X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FSearchMsg.pm;h=d43853a0ac6a5c5515a27950051ed02a1af3f65b;hb=dc03db7f4b5822b5763f0c078e6e9b8102cc5e17;hp=8d49ee207ab0b6e2ba2e5c03d13f39e2153278ec;hpb=1d236e649df10515bf042fa2283eef509648d9c9;p=public-inbox.git diff --git a/lib/PublicInbox/SearchMsg.pm b/lib/PublicInbox/SearchMsg.pm index 8d49ee20..d43853a0 100644 --- a/lib/PublicInbox/SearchMsg.pm +++ b/lib/PublicInbox/SearchMsg.pm @@ -1,27 +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; -use Email::Address qw//; -use Email::Simple qw//; -use POSIX qw//; -use Date::Parse qw/str2time/; -use PublicInbox::MID qw/mid_clean/; -use Encode qw/find_encoding/; -my $enc_utf8 = find_encoding('UTF-8'); -our $PFX2TERM_RE = undef; -use constant EPOCH_822 => 'Thu, 01 Jan 1970 00:00:00 +0000'; -use POSIX qw(strftime); +use PublicInbox::MID qw/mid_clean mid_mime/; +use PublicInbox::Address; +use PublicInbox::MsgTime qw(msg_timestamp msg_datestamp); sub new { my ($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 { @@ -29,32 +21,76 @@ sub wrap { bless { doc => $doc, mime => undef, mid => $mid }, $class; } +sub get { + my ($class, $head, $db, $mid) = @_; + my $doc_id = $head->get_docid; + my $doc = $db->get_document($doc_id); + load_expand(wrap($class, $doc, $mid)) +} + sub get_val ($$) { my ($doc, $col) = @_; Search::Xapian::sortable_unserialise($doc->get_value($col)); } +sub to_doc_data { + my ($self, $oid, $mid0) = @_; + $oid = '' unless defined $oid; + join("\n", + $self->subject, + $self->from, + $self->references, + $self->to, + $self->cc, + $oid, + $mid0, + $self->ds, + $self->{bytes}, + $self->{lines} + ); +} + + +sub load_from_data ($$) { + 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/, $_[1]); +} + +sub load_expand { + my ($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); + $self; +} + sub load_doc { my ($class, $doc) = @_; - my $data = $doc->get_data; - my $ts = get_val($doc, &PublicInbox::Search::TS); - $data = $enc_utf8->decode($data); - my ($subj, $from, $refs, $to, $cc) = split(/\n/, $data); - bless { - doc => $doc, - subject => $subj, - ts => $ts, - from => $from, - references => $refs, - to => $to, - cc => $cc, - }, $class; + my $self = bless { doc => $doc }, $class; + $self->load_expand; } # :bytes and :lines metadata in RFC 3977 -sub bytes ($) { get_val($_[0]->{doc}, &PublicInbox::Search::BYTES) } -sub lines ($) { get_val($_[0]->{doc}, &PublicInbox::Search::LINES) } -sub num ($) { get_val($_[0]->{doc}, &PublicInbox::Search::NUM) } +sub bytes ($) { $_[0]->{bytes} } +sub lines ($) { $_[0]->{lines} } +sub num ($) { $_[0]->{num} ||= _get_term_val($_[0], 'XNUM', qr/\AXNUM/) } sub __hdr ($$) { my ($self, $field) = @_; @@ -64,7 +100,8 @@ sub __hdr ($$) { my $mime = $self->{mime} or return; $val = $mime->header($field); $val = '' unless defined $val; - $val =~ tr/\t\r\n/ /; + $val =~ tr/\n/ /; + $val =~ tr/\r//d; $self->{$field} = $val; } @@ -72,22 +109,26 @@ sub subject ($) { __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 $date = __hdr($self, 'date'); - return $date if defined $date; - my $ts = $self->{ts}; - return unless defined $ts; - $self->{date} = strftime('%a, %d %b %Y %T %z', gmtime($ts)); + 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, $sec); + } sub from ($) { my ($self) = @_; my $from = __hdr($self, 'from'); if (defined $from && !defined $self->{from_name}) { - $from =~ tr/\t\r\n/ /; - my @from = Email::Address->parse($from); - $self->{from_name} = $from[0]->name; + my @n = PublicInbox::Address::names($from); + $self->{from_name} = join(', ', @n); } $from; } @@ -102,13 +143,12 @@ sub from_name { sub ts { my ($self) = @_; - $self->{ts} ||= eval { str2time($self->mime->header('Date')) } || 0; + $self->{ts} ||= eval { msg_timestamp($self->{mime}->header_obj) } || 0; } -sub to_doc_data { +sub ds { my ($self) = @_; - join("\n", $self->subject, $self->from, $self->references, - $self->to, $self->cc); + $self->{ds} ||= eval { msg_datestamp($self->{mime}->header_obj); } || 0; } sub references { @@ -117,61 +157,17 @@ sub references { defined $x ? $x : ''; } -sub ensure_metadata { - my ($self) = @_; +sub _get_term_val ($$$) { + my ($self, $pfx, $re) = @_; my $doc = $self->{doc}; my $end = $doc->termlist_end; - - unless (defined $PFX2TERM_RE) { - my $or = join('|', keys %PublicInbox::Search::PFX2TERM_RMAP); - $PFX2TERM_RE = qr/\A($or)/; - } - - while (my ($pfx, $field) = each %PublicInbox::Search::PFX2TERM_RMAP) { - # ideally we'd move this out of the loop: - my $i = $doc->termlist_begin; - - $i->skip_to($pfx); - if ($i != $end) { - my $val = $i->get_termname; - - if ($val =~ s/$PFX2TERM_RE//o) { - $self->{$field} = $val; - } - } + my $i = $doc->termlist_begin; + $i->skip_to($pfx); + if ($i != $end) { + my $val = $i->get_termname; + $val =~ s/$re// and return $val; } -} - -# for threading only -sub mini_mime { - my ($self) = @_; - $self->ensure_metadata; - my @hs = ( - 'Subject' => $self->subject, - 'X-PI-From' => $self->from_name, - ); - - my @h = ( - # prevent Email::Simple::Creator from running, - # this header is useless for threading as we use X-PI-TS - # for sorting and display: - 'Date' => EPOCH_822, - 'Message-ID' => "<$self->{mid}>", - 'X-PI-TS' => $self->ts, - ); - if (my $refs = $self->{references}) { - push @h, References => $refs; - } - my $mime = Email::MIME->create(header_str => \@hs, header => \@h); - my $h = $mime->header_obj; - - # set these headers manually since Encode::encode('MIME-Q', ...) - # will add spaces to long values when using header_str above. - - # drop useless headers Email::MIME set for us - $h->header_set('Date'); - $h->header_set('MIME-Version'); - $mime; + undef; } sub mid ($;$) { @@ -181,53 +177,13 @@ sub mid ($;$) { $self->{mid} = $mid; } elsif (my $rv = $self->{mid}) { $rv; + } elsif ($self->{doc}) { + $self->{mid} = _get_term_val($self, 'Q', qr/\AQ/); } else { - $self->ensure_metadata; # needed for ghosts - $self->{mid} ||= $self->_extract_mid; + $self->_extract_mid; # v1 w/o Xapian } } -sub _extract_mid { - my ($self) = @_; - - my $mid = $self->mime->header('Message-ID'); - defined $mid ? mid_clean($mid) : $mid; -} - -sub mime { - my ($self, $mime) = @_; - if (defined $mime) { - $self->{mime} = $mime; - } else { - # TODO load from git - $self->{mime}; - } -} - -sub doc_id { - my ($self, $doc_id) = @_; - if (defined $doc_id) { - $self->{doc_id} = $doc_id; - } else { - # TODO load from xapian - $self->{doc_id}; - } -} - -sub thread_id { - my ($self) = @_; - my $tid = $self->{thread}; - return $tid if defined $tid; - $self->ensure_metadata; - $self->{thread}; -} - -sub path { - my ($self) = @_; - my $path = $self->{path}; - return $path if defined $path; - $self->ensure_metadata; - $self->{path}; -} +sub _extract_mid { mid_clean(mid_mime($_[0]->{mime})) } 1;