[PATCH] search: index byte size of a message for IMAP search

[Eric Wong <e@yhbt.net>]

Searching for messages smaller than a certain size is allowed by
offlineimap(1), mbsync(1), and possibly other tools.  Maybe
public-inbox-watch will support it, too.

I don't see a reason to expose searching by size via WWW search
right now (but maybe in the future, I could be convinced to).

Note: we only store the byte-size of the message in git,
this is typically LF-only and we won't have the correct
size after CRLF conversion for NNTP or IMAP.

However, since most folks using tools like mbsync(1) and
offlineimap(1) would be on *nix systems where LF-only is
expected, I don't see the point of spending LoC or CPU cycles to
count bytes for CRLF on the wire.
---
 lib/PublicInbox/Search.pm    | 12 ++++++++----
 lib/PublicInbox/SearchIdx.pm |  2 ++
 t/search.t                   |  6 ++++++
 3 files changed, 16 insertions(+), 4 deletions(-)

diff --git a/lib/PublicInbox/Search.pm b/lib/PublicInbox/Search.pm
index cb669e8733e..f2d3b92dc82 100644
--- a/lib/PublicInbox/Search.pm
+++ b/lib/PublicInbox/Search.pm
@@ -5,12 +5,16 @@
 # Read-only search interface for use by the web and NNTP interfaces
 package PublicInbox::Search;
 use strict;
-use warnings;
 
 # values for searching
-use constant TS => 0;  # Received: header in Unix time
-use constant YYYYMMDD => 1; # Date: header for searching in the WWW UI
-use constant DT => 2; # Date: YYYYMMDDHHMMSS
+use constant {
+	TS => 0, # Received: header in Unix time (IMAP INTERNALDATE)
+	YYYYMMDD => 1, # Date: header for searching in the WWW UI
+	DT => 2, # Date: YYYYMMDDHHMMSS
+	BYTES => 3, # IMAP RFC822.SIZE
+	# TODO
+	# REPLYCNT => 4, # IMAP ANSWERED
+};
 
 use PublicInbox::Smsg;
 use PublicInbox::Over;
diff --git a/lib/PublicInbox/SearchIdx.pm b/lib/PublicInbox/SearchIdx.pm
index f10a9104e78..5c161b9accf 100644
--- a/lib/PublicInbox/SearchIdx.pm
+++ b/lib/PublicInbox/SearchIdx.pm
@@ -341,6 +341,7 @@ sub add_xapian ($$$$) {
 	add_val($doc, PublicInbox::Search::YYYYMMDD(), $yyyymmdd);
 	my $dt = strftime('%Y%m%d%H%M%S', @ds);
 	add_val($doc, PublicInbox::Search::DT(), $dt);
+	add_val($doc, PublicInbox::Search::BYTES(), $smsg->{bytes});
 
 	my $tg = term_generator($self);
 	$tg->set_document($doc);
@@ -388,6 +389,7 @@ sub add_message {
 
 	# v1 and tests only:
 	$smsg->populate($hdr, $self);
+	$smsg->{bytes} //= length($mime->as_string);
 
 	eval {
 		# order matters, overview stores every possible piece of
diff --git a/t/search.t b/t/search.t
index 6cf2bc2d6b4..cf3254169ca 100644
--- a/t/search.t
+++ b/t/search.t
@@ -318,6 +318,12 @@ $ibx->with_umask(sub {
 	foreach my $m ($mset->items) {
 		my $smsg = $ro->{over_ro}->get_art($m->get_docid);
 		like($smsg->{to}, qr/\blist\@example\.com\b/, 'to appears');
+		my $doc = $m->get_document;
+		my $col = PublicInbox::Search::BYTES();
+		my $bytes = PublicInbox::Smsg::get_val($doc, $col);
+		like($bytes, qr/\A[0-9]+\z/, '$bytes stored as digit');
+		ok($bytes > 0, '$bytes is > 0');
+		is($bytes, $smsg->{bytes}, 'bytes Xapian value matches Over');
 	}
 
 	$mset = $ro->query('tc:list@example.com', {mset => 1});