X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FLeiSavedSearch.pm;h=1d13aef6728a6a5ebcbcd942a93630f61872e1ed;hb=666dde69a3f64456321b7c701070712d6b85e359;hp=ed217cf2f47d4cd840c5b441e8f680a2d7453f77;hpb=7af47fb992e047f149b0b36c37e3a4984d34e139;p=public-inbox.git diff --git a/lib/PublicInbox/LeiSavedSearch.pm b/lib/PublicInbox/LeiSavedSearch.pm index ed217cf2..1d13aef6 100644 --- a/lib/PublicInbox/LeiSavedSearch.pm +++ b/lib/PublicInbox/LeiSavedSearch.pm @@ -6,6 +6,7 @@ package PublicInbox::LeiSavedSearch; use strict; use v5.10.1; use parent qw(PublicInbox::Lock); +use PublicInbox::Git; use PublicInbox::OverIdx; use PublicInbox::LeiSearch; use PublicInbox::Config; @@ -13,12 +14,14 @@ use PublicInbox::Spawn qw(run_die); use PublicInbox::ContentHash qw(git_sha); use PublicInbox::MID qw(mids_for_index); use Digest::SHA qw(sha256_hex); +our $LOCAL_PFX = qr!\A(?:maildir|mh|mbox.+|mmdf|v2):!i; # TODO: put in LeiToMail? # move this to PublicInbox::Config if other things use it: my %cquote = ("\n" => '\\n', "\t" => '\\t', "\b" => '\\b'); sub cquote_val ($) { # cf. git-config(1) my ($val) = @_; $val =~ s/([\n\t\b])/$cquote{$1}/g; + $val =~ s/\"/\\\"/g; $val; } @@ -27,27 +30,50 @@ sub BOOL_FIELDS () { qw(external local remote import-remote import-before threads) } -sub lss_dir_for ($$) { - my ($lei, $dstref) = @_; - my @n; +sub SINGLE_FIELDS () { qw(limit dedupe output) } + +sub lss_dir_for ($$;$) { + my ($lei, $dstref, $on_fs) = @_; + my $pfx; if ($$dstref =~ m,\Aimaps?://,i) { # already canonicalized require PublicInbox::URIimap; my $uri = PublicInbox::URIimap->new($$dstref)->canonical; $$dstref = $$uri; - @n = ($uri->mailbox); - } else { # basename + $pfx = $uri->mailbox; + } else { + # can't use Cwd::abs_path since dirname($$dstref) may not exist $$dstref = $lei->rel2abs($$dstref); - $$dstref .= '/' if -d $$dstref; $$dstref =~ tr!/!/!s; - @n = ($$dstref =~ m{([^/]+)/*\z}); + $pfx = $$dstref; + } + ($pfx) = ($pfx =~ m{([^/]+)/*\z}); # basename + my $lss_dir = $lei->share_path . '/saved-searches/'; + my $d = "$lss_dir$pfx-".sha256_hex($$dstref); + + # fall-back to looking up by st_ino + st_dev in case we're in + # a symlinked or bind-mounted path + if ($on_fs && !-d $d && -e $$dstref) { + my @cur = stat(_); + my $want = pack('dd', @cur[1,0]); # st_ino + st_dev + my ($c, $o, @st); + for my $g ("$pfx-*", '*') { + my @maybe = glob("$lss_dir$g/lei.saved-search"); + for my $f (@maybe) { + $c = $lei->cfg_dump($f) // next; + $o = $c->{'lei.q.output'} // next; + $o =~ s!$LOCAL_PFX!! or next; + @st = stat($o) or next; + next if pack('dd', @st[1,0]) ne $want; + $f =~ m!\A(.+?)/[^/]+\z! and return $1; + } + } } - push @n, sha256_hex($$dstref); - $lei->share_path . '/saved-searches/' . join('-', @n); + $d; } sub list { my ($lei, $pfx) = @_; - my $lss_dir = $lei->share_path.'/saved-searches/'; + my $lss_dir = $lei->share_path.'/saved-searches'; return () unless -d $lss_dir; # TODO: persist the cache? Use another format? my $f = $lei->cache_dir."/saved-tmp.$$.".time.'.config'; @@ -57,21 +83,21 @@ sub list { print $fh "\tpath = ", cquote_val($p), "\n"; } close $fh or die "close $f: $!"; - my $cfg = PublicInbox::Config->git_config_dump($f); + my $cfg = $lei->cfg_dump($f); unlink($f); - my $out = $cfg->get_all('lei.q.output') or return (); + my $out = $cfg ? $cfg->get_all('lei.q.output') : []; map {; - s!\A(?:maildir|mh|mbox.+|mmdf):!!i; + s!$LOCAL_PFX!!; $_; } @$out } -sub translate_dedupe ($$$) { - my ($self, $lei, $dd) = @_; - $dd //= 'content'; +sub translate_dedupe ($$) { + my ($self, $lei) = @_; + my $dd = $lei->{opt}->{dedupe} // 'content'; return 1 if $dd eq 'content'; # the default return $self->{"-dedupe_$dd"} = 1 if ($dd eq 'oid' || $dd eq 'mid'); - $lei->fail("--dedupe=$dd unsupported with --save"); + die("--dedupe=$dd requires --no-save\n"); } sub up { # updating existing saved search via "lei up" @@ -80,9 +106,9 @@ sub up { # updating existing saved search via "lei up" my $self = bless { ale => $lei->ale }, $cls; my $dir = $dst; output2lssdir($self, $lei, \$dir, \$f) or - return $lei->fail("--save was not used with $dst cwd=". - $lei->rel2abs('.')); - $self->{-cfg} = PublicInbox::Config->git_config_dump($f); + return die("--no-save was used with $dst cwd=". + $lei->rel2abs('.')."\n"); + $self->{-cfg} = $lei->cfg_dump($f) // return $lei->child_error; $self->{-ovf} = "$dir/over.sqlite3"; $self->{'-f'} = $f; $self->{lock_path} = "$self->{-f}.flock"; @@ -92,14 +118,20 @@ sub up { # updating existing saved search via "lei up" sub new { # new saved search "lei q --save" my ($cls, $lei) = @_; my $self = bless { ale => $lei->ale }, $cls; + require File::Path; my $dst = $lei->{ovv}->{dst}; + + # canonicalize away relative paths into the config + if ($lei->{ovv}->{fmt} eq 'maildir' && + $dst =~ m!(?:/*|\A)\.\.(?:/*|\z)! && !-d $dst) { + File::Path::make_path($dst); + $lei->{ovv}->{dst} = $dst = $lei->abs_path($dst); + } my $dir = lss_dir_for($lei, \$dst); - require File::Path; File::Path::make_path($dir); # raises on error $self->{-cfg} = {}; my $f = $self->{'-f'} = "$dir/lei.saved-search"; - my $dd = $lei->{opt}->{dedupe}; - translate_dedupe($self, $lei, $dd) or return; + translate_dedupe($self, $lei) or return; open my $fh, '>', $f or return $lei->fail("open $f: $!"); my $sq_dst = PublicInbox::Config::squote_maybe($dst); my $q = $lei->{mset_opt}->{q_raw} // die 'BUG: {q_raw} missing'; @@ -109,14 +141,14 @@ sub new { # new saved search "lei q --save" $q = "\tq = ".cquote_val($q); } $dst = "$lei->{ovv}->{fmt}:$dst" if $dst !~ m!\Aimaps?://!i; + $lei->{opt}->{output} = $dst; print $fh <{opt}->{$k} // next; for my $x (@$ary) { @@ -127,6 +159,14 @@ EOM my $val = $lei->{opt}->{$k} // next; print $fh "\t$k = ".($val ? 1 : 0)."\n"; } + for my $k (SINGLE_FIELDS) { + my $val = $lei->{opt}->{$k} // next; + print $fh "\t$k = $val\n"; + } + $lei->{opt}->{stdin} and print $fh <fail("close $f: $!"); $self->{lock_path} = "$self->{-f}.flock"; $self->{-ovf} = "$dir/over.sqlite3"; @@ -145,23 +185,24 @@ sub cfg_set { # called by LeiXSearch sub is_dup { my ($self, $eml, $smsg) = @_; my $oidx = $self->{oidx} // die 'BUG: no {oidx}'; - my $blob = $smsg ? $smsg->{blob} : undef; - my $lk = $self->lock_for_scope_fast; - return 1 if $blob && $oidx->blob_exists($blob); + my $lk; if ($self->{-dedupe_mid}) { + $lk //= $self->lock_for_scope_fast; for my $mid (@{mids_for_index($eml)}) { my ($id, $prv); return 1 if $oidx->next_by_mid($mid, \$id, \$prv); } } + my $blob = $smsg ? $smsg->{blob} : git_sha(1, $eml)->hexdigest; + $lk //= $self->lock_for_scope_fast; + return 1 if $oidx->blob_exists($blob); if (my $xoids = PublicInbox::LeiSearch::xoids_for($self, $eml, 1)) { for my $docid (values %$xoids) { $oidx->add_xref3($docid, -1, $blob, '.'); } $oidx->commit_lazy; if ($self->{-dedupe_oid}) { - $smsg->{blob} //= git_sha(1, $eml)->hexdigest; - exists $xoids->{$smsg->{blob}} ? 1 : undef; + exists $xoids->{$blob} ? 1 : undef; } else { 1; } @@ -172,11 +213,11 @@ sub is_dup { $smsg->{bytes} = 0; $smsg->populate($eml); } + $smsg->{blob} //= $blob; $oidx->begin_lazy; $smsg->{num} = $oidx->adj_counter('eidx_docid', '+'); - $smsg->{blob} //= git_sha(1, $eml)->hexdigest; $oidx->add_overview($eml, $smsg); - $oidx->add_xref3($smsg->{num}, -1, $smsg->{blob}, '.'); + $oidx->add_xref3($smsg->{num}, -1, $blob, '.'); $oidx->commit_lazy; undef; } @@ -200,16 +241,28 @@ sub prepare_dedupe { sub over { $_[0]->{oidx} } # for xoids_for -sub git { $_[0]->{ale}->git } +# don't use ale->git directly since is_dup is called inside +# ale->git->cat_async callbacks +sub git { $_[0]->{git} //= PublicInbox::Git->new($_[0]->{ale}->git->{git_dir}) } sub pause_dedupe { my ($self) = @_; - $self->{ale}->git->cleanup; + git($self)->cleanup; my $lockfh = delete $self->{lockfh}; # from lock_for_scope_fast; my $oidx = delete($self->{oidx}) // return; $oidx->commit_lazy; } +sub reset_dedupe { + my ($self) = @_; + prepare_dedupe($self); + my $lk = $self->lock_for_scope_fast; + for my $t (qw(xref3 over id2num)) { + $self->{oidx}->{dbh}->do("DELETE FROM $t"); + } + pause_dedupe($self); +} + sub mm { undef } sub altid_map { {} } @@ -221,10 +274,10 @@ sub cloneurl { [] } sub output2lssdir { my ($self, $lei, $dir_ref, $fn_ref) = @_; my $dst = $$dir_ref; # imap://$MAILBOX, /path/to/maildir, /path/to/mbox - my $dir = lss_dir_for($lei, \$dst); + my $dir = lss_dir_for($lei, \$dst, 1); my $f = "$dir/lei.saved-search"; if (-f $f && -r _) { - $self->{-cfg} = PublicInbox::Config->git_config_dump($f); + $self->{-cfg} = $lei->cfg_dump($f) // return; $$dir_ref = $dir; $$fn_ref = $f; return 1; @@ -232,6 +285,13 @@ sub output2lssdir { undef; } +# cf. LeiDedupe->has_entries +sub has_entries { + my $oidx = $_[0]->{oidx} // die 'BUG: no {oidx}'; + my @n = $oidx->{dbh}->selectrow_array('SELECT num FROM over LIMIT 1'); + scalar(@n) ? 1 : undef; +} + no warnings 'once'; *nntp_url = \&cloneurl; *base_url = \&PublicInbox::Inbox::base_url;