X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FLEI.pm;h=28fe0c836d7a3e842217b9bfd3cf63ac8ece2bf8;hb=b6eb866869609afef72c77f41507905828014673;hp=546fa773fc4c334cc83a27358ba74f016cf62dab;hpb=8e112fe221a74c2689c0a43576ed766deb0397e4;p=public-inbox.git diff --git a/lib/PublicInbox/LEI.pm b/lib/PublicInbox/LEI.pm index 546fa773..28fe0c83 100644 --- a/lib/PublicInbox/LEI.pm +++ b/lib/PublicInbox/LEI.pm @@ -28,14 +28,16 @@ use Time::HiRes qw(stat); # ctime comparisons for config cache use File::Path qw(mkpath); use File::Spec; our $quit = \&CORE::exit; -our ($current_lei, $errors_log, $listener, $oldset, $dir_idle); -my ($recv_cmd, $send_cmd); +our ($current_lei, $errors_log, $listener, $oldset, $dir_idle, + $recv_cmd, $send_cmd); my $GLP = Getopt::Long::Parser->new; $GLP->configure(qw(gnu_getopt no_ignore_case auto_abbrev)); my $GLP_PASS = Getopt::Long::Parser->new; $GLP_PASS->configure(qw(gnu_getopt no_ignore_case auto_abbrev pass_through)); our %PATH2CFG; # persistent for socket daemon +our $MDIR2CFGPATH; # /path/to/maildir => { /path/to/config => [ ino watches ] } +our %LIVE_SOCK; # "GLOB(0x....)" => $lei->{sock} # TBD: this is a documentation mechanism to show a subcommand # (may) pass options through to another command: @@ -93,6 +95,12 @@ sub rel2abs { # abs_path resolves symlinks in parent iff all parents exist sub abs_path { Cwd::abs_path($_[1]) // rel2abs(@_) } +sub canonpath_harder { + my $p = $_[-1]; # $_[0] may be self + $p = File::Spec->canonpath($p); + $p =~ m!(?:/*|\A)\.\.(?:/*|\z)! && -e $p ? Cwd::abs_path($p) : $p; +} + sub share_path ($) { # $HOME/.local/share/lei/$FOO my ($self) = @_; rel2abs($self, ($self->{env}->{XDG_DATA_HOME} // @@ -133,7 +141,7 @@ sub ale { sub index_opt { # TODO: drop underscore variants everywhere, they're undocumented qw(fsync|sync! jobs|j=i indexlevel|L=s compact - max_size|max-size=s sequential_shard|sequential-shard + max_size|max-size=s sequential-shard batch_size|batch-size=s skip-docdata) } @@ -166,12 +174,12 @@ our %CMD = ( # sorted in order of importance/use: 'q' => [ '--stdin|SEARCH_TERMS...', 'search for messages matching terms', 'stdin|', # /|\z/ must be first for lone dash @lxs_opt, - qw(save output|mfolder|o=s format|f=s dedupe|d=s threads|t+ + qw(save! output|mfolder|o=s format|f=s dedupe|d=s threads|t+ sort|s=s reverse|r offset=i pretty jobs|j=s globoff|g augment|a import-before! lock=s@ rsyncable alert=s@ mua=s verbose|v+ shared color! mail-sync!), @c_opt, opt_dash('limit|n=i', '[0-9]+') ], -'up' => [ 'OUTPUT|--all', 'update saved search', +'up' => [ 'OUTPUT...|--all', 'update saved search', qw(jobs|j=s lock=s@ alert=s@ mua=s verbose|v+ all:s), @c_opt ], 'lcat' => [ '--stdin|MSGID_OR_URL...', 'display local copy of message(s)', @@ -231,11 +239,12 @@ our %CMD = ( # sorted in order of importance/use: 'remove imported messages from IMAP, Maildirs, and MH', qw(exact! all jobs:i indexed), @c_opt ], -'add-watch' => [ 'LOCATION', 'watch for new messages and flag changes', - qw(import! kw! interval=s recursive|r - exclude=s include=s), @c_opt ], +'add-watch' => [ 'LOCATION...', 'watch for new messages and flag changes', + qw(poll-interval=s state=s recursive|r), @c_opt ], +'rm-watch' => [ 'LOCATION...', 'remove specified watch(es)', + qw(recursive|r), @c_opt ], 'ls-watch' => [ '[FILTER...]', 'list active watches with numbers and status', - qw(format|f=s z), @c_opt ], + qw(l z|0), @c_opt ], 'pause-watch' => [ '[WATCH_NUMBER_OR_FILTER]', qw(all local remote), @c_opt ], 'resume-watch' => [ '[WATCH_NUMBER_OR_FILTER]', qw(all local remote), @c_opt ], 'forget-watch' => [ '{WATCH_NUMBER|--prune}', 'stop and forget a watch', @@ -323,12 +332,13 @@ my %OPTDESC = ( 'path-a|a=s' => 'pre-image pathname associated with OID', 'path-b|b=s' => 'post-image pathname associated with OID', 'git-dir=s@' => 'additional git repository to scan', +'dir=s inspect' => 'specify a inboxdir, extindex topdir or Xapian shard', 'proxy=s' => [ 'PROTO://HOST[:PORT]', # shared with curl(1) "proxy for (e.g. `socks5h://0:9050')" ], 'torsocks=s' => ['VAL|auto|no|yes', 'whether or not to wrap git and curl commands with torsocks'], 'no-torsocks' => 'alias for --torsocks=no', -'save' => "save a search for `lei up'", +'save!' => "do not save a search for `lei up'", 'import-remote!' => 'do not memoize remote messages into local store', 'type=s' => [ 'any|mid|git', 'disambiguate type' ], @@ -368,7 +378,7 @@ my %OPTDESC = ( 'do not index messages larger than SIZE (default: infinity)' ], 'batch_size|batch-size=s' => [ 'SIZE', 'flush changes to OS after given number of bytes (default: 1m)' ], -'sequential_shard|sequential-shard' => +'sequential-shard' => 'index Xapian shards sequentially for slow storage', 'skip-docdata' => 'drop compatibility w/ public-inbox <1.6 to save ~1.5% space', @@ -390,6 +400,7 @@ my %OPTDESC = ( 'format|f=s ls-search' => ['OUT|json|jsonl|concatjson', 'listing output format' ], 'l ls-search' => 'long listing format', +'l ls-watch' => 'long listing format', 'l ls-mail-source' => 'long listing format', 'url ls-mail-source' => 'show full URL of newsgroup or IMAP folder', 'format|f=s ls-external' => $ls_format, @@ -434,7 +445,7 @@ my %CONFIG_KEYS = ( 'leistore.dir' => 'top-level storage location', ); -my @WQ_KEYS = qw(lxs l2m ikw pmd wq1); # internal workers +my @WQ_KEYS = qw(lxs l2m ikw pmd wq1 lne); # internal workers sub _drop_wq { my ($self) = @_; @@ -475,6 +486,12 @@ sub err ($;@) { sub qerr ($;@) { $_[0]->{opt}->{quiet} or err(shift, @_) } +sub qfin { # show message on finalization (LeiFinmsg) + my ($lei, $msg) = @_; + return if $lei->{opt}->{quiet}; + $lei->{fmsg} ? push(@{$lei->{fmsg}}, "$msg\n") : qerr($lei, $msg); +} + sub fail_handler ($;$$) { my ($lei, $code, $io) = @_; close($io) if $io; # needed to avoid warnings on SIGPIPE @@ -537,7 +554,7 @@ sub _lei_atfork_child { chdir '/' or die "chdir(/): $!"; close($_) for (grep(defined, delete @$self{qw(0 1 2 sock)})); if (my $cfg = $self->{cfg}) { - delete $cfg->{-lei_store}; + delete @$cfg{qw(-lei_store -watches -lei_note_event)}; } } else { # worker, Net::NNTP (Net::Cmd) uses STDERR directly open STDERR, '+>&='.fileno($self->{2}) or warn "open $!"; @@ -552,8 +569,11 @@ sub _lei_atfork_child { } close $listener if $listener; undef $listener; - undef $dir_idle; + $dir_idle->force_close if $dir_idle; %PATH2CFG = (); + $MDIR2CFGPATH = {}; + %LIVE_SOCK = (); + eval 'no warnings; undef $PublicInbox::LeiNoteEvent::to_flush'; undef $errors_log; $quit = \&CORE::exit; $self->{-eml_noisy} or # only "lei import" sets this atm @@ -762,11 +782,7 @@ sub dispatch { next if $d eq ''; # same as git(1) chdir $d or return fail($self, "cd $d: $!"); } - if (delete $self->{3}) { # update cwd for rel2abs - opendir my $dh, '.' or - return fail($self, "opendir . $!"); - $self->{3} = $dh; - } + open $self->{3}, '.' or return fail($self, "open . $!"); } $cb->($self, @argv); } elsif (grep(/\A-/, $cmd, @argv)) { # --help or -h only @@ -784,17 +800,19 @@ sub _lei_cfg ($;$) { my $f = _config_path($self); my @st = stat($f); my $cur_st = @st ? pack('dd', $st[10], $st[7]) : ''; # 10:ctime, 7:size - my ($sto, $sto_dir); + my ($sto, $sto_dir, $watches, $lne); if (my $cfg = $PATH2CFG{$f}) { # reuse existing object in common case return ($self->{cfg} = $cfg) if $cur_st eq $cfg->{-st}; - ($sto, $sto_dir) = @$cfg{qw(-lei_store leistore.dir)}; + ($sto, $sto_dir, $watches, $lne) = + @$cfg{qw(-lei_store leistore.dir -watches + -lei_note_event)}; } if (!@st) { unless ($creat) { delete $self->{cfg}; return bless {}, 'PublicInbox::Config'; } - my (undef, $cfg_dir, undef) = File::Spec->splitpath($f); + my ($cfg_dir) = ($f =~ m!(.*?/)[^/]+\z!); -d $cfg_dir or mkpath($cfg_dir) or die "mkpath($cfg_dir): $!\n"; open my $fh, '>>', $f or die "open($f): $!\n"; @st = stat($fh) or die "fstat($f): $!\n"; @@ -804,10 +822,12 @@ sub _lei_cfg ($;$) { my $cfg = PublicInbox::Config->git_config_dump($f); $cfg->{-st} = $cur_st; $cfg->{'-f'} = $f; - if ($sto && File::Spec->canonpath($sto_dir // store_path($self)) - eq File::Spec->canonpath($cfg->{'leistore.dir'} // + if ($sto && canonpath_harder($sto_dir // store_path($self)) + eq canonpath_harder($cfg->{'leistore.dir'} // store_path($self))) { $cfg->{-lei_store} = $sto; + $cfg->{-lei_note_event} = $lne; + $cfg->{-watches} = $watches if $watches; } if (scalar(keys %PATH2CFG) > 5) { # FIXME: use inotify/EVFILT_VNODE to detect unlinked configs @@ -816,11 +836,13 @@ sub _lei_cfg ($;$) { } } $self->{cfg} = $PATH2CFG{$f} = $cfg; + refresh_watches($self); + $cfg; } sub _lei_store ($;$) { my ($self, $creat) = @_; - my $cfg = _lei_cfg($self, $creat); + my $cfg = _lei_cfg($self, $creat) // return; $cfg->{-lei_store} //= do { require PublicInbox::LeiStore; my $dir = $cfg->{'leistore.dir'} // store_path($self); @@ -1098,6 +1120,7 @@ sub event_step { } die "unrecognized client signal: $buf"; } + _drop_wq($self); # EOF, client disconnected dclose($self); }; if (my $err = $@) { @@ -1128,6 +1151,53 @@ sub dump_and_clear_log { } } +sub cfg2lei ($) { + my ($cfg) = @_; + my $lei = bless { env => { %{$cfg->{-env}} } }, __PACKAGE__; + open($lei->{0}, '<&', \*STDIN) or die "dup 0: $!"; + open($lei->{1}, '>>&', \*STDOUT) or die "dup 1: $!"; + open($lei->{2}, '>>&', \*STDERR) or die "dup 2: $!"; + open($lei->{3}, '/') or die "open /: $!"; + chdir($lei->{3}) or die "chdir /': $!"; + my ($x, $y); + socketpair($x, $y, AF_UNIX, SOCK_SEQPACKET, 0) or die "socketpair: $!"; + $lei->{sock} = $x; + require PublicInbox::LeiSelfSocket; + PublicInbox::LeiSelfSocket->new($y); # adds to event loop + $lei; +} + +sub dir_idle_handler ($) { # PublicInbox::DirIdle callback + my ($ev) = @_; # Linux::Inotify2::Event or duck type + my $fn = $ev->fullname; + if ($fn =~ m!\A(.+)/(new|cur)/([^/]+)\z!) { # Maildir file + my ($mdir, $nc, $bn) = ($1, $2, $3); + $nc = '' if $ev->IN_DELETE; + for my $f (keys %{$MDIR2CFGPATH->{$mdir} // {}}) { + my $cfg = $PATH2CFG{$f} // next; + eval { + local %ENV = %{$cfg->{-env}}; + my $lei = cfg2lei($cfg); + $lei->dispatch('note-event', + "maildir:$mdir", $nc, $bn, $fn); + }; + warn "E note-event $f: $@\n" if $@; + } + } + if ($ev->can('cancel') && ($ev->IN_IGNORE || $ev->IN_UNMOUNT)) { + $ev->cancel; + } + if ($fn =~ m!\A(.+)/(?:new|cur)\z! && !-e $fn) { + delete $MDIR2CFGPATH->{$1}; + } + if (!-e $fn) { # config file or Maildir gone + for my $cfgpaths (values %$MDIR2CFGPATH) { + delete $cfgpaths->{$fn}; + } + delete $PATH2CFG{$fn}; + } +} + # lei(1) calls this when it can't connect sub lazy_start { my ($path, $errno, $narg) = @_; @@ -1149,7 +1219,6 @@ sub lazy_start { } umask(077) // die("umask(077): $!"); bind($listener, $addr) or die "bind($path): $!"; - listen($listener, 1024) or die "listen: $!"; $lk->lock_release; undef $lk; my @st = stat($path) or die "stat($path): $!"; @@ -1177,6 +1246,7 @@ sub lazy_start { return if $pid; $0 = "lei-daemon $path"; local %PATH2CFG; + local $MDIR2CFGPATH; $listener->blocking(0); my $exit_code; my $pil = PublicInbox::Listener->new($listener, \&accept_dispatch); @@ -1206,8 +1276,8 @@ sub lazy_start { local $SIG{PIPE} = 'IGNORE'; require PublicInbox::DirIdle; local $dir_idle = PublicInbox::DirIdle->new([$sock_dir], sub { - # just rely on wakeup ot hit PostLoopCallback set below - _dir_idle_handler(@_) if $_[0]->fullname ne $path; + # just rely on wakeup to hit PostLoopCallback set below + dir_idle_handler($_[0]) if $_[0]->fullname ne $path; }, 1); if ($sigfd) { undef $sigfd; # unref, already in DS::DescriptorMap @@ -1295,4 +1365,125 @@ sub wq_eof { # EOF callback for main daemon $wq1->wq_wait_old(\&wq_done_wait, $lei); } +sub watch_state_ok ($) { + my ($state) = $_[-1]; # $_[0] may be $self + $state =~ /\Apause|(?:import|index|tag)-(?:ro|rw)\z/; +} + +sub cancel_maildir_watch ($$) { + my ($d, $cfg_f) = @_; + my $w = delete $MDIR2CFGPATH->{$d}->{$cfg_f}; + scalar(keys %{$MDIR2CFGPATH->{$d}}) or + delete $MDIR2CFGPATH->{$d}; + for my $x (@{$w // []}) { $x->cancel } +} + +sub add_maildir_watch ($$) { + my ($d, $cfg_f) = @_; + if (!exists($MDIR2CFGPATH->{$d}->{$cfg_f})) { + my @w = $dir_idle->add_watches(["$d/cur", "$d/new"], 1); + push @{$MDIR2CFGPATH->{$d}->{$cfg_f}}, @w if @w; + } +} + +sub refresh_watches { + my ($lei) = @_; + my $cfg = _lei_cfg($lei) or return; + my $old = $cfg->{-watches}; + my $watches = $cfg->{-watches} //= {}; + my %seen; + my $cfg_f = $cfg->{'-f'}; + for my $w (grep(/\Awatch\..+\.state\z/, keys %$cfg)) { + my $url = substr($w, length('watch.'), -length('.state')); + require PublicInbox::LeiWatch; + $watches->{$url} //= PublicInbox::LeiWatch->new($url); + $seen{$url} = undef; + my $state = $cfg->get_1("watch.$url", 'state'); + if (!watch_state_ok($state)) { + $lei->err("watch.$url.state=$state not supported"); + next; + } + if ($url =~ /\Amaildir:(.+)/i) { + my $d = canonpath_harder($1); + if ($state eq 'pause') { + cancel_maildir_watch($d, $cfg_f); + } else { + add_maildir_watch($d, $cfg_f); + } + } else { # TODO: imap/nntp/jmap + $lei->child_error(1, "E: watch $url not supported, yet") + } + } + + # add all known Maildir folders as implicit watches + my $sto = $lei->_lei_store; + my $renames = 0; + if (my $lms = $sto ? $sto->search->lms : undef) { + for my $d ($lms->folders('maildir:')) { + substr($d, 0, length('maildir:')) = ''; + my $cd = canonpath_harder($d); + my $f = "maildir:$cd"; + + # fixup old bugs while we're iterating: + if ($d ne $cd) { + $sto->ipc_do('lms_rename_folder', + "maildir:$d", $f); + ++$renames; + } + next if $watches->{$f}; # may be set to pause + require PublicInbox::LeiWatch; + $watches->{$f} = PublicInbox::LeiWatch->new($f); + $seen{$f} = undef; + add_maildir_watch($cd, $cfg_f); + } + } + $lei->sto_done_request if $renames; + if ($old) { # cull old non-existent entries + for my $url (keys %$old) { + next if exists $seen{$url}; + delete $old->{$url}; + if ($url =~ /\Amaildir:(.+)/i) { + my $d = canonpath_harder($1); + cancel_maildir_watch($d, $cfg_f); + } else { # TODO: imap/nntp/jmap + $lei->child_error(1, "E: watch $url TODO"); + } + } + } + if (scalar keys %$watches) { + $cfg->{-env} //= { %{$lei->{env}}, PWD => '/' }; # for cfg2lei + } else { + delete $cfg->{-watches}; + } +} + +sub git_blob_id { + my ($lei, $eml) = @_; + ($lei->{sto} // _lei_store($lei, 1))->git_blob_id($eml); +} + +sub lms { # read-only LeiMailSync + my ($lei) = @_; + my $lse = $lei->{lse} // do { + my $sto = $lei->{sto} // _lei_store($lei); + $sto ? $sto->search : undef + }; + $lse ? $lse->lms : undef; +} + +sub sto_done_request { # only call this from lei-daemon process (not workers) + my ($lei, $sock) = @_; + if ($sock //= $lei->{sock}) { + $LIVE_SOCK{"$sock"} = $sock; + $lei->{sto}->ipc_do('done', "$sock"); # issue, async wait + } else { # forcibly wait + my $wait = $lei->{sto}->ipc_do('done'); + } +} + +sub sto_done_complete { # called in lei-daemon when LeiStore->done is complete + my ($sock_str) = @_; + delete $LIVE_SOCK{$sock_str}; # frees {sock} for waiting lei clients +} + 1;