X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FSearchMsg.pm;h=d43853a0ac6a5c5515a27950051ed02a1af3f65b;hb=dc03db7f4b5822b5763f0c078e6e9b8102cc5e17;hp=a9f3180b827e9c8aac8351671dce988d9b08459e;hpb=73214b02e581ef0a037d85a1ae37ff425654d7b3;p=public-inbox.git diff --git a/lib/PublicInbox/SearchMsg.pm b/lib/PublicInbox/SearchMsg.pm index a9f3180b..d43853a0 100644 --- a/lib/PublicInbox/SearchMsg.pm +++ b/lib/PublicInbox/SearchMsg.pm @@ -1,25 +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 mid_compress/; -use Encode qw/find_encoding/; -my $enc_utf8 = find_encoding('UTF-8'); -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 { @@ -27,44 +21,116 @@ 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; - $data = $enc_utf8->decode($data); - my ($mid, $subj, $from, $date, $refs) = split(/\n/, $data); - bless { - doc => $doc, - mid => $mid, - subject => $subj, - date => $date, - from_name => $from, - references_sorted => $refs, - }, $class; -} - -sub subject { + 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 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 $subj = $self->{subject}; - return $subj if defined $subj; - $subj = $self->{mime}->header('Subject'); - $subj = '' unless defined $subj; - $subj =~ tr/\n/ /; - $self->{subject} = $subj; + 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 { +sub from ($) { my ($self) = @_; - my $from = $self->mime->header('From') || ''; - my @from; - - if ($from) { - $from =~ tr/\n/ /; - @from = Email::Address->parse($from); - $self->{from} = $from[0]; - $from = $from[0]->name; + my $from = __hdr($self, 'from'); + if (defined $from && !defined $self->{from_name}) { + my @n = PublicInbox::Address::names($from); + $self->{from_name} = join(', ', @n); } - $self->{from_name} = $from; - $self->{from}; + $from; } sub from_name { @@ -77,127 +143,47 @@ sub from_name { sub ts { my ($self) = @_; - my $ts = $self->{ts}; - return $ts if $ts; - $self->{ts} = eval { - str2time($self->date || $self->mime->header('Date')) - } || 0; -} - -sub date { - my ($self) = @_; - my $date = $self->{date}; - return $date if $date; - my $ts = eval { str2time($self->mime->header('Date')) }; - $self->{date} = POSIX::strftime('%Y-%m-%d %H:%M', gmtime($ts)); + $self->{ts} ||= eval { msg_timestamp($self->{mime}->header_obj) } || 0; } -sub to_doc_data { +sub ds { my ($self) = @_; - $self->mid . "\n" . - PublicInbox::Search::subject_summary($self->subject) . "\n" . - $self->from_name . "\n". - $self->date . "\n" . - $self->references_sorted; + $self->{ds} ||= eval { msg_datestamp($self->{mime}->header_obj); } || 0; } -sub references_sorted { +sub references { my ($self) = @_; - my $x = $self->{references_sorted}; + my $x = $self->{references}; defined $x ? $x : ''; } -sub ensure_metadata { - my ($self) = @_; +sub _get_term_val ($$$) { + my ($self, $pfx, $re) = @_; my $doc = $self->{doc}; - my $i = $doc->termlist_begin; my $end = $doc->termlist_end; - - unless (defined $PFX2TERM_RE) { - my $or = join('|', keys %PublicInbox::Search::PFX2TERM_RMAP); - $PFX2TERM_RE = qr/\A($or)/; - } - - for (; $i != $end; $i->inc) { + my $i = $doc->termlist_begin; + $i->skip_to($pfx); + if ($i != $end) { my $val = $i->get_termname; - - if ($val =~ s/$PFX2TERM_RE//o) { - my $field = $PublicInbox::Search::PFX2TERM_RMAP{$1}; - $self->{$field} = $val; - } + $val =~ s/$re// and return $val; } + undef; } -# for threading only -sub mini_mime { - my ($self) = @_; - $self->ensure_metadata; - my @h = ( - Subject => $self->subject, - 'X-PI-From' => $self->from_name, - 'X-PI-Date' => $self->date, - 'X-PI-TS' => $self->ts, - 'Message-ID' => "<$self->{mid}>", - ); - - my $refs = $self->{references_sorted}; - my $mime = Email::MIME->create(header_str => \@h); - $mime->header_set('References', $refs) if (defined $refs); - $mime; -} - -sub mid { +sub mid ($;$) { my ($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->ensure_metadata; # needed for ghosts - $self->{mid} ||= $self->_extract_mid; - } -} - -sub _extract_mid { - my ($self) = @_; - - my $mid = $self->mime->header('Message-ID'); - $mid ? mid_compress(mid_clean($mid)) : $mid; -} - -sub mime { - my ($self, $mime) = @_; - if (defined $mime) { - $self->{mime} = $mime; - } else { - # TODO load from git - $self->{mime}; + $self->_extract_mid; # v1 w/o Xapian } } -sub doc_id { - my ($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;