X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FSearch.pm;h=593040a8a285340a69465d8bd27fba357387f28a;hb=365d58c295a12d38cc92be3b274919519bedaa40;hp=86a6ad674b3e1fbfdb70d3a04d8dac40d4fae881;hpb=7909c5f7439777e3b3643f14224b27f0a8f8fb07;p=public-inbox.git diff --git a/lib/PublicInbox/Search.pm b/lib/PublicInbox/Search.pm index 86a6ad67..593040a8 100644 --- a/lib/PublicInbox/Search.pm +++ b/lib/PublicInbox/Search.pm @@ -5,12 +5,49 @@ # Read-only search interface for use by the web and NNTP interfaces package PublicInbox::Search; use strict; -use warnings; -# values for searching -use constant TS => 0; # Received: header in Unix time -use constant YYYYMMDD => 1; # Date: header for searching in the WWW UI -use constant DT => 2; # Date: YYYYMMDDHHMMSS +# values for searching, changing the numeric value breaks +# compatibility with old indices (so don't change them it) +use constant { + 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 + + # TODO + # THREADID => ? + # REPLYCNT => ?, # IMAP ANSWERED + + # SCHEMA_VERSION history + # 0 - initial + # 1 - subject_path is lower-cased + # 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 and UID values + SCHEMA_VERSION => 15, +}; use PublicInbox::Smsg; use PublicInbox::Over; @@ -56,32 +93,11 @@ sub load_xapian () { # a prefix common in patch emails our $LANG = 'english'; -use constant { - # SCHEMA_VERSION history - # 0 - initial - # 1 - subject_path is lower-cased - # 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 - SCHEMA_VERSION => 15, -}; - +# note: the non-X term prefix allocations are shared with +# Xapian omega, see xapian-applications/omega/docs/termprefixes.rst my %bool_pfx_external = ( 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', @@ -92,6 +108,7 @@ my %prob_prefix = ( # 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', @@ -115,6 +132,9 @@ my %prob_prefix = ( ); # 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 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', @@ -148,15 +169,10 @@ chomp @HELP; sub xdir ($;$) { my ($self, $rdonly) = @_; - if ($self->{ibx_ver} == 1) { - "$self->{inboxdir}/public-inbox/xapian" . SCHEMA_VERSION; - } else { - my $dir = "$self->{inboxdir}/xap" . SCHEMA_VERSION; - return $dir if $rdonly; - - my $shard = $self->{shard}; - defined $shard or die "shard not given"; - $dir .= "/$shard"; + if ($rdonly || !defined($self->{shard})) { + $self->{xpfx}; + } else { # v2 only: + "$self->{xpfx}/$self->{shard}"; } } @@ -166,16 +182,24 @@ sub _xdb ($) { my ($xdb, $slow_phrase); my $qpf = \($self->{qp_flags} ||= $QP_FLAGS); if ($self->{ibx_ver} >= 2) { - foreach my $shard (<$dir/*>) { - -d $shard && $shard =~ m!/[0-9]+\z! or next; - my $sub = $X{Database}->new($shard); - if ($xdb) { - $xdb->add_database($sub); - } else { - $xdb = $sub; + 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; } - $slow_phrase ||= -f "$shard/iamchert"; } + $self->{nshard} = scalar(@xdb); + $xdb = shift @xdb; + $xdb->add_database($_) for @xdb; } else { $slow_phrase = -f "$dir/iamchert"; $xdb = $X{Database}->new($dir); @@ -184,6 +208,15 @@ sub _xdb ($) { $xdb; } +# v2 Xapian docids don't conflict, so they're identical to +# NNTP article numbers and IMAP UIDs. +# https://trac.xapian.org/wiki/FAQ/MultiDatabaseDocumentID +sub mdocid { + my ($nshard, $mitem) = @_; + my $docid = $mitem->get_docid; + int(($docid - 1) / $nshard) + 1; +} + sub xdb ($) { my ($self) = @_; $self->{xdb} ||= do { @@ -192,14 +225,24 @@ sub xdb ($) { }; } +sub xpfx_init ($) { + my ($self) = @_; + if ($self->{ibx_ver} == 1) { + $self->{xpfx} .= '/public-inbox/xapian' . SCHEMA_VERSION; + } else { + $self->{xpfx} .= '/xap'.SCHEMA_VERSION; + } +} + sub new { my ($class, $ibx) = @_; ref $ibx or die "BUG: expected PublicInbox::Inbox object: $ibx"; my $self = bless { - inboxdir => $ibx->{inboxdir}, + 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; @@ -267,7 +310,7 @@ sub _enquire_once { # retry_reopen callback $enquire->set_query($query); $opts ||= {}; my $desc = !$opts->{asc}; - if (($opts->{mset} || 0) == 2) { + if (($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}) { @@ -306,6 +349,11 @@ sub qp { $qp->add_valuerangeprocessor($nvrp->new(YYYYMMDD, 'd:')); $qp->add_valuerangeprocessor($nvrp->new(DT, 'dt:')); + # for IMAP, undocumented for WWW and may be split off go away + $qp->add_valuerangeprocessor($nvrp->new(BYTES, 'bytes:')); + $qp->add_valuerangeprocessor($nvrp->new(TS, 'ts:')); + $qp->add_valuerangeprocessor($nvrp->new(UID, 'uid:')); + while (my ($name, $prefix) = each %bool_pfx_external) { $qp->add_boolean_prefix($name, $_) foreach split(/ /, $prefix); }