X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FMID.pm;h=c82e84013ec0d68730423ab4b691074c5b690dc5;hb=1cb230c4d47c8ef5d03f0b8de2f8ad93c303d816;hp=1b03fdce72e1333132f0884371754b290ac6ed24;hpb=1b41e3f4500541667b3b205f3d8683b4e1532384;p=public-inbox.git

diff --git a/lib/PublicInbox/MID.pm b/lib/PublicInbox/MID.pm
index 1b03fdce..c82e8401 100644
--- a/lib/PublicInbox/MID.pm
+++ b/lib/PublicInbox/MID.pm
@@ -1,12 +1,19 @@
-# Copyright (C) 2015, all contributors <meta@public-inbox.org>
-# License: AGPLv3 or later (https://www.gnu.org/licenses/agpl-3.0.txt)
+# Copyright (C) 2015-2018 all contributors <meta@public-inbox.org>
+# License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
+#
+# Various Message-ID-related functions.
 package PublicInbox::MID;
 use strict;
 use warnings;
 use base qw/Exporter/;
-our @EXPORT_OK = qw/mid_clean mid_compress mid2path/;
+our @EXPORT_OK = qw/mid_clean id_compress mid2path mid_mime mid_escape MID_ESC
+	mids references/;
+use URI::Escape qw(uri_escape_utf8);
 use Digest::SHA qw/sha1_hex/;
-use constant MID_MAX => 40; # SHA-1 hex length
+use constant {
+	MID_MAX => 40, # SHA-1 hex length # TODO: get rid of this
+	MAX_MID_SIZE => 244, # max term size (Xapian limitation) - length('Q')
+};
 
 sub mid_clean {
 	my ($mid) = @_;
@@ -19,18 +26,14 @@ sub mid_clean {
 }
 
 # this is idempotent
-sub mid_compress {
-	my ($mid, $force) = @_;
-
-	# XXX dirty hack! FIXME!
-	# Some HTTP servers (apache2 2.2.22-13+deb7u5 on my system)
-	# apparently do not handle "%25" in the URL path component correctly.
-	# I'm not yet sure if it's something weird with my rewrite rules
-	# or what; will need to debug...
-	return sha1_hex($mid) if (index($mid, '%') >= 0);
-
-	return $mid if (!$force && length($mid) <= MID_MAX);
-	sha1_hex($mid);
+sub id_compress {
+	my ($id, $force) = @_;
+
+	if ($force || $id =~ /[^\w\-]/ || length($id) > MID_MAX) {
+		utf8::encode($id);
+		return sha1_hex($id);
+	}
+	$id;
 }
 
 sub mid2path {
@@ -39,10 +42,64 @@ sub mid2path {
 
 	unless (defined $x38) {
 		# compatibility with old links (or short Message-IDs :)
-		$mid = sha1_hex(mid_clean($mid));
+		$mid = mid_clean($mid);
+		utf8::encode($mid);
+		$mid = sha1_hex($mid);
 		($x2, $x38) = ($mid =~ /\A([a-f0-9]{2})([a-f0-9]{38})\z/);
 	}
 	"$x2/$x38";
 }
 
+# Only for v1 code paths:
+sub mid_mime ($) { mids($_[0]->header_obj)->[0] }
+
+sub mids ($) {
+	my ($hdr) = @_;
+	my @mids;
+	my @v = $hdr->header_raw('Message-Id');
+	foreach my $v (@v) {
+		my @cur = ($v =~ /<([^>]+)>/sg);
+		if (@cur) {
+			push(@mids, @cur);
+		} else {
+			push(@mids, $v);
+		}
+	}
+	uniq_mids(\@mids);
+}
+
+# last References should be IRT, but some mail clients do things
+# out of order, so trust IRT over References iff IRT exists
+sub references ($) {
+	my ($hdr) = @_;
+	my @mids;
+	foreach my $f (qw(References In-Reply-To)) {
+		my @v = $hdr->header_raw($f);
+		foreach my $v (@v) {
+			push(@mids, ($v =~ /<([^>]+)>/sg));
+		}
+	}
+	uniq_mids(\@mids);
+}
+
+sub uniq_mids ($) {
+	my ($mids) = @_;
+	my @ret;
+	my %seen;
+	foreach my $mid (@$mids) {
+		if (length($mid) > MAX_MID_SIZE) {
+			warn "Message-ID: <$mid> too long, truncating\n";
+			$mid = substr($mid, 0, MAX_MID_SIZE);
+		}
+		next if $seen{$mid};
+		push @ret, $mid;
+		$seen{$mid} = 1;
+	}
+	\@ret;
+}
+
+# RFC3986, section 3.3:
+sub MID_ESC () { '^A-Za-z0-9\-\._~!\$\&\';\(\)\*\+,;=:@' }
+sub mid_escape ($) { uri_escape_utf8($_[0], MID_ESC) }
+
 1;