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;
9 sub prep_ext { # externals_each callback
10 my ($lxs, $exclude, $loc) = @_;
11 $lxs->prepare_external($loc) unless $exclude->{$loc};
14 sub qstr_add { # for --stdin
15 my ($self) = @_; # $_[1] = $rbuf
17 return eval { $self->{lxs}->do_query($self) } if $_[1] eq '';
18 $self->{mset_opt}->{qstr} .= $_[1];
20 $self->fail("error reading stdin: $!");
24 # the main "lei q SEARCH_TERMS" method
26 my ($self, @argv) = @_;
27 require PublicInbox::LeiXSearch;
28 require PublicInbox::LeiOverview;
29 require PublicInbox::V2Writable;
30 PublicInbox::Config->json; # preload before forking
31 my $opt = $self->{opt};
32 # prepare any number of LeiXSearch || LeiSearch || Inbox || URL
33 my $lxs = $self->{lxs} = PublicInbox::LeiXSearch->new;
34 my @only = @{$opt->{only} // []};
35 # --local is enabled by default unless --only is used
36 # we'll allow "--only $LOCATION --local"
37 if ($opt->{'local'} //= scalar(@only) ? 0 : 1) {
38 my $sto = $self->_lei_store(1);
39 $lxs->prepare_external($sto->search);
43 my @loc = $self->get_externals($loc) or return;
44 $lxs->prepare_external($_) for @loc;
47 for my $loc (@{$opt->{include} // []}) {
48 my @loc = $self->get_externals($loc) or return;
49 $lxs->prepare_external($_) for @loc;
51 # --external is enabled by default, but allow --no-external
52 if ($opt->{external} //= 1) {
54 for my $loc (@{$opt->{exclude} // []}) {
55 my @l = $self->get_externals($loc, 1) or return;
58 my $ne = $self->externals_each(\&prep_ext, $lxs, \%x);
59 $opt->{remote} //= !($lxs->locals - $opt->{'local'});
60 if ($opt->{'local'}) {
61 delete($lxs->{remotes}) if !$opt->{remote};
63 delete($lxs->{locals});
67 unless ($lxs->locals || $lxs->remotes) {
68 return $self->fail('no local or remote inboxes to search');
70 my ($xj, $mj) = split(/,/, $opt->{jobs} // '');
71 if (defined($xj) && $xj ne '' && $xj !~ /\A[1-9][0-9]*\z/) {
72 return $self->fail("`$xj' search jobs must be >= 1");
74 $xj ||= $lxs->concurrency($opt); # allow: "--jobs ,$WRITER_ONLY"
75 my $nproc = $lxs->detect_nproc; # don't memoize, schedtool(1) exists
76 $xj = $nproc if $xj > $nproc;
78 if (defined($mj) && $mj !~ /\A[1-9][0-9]*\z/) {
79 return $self->fail("`$mj' writer jobs must be >= 1");
81 $self->{l2m}->{jobs} = ($mj // $nproc) if $self->{l2m};
82 PublicInbox::LeiOverview->new($self) or return;
84 my %mset_opt = map { $_ => $opt->{$_} } qw(threads limit offset);
85 $mset_opt{asc} = $opt->{'reverse'} ? 1 : 0;
86 $mset_opt{limit} //= 10000;
87 if (defined(my $sort = $opt->{'sort'})) {
88 if ($sort eq 'relevance') {
89 $mset_opt{relevance} = 1;
90 } elsif ($sort eq 'docid') {
91 $mset_opt{relevance} = $mset_opt{asc} ? -1 : -2;
92 } elsif ($sort =~ /\Areceived(?:-?[aA]t)?\z/) {
95 die "unrecognized --sort=$sort\n";
98 # descending docid order
99 $mset_opt{relevance} //= -2 if $opt->{threads};
100 $self->{mset_opt} = \%mset_opt;
103 return $self->fail(<<'') if @argv;
104 no query allowed on command-line with --stdin
106 require PublicInbox::InputPipe;
107 PublicInbox::InputPipe::consume($self->{0}, \&qstr_add, $self);
110 # Consider spaces in argv to be for phrase search in Xapian.
111 # In other words, the users should need only care about
112 # normal shell quotes and not have to learn Xapian quoting.
113 $mset_opt{qstr} = join(' ', map {;
114 /\s/ ? (s/\A(\w+:)// ? qq{$1"$_"} : qq{"$_"}) : $_
116 $lxs->do_query($self);
119 # shell completion helper called by lei__complete
121 my ($self, @argv) = @_;
122 my $ext = qr/\A(?:-I|(?:--(?:include|exclude|only)))\z/;
125 if ($argv[-1] =~ $ext) {
126 my @c = $self->_complete_forget_external(@cur);
127 # try basename match:
128 if (scalar(@cur) == 1 && index($cur[0], '/') < 0) {
129 my $all = $self->externals_each;
131 for my $loc (keys %$all) {
132 my $bn = (split(m!/!, $loc))[-1];
136 $bn{$_} == 1 && /\A\Q$cur[0]/
141 unshift(@cur, pop @argv);
146 # Stuff we may pass through to curl (as of 7.64.0), see curl manpage for
147 # details, so most options which make sense for HTTP/HTTPS (including proxy
148 # support for Tor and other methods of getting past weird networks).
149 # Most of these are untested by us, some may not make sense for our use case
150 # and typos below are likely.
151 # n.b. some short options (-$NUMBER) are not supported since they conflict
152 # with other "lei q" switches.
153 # FIXME: Getopt::Long doesn't easily let us support support options with
154 # '.' in them (e.g. --http1.1)
155 # TODO: should we depend on "-c http.*" options for things which have
156 # analogues in git(1)? that would reduce likelyhood of conflicts with
157 # our other CLI options
158 # Note: some names are renamed to avoid potential conflicts,
159 # see %lei2mail in lib/PublicInbox/LeiCurl.pm
162 abstract-unix-socket=s anyauth basic cacert=s capath=s
163 cert-status cert-type cert=s ciphers=s
164 connect-timeout=s connect-to=s cookie-jar=s cookie=s crlfile=s
165 digest disable dns-interface=s dns-ipv4-addr=s dns-ipv6-addr=s
166 dns-servers=s doh-url=s egd-file=s engine=s false-start
167 happy-eyeballs-timeout-ms=s haproxy-protocol header=s@
168 http2-prior-knowledge http2 insecure
169 interface=s ipv4 ipv6 junk-session-cookies
170 key-type=s key=s limit-rate=s local-port=s location-trusted location
171 max-redirs=i max-time=s negotiate netrc-file=s netrc-optional netrc
172 no-alpn no-buffer no-npn no-sessionid noproxy=s ntlm-wb ntlm
173 pass=s pinnedpubkey=s post301 post302 post303 preproxy=s
174 proxy-anyauth proxy-basic proxy-cacert=s proxy-capath=s
175 proxy-cert-type=s proxy-cert=s proxy-ciphers=s proxy-crlfile=s
176 proxy-digest proxy-header=s@ proxy-insecure
177 proxy-key-type=s proxy-key proxy-negotiate proxy-ntlm proxy-pass=s
178 proxy-pinnedpubkey=s proxy-service-name=s proxy-ssl-allow-beast
179 proxy-tls13-ciphers=s proxy-tlsauthtype=s proxy-tlspassword=s
180 proxy-tlsuser=s proxy-tlsv1 proxy-user=s proxy=s
181 proxytunnel=s pubkey=s random-file=s referer=s resolve=s
182 retry-connrefused retry-delay=s retry-max-time=s retry=i
183 sasl-ir service-name=s socks4=s socks4a=s socks5-basic
184 socks5-gssapi-service-name=s socks5-gssapi socks5-hostname=s socks5=s
185 speed-limit speed-type ssl-allow-beast sslv2 sslv3
186 suppress-connect-headers tcp-fastopen tls-max=s
187 tls13-ciphers=s tlsauthtype=s tlspassword=s tlsuser=s
188 tlsv1 trace-ascii=s trace-time trace=s
189 unix-socket=s user-agent=s user=s