# Copyright (C) 2015 all contributors <meta@public-inbox.org>
# 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
+#
+# Indexes mail with Xapian and our (SQLite-based) ::Msgmap for use
+# with the web and NNTP interfaces. This index maintains thread
+# relationships for use by Mail::Thread. This writes to the search
+# index.
package PublicInbox::SearchIdx;
use strict;
use warnings;
+use Fcntl qw(:flock :DEFAULT);
+use Email::MIME;
+use Email::MIME::ContentType;
+$Email::MIME::ContentType::STRICT_PARAMS = 0;
use base qw(PublicInbox::Search);
-use PublicInbox::MID qw/mid_clean mid_compress/;
+use PublicInbox::MID qw/mid_clean id_compress mid_mime/;
+use PublicInbox::MsgIter;
+require PublicInbox::Git;
*xpfx = *PublicInbox::Search::xpfx;
+use constant MAX_MID_SIZE => 244; # max term size - 1 in Xapian
use constant {
PERM_UMASK => 0,
OLD_PERM_GROUP => 1,
sub new {
my ($class, $git_dir, $writable) = @_;
- my $dir = $class->xdir($git_dir);
+ my $dir = PublicInbox::Search->xdir($git_dir);
require Search::Xapian::WritableDatabase;
my $flag = Search::Xapian::DB_OPEN;
my $self = bless { git_dir => $git_dir }, $class;
my $perm = $self->_git_config_perm;
my $umask = _umask_for($perm);
$self->{umask} = $umask;
+ $self->{lock_path} = "$git_dir/ssoma.lock";
$self->{xdb} = $self->with_umask(sub {
if ($writable == 1) {
require File::Path;
File::Path::mkpath($dir);
+ $self->{batch_size} = 100;
$flag = Search::Xapian::DB_CREATE_OR_OPEN;
+ _lock_acquire($self);
}
Search::Xapian::WritableDatabase->new($dir, $flag);
});
$self;
}
+sub _xdb_release {
+ my ($self) = @_;
+ my $xdb = delete $self->{xdb};
+ $xdb->commit_transaction;
+ $xdb->close;
+ _lock_release($self);
+}
+
+sub _xdb_acquire {
+ my ($self) = @_;
+ _lock_acquire($self);
+ my $dir = PublicInbox::Search->xdir($self->{git_dir});
+ my $flag = Search::Xapian::DB_OPEN;
+ my $xdb = Search::Xapian::WritableDatabase->new($dir, $flag);
+ $xdb->begin_transaction;
+ $self->{xdb} = $xdb;
+}
+
+sub _lock_acquire {
+ my ($self) = @_;
+ sysopen(my $lockfh, $self->{lock_path}, O_WRONLY|O_CREAT) or
+ die "failed to open lock $self->{lock_path}: $!\n";
+ flock($lockfh, LOCK_EX) or die "lock failed: $!\n";
+ $self->{lockfh} = $lockfh;
+}
+
+sub _lock_release {
+ my ($self) = @_;
+ my $lockfh = delete $self->{lockfh};
+ flock($lockfh, LOCK_UN) or die "unlock failed: $!\n";
+ close $lockfh or die "close failed: $!\n";
+}
+
+sub add_val {
+ my ($doc, $col, $num) = @_;
+ $num = Search::Xapian::sortable_serialise($num);
+ $doc->add_value($col, $num);
+}
+
sub add_message {
- my ($self, $mime) = @_; # mime = Email::MIME object
+ my ($self, $mime, $bytes, $num) = @_; # mime = Email::MIME object
my $db = $self->{xdb};
- my $doc_id;
- my $mid = mid_clean($mime->header('Message-ID'));
- my $was_ghost = 0;
+ my ($doc_id, $old_tid);
+ my $mid = mid_clean(mid_mime($mime));
my $ct_msg = $mime->header('Content-Type') || 'text/plain';
eval {
+ die 'Message-ID too long' if length($mid) > MAX_MID_SIZE;
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);
+ $doc_id = $smsg->doc_id;
+ $old_tid = $smsg->thread_id;
}
+ $smsg = PublicInbox::SearchMsg->new($mime);
+ my $doc = $smsg->{doc};
+ $doc->add_term(xpfx('mid') . $mid);
my $subj = $smsg->subject;
-
if ($subj ne '') {
- $doc->add_term(xpfx('subject') . $subj);
-
my $path = $self->subject_path($subj);
- $doc->add_term(xpfx('path') . mid_compress($path));
+ $doc->add_term(xpfx('path') . id_compress($path));
}
- my $ts = Search::Xapian::sortable_serialise($smsg->ts);
- $doc->add_value(PublicInbox::Search::TS, $ts);
+ add_val($doc, &PublicInbox::Search::TS, $smsg->ts);
+
+ defined($num) and
+ add_val($doc, &PublicInbox::Search::NUM, $num);
+
+ defined($bytes) and
+ add_val($doc, &PublicInbox::Search::BYTES, $bytes);
+
+ add_val($doc, &PublicInbox::Search::LINES,
+ $mime->body_raw =~ tr!\n!\n!);
my $tg = $self->term_generator;
$tg->index_text($subj) if $subj;
$tg->increase_termpos;
- $tg->index_text($smsg->from->format);
+ $tg->index_text($smsg->from);
$tg->increase_termpos;
- $mime->walk_parts(sub {
- my ($part) = @_;
- return if $part->subparts; # walk_parts already recurses
+ msg_iter($mime, sub {
+ my ($part, $depth, @idx) = @{$_[0]};
my $ct = $part->content_type || $ct_msg;
# account for filter bugs...
}
});
- if ($was_ghost) {
- $doc_id = $smsg->doc_id;
- $self->link_message($smsg, 0);
- $doc->set_data($smsg->to_doc_data);
+ link_message($self, $smsg, $old_tid);
+ $doc->set_data($smsg->to_doc_data);
+ if (defined $doc_id) {
$db->replace_document($doc_id, $doc);
} else {
- $self->link_message($smsg, 0);
- $doc->set_data($smsg->to_doc_data);
$doc_id = $db->add_document($doc);
}
};
$self->{term_generator} = $tg;
}
-sub next_doc_id { $_[0]->{xdb}->get_lastdocid + 1 }
-
# increments last_thread_id counter
# returns a 64-bit integer represented as a hex string
sub next_thread_id {
}
sub link_message {
- my ($self, $smsg, $is_ghost) = @_;
-
- if ($is_ghost) {
- $smsg->ensure_metadata;
- } else {
- $self->link_message_to_parents($smsg);
- }
-}
-
-sub link_message_to_parents {
- my ($self, $smsg) = @_;
+ my ($self, $smsg, $old_tid) = @_;
my $doc = $smsg->{doc};
my $mid = $smsg->mid;
my $mime = $smsg->mime;
- my $refs = $mime->header('References');
+ my $hdr = $mime->header_obj;
+ my $refs = $hdr->header_raw('References');
my @refs = $refs ? ($refs =~ /<([^>]+)>/g) : ();
- if (my $irt = $mime->header('In-Reply-To')) {
+ if (my $irt = $hdr->header_raw('In-Reply-To')) {
# last References should be $irt
# we will de-dupe later
push @refs, mid_clean($irt);
# prevent circular references via References: here:
foreach my $ref (@orig_refs) {
+ if (length($ref) > MAX_MID_SIZE) {
+ warn "References: <$ref> too long, ignoring\n";
+ }
next if $uniq{$ref};
$uniq{$ref} = 1;
push @refs, $ref;
}
}
if (@refs) {
- $smsg->{references_sorted} = '<'.join('><', @refs).'>';
+ $smsg->{references} = '<'.join('> <', @refs).'>';
# first ref *should* be the thread root,
# but we can never trust clients to do the right thing
my $ref = shift @refs;
$tid = $self->_resolve_mid_to_tid($ref);
+ $self->merge_threads($tid, $old_tid) if defined $old_tid;
# the rest of the refs should point to this tid:
foreach $ref (@refs) {
my $ptid = $self->_resolve_mid_to_tid($ref);
- if ($tid ne $ptid) {
- $self->merge_threads($tid, $ptid);
- }
+ merge_threads($self, $tid, $ptid);
}
} else {
$tid = $self->next_thread_id;
}
sub index_blob {
- my ($self, $git, $mime) = @_;
- $self->add_message($mime);
+ my ($self, $git, $mime, $bytes, $num) = @_;
+ $self->add_message($mime, $bytes, $num);
}
sub unindex_blob {
my ($self, $git, $mime) = @_;
- my $mid = mid_clean($mime->header('Message-ID'));
+ my $mid = eval { mid_clean(mid_mime($mime)) };
$self->remove_message($mid) if defined $mid;
}
sub index_mm {
my ($self, $git, $mime) = @_;
- $self->{mm}->mid_insert(mid_clean($mime->header('Message-ID')));
+ $self->{mm}->mid_insert(mid_clean(mid_mime($mime)));
}
sub unindex_mm {
my ($self, $git, $mime) = @_;
- $self->{mm}->mid_delete(mid_clean($mime->header('Message-ID')));
+ $self->{mm}->mid_delete(mid_clean(mid_mime($mime)));
}
-sub index_both {
+sub index_mm2 {
+ my ($self, $git, $mime, $bytes) = @_;
+ my $num = $self->{mm}->num_for(mid_clean(mid_mime($mime)));
+ index_blob($self, $git, $mime, $bytes, $num);
+}
+
+sub unindex_mm2 {
my ($self, $git, $mime) = @_;
- index_blob($self, $git, $mime);
- index_mm($self, $git, $mime);
+ $self->{mm}->mid_delete(mid_clean(mid_mime($mime)));
+ unindex_blob($self, $git, $mime);
+}
+
+sub index_both {
+ my ($self, $git, $mime, $bytes) = @_;
+ my $num = index_mm($self, $git, $mime);
+ index_blob($self, $git, $mime, $bytes, $num);
}
sub unindex_both {
}
sub do_cat_mail {
- my ($git, $blob) = @_;
+ my ($git, $blob, $sizeref) = @_;
my $mime = eval {
- my $str = $git->cat_file($blob);
+ my $str = $git->cat_file($blob, $sizeref);
+ # fixup bugs from import:
+ $$str =~ s/\A[\r\n]*From [^\r\n]*\r?\n//s;
Email::MIME->new($str);
};
$@ ? undef : $mime;
}
sub index_sync {
- my ($self, $head) = @_;
- $self->with_umask(sub { $self->_index_sync($head) });
+ my ($self, $opts) = @_;
+ with_umask($self, sub { $self->_index_sync($opts) });
}
sub rlog {
- my ($self, $range, $add_cb, $del_cb) = @_;
+ my ($self, $range, $add_cb, $del_cb, $batch_cb) = @_;
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}$!;
- my $git_dir = $self->{git_dir};
- require PublicInbox::GitCatFile;
- my $git = PublicInbox::GitCatFile->new($git_dir);
- my @cmd = ('git', "--git-dir=$git_dir", "log",
- qw/--reverse --no-notes --no-color --raw -r --no-abbrev/,
- $range);
+ my $git = PublicInbox::Git->new($self->{git_dir});
+ my $log = $git->popen(qw/log --reverse --no-notes --no-color
+ --raw -r --no-abbrev/, $range);
my $latest;
- my $pid = open(my $log, '-|', @cmd) or
- die('open` '.join(' ', @cmd) . " pipe failed: $!\n");
- while (my $line = <$log>) {
+ my $bytes;
+ my $max = $self->{batch_size}; # may be undef
+ local $/ = "\n";
+ my $line;
+ while (defined($line = <$log>)) {
if ($line =~ /$addmsg/o) {
- my $mime = do_cat_mail($git, $1) or next;
- $add_cb->($self, $git, $mime);
+ my $mime = do_cat_mail($git, $1, \$bytes) or next;
+ $add_cb->($self, $git, $mime, $bytes);
} elsif ($line =~ /$delmsg/o) {
my $mime = do_cat_mail($git, $1) or next;
$del_cb->($self, $git, $mime);
} elsif ($line =~ /^commit ($h40)/o) {
+ if (defined $max && --$max <= 0) {
+ $max = $self->{batch_size};
+ $batch_cb->($latest, 1);
+ }
$latest = $1;
}
}
- close $log;
- $latest;
+ $batch_cb->($latest, 0);
}
# indexes all unindexed messages
sub _index_sync {
- my ($self, $head) = @_;
- my $db = $self->{xdb};
- $head ||= 'HEAD';
+ my ($self, $opts) = @_;
+ my $head = 'HEAD';
my $mm = $self->{mm} = eval {
require PublicInbox::Msgmap;
PublicInbox::Msgmap->new($self->{git_dir}, 1);
};
+ my $xdb = $self->{xdb};
+ $xdb->begin_transaction;
+ my $reindex = $opts->{reindex};
+ my $mkey = $reindex ? undef : 'last_commit';
+ my $lx = $reindex ? '' : $xdb->get_metadata('last_commit');
+ my $dbh;
+ my $cb = sub {
+ my ($commit, $more) = @_;
+ $xdb->set_metadata($mkey, $commit) if $mkey && $commit;
+ if ($dbh) {
+ $mm->last_commit($commit) if $commit;
+ $dbh->commit;
+ }
+ _xdb_release($self);
+ # let another process do some work...
+ $dbh->begin_work if $dbh && $more;
+ $xdb = _xdb_acquire($self);
+ };
- $db->begin_transaction;
- my $lx = $db->get_metadata('last_commit');
my $range = $lx eq '' ? $head : "$lx..$head";
if ($mm) {
- $mm->{dbh}->begin_work;
+ $dbh = $mm->{dbh};
+ $dbh->begin_work;
my $lm = $mm->last_commit || '';
if ($lm eq $lx) {
# Common case is the indexes are synced,
# we only need to run git-log once:
- $lx = $self->rlog($range, *index_both, *unindex_both);
- $mm->{dbh}->commit;
- if (defined $lx) {
- $db->set_metadata('last_commit', $lx);
- $mm->last_commit($lx);
- }
+ rlog($self, $range, *index_both, *unindex_both, $cb);
} else {
- # dumb case, msgmap and xapian are out-of-sync
- # do not care for performance:
+ # Uncommon case, msgmap and xapian are out-of-sync
+ # do not care for performance (but git is fast :>)
+ # This happens if we have to reindex Xapian since
+ # msgmap is a frozen format and our Xapian format
+ # is evolving.
my $r = $lm eq '' ? $head : "$lm..$head";
- $lm = $self->rlog($r, *index_mm, *unindex_mm);
- $mm->{dbh}->commit;
- $mm->last_commit($lm) if defined $lm;
- goto xapian_only;
+ # first, ensure msgmap is up-to-date:
+ $mkey = undef; # ignore xapian, for now
+ rlog($self, $r, *index_mm, *unindex_mm, $cb);
+
+ # now deal with Xapian
+ $mkey = 'last_commit' unless $reindex;
+ $dbh = undef;
+ rlog($self, $range, *index_mm2, *unindex_mm2, $cb);
}
} else {
# user didn't install DBD::SQLite and DBI
-xapian_only:
- $lx = $self->rlog($range, *index_blob, *unindex_blob);
- $db->set_metadata('last_commit', $lx) if defined $lx;
- }
- if ($@) {
- $db->cancel_transaction;
- $mm->{dbh}->rollback if $mm;
- } else {
- $db->commit_transaction;
+ rlog($self, $range, *index_blob, *unindex_blob, $cb);
}
}
}
sub create_ghost {
- my ($self, $mid, $tid) = @_;
-
- $tid = $self->next_thread_id unless defined $tid;
+ my ($self, $mid) = @_;
+ my $tid = $self->next_thread_id;
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) = @_;
+ return if $winner_tid == $loser_tid;
my ($head, $tail) = $self->find_doc_ids('thread', $loser_tid);
my $thread_pfx = xpfx('thread');
my $db = $self->{xdb};
sub _read_git_config_perm {
my ($self) = @_;
- my @cmd = ('git', "--git-dir=$self->{git_dir}",
- qw(config core.sharedRepository));
- my $pid = open(my $fh, '-|', @cmd) or
- die('open `'.join(' ', @cmd) . " pipe failed: $!\n");
+ my @cmd = qw(config core.sharedRepository);
+ my $fh = PublicInbox::Git->new($self->{git_dir})->popen(@cmd);
+ local $/ = "\n";
my $perm = <$fh>;
- close $fh;
chomp $perm if defined $perm;
$perm;
}
$rv;
}
+sub DESTROY {
+ # order matters for unlocking
+ $_[0]->{xdb} = undef;
+ $_[0]->{lockfh} = undef;
+}
+
1;