X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FSearch.pm;h=24cb26670754b2f4a21b62e7d22c3cdc76098a5a;hb=6e0659d2ae455fb28d433fd8c71b563210a3997a;hp=862ed6ddd41b9f5c282bc19efac0400fbcea4111;hpb=5c3f7d5f9a03056da6740af523f22903880cc9ec;p=public-inbox.git diff --git a/lib/PublicInbox/Search.pm b/lib/PublicInbox/Search.pm index 862ed6dd..b739faf1 100644 --- a/lib/PublicInbox/Search.pm +++ b/lib/PublicInbox/Search.pm @@ -1,550 +1,433 @@ -# Copyright (C) 2015, all contributors -# License: AGPLv3 or later (https://www.gnu.org/licenses/agpl-3.0.txt) +# Copyright (C) 2015-2020 all contributors +# License: AGPL-3.0+ # 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; -use Search::Xapian qw/:standard/; -require PublicInbox::View; -use Email::MIME; -use PublicInbox::MID qw/mid_clean mid_compressed/; +use parent qw(Exporter); +our @EXPORT_OK = qw(mdocid); +# values for searching, changing the numeric value breaks +# compatibility with old indices (so don't change them it) use constant { - TS => 0, + TS => 0, # Received: header in Unix time (IMAP INTERNALDATE) + YYYYMMDD => 1, # Date: header for searching in the WWW UI + DT => 2, # Date: YYYYMMDDHHMMSS + + # added for public-inbox 1.6.0+ + BYTES => 3, # IMAP RFC822.SIZE + UID => 4, # IMAP UID == NNTP article number == Xapian docid + THREADID => 5, # RFC 8474, RFC 8621 + + # TODO + # REPLYCNT => ?, # IMAP ANSWERED + # SCHEMA_VERSION history # 0 - initial # 1 - subject_path is lower-cased - # 2 - subject_path is mid_compressed in the index, only - SCHEMA_VERSION => 2, - LANG => 'english', - QP_FLAGS => FLAG_PHRASE|FLAG_BOOLEAN|FLAG_LOVEHATE|FLAG_WILDCARD, + # 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 + # 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 + # 15 - see public-inbox-v2-format(5) + # further bumps likely unnecessary, we'll suggest in-place + # "--reindex" use for further fixes and tweaks: + # + # public-inbox v1.5.0 adds (still SCHEMA_VERSION=15): + # * "lid:" and "l:" for List-Id searches + # + # v1.6.0 adds BYTES, UID and THREADID values + SCHEMA_VERSION => 15, }; -# setup prefixes -my %bool_pfx_internal = ( - type => 'T', # "mail" or "ghost" - mid => 'Q', # uniQue id (Message-ID or mid_compressed) -); +use PublicInbox::Smsg; +use PublicInbox::Over; +my $QP_FLAGS; +our %X = map { $_ => 0 } qw(BoolWeight Database Enquire QueryParser Stem); +our $Xap; # 'Search::Xapian' or 'Xapian' +my $NVRP; # '$Xap::'.('NumberValueRangeProcessor' or 'NumberRangeProcessor') +my $ENQ_ASCENDING; + +sub load_xapian () { + return 1 if defined $Xap; + # n.b. PI_XAPIAN is intended for development use only. We still + # favor Search::Xapian since that's what's available in current + # Debian stable (10.x) and derived distros. + for my $x (($ENV{PI_XAPIAN} // 'Search::Xapian'), 'Xapian') { + eval "require $x"; + next if $@; + + $x->import(qw(:standard)); + $Xap = $x; + + # `version_string' was added in Xapian 1.1 + my $xver = eval('v'.eval($x.'::version_string()')) // + eval('v'.eval($x.'::xapian_version_string()')); + + # NumberRangeProcessor was added in Xapian 1.3.6, + # NumberValueRangeProcessor was removed for 1.5.0+, + # favor the older /Value/ variant since that's what our + # (currently) preferred Search::Xapian supports + $NVRP = $x.'::'.($x eq 'Xapian' && $xver ge v1.5 ? + 'NumberRangeProcessor' : 'NumberValueRangeProcessor'); + $X{$_} = $Xap.'::'.$_ for (keys %X); + + # ENQ_ASCENDING doesn't seem exported by SWIG Xapian.pm, + # so lets hope this part of the ABI is stable because it's + # just an integer: + $ENQ_ASCENDING = $x eq 'Xapian' ? + 1 : Search::Xapian::ENQ_ASCENDING(); + + # for Smsg: + *PublicInbox::Smsg::sortable_unserialise = + $Xap.'::sortable_unserialise'; + # n.b. FLAG_PURE_NOT is expensive not suitable for a public + # website as it could become a denial-of-service vector + # FLAG_PHRASE also seems to cause performance problems chert + # (and probably earlier Xapian DBs). glass seems fine... + # TODO: make this an option, maybe? + # or make indexlevel=medium as default + $QP_FLAGS = FLAG_PHRASE() | FLAG_BOOLEAN() | FLAG_LOVEHATE() | + FLAG_WILDCARD(); + return 1; + } + undef; +} + +# This is English-only, everything else is non-standard and may be confused as +# a prefix common in patch emails +our $LANG = 'english'; +# note: the non-X term prefix allocations are shared with +# Xapian omega, see xapian-applications/omega/docs/termprefixes.rst my %bool_pfx_external = ( - path => 'XPATH', - thread => 'G', # newsGroup (or similar entity - e.g. a web forum name) - references => 'XREFS', - inreplyto => 'XIRT', + mid => 'Q', # Message-ID (full/exact), this is mostly uniQue + lid => 'G', # newsGroup (or similar entity), just inside <> + dfpre => 'XDFPRE', + dfpost => 'XDFPOST', + dfblob => 'XDFPRE XDFPOST', ); +my $non_quoted_body = 'XNQ XDFN XDFA XDFB XDFHH XDFCTX XDFPRE XDFPOST'; my %prob_prefix = ( - subject => 'S', + # for mairix compatibility + s => 'S', + m => 'XM', # 'mid:' (bool) is exact, 'm:' (prob) can do partial + l => 'XL', # 'lid:' (bool) is exact, 'l:' (prob) can do partial + f => 'A', + t => 'XTO', + tc => 'XTO XCC', + c => 'XCC', + tcf => 'XTO XCC A', + a => 'XTO XCC A', + b => $non_quoted_body . ' XQUOT', + bs => $non_quoted_body . ' XQUOT S', + n => 'XFN', + + q => 'XQUOT', + nq => $non_quoted_body, + dfn => 'XDFN', + dfa => 'XDFA', + dfb => 'XDFB', + dfhh => 'XDFHH', + dfctx => 'XDFCTX', + + # default: + '' => 'XM S A XQUOT XFN ' . $non_quoted_body, ); -my %all_pfx = (%bool_pfx_internal, %bool_pfx_external, %prob_prefix); - -sub xpfx { $all_pfx{$_[0]} } - -our %PFX2TERM_RMAP; -while (my ($k, $v) = each %all_pfx) { - next if $prob_prefix{$k}; - $PFX2TERM_RMAP{$v} = $k; +# not documenting m: and mid: for now, the using the URLs works w/o Xapian +# not documenting lid: for now, either, it is probably redundant with l:, +# especially since we don't offer boolean searches for To/Cc/From +# headers, either +our @HELP = ( + 's:' => 'match within Subject e.g. s:"a quick brown fox"', + 'd:' => < < 'match within message body, including text attachments', + 'nq:' => 'match non-quoted text within message body', + 'q:' => 'match quoted text within message body', + 'n:' => 'match filename of attachment(s)', + 't:' => 'match within the To header', + 'c:' => 'match within the Cc header', + 'f:' => 'match within the From header', + 'a:' => 'match within the To, Cc, and From headers', + 'tc:' => 'match within the To and Cc headers', + 'l:' => 'match contents of the List-Id header', + 'bs:' => 'match within the Subject and body', + 'dfn:' => 'match filename from diff', + 'dfa:' => 'match diff removed (-) lines', + 'dfb:' => 'match diff added (+) lines', + 'dfhh:' => 'match diff hunk header context (usually a function name)', + 'dfctx:' => 'match diff context lines', + 'dfpre:' => 'match pre-image git blob ID', + 'dfpost:' => 'match post-image git blob ID', + 'dfblob:' => 'match either pre or post-image git blob ID', +); +chomp @HELP; + +sub xdir ($;$) { + my ($self, $rdonly) = @_; + if ($rdonly || !defined($self->{shard})) { + $self->{xpfx}; + } else { # v2 only: + "$self->{xpfx}/$self->{shard}"; + } } -my $mail_query = Search::Xapian::Query->new(xpfx('type') . 'mail'); - -sub new { - my ($class, $git_dir, $writable) = @_; - # allow concurrent versions for easier rollback: - my $dir = "$git_dir/public-inbox/xapian" . SCHEMA_VERSION; - my $db; - - if ($writable) { # not used by the WWW interface - require Search::Xapian::WritableDatabase; - require File::Path; - File::Path::mkpath($dir); - $db = Search::Xapian::WritableDatabase->new($dir, - Search::Xapian::DB_CREATE_OR_OPEN); +sub _xdb ($) { + my ($self) = @_; + my $dir = xdir($self, 1); + my ($xdb, $slow_phrase); + my $qpf = \($self->{qp_flags} ||= $QP_FLAGS); + if ($self->{ibx_ver} >= 2) { + my @xdb; + opendir(my $dh, $dir) or return; # not initialized yet + + # We need numeric sorting so shard[0] is first for reading + # Xapian metadata, if needed + for (sort { $a <=> $b } grep(/\A[0-9]+\z/, readdir($dh))) { + my $shard_dir = "$dir/$_"; + if (-d $shard_dir && -r _) { + push @xdb, $X{Database}->new($shard_dir); + $slow_phrase ||= -f "$shard_dir/iamchert"; + } else { # gaps from missing epochs throw off mdocid() + warn "E: $shard_dir missing or unreadable\n"; + return; + } + } + $self->{nshard} = scalar(@xdb); + $xdb = shift @xdb; + $xdb->add_database($_) for @xdb; } else { - $db = Search::Xapian::Database->new($dir); + $slow_phrase = -f "$dir/iamchert"; + $xdb = $X{Database}->new($dir); } - bless { xdb => $db, git_dir => $git_dir }, $class; + $$qpf |= FLAG_PHRASE() unless $slow_phrase; + $xdb; } -sub reopen { $_[0]->{xdb}->reopen } - -sub add_message { - my ($self, $mime) = @_; # mime = Email::MIME object - my $db = $self->{xdb}; - - my $doc_id; - my $mid = mid_clean($mime->header_obj->header_raw('Message-ID')); - $mid = mid_compressed($mid); - my $was_ghost = 0; - my $ct_msg = $mime->header('Content-Type') || 'text/plain'; - my $enc_msg = PublicInbox::View::enc_for($ct_msg); - - $db->begin_transaction; - eval { - my $smsg = $self->lookup_message($mid); - my $doc; - - if ($smsg) { - $smsg->ensure_metadata; - # convert a ghost to a regular message - # it will also clobber any existing regular message - $smsg->mime($mime); - $doc = $smsg->{doc}; - - my $type = xpfx('type'); - eval { - $doc->remove_term($type . 'ghost'); - $was_ghost = 1; - }; - - # probably does not exist: - eval { $doc->remove_term($type . 'mail') }; - $doc->add_term($type . 'mail'); - } else { - $smsg = PublicInbox::SearchMsg->new($mime); - $doc = $smsg->{doc}; - $doc->add_term(xpfx('mid') . $mid); - } - - my $subj = $smsg->subject; - - if (length $subj) { - $doc->add_term(xpfx('subject') . $subj); - - my $path = subject_path($subj); - $doc->add_term(xpfx('path') . mid_compressed($path)); - } - - my $from = $smsg->from_name; - my $date = $smsg->date; - my $ts = Search::Xapian::sortable_serialise($smsg->ts); - $doc->add_value(PublicInbox::Search::TS, $ts); - - $doc->set_data($smsg->to_doc_data); - - my $tg = $self->term_generator; - - $tg->set_document($doc); - $tg->index_text($subj, 1, 'S') if $subj; - $tg->increase_termpos; - $tg->index_text($subj) if $subj; - $tg->increase_termpos; - - $tg->index_text($smsg->from->format); - $tg->increase_termpos; - - $mime->walk_parts(sub { - my ($part) = @_; - return if $part->subparts; # walk_parts already recurses - my $ct = $part->content_type || $ct_msg; - - # account for filter bugs... - $ct =~ m!\btext/plain\b!i or return; +# v2 Xapian docids don't conflict, so they're identical to +# NNTP article numbers and IMAP UIDs. +# https://trac.xapian.org/wiki/FAQ/MultiDatabaseDocumentID +sub mdocid { + my ($nshard, $mitem) = @_; + my $docid = $mitem->get_docid; + int(($docid - 1) / $nshard) + 1; +} - my $enc = PublicInbox::View::enc_for($ct, $enc_msg); - my (@orig, @quot); - foreach my $l (split(/\n/, $enc->decode($part->body))) { - if ($l =~ /^\s*>/) { - push @quot, $l; - } else { - push @orig, $l; - } - } - if (@quot) { - $tg->index_text(join("\n", @quot), 0); - $tg->increase_termpos; - } - if (@orig) { - $tg->index_text(join("\n", @orig)); - $tg->increase_termpos; - } - }); +sub mset_to_artnums { + my ($self, $mset) = @_; + my $nshard = $self->{nshard} // 1; + [ map { mdocid($nshard, $_) } $mset->items ]; +} - if ($was_ghost) { - $doc_id = $smsg->doc_id; - $self->link_message($smsg, 0); - $db->replace_document($doc_id, $doc); - } else { - $self->link_message($smsg, 0); - $doc_id = $db->add_document($doc); - } +sub xdb ($) { + my ($self) = @_; + $self->{xdb} ||= do { + load_xapian(); + _xdb($self); }; +} - if ($@) { - warn "failed to index message <$mid>: $@\n"; - $db->cancel_transaction; +sub xpfx_init ($) { + my ($self) = @_; + if ($self->{ibx_ver} == 1) { + $self->{xpfx} .= '/public-inbox/xapian' . SCHEMA_VERSION; } else { - $db->commit_transaction; + $self->{xpfx} .= '/xap'.SCHEMA_VERSION; } - $doc_id; } -# returns deleted doc_id on success, undef on missing -sub remove_message { - my ($self, $mid) = @_; - my $db = $self->{xdb}; - my $doc_id; - $mid = mid_clean($mid); - $mid = mid_compressed($mid); - - $db->begin_transaction; - eval { - $doc_id = $self->find_unique_doc_id('mid', $mid); - $db->delete_document($doc_id) if defined $doc_id; - }; +sub new { + my ($class, $ibx) = @_; + ref $ibx or die "BUG: expected PublicInbox::Inbox object: $ibx"; + my $self = bless { + xpfx => $ibx->{inboxdir}, # for xpfx_init + altid => $ibx->{altid}, + ibx_ver => $ibx->version, + }, $class; + xpfx_init($self); + my $dir = xdir($self, 1); + $self->{over_ro} = PublicInbox::Over->new("$dir/over.sqlite3"); + $self; +} - if ($@) { - warn "failed to remove message <$mid>: $@\n"; - $db->cancel_transaction; - } else { - $db->commit_transaction; +sub reopen { + my ($self) = @_; + if (my $xdb = $self->{xdb}) { + $xdb->reopen; } - $doc_id; + $self; # make chaining easier } # read-only sub query { my ($self, $query_string, $opts) = @_; - my $query = $self->qp->parse_query($query_string, QP_FLAGS); - - $query = Search::Xapian::Query->new(OP_AND, $mail_query, $query); - $self->do_enquire($query, $opts); + $opts ||= {}; + if ($query_string eq '' && !$opts->{mset}) { + $self->{over_ro}->recent($opts); + } else { + my $qp = $self->{qp} //= qparse_new($self); + my $qp_flags = $self->{qp_flags}; + my $query = $qp->parse_query($query_string, $qp_flags); + $opts->{relevance} = 1 unless exists $opts->{relevance}; + _do_enquire($self, $query, $opts); + } } -sub get_subject_path { - my ($self, $path, $opts) = @_; - my $query = $self->qp->parse_query("path:".mid_compressed($path), 0); - $self->do_enquire($query); +sub retry_reopen { + my ($self, $cb, $arg) = @_; + for my $i (1..10) { + if (wantarray) { + my @ret; + eval { @ret = $cb->($arg) }; + return @ret unless $@; + } else { + my $ret; + eval { $ret = $cb->($arg) }; + return $ret unless $@; + } + # Exception: The revision being read has been discarded - + # you should call Xapian::Database::reopen() + if (ref($@) =~ /\bDatabaseModifiedError\b/) { + warn "reopen try #$i on $@\n"; + reopen($self); + } else { + # let caller decide how to spew, because ExtMsg queries + # get wonky and trigger: + # "something terrible happened at .../Xapian/Enquire.pm" + die; + } + } + die "Too many Xapian database modifications in progress\n"; } -# 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); +sub _do_enquire { + my ($self, $query, $opts) = @_; + retry_reopen($self, \&_enquire_once, [ $self, $query, $opts ]); } -sub get_thread { - my ($self, $mid, $opts) = @_; - my $smsg = eval { $self->lookup_message($mid) }; - - return { count => 0, msgs => [] } unless $smsg; - my $qp = $self->qp; - my $qtid = $qp->parse_query('thread:'.$smsg->thread_id); - my $qsub = $qp->parse_query('path:'.mid_compressed($smsg->path)); - my $query = Search::Xapian::Query->new(OP_OR, $qtid, $qsub); - $self->do_enquire($query); +# returns true if all docs have the THREADID value +sub has_threadid ($) { + my ($self) = @_; + (xdb($self)->get_metadata('has_threadid') // '') eq '1'; } -# private subs below - -sub do_enquire { - my ($self, $query, $opts) = @_; - my $enquire = $self->enquire; - +sub _enquire_once { # retry_reopen callback + my ($self, $query, $opts) = @{$_[0]}; + my $xdb = xdb($self); + my $enquire = $X{Enquire}->new($xdb); $enquire->set_query($query); - $enquire->set_sort_by_relevance_then_value(TS, 0); $opts ||= {}; + my $desc = !$opts->{asc}; + if (($opts->{mset} || 0) == 2) { # mset == 2: ORDER BY docid/UID + $enquire->set_docid_order($ENQ_ASCENDING); + $enquire->set_weighting_scheme($X{BoolWeight}->new); + } elsif ($opts->{relevance}) { + $enquire->set_sort_by_relevance_then_value(TS, $desc); + } else { + $enquire->set_sort_by_value_then_relevance(TS, $desc); + } + + # `mairix -t / --threads' or JMAP collapseThreads + if ($opts->{thread} && has_threadid($self)) { + $enquire->set_collapse_key(THREADID); + } + my $offset = $opts->{offset} || 0; my $limit = $opts->{limit} || 50; my $mset = $enquire->get_mset($offset, $limit); - my @msgs = map { - PublicInbox::SearchMsg->load_doc($_->get_document); - } $mset->items; - - { count => $mset->get_matches_estimated, msgs => \@msgs } + return $mset if $opts->{mset}; + my $nshard = $self->{nshard} // 1; + my $i = 0; + my %order = map { mdocid($nshard, $_) => ++$i } $mset->items; + my @msgs = sort { + $order{$a->{num}} <=> $order{$b->{num}} + } @{$self->{over_ro}->get_all(keys %order)}; + wantarray ? ($mset->get_matches_estimated, \@msgs) : \@msgs; } # read-write -sub stemmer { Search::Xapian::Stem->new(LANG) } +sub stemmer { $X{Stem}->new($LANG) } # read-only -sub qp { +sub qparse_new ($) { my ($self) = @_; - my $qp = $self->{query_parser}; - return $qp if $qp; - - # new parser - $qp = Search::Xapian::QueryParser->new; - $qp->set_default_op(OP_AND); - $qp->set_database($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); + my $xdb = xdb($self); + my $qp = $X{QueryParser}->new; + $qp->set_default_op(OP_AND()); + $qp->set_database($xdb); + $qp->set_stemmer(stemmer($self)); + $qp->set_stemming_strategy(STEM_SOME()); + my $cb = $qp->can('set_max_wildcard_expansion') // + $qp->can('set_max_expansion'); # Xapian 1.5.0+ + $cb->($qp, 100); + $cb = $qp->can('add_valuerangeprocessor') // + $qp->can('add_rangeprocessor'); # Xapian 1.5.0+ + $cb->($qp, $NVRP->new(YYYYMMDD, 'd:')); + $cb->($qp, $NVRP->new(DT, 'dt:')); + + # for IMAP, undocumented for WWW and may be split off go away + $cb->($qp, $NVRP->new(BYTES, 'bytes:')); + $cb->($qp, $NVRP->new(TS, 'ts:')); + $cb->($qp, $NVRP->new(UID, 'uid:')); while (my ($name, $prefix) = each %bool_pfx_external) { - $qp->add_boolean_prefix($name, $prefix); - } - - while (my ($name, $prefix) = each %prob_prefix) { - $qp->add_prefix($name, $prefix); - } - - $self->{query_parser} = $qp; -} - -sub term_generator { # write-only - my ($self) = @_; - - my $tg = $self->{term_generator}; - return $tg if $tg; - - $tg = Search::Xapian::TermGenerator->new; - $tg->set_stemmer($self->stemmer); - - $self->{term_generator} = $tg; -} - -sub next_doc_id { $_[0]->{xdb}->get_lastdocid + 1 } - -# increments last_thread_id counter -# returns a 64-bit integer represented as a hex string -sub next_thread_id { - my ($self) = @_; - my $db = $self->{xdb}; - my $last_thread_id = int($db->get_metadata('last_thread_id') || 0); - - $db->set_metadata('last_thread_id', ++$last_thread_id); - - $last_thread_id; -} - -sub ts_range_processor { - $_[0]->{tsrp} ||= Search::Xapian::NumberValueRangeProcessor->new(TS); -} - -sub date_range_processor { - $_[0]->{drp} ||= Search::Xapian::DateValueRangeProcessor->new(TS); -} - -sub link_message { - my ($self, $smsg, $is_ghost) = @_; - - if ($is_ghost) { - $smsg->ensure_metadata; - } else { - $self->link_message_to_parents($smsg); + $qp->add_boolean_prefix($name, $_) foreach split(/ /, $prefix); } -} -sub link_message_to_parents { - my ($self, $smsg) = @_; - my $doc = $smsg->{doc}; - my $mid = mid_compressed($smsg->mid); - my $mime = $smsg->mime; - my $refs = $mime->header_obj->header_raw('References'); - my @refs = $refs ? ($refs =~ /<([^>]+)>/g) : (); - my $irt = $mime->header_obj->header_raw('In-Reply-To'); - if ($irt) { - if ($irt =~ /<([^>]+)>/) { - $irt = $1; + # 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) { + # $_ = 'serial:gmane:/path/to/gmane.msgmap.sqlite3' + # note: Xapian supports multibyte UTF-8, /^[0-9]+$/, + # and '_' with prefixes matching \w+ + /\Aserial:(\w+):/ or next; + my $pfx = $1; + push @$user_pfx, "$pfx:", < XGMANE + $qp->add_boolean_prefix($pfx, 'X'.uc($pfx)); } - push @refs, $irt; + chomp @$user_pfx; } - my $tid; - if (@refs) { - @refs = map { mid_compressed($_) } @refs; - my %uniq; - @refs = grep { !$uniq{$_}++ } @refs; # uniq - - $doc->add_term(xpfx('inreplyto') . $refs[-1]); - - my $ref_pfx = xpfx('references'); - - # first ref *should* be the thread root, - # but we can never trust clients to do the right thing - my $ref = shift @refs; - $doc->add_term($ref_pfx . $ref); - $tid = $self->_resolve_mid_to_tid($ref); - - # the rest of the refs should point to this tid: - foreach $ref (@refs) { - $doc->add_term($ref_pfx . $ref); - my $ptid = $self->_resolve_mid_to_tid($ref); - if ($tid ne $ptid) { - $self->merge_threads($tid, $ptid); - } - } - } else { - $tid = $self->next_thread_id; - } - $doc->add_term(xpfx('thread') . $tid); -} - -sub lookup_message { - my ($self, $mid) = @_; - $mid = mid_clean($mid); - $mid = mid_compressed($mid); - - my $doc_id = $self->find_unique_doc_id('mid', $mid); - my $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); - } - $smsg; -} - -sub find_unique_doc_id { - my ($self, $term, $value) = @_; - - my ($begin, $end) = $self->find_doc_ids($term, $value); - - return undef if $begin->equal($end); # not found - - my $rv = $begin->get_docid; - - # sanity check - $begin->inc; - $begin->equal($end) or die "Term '$term:$value' 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 $db = $self->{xdb}; - - ($db->postlist_begin($term), $db->postlist_end($term)); -} - -# this will create a ghost as necessary -sub _resolve_mid_to_tid { - my ($self, $mid) = @_; - - my $smsg = $self->lookup_message($mid) || $self->create_ghost($mid); - $smsg->thread_id; -} - -sub create_ghost { - my ($self, $mid, $tid) = @_; - - $mid = mid_compressed($mid); - $tid = $self->next_thread_id unless defined $tid; - - my $doc = Search::Xapian::Document->new; - $doc->add_term(xpfx('mid') . $mid); - $doc->add_term(xpfx('thread') . $tid); - $doc->add_term(xpfx('type') . 'ghost'); - - my $smsg = PublicInbox::SearchMsg->wrap($doc, $mid); - $self->link_message($smsg, 1); - $self->{xdb}->add_document($doc); - - $smsg; -} - -sub merge_threads { - my ($self, $winner_tid, $loser_tid) = @_; - my ($head, $tail) = $self->find_doc_ids('thread', $loser_tid); - my $thread_pfx = xpfx('thread'); - my $db = $self->{xdb}; - - for (; $head != $tail; $head->inc) { - my $docid = $head->get_docid; - my $doc = $db->get_document($docid); - $doc->remove_term($thread_pfx . $loser_tid); - $doc->add_term($thread_pfx . $winner_tid); - $db->replace_document($docid, $doc); + while (my ($name, $prefix) = each %prob_prefix) { + $qp->add_prefix($name, $_) foreach split(/ /, $prefix); } + $qp; } -# normalize subjects so they are suitable as pathnames for URLs -sub subject_path { - my $subj = pop; - - $subj =~ s/\A\s+//; - $subj =~ s/\s+\z//; - $subj =~ s/^(?:re|aw):\s*//i; # remove reply prefix (aw: German) - $subj =~ s![^a-zA-Z0-9_\.~/\-]+!_!g; - lc($subj); -} - -sub do_cat_mail { - my ($git, $blob) = @_; - my $mime = eval { - my $str = $git->cat_file($blob); - Email::MIME->new($str); - }; - $@ ? undef : $mime; -} - -sub index_blob { - my ($self, $git, $blob) = @_; - my $mime = do_cat_mail($git, $blob) or return; - eval { $self->add_message($mime) }; - warn "W: index_blob $blob: $@\n" if $@; -} - -sub unindex_blob { - my ($self, $git, $blob) = @_; - my $mime = do_cat_mail($git, $blob) or return; - my $mid = $mime->header_obj->header_raw('Message-ID'); - eval { $self->remove_message($mid) } if defined $mid; - warn "W: unindex_blob $blob: $@\n" if $@; -} - -sub enquire { +sub help { my ($self) = @_; - $self->{enquire} ||= Search::Xapian::Enquire->new($self->{xdb}); -} - -# indexes all unindexed messages -sub index_sync { - my ($self, $git) = @_; - my $db = $self->{xdb}; - my $latest = $db->get_metadata('last_commit'); - my $range = length $latest ? "$latest..HEAD" : 'HEAD'; - $latest = undef; - - my $hex = '[a-f0-9]'; - my $h40 = $hex .'{40}'; - my $addmsg = qr!^:000000 100644 \S+ ($h40) A\t${hex}{2}/${hex}{38}$!; - my $delmsg = qr!^:100644 000000 ($h40) \S+ D\t${hex}{2}/${hex}{38}$!; - - # get indexed messages - my @cmd = ('git', "--git-dir=$git->{git_dir}", "log", - qw/--reverse --no-notes --no-color --raw -r --no-abbrev/, - $range); - - my $pid = open(my $log, '-|', @cmd) or - die('open` '.join(' ', @cmd) . " pipe failed: $!\n"); - my $last; - while (my $line = <$log>) { - if ($line =~ /$addmsg/o) { - $self->index_blob($git, $1); - } elsif ($line =~ /$delmsg/o) { - $self->unindex_blob($git, $1); - } elsif ($line =~ /^commit ($h40)/o) { - my $commit = $1; - if (defined $latest) { - $db->set_metadata('last_commit', $latest) - } - $latest = $commit; - } + $self->{qp} //= qparse_new($self); # parse altids + my @ret = @HELP; + if (my $user_pfx = $self->{-user_pfx}) { + push @ret, @$user_pfx; } - close $log; - $db->set_metadata('last_commit', $latest) if defined $latest; + \@ret; } 1;