-Count not numeric');
+                my @v = $hdr->header_raw('X-Mail-Count'); # old host only
+                if (@v) {
+                        for (@v) {
+                                /\A\s*([0-9]+)\s*\z/ or next;
+                                $n = $1;
+                                last;
+                        }
+                } else { # new host: nue.mailmanlists.eu
+                        for ($hdr->header_str('Subject')) {
+                                /\A\[ruby-[^:]+:([0-9]+)\]/ or next;
+                                $n = $1;
+                                last;
+                        }
                 }
+                $n // return $self->REJECT('could not get count not numeric');
                 foreach my $mid (@$mids) {
                         my $r = $altid->mm_alt->mid_set($n, $mid);
                         next if $r == 0;
diff --git a/lib/PublicInbox/Filter/SubjectTag.pm b/lib/PublicInbox/Filter/SubjectTag.pm
index aca6688b..ecedf666 100644
--- a/lib/PublicInbox/Filter/SubjectTag.pm
+++ b/lib/PublicInbox/Filter/SubjectTag.pm
@@ -1,4 +1,4 @@
-# Copyright (C) 2017-2020 all contributors <meta@public-inbox.org>
+# Copyright (C) 2017-2021 all contributors <meta@public-inbox.org>
 # License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
 
 # Filter for various [tags] in subjects
diff --git a/lib/PublicInbox/Filter/Vger.pm b/lib/PublicInbox/Filter/Vger.pm
index e746238c..5b3c0277 100644
--- a/lib/PublicInbox/Filter/Vger.pm
+++ b/lib/PublicInbox/Filter/Vger.pm
@@ -1,11 +1,11 @@
-# Copyright (C) 2016-2020 all contributors <meta@public-inbox.org>
+# Copyright (C) 2016-2021 all contributors <meta@public-inbox.org>
 # License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
 
 # Filter for vger.kernel.org list trailer
 package PublicInbox::Filter::Vger;
 use base qw(PublicInbox::Filter::Base);
 use strict;
-use warnings;
+use PublicInbox::Eml;
 
 my $l0 = qr/-+/; # older messages only had one '-'
 my $l1 =
@@ -24,8 +24,8 @@ sub scrub {
         # the vger appender seems to only work on the raw string,
         # so in multipart (e.g. GPG-signed) messages, the list trailer
         # becomes invisible to MIME-aware email clients.
-        if ($s =~ s/$l0\n$l1\n$l2\n$l3\n($l4\n)?\z//os) {
-                $mime = PublicInbox::MIME->new(\$s);
+        if ($s =~ s/$l0\n$l1\n$l2\n$l3\n(?:$l4\n)?\n*\z//os) {
+                $mime = PublicInbox::Eml->new(\$s);
         }
         $self->ACCEPT($mime);
 }