-# Copyright (C) 2015-2018 all contributors <meta@public-inbox.org>
+# Copyright (C) 2015-2020 all contributors <meta@public-inbox.org>
# License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
#
# Displays search results for the web interface
package PublicInbox::SearchView;
use strict;
-use warnings;
-use URI::Escape qw(uri_unescape uri_escape);
-use PublicInbox::SearchMsg;
-use PublicInbox::Hval qw/ascii_html obfuscate_addrs/;
+use v5.10.1;
+use List::Util qw(min max);
+use URI::Escape qw(uri_unescape);
+use PublicInbox::Smsg;
+use PublicInbox::Hval qw(ascii_html obfuscate_addrs mid_href fmt_ts);
use PublicInbox::View;
use PublicInbox::WwwAtomStream;
-use PublicInbox::MID qw(MID_ESC);
-use PublicInbox::MIME;
-require PublicInbox::Git;
-require PublicInbox::SearchThread;
-our $LIM = 200;
-
-sub noop {}
+use PublicInbox::WwwStream qw(html_oneshot);
+use PublicInbox::SearchThread;
+use PublicInbox::SearchQuery;
+use PublicInbox::Search qw(mdocid);
+my %rmap_inc;
sub mbox_results {
my ($ctx) = @_;
my $q = PublicInbox::SearchQuery->new($ctx->{qp});
- my $x = $q->{x};
+ if ($ctx->{env}->{'psgi.input'}->read(my $buf, 3)) {
+ $q->{t} = 1 if $buf =~ /\Ax=[^0]/;
+ }
require PublicInbox::Mbox;
- return PublicInbox::Mbox::mbox_all($ctx, $q->{'q'}) if $x eq 'm';
- sres_top_html($ctx);
+ $q->{x} eq 'm' ? PublicInbox::Mbox::mbox_all($ctx, $q) :
+ sres_top_html($ctx);
}
sub sres_top_html {
my ($ctx) = @_;
+ my $srch = $ctx->{-inbox}->search or
+ return PublicInbox::WWW::need($ctx, 'Search');
my $q = PublicInbox::SearchQuery->new($ctx->{qp});
my $x = $q->{x};
my $query = $q->{'q'};
+ my $o = $q->{o};
+ my $asc;
+ if ($o < 0) {
+ $asc = 1;
+ $o = -($o + 1); # so [-1] is the last element, like Perl lists
+ }
my $code = 200;
# double the limit for expanded views:
my $opts = {
limit => $q->{l},
- offset => $q->{o},
+ offset => $o,
mset => 1,
relevance => $q->{r},
+ thread => $q->{t},
+ asc => $asc,
};
- my ($mset, $total, $err, $cb);
+ my ($mset, $total, $err, $html);
retry:
eval {
- $mset = $ctx->{srch}->query($query, $opts);
+ $mset = $srch->query($query, $opts);
$total = $mset->get_matches_estimated;
};
$err = $@;
ctx_prepare($q, $ctx);
if ($err) {
$code = 400;
- $ctx->{-html_tip} = '<pre>'.err_txt($ctx, $err).'</pre><hr>';
- $cb = *noop;
+ $html = '<pre>'.err_txt($ctx, $err).'</pre><hr>';
} elsif ($total == 0) {
if (defined($ctx->{-uxs_retried})) {
# undo retry damage:
goto retry;
}
$code = 404;
- $ctx->{-html_tip} = "<pre>\n[No results found]</pre><hr>";
- $cb = *noop;
+ $html = "<pre>\n[No results found]</pre><hr>";
} else {
return adump($_[0], $mset, $q, $ctx) if $x eq 'A';
$ctx->{-html_tip} = search_nav_top($mset, $q, $ctx);
- if ($x eq 't') {
- $cb = mset_thread($ctx, $mset, $q);
- } else {
- $cb = mset_summary($ctx, $mset, $q);
- }
- }
- PublicInbox::WwwStream->response($ctx, $code, $cb);
-}
-
-# allow undef for individual doc loads...
-sub load_doc_retry {
- my ($srch, $mitem) = @_;
-
- eval {
- $srch->retry_reopen(sub {
- PublicInbox::SearchMsg->load_doc($mitem->get_document)
- });
+ return mset_thread($ctx, $mset, $q) if $x eq 't';
+ mset_summary($ctx, $mset, $q); # appends to {-html_tip}
+ $html = '';
}
+ html_oneshot($ctx, $code);
}
# display non-nested search results similar to what users expect from
my $pad = length("$total");
my $pfx = ' ' x $pad;
my $res = \($ctx->{-html_tip});
- my $srch = $ctx->{srch};
my $ibx = $ctx->{-inbox};
my $obfs_ibx = $ibx->{obfuscate} ? $ibx : undef;
+ my @nums = @{$ibx->search->mset_to_artnums($mset)};
+ my %num2msg = map { $_->{num} => $_ } @{$ibx->over->get_all(@nums)};
+ my ($min, $max);
+
foreach my $m ($mset->items) {
my $rank = sprintf("%${pad}d", $m->get_rank + 1);
- my $pct = $m->get_percent;
- my $smsg = load_doc_retry($srch, $m);
- unless ($smsg) {
+ my $pct = get_pct($m);
+ my $num = shift @nums;
+ my $smsg = delete($num2msg{$num}) or do {
eval {
- $m = "$m ".$m->get_docid . " expired\n";
+ $m = "$m $num expired\n";
$ctx->{env}->{'psgi.errors'}->print($m);
};
next;
- }
- my $s = ascii_html($smsg->subject);
- my $f = ascii_html($smsg->from_name);
+ };
+ $ctx->{-t_max} //= $smsg->{ts};
+
+ # only when sorting by relevance, ->items is always
+ # ordered descending:
+ $max //= $pct;
+ $min = $pct;
+
+ my $s = ascii_html($smsg->{subject});
+ my $f = ascii_html($smsg->{from_name});
if ($obfs_ibx) {
obfuscate_addrs($obfs_ibx, $s);
obfuscate_addrs($obfs_ibx, $f);
}
- my $date = PublicInbox::View::fmt_ts($smsg->ds);
- my $mid = PublicInbox::Hval->new_msgid($smsg->mid)->{href};
+ my $date = fmt_ts($smsg->{ds});
+ my $mid = mid_href($smsg->{mid});
$s = '(no subject)' if $s eq '';
$$res .= qq{$rank. <b><a\nhref="$mid/">}.
$s . "</a></b>\n";
$$res .= "$pfx - by $f @ $date UTC [$pct%]\n\n";
}
+ if ($q->{r}) { # for descriptions in search_nav_bot
+ $q->{-min_pct} = $min;
+ $q->{-max_pct} = $max;
+ }
$$res .= search_nav_bot($mset, $q);
- *noop;
+ undef;
+}
+
+# shorten "/full/path/to/Foo/Bar.pm" to "Foo/Bar.pm" so error
+# messages don't reveal FS layout info in case people use non-standard
+# installation paths
+sub path2inc ($) {
+ my $full = $_[0];
+ if (my $short = $rmap_inc{$full}) {
+ return $short;
+ } elsif (!scalar(keys %rmap_inc) && -e $full) {
+ %rmap_inc = map {; "$INC{$_}" => $_ } keys %INC;
+ # fall back to basename as last resort
+ $rmap_inc{$full} // (split('/', $full))[-1];
+ } else {
+ $full;
+ }
}
sub err_txt {
my ($ctx, $err) = @_;
my $u = $ctx->{-inbox}->base_url($ctx->{env}) . '_/text/help/';
$err =~ s/^\s*Exception:\s*//; # bad word to show users :P
+ $err =~ s!(\S+)!path2inc($1)!sge;
$err = ascii_html($err);
"\nBad query: <b>$err</b>\n" .
qq{See <a\nhref="$u">$u</a> for help on using search};
sub search_nav_top {
my ($mset, $q, $ctx) = @_;
- my $m = $q->qs_html(x => 'm', r => undef);
+ my $m = $q->qs_html(x => 'm', r => undef, t => undef);
my $rv = qq{<form\naction="?$m"\nmethod="post"><pre>};
my $initial_q = $ctx->{-uxs_retried};
if (defined $initial_q) {
}
my $A = $q->qs_html(x => 'A', r => undef);
$rv .= qq{|<a\nhref="?$A">Atom feed</a>]};
- $rv .= qq{\n\t\t\t\t\t\tdownload: };
- $rv .= qq{<input\ntype=submit\nvalue="mbox.gz"/></pre></form><pre>};
+ if ($ctx->{-inbox}->search->has_threadid) {
+ $rv .= qq{\n\t\t\tdownload mbox.gz: } .
+ # we set name=z w/o using it since it seems required for
+ # lynx (but works fine for w3m).
+ qq{<input\ntype=submit\nname=z\n} .
+ q{value="results only"/>} .
+ qq{|<input\ntype=submit\nname=x\n} .
+ q{value="full threads"/>};
+ } else { # BOFH needs to --reindex
+ $rv .= qq{\n\t\t\t\t\t\tdownload: } .
+ qq{<input\ntype=submit\nname=z\nvalue="mbox.gz"/>}
+ }
+ $rv .= qq{</pre></form><pre>};
}
sub search_nav_bot {
my ($mset, $q) = @_;
my $total = $mset->get_matches_estimated;
- my $o = $q->{o};
my $l = $q->{l};
- my $end = $o + $mset->size;
- my $beg = $o + 1;
my $rv = '</pre><hr><pre id=t>';
+ my $o = $q->{o};
+ my $off = $o < 0 ? -($o + 1) : $o;
+ my $end = $off + $mset->size;
+ my $beg = $off + 1;
+
if ($beg <= $end) {
- $rv .= "Results $beg-$end of $total";
- $rv .= ' (estimated)' if $end != $total;
+ my $approx = $end == $total ? '' : '~';
+ $rv .= "Results $beg-$end of $approx$total";
} else {
$rv .= "No more results, only $total";
}
- my $n = $o + $l;
+ my ($next, $join, $prev, $nd, $pd);
- if ($n < $total) {
- my $qs = $q->qs_html(o => $n, l => $l);
- $rv .= qq{ <a\nhref="?$qs"\nrel=next>next</a>}
- }
- if ($o > 0) {
- $rv .= $n < $total ? '/' : ' ';
- my $p = $o - $l;
- my $qs = $q->qs_html(o => ($p > 0 ? $p : 0));
- $rv .= qq{<a\nhref="?$qs"\nrel=prev>prev</a>};
+ if ($o >= 0) { # sort descending
+ my $n = $o + $l;
+ if ($n < $total) {
+ $next = $q->qs_html(o => $n, l => $l);
+ $nd = $q->{r} ? "[<= $q->{-min_pct}%]" : '(older)';
+ }
+ if ($o > 0) {
+ $join = $n < $total ? ' | ' : "\t";
+ my $p = $o - $l;
+ $prev = $q->qs_html(o => ($p > 0 ? $p : 0));
+ $pd = $q->{r} ? "[>= $q->{-max_pct}%]" : '(newer)';
+ }
+ } else { # o < 0, sort ascending
+ my $n = $o - $l;
+
+ if (-$n < $total) {
+ $next = $q->qs_html(o => $n, l => $l);
+ $nd = $q->{r} ? "[<= $q->{-min_pct}%]" : '(newer)';
+ }
+ if ($o < -1) {
+ $join = -$n < $total ? ' | ' : "\t";
+ my $p = $o + $l;
+ $prev = $q->qs_html(o => ($p < 0 ? $p : 0));
+ $pd = $q->{r} ? "[>= $q->{-max_pct}%]" : '(older)';
+ }
}
- $rv .= '</pre>';
+
+ $rv .= qq{ <a\nhref="?$next"\nrel=next>next $nd</a>} if $next;
+ $rv .= $join if $join;
+ $rv .= qq{<a\nhref="?$prev"\nrel=prev>prev $pd</a>} if $prev;
+
+ my $rev = $q->qs_html(o => $o < 0 ? 0 : -1);
+ $rv .= qq{ | <a\nhref="?$rev">reverse</a></pre>};
}
sub sort_relevance {
- my ($pct) = @_;
- sub {
- [ sort { (eval { $pct->{$b->topmost->{id}} } || 0)
- <=>
- (eval { $pct->{$a->topmost->{id}} } || 0)
- } @{$_[0]} ] };
+ [ sort {
+ (eval { $b->topmost->{pct} } // 0) <=>
+ (eval { $a->topmost->{pct} } // 0)
+ } @{$_[0]} ]
+}
+
+sub get_pct ($) {
+ # Capped at "99%" since "100%" takes an extra column in the
+ # thread skeleton view. <xapian/mset.h> says the value isn't
+ # very meaningful, anyways.
+ my $n = $_[0]->get_percent;
+ $n > 99 ? 99 : $n;
}
sub mset_thread {
my ($ctx, $mset, $q) = @_;
- my %pct;
- my $srch = $ctx->{srch};
- my $msgs = $srch->retry_reopen(sub { [ map {
- my $i = $_;
- my $smsg = PublicInbox::SearchMsg->load_doc($i->get_document);
- $pct{$smsg->mid} = $i->get_percent;
- $smsg;
- } ($mset->items) ]});
+ my $ibx = $ctx->{-inbox};
+ my $nshard = $ibx->search->{nshard} // 1;
+ my %pct = map { mdocid($nshard, $_) => get_pct($_) } $mset->items;
+ my $msgs = $ibx->over->get_all(keys %pct);
+ $_->{pct} = $pct{$_->{num}} for @$msgs;
my $r = $q->{r};
+ if ($r) { # for descriptions in search_nav_bot
+ my @pct = values %pct;
+ $q->{-min_pct} = min(@pct);
+ $q->{-max_pct} = max(@pct);
+ }
my $rootset = PublicInbox::SearchThread::thread($msgs,
- $r ? sort_relevance(\%pct) : *PublicInbox::View::sort_ds,
+ $r ? \&sort_relevance : \&PublicInbox::View::sort_ds,
$ctx);
my $skel = search_nav_bot($mset, $q). "<pre>";
- my $ibx = $ctx->{-inbox};
$ctx->{-upfx} = '';
$ctx->{anchor_idx} = 1;
$ctx->{cur_level} = 0;
- $ctx->{dst} = \$skel;
+ $ctx->{skel} = \$skel;
$ctx->{mapping} = {};
- $ctx->{pct} = \%pct;
+ $ctx->{searchview} = 1;
$ctx->{prev_attr} = '';
$ctx->{prev_level} = 0;
$ctx->{s_nr} = scalar(@$msgs).'+ results';
# reduce hash lookups in skel_dump
- $ctx->{-obfuscate} = $ctx->{-inbox}->{obfuscate};
+ $ctx->{-obfs_ibx} = $ibx->{obfuscate} ? $ibx : undef;
PublicInbox::View::walk_thread($rootset, $ctx,
- *PublicInbox::View::pre_thread);
+ \&PublicInbox::View::pre_thread);
+
+ # link $INBOX_DIR/description text to "recent" view around
+ # the newest message in this result set:
+ $ctx->{-t_max} = max(map { delete $_->{ts} } @$msgs);
@$msgs = reverse @$msgs if $r;
- sub {
- return unless $msgs;
- my $smsg;
- while (my $m = pop @$msgs) {
- $smsg = $ibx->smsg_mime($m) and last;
- }
- if ($smsg) {
- return PublicInbox::View::index_entry($smsg, $ctx,
- scalar @$msgs);
- }
- $msgs = undef;
- $skel .= "\n</pre>";
- };
+ $ctx->{msgs} = $msgs;
+ PublicInbox::WwwStream::aresponse($ctx, 200, \&mset_thread_i);
+}
+
+# callback for PublicInbox::WwwStream::getline
+sub mset_thread_i {
+ my ($ctx, $eml) = @_;
+ $ctx->zmore($ctx->html_top) if exists $ctx->{-html_tip};
+ $eml and return PublicInbox::View::eml_entry($ctx, $eml);
+ my $smsg = shift @{$ctx->{msgs}} or
+ $ctx->zmore(${delete($ctx->{skel})});
+ $smsg;
}
sub ctx_prepare {
sub adump {
my ($cb, $mset, $q, $ctx) = @_;
- my $ibx = $ctx->{-inbox};
- my @items = $mset->items;
- $ctx->{search_query} = $q;
- my $srch = $ctx->{srch};
- PublicInbox::WwwAtomStream->response($ctx, 200, sub {
- while (my $x = shift @items) {
- $x = load_doc_retry($srch, $x);
- $x = $ibx->smsg_mime($x) and return $x;
- }
- return undef;
- });
+ $ctx->{ids} = $ctx->{-inbox}->search->mset_to_artnums($mset);
+ $ctx->{search_query} = $q; # used by WwwAtomStream::atom_header
+ PublicInbox::WwwAtomStream->response($ctx, 200, \&adump_i);
}
-package PublicInbox::SearchQuery;
-use strict;
-use warnings;
-use URI::Escape qw(uri_escape);
-use PublicInbox::Hval;
-use PublicInbox::MID qw(MID_ESC);
-
-sub new {
- my ($class, $qp) = @_;
-
- my $r = $qp->{r};
- my ($l) = (($qp->{l} || '') =~ /(\d+)/);
- $l = $LIM if !$l || $l > $LIM;
- bless {
- q => $qp->{'q'},
- x => $qp->{x} || '',
- o => (($qp->{o} || '0') =~ /(\d+)/),
- l => $l,
- r => (defined $r && $r ne '0'),
- }, $class;
-}
-
-sub qs_html {
- my ($self, %over) = @_;
-
- if (keys %over) {
- my $tmp = bless { %$self }, ref($self);
- foreach my $k (keys %over) {
- $tmp->{$k} = $over{$k};
- }
- $self = $tmp;
- }
-
- my $q = uri_escape($self->{'q'}, MID_ESC);
- $q =~ s/%20/+/g; # improve URL readability
- my $qs = "q=$q";
-
- if (my $o = $self->{o}) { # ignore o == 0
- $qs .= "&o=$o";
- }
- if (my $l = $self->{l}) {
- $qs .= "&l=$l" unless $l == $LIM;
- }
- if (my $r = $self->{r}) {
- $qs .= "&r";
- }
- if (my $x = $self->{x}) {
- $qs .= "&x=$x" if ($x eq 't' || $x eq 'A' || $x eq 'm');
+# callback for PublicInbox::WwwAtomStream::getline
+sub adump_i {
+ my ($ctx) = @_;
+ while (my $num = shift @{$ctx->{ids}}) {
+ my $smsg = eval { $ctx->{-inbox}->over->get_art($num) } or next;
+ return $smsg;
}
- $qs;
}
1;