[PATCH] extract redundant Message-ID handling code

[Eric Wong <e@80x24.org>]

Quit repeating ourselves and use a common MID module
instead.
---
 lib/PublicInbox/Hval.pm   | 13 +++----------
 lib/PublicInbox/MID.pm    | 24 ++++++++++++++++++++++++
 lib/PublicInbox/Search.pm | 31 ++++++-------------------------
 lib/PublicInbox/View.pm   |  8 +++-----
 4 files changed, 36 insertions(+), 40 deletions(-)
 create mode 100644 lib/PublicInbox/MID.pm

diff --git a/lib/PublicInbox/Hval.pm b/lib/PublicInbox/Hval.pm
index 68f8954..d8b31c8 100644
--- a/lib/PublicInbox/Hval.pm
+++ b/lib/PublicInbox/Hval.pm
@@ -8,6 +8,7 @@ use warnings;
 use fields qw(raw href);
 use Encode qw(find_encoding);
 use URI::Escape qw(uri_escape_utf8);
+use PublicInbox::MID qw/mid_clean mid_compressed/;
 
 my $enc_ascii = find_encoding('us-ascii');
 
@@ -25,16 +26,8 @@ sub new {
 
 sub new_msgid {
 	my ($class, $msgid) = @_;
-	$msgid =~ s/\A\s*<?//;
-	$msgid =~ s/>?\s*\z//;
-
-	if (length($msgid) <= 40) {
-		$class->new($msgid);
-	} else {
-		require Digest::SHA;
-		my $hex = Digest::SHA::sha1_hex($msgid);
-		$class->new($msgid, $hex);
-	}
+	$msgid = mid_clean($msgid);
+	$class->new($msgid, mid_compressed($msgid));
 }
 
 sub new_oneline {
diff --git a/lib/PublicInbox/MID.pm b/lib/PublicInbox/MID.pm
new file mode 100644
index 0000000..b56ce03
--- /dev/null
+++ b/lib/PublicInbox/MID.pm
@@ -0,0 +1,24 @@
+# Copyright (C) 2015, all contributors <meta@public-inbox.org>
+# License: AGPLv3 or later (https://www.gnu.org/licenses/agpl-3.0.txt)
+use base qw/Exporter/;
+our @EXPORT = qw/mid_clean mid_compressed/;
+use Digest::SHA qw/sha1_hex/;
+use constant MID_MAX => 40; # SHA-1 hex length
+
+sub mid_clean {
+	my ($mid) = @_;
+	defined($mid) or die "no Message-ID";
+	# MDA->precheck did more checking for us
+	$mid =~ s/\A\s*<?//;
+	$mid =~ s/>?\s*\z//;
+	$mid;
+}
+
+# this is idempotent
+sub mid_compressed {
+	my ($mid) = @_;
+	return $mid if (length($mid) <= MID_MAX);
+	sha1_hex($mid);
+}
+
+1;
diff --git a/lib/PublicInbox/Search.pm b/lib/PublicInbox/Search.pm
index fe4984e..328c9a2 100644
--- a/lib/PublicInbox/Search.pm
+++ b/lib/PublicInbox/Search.pm
@@ -6,25 +6,22 @@ use strict;
 use warnings;
 use PublicInbox::SearchMsg;
 use base qw/Exporter/;
-use Digest::SHA qw//;
 use Search::Xapian qw/:standard/;
 require PublicInbox::View;
 use Date::Parse qw/str2time/;
 use POSIX qw//;
 use Email::MIME;
+use PublicInbox::MID qw/mid_clean mid_compressed/;
 
-our @EXPORT = qw/xpfx mid_compressed/;
+our @EXPORT = qw/xpfx/;
 
 use constant {
 	TS => 0,
-	SHA1HEX_LEN => 40,
 	SCHEMA_VERSION => 0,
 	LANG => 'english',
 	QP_FLAGS => FLAG_PHRASE|FLAG_BOOLEAN|FLAG_LOVEHATE|FLAG_WILDCARD,
 };
 
-use constant MID_MAX => SHA1HEX_LEN;
-
 # setup prefixes
 my %bool_pfx_internal = (
 	type => 'T', # "mail" or "ghost"
@@ -54,13 +51,6 @@ while (my ($k, $v) = each %all_pfx) {
 
 my $mail_query = Search::Xapian::Query->new(xpfx('type') . 'mail');
 
-# this is idempotent
-sub mid_compressed {
-	my ($mid) = @_;
-	return $mid if (length($mid) <= MID_MAX);
-	Digest::SHA::sha1_hex($mid);
-}
-
 sub new {
 	my ($class, $git_dir, $writable) = @_;
 	# allow concurrent versions for easier rollback:
@@ -86,7 +76,7 @@ sub add_message {
 	my $db = $self->{xdb};
 
 	my $doc_id;
-	my $mid = clean_mid($mime->header('Message-ID'));
+	my $mid = mid_clean($mime->header('Message-ID'));
 	$mid = mid_compressed($mid);
 	my $was_ghost = 0;
 	my $ct_msg = $mime->header('Content-Type') || 'text/plain';
@@ -211,7 +201,7 @@ sub remove_message {
 	my ($self, $mid) = @_;
 	my $db = $self->{xdb};
 	my $doc_id;
-	$mid = clean_mid($mid);
+	$mid = mid_clean($mid);
 	$mid = mid_compressed($mid);
 
 	$db->begin_transaction;
@@ -242,7 +232,7 @@ sub query {
 # given a message ID, get replies to a message
 sub get_replies {
 	my ($self, $mid, $opts) = @_;
-	$mid = clean_mid($mid);
+	$mid = mid_clean($mid);
 	$mid = mid_compressed($mid);
 	my $qp = $self->qp;
 	my $irt = $qp->parse_query("inreplyto:$mid", 0);
@@ -345,15 +335,6 @@ sub date_range_processor {
 	$_[0]->{drp} ||= Search::Xapian::DateValueRangeProcessor->new(TS);
 }
 
-sub clean_mid {
-	my ($mid) = @_;
-	defined($mid) or die "no Message-ID";
-	# MDA->precheck did more checking for us
-	$mid =~ s/\A\s*<?//;
-	$mid =~ s/>?\s*\z//;
-	$mid;
-}
-
 sub link_message {
 	my ($self, $smsg, $is_ghost) = @_;
 
@@ -411,7 +392,7 @@ sub link_message_to_parents {
 
 sub lookup_message {
 	my ($self, $mid) = @_;
-	$mid = clean_mid($mid);
+	$mid = mid_clean($mid);
 	$mid = mid_compressed($mid);
 
 	my $doc_id = $self->find_unique_doc_id('mid', $mid);
diff --git a/lib/PublicInbox/View.pm b/lib/PublicInbox/View.pm
index 30759a3..c2dbb7e 100644
--- a/lib/PublicInbox/View.pm
+++ b/lib/PublicInbox/View.pm
@@ -3,11 +3,12 @@
 package PublicInbox::View;
 use strict;
 use warnings;
-use PublicInbox::Hval;
 use URI::Escape qw/uri_escape_utf8/;
 use Encode qw/find_encoding/;
 use Encode::MIME::Header;
 use Email::MIME::ContentType qw/parse_content_type/;
+use PublicInbox::Hval;
+use PublicInbox::MID qw/mid_clean mid_compressed/;
 require POSIX;
 
 # TODO: make these constants tunable
@@ -366,12 +367,9 @@ sub linkify_refs {
 	} @_);
 }
 
-require Digest::SHA;
 sub anchor_for {
 	my ($msgid) = @_;
-	$msgid =~ s/\A\s*<?//;
-	$msgid =~ s/>?\s*\z//;
-	'm' . Digest::SHA::sha1_hex($msgid);
+	'm' . mid_compressed(mid_clean($msgid));
 }
 
 1;
-- 
EW