X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FSearchMsg.pm;h=d43853a0ac6a5c5515a27950051ed02a1af3f65b;hb=dc03db7f4b5822b5763f0c078e6e9b8102cc5e17;hp=920804ac65d70084c21d3400e1452f1497e89e95;hpb=675494c58ef7a39a92c79cbf02975b9da3991c0b;p=public-inbox.git diff --git a/lib/PublicInbox/SearchMsg.pm b/lib/PublicInbox/SearchMsg.pm index 920804ac..d43853a0 100644 --- a/lib/PublicInbox/SearchMsg.pm +++ b/lib/PublicInbox/SearchMsg.pm @@ -1,18 +1,19 @@ -# Copyright (C) 2015, all contributors -# License: GPLv3 or later (https://www.gnu.org/licenses/agpl-3.0.txt) +# Copyright (C) 2015-2018 all contributors +# License: AGPL-3.0+ # based on notmuch, but with no concept of folders, files or flags +# +# Wraps a document inside our Xapian search index. package PublicInbox::SearchMsg; use strict; use warnings; -use Search::Xapian; -our $PFX2TERM_RE = undef; +use PublicInbox::MID qw/mid_clean mid_mime/; +use PublicInbox::Address; +use PublicInbox::MsgTime qw(msg_timestamp msg_datestamp); sub new { my ($class, $mime) = @_; my $doc = Search::Xapian::Document->new; - $doc->add_term(PublicInbox::Search::xpfx('type') . 'mail'); - - bless { type => 'mail', doc => $doc, mime => $mime }, $class; + bless { doc => $doc, mime => $mime }, $class; } sub wrap { @@ -20,86 +21,169 @@ sub wrap { bless { doc => $doc, mime => undef, mid => $mid }, $class; } -sub ensure_metadata { - my ($self) = @_; - my $doc = $self->{doc}; - my $i = $doc->termlist_begin; - my $end = $doc->termlist_end; +sub get { + my ($class, $head, $db, $mid) = @_; + my $doc_id = $head->get_docid; + my $doc = $db->get_document($doc_id); + load_expand(wrap($class, $doc, $mid)) +} - unless (defined $PFX2TERM_RE) { - my $or = join('|', keys %PublicInbox::Search::PFX2TERM_RMAP); - $PFX2TERM_RE = qr/\A($or)/; - } +sub get_val ($$) { + my ($doc, $col) = @_; + Search::Xapian::sortable_unserialise($doc->get_value($col)); +} - for (; $i != $end; $i->inc) { - my $val = $i->get_termname; +sub to_doc_data { + my ($self, $oid, $mid0) = @_; + $oid = '' unless defined $oid; + join("\n", + $self->subject, + $self->from, + $self->references, + $self->to, + $self->cc, + $oid, + $mid0, + $self->ds, + $self->{bytes}, + $self->{lines} + ); +} - if ($val =~ s/$PFX2TERM_RE//o) { - my $field = $PublicInbox::Search::PFX2TERM_RMAP{$1}; - if ($field eq 'references') { - my $refs = $self->{references} ||= []; - push @$refs, $val; - } else { - $self->{$field} = $val; - } - } - } + +sub load_from_data ($$) { + my ($self) = $_[0]; # data = $_[1] + ( + $self->{subject}, + $self->{from}, + $self->{references}, + + # To: and Cc: are stored to optimize HDR/XHDR in NNTP since + # some NNTP clients will use that for message displays. + $self->{to}, + $self->{cc}, + + $self->{blob}, + $self->{mid}, + $self->{ds}, + $self->{bytes}, + $self->{lines} + ) = split(/\n/, $_[1]); } -sub mid { - my ($self, $mid) = @_; +sub load_expand { + my ($self) = @_; + my $doc = $self->{doc}; + my $data = $doc->get_data or return; + $self->{ts} = get_val($doc, &PublicInbox::Search::TS); + utf8::decode($data); + load_from_data($self, $data); + $self; +} - if (defined $mid) { - $self->{mid} = $mid; - } else { - $self->{mid} ||= $self->_extract_mid; - } +sub load_doc { + my ($class, $doc) = @_; + my $self = bless { doc => $doc }, $class; + $self->load_expand; +} + +# :bytes and :lines metadata in RFC 3977 +sub bytes ($) { $_[0]->{bytes} } +sub lines ($) { $_[0]->{lines} } +sub num ($) { $_[0]->{num} ||= _get_term_val($_[0], 'XNUM', qr/\AXNUM/) } + +sub __hdr ($$) { + my ($self, $field) = @_; + my $val = $self->{$field}; + return $val if defined $val; + + my $mime = $self->{mime} or return; + $val = $mime->header($field); + $val = '' unless defined $val; + $val =~ tr/\n/ /; + $val =~ tr/\r//d; + $self->{$field} = $val; } -sub _extract_mid { +sub subject ($) { __hdr($_[0], 'subject') } +sub to ($) { __hdr($_[0], 'to') } +sub cc ($) { __hdr($_[0], 'cc') } + +# no strftime, that is locale-dependent and not for RFC822 +my @DoW = qw(Sun Mon Tue Wed Thu Fri Sat); +my @MoY = qw(Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec); + +sub date ($) { my ($self) = @_; + my $ds = $self->{ds}; + return unless defined $ds; + my ($sec, $min, $hour, $mday, $mon, $year, $wday) = gmtime($ds); + "$DoW[$wday], " . sprintf("%02d $MoY[$mon] %04d %02d:%02d:%02d +0000", + $mday, $year+1900, $hour, $min, $sec); - my $mid = $self->mime->header('Message-ID'); - if ($mid && $mid =~ /<([^>]+)>/) { - return $1; - } - return $mid; } -sub mime { - my ($self, $mime) = @_; - if (defined $mime) { - $self->{mime} = $mime; - } else { - # TODO load from git - $self->{mime}; +sub from ($) { + my ($self) = @_; + my $from = __hdr($self, 'from'); + if (defined $from && !defined $self->{from_name}) { + my @n = PublicInbox::Address::names($from); + $self->{from_name} = join(', ', @n); } + $from; } -sub doc_id { - my ($self, $doc_id) = @_; - if (defined $doc_id) { - $self->{doc_id} = $doc_id; - } else { - # TODO load from xapian - $self->{doc_id}; - } +sub from_name { + my ($self) = @_; + my $from_name = $self->{from_name}; + return $from_name if defined $from_name; + $self->from; + $self->{from_name}; +} + +sub ts { + my ($self) = @_; + $self->{ts} ||= eval { msg_timestamp($self->{mime}->header_obj) } || 0; } -sub thread_id { +sub ds { my ($self) = @_; - my $tid = $self->{thread}; - return $tid if defined $tid; - $self->ensure_metadata; - $self->{thread}; + $self->{ds} ||= eval { msg_datestamp($self->{mime}->header_obj); } || 0; } -sub path { +sub references { my ($self) = @_; - my $path = $self->{path}; - return $path if defined $path; - $self->ensure_metadata; - $self->{path}; + my $x = $self->{references}; + defined $x ? $x : ''; } +sub _get_term_val ($$$) { + my ($self, $pfx, $re) = @_; + my $doc = $self->{doc}; + my $end = $doc->termlist_end; + my $i = $doc->termlist_begin; + $i->skip_to($pfx); + if ($i != $end) { + my $val = $i->get_termname; + $val =~ s/$re// and return $val; + } + undef; +} + +sub mid ($;$) { + my ($self, $mid) = @_; + + if (defined $mid) { + $self->{mid} = $mid; + } elsif (my $rv = $self->{mid}) { + $rv; + } elsif ($self->{doc}) { + $self->{mid} = _get_term_val($self, 'Q', qr/\AQ/); + } else { + $self->_extract_mid; # v1 w/o Xapian + } +} + +sub _extract_mid { mid_clean(mid_mime($_[0]->{mime})) } + 1;