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;
10 sub prep_ext { # externals_each callback
11 my ($lxs, $exclude, $loc) = @_;
12 $lxs->prepare_external($loc) unless $exclude->{$loc};
18 require PublicInbox::LeiAuth;
19 $self->{auth} = PublicInbox::LeiAuth->new
21 $self->{lxs}->do_query($self);
24 sub qstr_add { # PublicInbox::InputPipe::consume callback for --stdin
25 my ($self) = @_; # $_[1] = $rbuf
27 $_[1] eq '' and return eval {
28 my $lse = delete $self->{lse};
29 $lse->query_approxidate($lse->git,
30 $self->{mset_opt}->{qstr});
33 $self->{mset_opt}->{qstr} .= $_[1];
35 $self->fail("error reading stdin: $!");
39 # the main "lei q SEARCH_TERMS" method
41 my ($self, @argv) = @_;
42 require PublicInbox::LeiXSearch;
43 require PublicInbox::LeiOverview;
44 require PublicInbox::V2Writable;
45 PublicInbox::Config->json; # preload before forking
46 my $opt = $self->{opt};
47 # prepare any number of LeiXSearch || LeiSearch || Inbox || URL
48 my $lxs = $self->{lxs} = PublicInbox::LeiXSearch->new;
49 my @only = @{$opt->{only} // []};
50 # --local is enabled by default unless --only is used
51 # we'll allow "--only $LOCATION --local"
52 my $sto = $self->_lei_store(1);
53 my $lse = $sto->search;
54 if (($opt->{'import-remote'} //= 1) |
55 ($opt->{'import-augment'} //= 1)) {
56 $sto->write_prepare($self);
58 if ($opt->{'local'} //= scalar(@only) ? 0 : 1) {
59 $lxs->prepare_external($lse);
63 my @loc = $self->get_externals($loc) or return;
64 $lxs->prepare_external($_) for @loc;
67 for my $loc (@{$opt->{include} // []}) {
68 my @loc = $self->get_externals($loc) or return;
69 $lxs->prepare_external($_) for @loc;
71 # --external is enabled by default, but allow --no-external
72 if ($opt->{external} //= 1) {
74 for my $loc (@{$opt->{exclude} // []}) {
75 my @l = $self->get_externals($loc, 1) or return;
78 my $ne = $self->externals_each(\&prep_ext, $lxs, \%x);
79 $opt->{remote} //= !($lxs->locals - $opt->{'local'});
80 if ($opt->{'local'}) {
81 delete($lxs->{remotes}) if !$opt->{remote};
83 delete($lxs->{locals});
87 unless ($lxs->locals || $lxs->remotes) {
88 return $self->fail('no local or remote inboxes to search');
90 my ($xj, $mj) = split(/,/, $opt->{jobs} // '');
91 if (defined($xj) && $xj ne '' && $xj !~ /\A[1-9][0-9]*\z/) {
92 return $self->fail("`$xj' search jobs must be >= 1");
94 $xj ||= $lxs->concurrency($opt); # allow: "--jobs ,$WRITER_ONLY"
95 my $nproc = $lxs->detect_nproc // 1; # don't memoize, schedtool(1) exists
96 $xj = $nproc if $xj > $nproc;
97 $lxs->{-wq_nr_workers} = $xj;
98 if (defined($mj) && $mj !~ /\A[1-9][0-9]*\z/) {
99 return $self->fail("`$mj' writer jobs must be >= 1");
101 PublicInbox::LeiOverview->new($self) or return;
102 $self->{l2m} and $self->{l2m}->{-wq_nr_workers} = $mj // do {
103 $mj = POSIX::lround($nproc * 3 / 4); # keep some CPU for git
107 my %mset_opt = map { $_ => $opt->{$_} } qw(threads limit offset);
108 $mset_opt{asc} = $opt->{'reverse'} ? 1 : 0;
109 $mset_opt{limit} //= 10000;
110 if (defined(my $sort = $opt->{'sort'})) {
111 if ($sort eq 'relevance') {
112 $mset_opt{relevance} = 1;
113 } elsif ($sort eq 'docid') {
114 $mset_opt{relevance} = $mset_opt{asc} ? -1 : -2;
115 } elsif ($sort =~ /\Areceived(?:-?[aA]t)?\z/) {
118 die "unrecognized --sort=$sort\n";
121 # descending docid order
122 $mset_opt{relevance} //= -2 if $opt->{threads};
123 $self->{mset_opt} = \%mset_opt;
126 return $self->fail(<<'') if @argv;
127 no query allowed on command-line with --stdin
129 require PublicInbox::InputPipe;
130 $self->{lse} = $lse; # for query_approxidate
131 PublicInbox::InputPipe::consume($self->{0}, \&qstr_add, $self);
134 $mset_opt{qstr} = $lse->query_argv_to_string($lse->git, \@argv);
138 # shell completion helper called by lei__complete
140 my ($self, @argv) = @_;
141 my $ext = qr/\A(?:-I|(?:--(?:include|exclude|only)))\z/;
144 if ($argv[-1] =~ $ext) {
145 my @c = $self->_complete_forget_external(@cur);
146 # try basename match:
147 if (scalar(@cur) == 1 && index($cur[0], '/') < 0) {
148 my $all = $self->externals_each;
150 for my $loc (keys %$all) {
151 my $bn = (split(m!/!, $loc))[-1];
155 $bn{$_} == 1 && /\A\Q$cur[0]/
160 unshift(@cur, pop @argv);
165 # Stuff we may pass through to curl (as of 7.64.0), see curl manpage for
166 # details, so most options which make sense for HTTP/HTTPS (including proxy
167 # support for Tor and other methods of getting past weird networks).
168 # Most of these are untested by us, some may not make sense for our use case
169 # and typos below are likely.
170 # n.b. some short options (-$NUMBER) are not supported since they conflict
171 # with other "lei q" switches.
172 # FIXME: Getopt::Long doesn't easily let us support support options with
173 # '.' in them (e.g. --http1.1)
174 # TODO: should we depend on "-c http.*" options for things which have
175 # analogues in git(1)? that would reduce likelyhood of conflicts with
176 # our other CLI options
177 # Note: some names are renamed to avoid potential conflicts,
178 # see %lei2mail in lib/PublicInbox/LeiCurl.pm
181 abstract-unix-socket=s anyauth basic cacert=s capath=s
182 cert-status cert-type cert=s ciphers=s
183 connect-timeout=s connect-to=s cookie-jar=s cookie=s crlfile=s
184 digest disable dns-interface=s dns-ipv4-addr=s dns-ipv6-addr=s
185 dns-servers=s doh-url=s egd-file=s engine=s false-start
186 happy-eyeballs-timeout-ms=s haproxy-protocol header=s@
187 http2-prior-knowledge http2 insecure
188 interface=s ipv4 ipv6 junk-session-cookies
189 key-type=s key=s limit-rate=s local-port=s location-trusted location
190 max-redirs=i max-time=s negotiate netrc-file=s netrc-optional netrc
191 no-alpn no-buffer no-npn no-sessionid noproxy=s ntlm-wb ntlm
192 pass=s pinnedpubkey=s post301 post302 post303 preproxy=s
193 proxy-anyauth proxy-basic proxy-cacert=s proxy-capath=s
194 proxy-cert-type=s proxy-cert=s proxy-ciphers=s proxy-crlfile=s
195 proxy-digest proxy-header=s@ proxy-insecure
196 proxy-key-type=s proxy-key proxy-negotiate proxy-ntlm proxy-pass=s
197 proxy-pinnedpubkey=s proxy-service-name=s proxy-ssl-allow-beast
198 proxy-tls13-ciphers=s proxy-tlsauthtype=s proxy-tlspassword=s
199 proxy-tlsuser=s proxy-tlsv1 proxy-user=s proxy=s
200 proxytunnel=s pubkey=s random-file=s referer=s resolve=s
201 retry-connrefused retry-delay=s retry-max-time=s retry=i
202 sasl-ir service-name=s socks4=s socks4a=s socks5-basic
203 socks5-gssapi-service-name=s socks5-gssapi socks5-hostname=s socks5=s
204 speed-limit speed-type ssl-allow-beast sslv2 sslv3
205 suppress-connect-headers tcp-fastopen tls-max=s
206 tls13-ciphers=s tlsauthtype=s tlspassword=s tlsuser=s
207 tlsv1 trace-ascii=s trace-time trace=s
208 unix-socket=s user-agent=s user=s