X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FMID.pm;h=2c9822f478da32879866065c21a1f1f247297827;hb=3d41aa23f35501ca92aab8aa42980fa73f7fa74f;hp=677a9d59e2f4727937eaf815509ee6e59ec2bf50;hpb=f76f265a851944b5dedcc3be5f3b5224b6ebda89;p=public-inbox.git
diff --git a/lib/PublicInbox/MID.pm b/lib/PublicInbox/MID.pm
index 677a9d59..2c9822f4 100644
--- a/lib/PublicInbox/MID.pm
+++ b/lib/PublicInbox/MID.pm
@@ -1,10 +1,13 @@
-# Copyright (C) 2015 all contributors
-# License: AGPLv3 or later (https://www.gnu.org/licenses/agpl-3.0.txt)
+# Copyright (C) 2015-2018 all contributors
+# License: AGPL-3.0+
+#
+# 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/;
+use URI::Escape qw(uri_escape_utf8);
use Digest::SHA qw/sha1_hex/;
use constant MID_MAX => 40; # SHA-1 hex length
@@ -19,18 +22,14 @@ 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 +38,18 @@ 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";
}
+sub mid_mime ($) { $_[0]->header_obj->header_raw('Message-ID') }
+
+# RFC3986, section 3.3:
+sub MID_ESC () { '^A-Za-z0-9\-\._~!\$\&\';\(\)\*\+,;=:@' }
+sub mid_escape ($) { uri_escape_utf8($_[0], MID_ESC) }
+
1;