X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FSearch.pm;h=25ab8d52718950482847f048631820be4dec98b0;hb=38c481a5438593cff686709493a70b8a6b3033d1;hp=c61d4cf649559d5c46bf081041e9a459fb54f5ec;hpb=746323cbde3f5da5d66efe742b30894b0d806ffe;p=public-inbox.git diff --git a/lib/PublicInbox/Search.pm b/lib/PublicInbox/Search.pm index c61d4cf6..25ab8d52 100644 --- a/lib/PublicInbox/Search.pm +++ b/lib/PublicInbox/Search.pm @@ -1,13 +1,23 @@ -# Copyright (C) 2015, all contributors +# Copyright (C) 2015 all contributors # License: AGPLv3 or later (https://www.gnu.org/licenses/agpl-3.0.txt) # based on notmuch, but with no concept of folders, files or flags +# +# Read-only search interface for use by the web and NNTP interfaces package PublicInbox::Search; use strict; use warnings; -use PublicInbox::SearchMsg; + +# values for searching +use constant TS => 0; # timestamp +use constant NUM => 1; # NNTP article number +use constant BYTES => 2; # :bytes as defined in RFC 3977 +use constant LINES => 3; # :lines as defined in RFC 3977 +use constant YYYYMMDD => 4; # for searching in the WWW UI + use Search::Xapian qw/:standard/; -use Email::MIME; -use PublicInbox::MID qw/mid_clean mid_compressed/; +use PublicInbox::SearchMsg; +use PublicInbox::MIME; +use PublicInbox::MID qw/mid_clean id_compress/; # This is English-only, everything else is non-standard and may be confused as # a prefix common in patch emails @@ -15,47 +25,99 @@ our $REPLY_RE = qr/^re:\s+/i; our $LANG = 'english'; use constant { - TS => 0, # SCHEMA_VERSION history # 0 - initial # 1 - subject_path is lower-cased - # 2 - subject_path is mid_compressed in the index, only + # 2 - subject_path is id_compress in the index, only # 3 - message-ID is compressed if it includes '%' (hack!) # 4 - change "Re: " normalization, avoid circular Reference ghosts # 5 - subject_path drops trailing '.' # 6 - preserve References: order in document data - SCHEMA_VERSION => 6, + # 7 - remove references and inreplyto terms + # 8 - remove redundant/unneeded document data + # 9 - disable Message-ID compression (SHA-1) + # 10 - optimize doc for NNTP overviews + # 11 - merge threads when vivifying ghosts + # 12 - change YYYYMMDD value column to numeric + # 13 - fix threading for empty References/In-Reply-To + # (commit 83425ef12e4b65cdcecd11ddcb38175d4a91d5a0) + # 14 - fix ghost root vivification + SCHEMA_VERSION => 14, + + # n.b. FLAG_PURE_NOT is expensive not suitable for a public website + # as it could become a denial-of-service vector QP_FLAGS => FLAG_PHRASE|FLAG_BOOLEAN|FLAG_LOVEHATE|FLAG_WILDCARD, }; # setup prefixes my %bool_pfx_internal = ( type => 'T', # "mail" or "ghost" - mid => 'Q', # uniQue id (Message-ID or mid_compressed) + thread => 'G', # newsGroup (or similar entity - e.g. a web forum name) ); my %bool_pfx_external = ( - path => 'XPATH', - thread => 'G', # newsGroup (or similar entity - e.g. a web forum name) - references => 'XREFS', - inreplyto => 'XIRT', + mid => 'Q', # uniQue id (Message-ID) ); my %prob_prefix = ( - subject => 'S', + # for mairix compatibility + s => 'S', + m => 'XMID', # 'mid:' (bool) is exact, 'm:' (prob) can do partial + f => 'A', + t => 'XTO', + tc => 'XTO XCC', + c => 'XCC', + tcf => 'XTO XCC A', + a => 'XTO XCC A', + b => 'XNQ XQUOT', + bs => 'XNQ XQUOT S', + n => 'XFN', + + q => 'XQUOT', + nq => 'XNQ', + dfn => 'XDFN', + dfa => 'XDFA', + dfb => 'XDFB', + dfhh => 'XDFHH', + dfctx => 'XDFCTX', + dfpre => 'XDFPRE', + dfpost => 'XDFPOST', + dfblob => 'XDFPRE XDFPOST', + + # default: + '' => 'XMID S A XNQ XQUOT XFN', ); -my %all_pfx = (%bool_pfx_internal, %bool_pfx_external, %prob_prefix); - -sub xpfx { $all_pfx{$_[0]} } - -our %PFX2TERM_RMAP; -my %meta_pfx = (mid => 1, thread => 1, path => 1, type => 1); -while (my ($k, $v) = each %all_pfx) { - $PFX2TERM_RMAP{$v} = $k if $meta_pfx{$k}; -} +# not documenting m: and mid: for now, the using the URLs works w/o Xapian +our @HELP = ( + 's:' => 'match within Subject e.g. s:"a quick brown fox"', + 'd:' => < 'match within message body, including text attachments', + 'nq:' => 'match non-quoted text within message body', + 'q:' => 'match quoted text within message body', + 'n:' => 'match filename of attachment(s)', + 't:' => 'match within the To header', + 'c:' => 'match within the Cc header', + 'f:' => 'match within the From header', + 'a:' => 'match within the To, Cc, and From headers', + 'tc:' => 'match within the To and Cc headers', + 'bs:' => 'match within the Subject and body', + 'dfn:' => 'match filename from diff', + 'dfa:' => 'match diff removed (-) lines', + 'dfb:' => 'match diff added (+) lines', + 'dfhh:' => 'match diff hunk header context (usually a function name)', + 'dfctx:' => 'match diff context lines', + 'dfpre:' => 'match pre-image git blob ID', + 'dfpost:' => 'match post-image git blob ID', + 'dfblob:' => 'match either pre or post-image git blob ID', +); +chomp @HELP; -my $mail_query = Search::Xapian::Query->new(xpfx('type') . 'mail'); +my $mail_query = Search::Xapian::Query->new('T' . 'mail'); sub xdir { my (undef, $git_dir) = @_; @@ -63,10 +125,10 @@ sub xdir { } sub new { - my ($class, $git_dir) = @_; + my ($class, $git_dir, $altid) = @_; my $dir = $class->xdir($git_dir); my $db = Search::Xapian::Database->new($dir); - bless { xdb => $db, git_dir => $git_dir }, $class; + bless { xdb => $db, git_dir => $git_dir, altid => $altid }, $class; } sub reopen { $_[0]->{xdb}->reopen } @@ -74,27 +136,15 @@ sub reopen { $_[0]->{xdb}->reopen } # read-only sub query { my ($self, $query_string, $opts) = @_; - my $query = $self->qp->parse_query($query_string, QP_FLAGS); - - $self->do_enquire($query, $opts); -} + my $query; -sub get_subject_path { - my ($self, $path, $opts) = @_; - my $query = $self->qp->parse_query("path:".mid_compressed($path), 0); - $self->do_enquire($query, $opts); -} + $opts ||= {}; + unless ($query_string eq '') { + $query = $self->qp->parse_query($query_string, QP_FLAGS); + $opts->{relevance} = 1 unless exists $opts->{relevance}; + } -# given a message ID, get followups to a message -sub get_followups { - my ($self, $mid, $opts) = @_; - $mid = mid_clean($mid); - $mid = mid_compressed($mid); - my $qp = $self->qp; - my $irt = $qp->parse_query("inreplyto:$mid", 0); - my $ref = $qp->parse_query("references:$mid", 0); - my $query = Search::Xapian::Query->new(OP_OR, $irt, $ref); - $self->do_enquire($query, $opts); + _do_enquire($self, $query, $opts); } sub get_thread { @@ -102,26 +152,66 @@ sub get_thread { my $smsg = eval { $self->lookup_message($mid) }; return { total => 0, msgs => [] } unless $smsg; - my $qp = $self->qp; - my $qtid = $qp->parse_query('thread:'.$smsg->thread_id, 0); - my $qsub = $qp->parse_query('path:'.mid_compressed($smsg->path), 0); - my $query = Search::Xapian::Query->new(OP_OR, $qtid, $qsub); - $self->do_enquire($query, $opts); + my $qtid = Search::Xapian::Query->new('G' . $smsg->thread_id); + my $path = $smsg->path; + if (defined $path && $path ne '') { + my $path = id_compress($smsg->path); + my $qsub = Search::Xapian::Query->new('XPATH' . $path); + $qtid = Search::Xapian::Query->new(OP_OR, $qtid, $qsub); + } + $opts ||= {}; + $opts->{limit} ||= 1000; + + # always sort threads by timestamp, this makes life easier + # for the threading algorithm (in SearchThread.pm) + $opts->{asc} = 1; + + _do_enquire($self, $qtid, $opts); +} + +sub retry_reopen { + my ($self, $cb) = @_; + my $ret; + for (1..10) { + eval { $ret = $cb->() }; + return $ret unless $@; + # Exception: The revision being read has been discarded - + # you should call Xapian::Database::reopen() + if (ref($@) eq 'Search::Xapian::DatabaseModifiedError') { + reopen($self); + } else { + die; + } + } } -# private subs below +sub _do_enquire { + my ($self, $query, $opts) = @_; + retry_reopen($self, sub { _enquire_once($self, $query, $opts) }); +} -sub do_enquire { +sub _enquire_once { my ($self, $query, $opts) = @_; my $enquire = $self->enquire; - - $query = Search::Xapian::Query->new(OP_AND, $query, $mail_query); + if (defined $query) { + $query = Search::Xapian::Query->new(OP_AND,$query,$mail_query); + } else { + $query = $mail_query; + } $enquire->set_query($query); - $enquire->set_sort_by_relevance_then_value(TS, 0); $opts ||= {}; + my $desc = !$opts->{asc}; + if ($opts->{relevance}) { + $enquire->set_sort_by_relevance_then_value(TS, $desc); + } elsif ($opts->{num}) { + $enquire->set_sort_by_value(NUM, 0); + } else { + $enquire->set_sort_by_value_then_relevance(TS, $desc); + } my $offset = $opts->{offset} || 0; my $limit = $opts->{limit} || 50; my $mset = $enquire->get_mset($offset, $limit); + return $mset if $opts->{mset}; my @msgs = map { PublicInbox::SearchMsg->load_doc($_->get_document); } $mset->items; @@ -145,48 +235,79 @@ sub qp { $qp->set_database($self->{xdb}); $qp->set_stemmer($self->stemmer); $qp->set_stemming_strategy(STEM_SOME); - $qp->add_valuerangeprocessor($self->ts_range_processor); - $qp->add_valuerangeprocessor($self->date_range_processor); + $qp->add_valuerangeprocessor( + Search::Xapian::NumberValueRangeProcessor->new(YYYYMMDD, 'd:')); while (my ($name, $prefix) = each %bool_pfx_external) { $qp->add_boolean_prefix($name, $prefix); } + # we do not actually create AltId objects, + # just parse the spec to avoid the extra DB handles for now. + if (my $altid = $self->{altid}) { + my $user_pfx = $self->{-user_pfx} ||= []; + for (@$altid) { + # $_ = 'serial:gmane:/path/to/gmane.msgmap.sqlite3' + /\Aserial:(\w+):/ or next; + my $pfx = $1; + push @$user_pfx, "$pfx:", < XGMANE + $qp->add_boolean_prefix($pfx, 'X'.uc($pfx)); + } + chomp @$user_pfx; + } + while (my ($name, $prefix) = each %prob_prefix) { - $qp->add_prefix($name, $prefix); + $qp->add_prefix($name, $_) foreach split(/ /, $prefix); } $self->{query_parser} = $qp; } -sub ts_range_processor { - $_[0]->{tsrp} ||= Search::Xapian::NumberValueRangeProcessor->new(TS); +sub num_range_processor { + $_[0]->{nrp} ||= Search::Xapian::NumberValueRangeProcessor->new(NUM); } -sub date_range_processor { - $_[0]->{drp} ||= Search::Xapian::DateValueRangeProcessor->new(TS); +# only used for NNTP server +sub query_xover { + my ($self, $beg, $end, $offset) = @_; + my $qp = Search::Xapian::QueryParser->new; + $qp->set_database($self->{xdb}); + $qp->add_valuerangeprocessor($self->num_range_processor); + my $query = $qp->parse_query("$beg..$end", QP_FLAGS); + + _do_enquire($self, $query, {num => 1, limit => 200, offset => $offset}); } sub lookup_message { my ($self, $mid) = @_; $mid = mid_clean($mid); - $mid = mid_compressed($mid); - my $doc_id = $self->find_unique_doc_id('mid', $mid); + my $doc_id = $self->find_unique_doc_id('Q' . $mid); my $smsg; if (defined $doc_id) { # raises on error: my $doc = $self->{xdb}->get_document($doc_id); $smsg = PublicInbox::SearchMsg->wrap($doc, $mid); - $smsg->doc_id($doc_id); + $smsg->{doc_id} = $doc_id; } $smsg; } +sub lookup_mail { # no ghosts! + my ($self, $mid) = @_; + retry_reopen($self, sub { + my $smsg = lookup_message($self, $mid) or return; + $smsg->load_expand; + }); +} + sub find_unique_doc_id { - my ($self, $term, $value) = @_; + my ($self, $termval) = @_; - my ($begin, $end) = $self->find_doc_ids($term, $value); + my ($begin, $end) = $self->find_doc_ids($termval); return undef if $begin->equal($end); # not found @@ -194,26 +315,20 @@ sub find_unique_doc_id { # sanity check $begin->inc; - $begin->equal($end) or die "Term '$term:$value' is not unique\n"; + $begin->equal($end) or die "Term '$termval' is not unique\n"; $rv; } # returns begin and end PostingIterator sub find_doc_ids { - my ($self, $term, $value) = @_; - - $self->find_doc_ids_for_term(xpfx($term) . $value); -} - -# returns begin and end PostingIterator -sub find_doc_ids_for_term { - my ($self, $term) = @_; + my ($self, $termval) = @_; my $db = $self->{xdb}; - ($db->postlist_begin($term), $db->postlist_end($term)); + ($db->postlist_begin($termval), $db->postlist_end($termval)); } # normalize subjects so they are suitable as pathnames for URLs +# XXX: consider for removal sub subject_path { my $subj = pop; $subj = subject_normalized($subj); @@ -236,4 +351,14 @@ sub enquire { $self->{enquire} ||= Search::Xapian::Enquire->new($self->{xdb}); } +sub help { + my ($self) = @_; + $self->qp; # parse altids + my @ret = @HELP; + if (my $user_pfx = $self->{-user_pfx}) { + push @ret, @$user_pfx; + } + \@ret; +} + 1;