1 # Copyright (C) 2021 all contributors <meta@public-inbox.org>
2 # License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
4 # handles lei <q|ls-query|rm-query|mv-query> commands
5 package PublicInbox::LeiQuery;
8 use PublicInbox::DS qw(dwaitpid);
10 sub prep_ext { # externals_each callback
11 my ($lxs, $exclude, $loc) = @_;
12 $lxs->prepare_external($loc) unless $exclude->{$loc};
15 sub qstr_add { # for --stdin
16 my ($self) = @_; # $_[1] = $rbuf
18 return eval { $self->{lxs}->do_query($self) } if $_[1] eq '';
19 $self->{mset_opt}->{qstr} .= $_[1];
21 $self->fail("error reading stdin: $!");
25 # the main "lei q SEARCH_TERMS" method
27 my ($self, @argv) = @_;
28 require PublicInbox::LeiXSearch;
29 require PublicInbox::LeiOverview;
30 require PublicInbox::V2Writable;
31 PublicInbox::Config->json; # preload before forking
32 my $opt = $self->{opt};
33 # prepare any number of LeiXSearch || LeiSearch || Inbox || URL
34 my $lxs = $self->{lxs} = PublicInbox::LeiXSearch->new;
35 my @only = @{$opt->{only} // []};
36 # --local is enabled by default unless --only is used
37 # we'll allow "--only $LOCATION --local"
38 if ($opt->{'local'} //= scalar(@only) ? 0 : 1) {
39 my $sto = $self->_lei_store(1);
40 $lxs->prepare_external($sto->search);
44 my @loc = $self->get_externals($loc) or return;
45 $lxs->prepare_external($_) for @loc;
48 for my $loc (@{$opt->{include} // []}) {
49 my @loc = $self->get_externals($loc) or return;
50 $lxs->prepare_external($_) for @loc;
52 # --external is enabled by default, but allow --no-external
53 if ($opt->{external} //= 1) {
55 for my $loc (@{$opt->{exclude} // []}) {
56 my @l = $self->get_externals($loc, 1) or return;
59 my $ne = $self->externals_each(\&prep_ext, $lxs, \%x);
60 $opt->{remote} //= !($lxs->locals - $opt->{'local'});
61 if ($opt->{'local'}) {
62 delete($lxs->{remotes}) if !$opt->{remote};
64 delete($lxs->{locals});
68 unless ($lxs->locals || $lxs->remotes) {
69 return $self->fail('no local or remote inboxes to search');
71 my ($xj, $mj) = split(/,/, $opt->{jobs} // '');
72 if (defined($xj) && $xj ne '' && $xj !~ /\A[1-9][0-9]*\z/) {
73 return $self->fail("`$xj' search jobs must be >= 1");
75 $xj ||= $lxs->concurrency($opt); # allow: "--jobs ,$WRITER_ONLY"
76 my $nproc = $lxs->detect_nproc; # don't memoize, schedtool(1) exists
77 $xj = $nproc if $xj > $nproc;
78 PublicInbox::LeiOverview->new($self) or return;
79 $self->atfork_prepare_wq($lxs);
80 $lxs->wq_workers_start('lei_xsearch', $xj, $self->oldset);
81 delete $lxs->{-ipc_atfork_child_close};
82 if (my $l2m = $self->{l2m}) {
83 if (defined($mj) && $mj !~ /\A[1-9][0-9]*\z/) {
84 return $self->fail("`$mj' writer jobs must be >= 1");
87 $self->atfork_prepare_wq($l2m);
88 $l2m->wq_workers_start('lei2mail', $mj, $self->oldset);
89 delete $l2m->{-ipc_atfork_child_close};
92 # no forking workers after this
94 my %mset_opt = map { $_ => $opt->{$_} } qw(thread limit offset);
95 $mset_opt{asc} = $opt->{'reverse'} ? 1 : 0;
96 $mset_opt{limit} //= 10000;
97 if (defined(my $sort = $opt->{'sort'})) {
98 if ($sort eq 'relevance') {
99 $mset_opt{relevance} = 1;
100 } elsif ($sort eq 'docid') {
101 $mset_opt{relevance} = $mset_opt{asc} ? -1 : -2;
102 } elsif ($sort =~ /\Areceived(?:-?[aA]t)?\z/) {
105 die "unrecognized --sort=$sort\n";
108 # descending docid order
109 $mset_opt{relevance} //= -2 if $opt->{thread};
110 $self->{mset_opt} = \%mset_opt;
113 return $self->fail(<<'') if @argv;
114 no query allowed on command-line with --stdin
116 require PublicInbox::InputPipe;
117 PublicInbox::InputPipe::consume($self->{0}, \&qstr_add, $self);
120 # Consider spaces in argv to be for phrase search in Xapian.
121 # In other words, the users should need only care about
122 # normal shell quotes and not have to learn Xapian quoting.
123 $mset_opt{qstr} = join(' ', map {;
124 /\s/ ? (s/\A(\w+:)// ? qq{$1"$_"} : qq{"$_"}) : $_
126 $lxs->do_query($self);
129 # shell completion helper called by lei__complete
131 my ($self, @argv) = @_;
132 my $ext = qr/\A(?:-I|(?:--(?:include|exclude|only)))\z/;
135 if ($argv[-1] =~ $ext) {
136 my @c = $self->_complete_forget_external(@cur);
137 # try basename match:
138 if (scalar(@cur) == 1 && index($cur[0], '/') < 0) {
139 my $all = $self->externals_each;
141 for my $loc (keys %$all) {
142 my $bn = (split(m!/!, $loc))[-1];
146 $bn{$_} == 1 && /\A\Q$cur[0]/
151 unshift(@cur, pop @argv);
156 # Stuff we may pass through to curl (as of 7.64.0), see curl manpage for
157 # details, so most options which make sense for HTTP/HTTPS (including proxy
158 # support for Tor and other methods of getting past weird networks).
159 # Most of these are untested by us, some may not make sense for our use case
160 # and typos below are likely.
161 # n.b. some short options (-$NUMBER) are not supported since they conflict
162 # with other "lei q" switches.
163 # FIXME: Getopt::Long doesn't easily let us support support options with
164 # '.' in them (e.g. --http1.1)
166 abstract-unix-socket=s anyauth basic cacert=s capath=s
167 cert-status cert-type cert|E=s ciphers=s config|K=s@
168 connect-timeout=s connect-to=s cookie-jar|c=s cookie|b=s crlfile=s
169 digest disable dns-interface=s dns-ipv4-addr=s dns-ipv6-addr=s
170 dns-servers=s doh-url=s egd-file=s engine=s false-start
171 happy-eyeballs-timeout-ms=s haproxy-protocol header|H=s@
172 http2-prior-knowledge http2 insecure|k
173 interface=s ipv4 ipv6 junk-session-cookies
174 key-type=s key=s limit-rate=s local-port=s location-trusted location|L
175 max-redirs=i max-time=s negotiate netrc-file=s netrc-optional netrc
176 no-alpn no-buffer|N no-npn no-sessionid noproxy=s ntlm-wb ntlm
177 pass=s pinnedpubkey=s post301 post302 post303 preproxy=s
178 proxy-anyauth proxy-basic proxy-cacert=s proxy-capath=s
179 proxy-cert-type=s proxy-cert=s proxy-ciphers=s proxy-crlfile=s
180 proxy-digest proxy-header=s@ proxy-insecure
181 proxy-key-type=s proxy-key proxy-negotiate proxy-ntlm proxy-pass=s
182 proxy-pinnedpubkey=s proxy-service-name=s proxy-ssl-allow-beast
183 proxy-tls13-ciphers=s proxy-tlsauthtype=s proxy-tlspassword=s
184 proxy-tlsuser=s proxy-tlsv1 proxy-user|U=s proxy=s
185 proxytunnel=s pubkey=s random-file=s referer=s resolve=s
186 retry-connrefused retry-delay=s retry-max-time=s retry=i
187 sasl-ir service-name=s socks4=s socks4a=s socks5-basic
188 socks5-gssapi-service-name=s socks5-gssapi socks5-hostname=s socks5=s
189 speed-limit|Y speed-type|y ssl-allow-beast sslv2 sslv3
190 suppress-connect-headers tcp-fastopen tls-max=s
191 tls13-ciphers=s tlsauthtype=s tlspassword=s tlsuser=s
192 tlsv1 trace-ascii=s trace-time trace=s
193 unix-socket=s user-agent|A=s user|u=s