$qh - search results

X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FSearchView.pm;h=82b97f32ef6aecdf54c18f8f436cab2437d233db;hb=55b707d788ce13696e4411389583e720ea6dab01;hp=960049f8935cf01e011ef3983a5049598f8dc073;hpb=6f9221f76521c8c55356a744bd31dc1fee753b93;p=public-inbox.git diff --git a/lib/PublicInbox/SearchView.pm b/lib/PublicInbox/SearchView.pm index 960049f8..0d2e71fd 100644 --- a/lib/PublicInbox/SearchView.pm +++ b/lib/PublicInbox/SearchView.pm @@ -1,101 +1,397 @@ -# Copyright (C) 2015 all contributors -# License: AGPLv3 or later (https://www.gnu.org/licenses/agpl-3.0.txt) +# Copyright (C) 2015-2019 all contributors +# License: AGPL-3.0+ +# +# 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; +use PublicInbox::Hval qw/ascii_html obfuscate_addrs/; use PublicInbox::View; -use POSIX qw/strftime/; -our $LIM = 25; +use PublicInbox::WwwAtomStream; +use PublicInbox::SearchThread; +our $LIM = 200; +my %rmap_inc; -sub sres_top_html { - my ($ctx, $q) = @_; - my $cgi = $ctx->{cgi}; - # $q ||= $cgi->param('q'); - my $o = int($cgi->param('o') || 0); - my $r = $cgi->param('r'); - $r = (defined $r && $r ne '0'); - my $opts = { limit => $LIM, offset => $o, mset => 1, relevance => $r }; - my $mset = $ctx->{srch}->query($q, $opts); - my $total = $mset->get_matches_estimated; - my $query = PublicInbox::Hval->new_oneline($q); - my $qh = $query->as_html; - my $res = "$qh - search results" . - qq{} . - qq{}; +sub noop {} - $res .= qq{} if $r; +sub mbox_results { + my ($ctx) = @_; + my $q = PublicInbox::SearchQuery->new($ctx->{qp}); + my $x = $q->{x}; + require PublicInbox::Mbox; + return PublicInbox::Mbox::mbox_all($ctx, $q->{'q'}) if $x eq 'm'; + sres_top_html($ctx); +} - $res .= qq{} . - PublicInbox::View::PRE_WRAP; +sub sres_top_html { + my ($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 $foot = $ctx->{footer} || ''; - $foot = qq{Back to index.}; - if ($total == 0) { - $res .= "\n\n[No results found]

$foot";
+	my $code = 200;
+	# double the limit for expanded views:
+	my $opts = {
+		limit => $q->{l},
+		offset => $o,
+		mset => 1,
+		relevance => $q->{r},
+		asc => $asc,
+	};
+	my ($mset, $total, $err, $cb);
+retry:
+	eval {
+		$mset = $srch->query($query, $opts);
+		$total = $mset->get_matches_estimated;
+	};
+	$err = $@;
+	ctx_prepare($q, $ctx);
+	if ($err) {
+		$code = 400;
+		$ctx->{-html_tip} = ''.err_txt($ctx, $err).'
';
+		$cb = *noop;
+	} elsif ($total == 0) {
+		if (defined($ctx->{-uxs_retried})) {
+			# undo retry damage:
+			$q->{'q'} = $ctx->{-uxs_retried};
+		} elsif (index($q->{'q'}, '%') >= 0) {
+			$ctx->{-uxs_retried} = $q->{'q'};
+			$q->{'q'} = uri_unescape($q->{'q'});
+			goto retry;
+		}
+		$code = 404;
+		$ctx->{-html_tip} = "\n[No results found]
";
+		$cb = *noop;
 	} else {
-		$q = $query->as_href;
-		$q =~ s/%20/+/g; # improve URL readability
-		my $qp = "?q=$q";
-		$qp .= "&o=$o" if $o;
-
-		$res .= "Search results ordered by [";
-		if ($r) {
-			$res .= qq{date|relevance};
+		return adump($_[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 {
-			$qp .= '&r';
-			$res .= qq{date|relevance};
+			$cb = mset_summary($ctx, $mset, $q);
 		}
-		$res .= "]\n\n";
+	}
+	PublicInbox::WwwStream->response($ctx, $code, $cb);
+}
 
-		dump_mset(\$res, $mset);
-		my $nr = scalar $mset->items;
-		my $end = $o + $nr;
-		my $beg = $o + 1;
-		$res .= "
";
-		$res .= "Results $beg-$end of $total";
+# display non-nested search results similar to what users expect from
+# regular WWW search engines:
+sub mset_summary {
+	my ($ctx, $mset, $q) = @_;
 
-		my $n = $o + $LIM;
+	my $total = $mset->get_matches_estimated;
+	my $pad = length("$total");
+	my $pfx = ' ' x $pad;
+	my $res = \($ctx->{-html_tip});
+	my $ibx = $ctx->{-inbox};
+	my $srch = $ibx->search;
+	my $obfs_ibx = $ibx->{obfuscate} ? $ibx : undef;
+	foreach my $m ($mset->items) {
+		my $rank = sprintf("%${pad}d", $m->get_rank + 1);
+		my $pct = get_pct($m);
+		my $smsg = PublicInbox::SearchMsg::from_mitem($m, $srch);
+		unless ($smsg) {
+			eval {
+				$m = "$m ".$m->get_docid . " expired\n";
+				$ctx->{env}->{'psgi.errors'}->print($m);
+			};
+			next;
+		}
+		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};
+		$s = '(no subject)' if $s eq '';
+		$$res .= qq{$rank. }.
+			$s . "\n";
+		$$res .= "$pfx  - by $f @ $date UTC [$pct%]\n\n";
+	}
+	$$res .= search_nav_bot($mset, $q);
+	*noop;
+}
+
+# shorten "/full/path/to/Foo/Bar.pm" to "Foo/Bar.pm" so error
+# messages don't reveal FS layout info in case people use non-standard
+# installation paths
+sub path2inc ($) {
+	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: $err\n" .
+		qq{See $u for help on using search};
+}
+
+sub search_nav_top {
+	my ($mset, $q, $ctx) = @_;
+	my $m = $q->qs_html(x => 'm', r => undef);
+	my $rv = qq{};
+	my $initial_q = $ctx->{-uxs_retried};
+	if (defined $initial_q) {
+		my $rewritten = $q->{'q'};
+		utf8::decode($initial_q);
+		utf8::decode($rewritten);
+		$initial_q = ascii_html($initial_q);
+		$rewritten = ascii_html($rewritten);
+		$rv .= " Warning: Initial query:\n $initial_q\n";
+		$rv .= " returned no results, used:\n";
+		$rv .= " $rewritten\n instead\n\n";
+	}
+
+	$rv .= 'Search results ordered by [';
+	if ($q->{r}) {
+		my $d = $q->qs_html(r => 0);
+		$rv .= qq{date|relevance};
+	} else {
+		my $d = $q->qs_html(r => 1);
+		$rv .= qq{date|relevance};
+	}
+
+	$rv .= ']  view[';
+
+	my $x = $q->{x};
+	if ($x eq '') {
+		my $t = $q->qs_html(x => 't');
+		$rv .= qq{summary|nested}
+	} elsif ($q->{x} eq 't') {
+		my $s = $q->qs_html(x => '');
+		$rv .= qq{summary|nested};
+	}
+	my $A = $q->qs_html(x => 'A', r => undef);
+	$rv .= qq{|Atom feed]};
+	$rv .= qq{\n\t\t\t\t\t\tdownload: };
+	$rv .= qq{
};
+}
+
+sub search_nav_bot {
+	my ($mset, $q) = @_;
+	my $total = $mset->get_matches_estimated;
+	my $l = $q->{l};
+	my $rv = '
';
+	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;
+	} else {
+		$rv .= "No more results, only $total";
+	}
+	my ($next, $join, $prev);
+
+	if ($o >= 0) { # sort descending
+		my $n = $o + $l;
 		if ($n < $total) {
-			$qp = "q=$q&o=$n";
-			$qp .= "&r" if $r;
-			$res .= qq{, next}
+			$next = $q->qs_html(o => $n, l => $l);
 		}
 		if ($o > 0) {
-			$res .= $n < $total ? '/' : ',      ';
-			my $p = $o - $LIM;
-			$qp = "q=$q";
-			$qp .= "&o=$p" if $p > 0;
-			$qp .= "&r" if $r;
-			$res .= qq{prev};
+			$join = $n < $total ? '/' : '       ';
+			my $p = $o - $l;
+			$prev = $q->qs_html(o => ($p > 0 ? $p : 0));
+		}
+	} else { # o < 0, sort ascending
+		my $n = $o - $l;
+
+		if (-$n < $total) {
+			$next = $q->qs_html(o => $n, l => $l);
+		}
+		if ($o < -1) {
+			$join = -$n < $total ? '/' : '       ';
+			my $p = $o + $l;
+			$prev = $q->qs_html(o => ($p < 0 ? $p : 0));
 		}
-		$res .= "\n\n" . $foot;
 	}
 
-	$res .= "";
-	[200, ['Content-Type'=>'text/html; charset=UTF-8'], [$res]];
+	$rv .= qq{  next} if $next;
+	$rv .= $join if $join;
+	$rv .= qq{prev} if $prev;
+
+	my $rev = $q->qs_html(o => $o < 0 ? 0 : -1);
+	$rv .= qq{ | reverse results};
 }
 
-sub dump_mset {
-	my ($res, $mset) = @_;
+sub sort_relevance {
+	[ sort {
+		(eval { $b->topmost->{smsg}->{pct} } // 0) <=>
+		(eval { $a->topmost->{smsg}->{pct} } // 0)
+	} @{$_[0]} ]
+}
 
-	my $total = $mset->get_matches_estimated;
-	my $pad = length("$total");
-	my $pfx = ' ' x $pad;
-	foreach my $m ($mset->items) {
-		my $rank = sprintf("%${pad}d", $m->get_rank + 1);
-		my $pct = $m->get_percent;
-		my $smsg = PublicInbox::SearchMsg->load_doc($m->get_document);
-		my $s = PublicInbox::Hval->new_oneline($smsg->subject);
-		my $f = $smsg->from_name;
-		$f = PublicInbox::Hval->new_oneline($f)->as_html;
-		my $d = strftime('%Y-%m-%d %H:%M', gmtime($smsg->ts));
-		my $mid = PublicInbox::Hval->new_msgid($smsg->mid)->as_href;
-		$$res .= qq{$rank. }.
-			$s->as_html . "\n";
-		$$res .= "$pfx  - by $f @ $d UTC [$pct%]\n\n";
+sub get_pct ($) {
+	# Capped at "99%" since "100%" takes an extra column in the
+	# thread skeleton view.   says the value isn't
+	# very meaningful, anyways.
+	my $n = $_[0]->get_percent;
+	$n > 99 ? 99 : $n;
+}
+
+sub load_msgs {
+	my ($mset) = @_;
+	[ map {
+		my $mi = $_;
+		my $smsg = PublicInbox::SearchMsg::from_mitem($mi);
+		$smsg->{pct} = get_pct($mi);
+		$smsg;
+	} ($mset->items) ]
+}
+
+sub mset_thread {
+	my ($ctx, $mset, $q) = @_;
+	my $msgs = $ctx->{-inbox}->search->retry_reopen(\&load_msgs, $mset);
+	my $r = $q->{r};
+	my $rootset = PublicInbox::SearchThread::thread($msgs,
+		$r ? \&sort_relevance : \&PublicInbox::View::sort_ds,
+		$ctx);
+	my $skel = search_nav_bot($mset, $q). "";
+	$ctx->{-upfx} = '';
+	$ctx->{anchor_idx} = 1;
+	$ctx->{cur_level} = 0;
+	$ctx->{dst} = \$skel;
+	$ctx->{mapping} = {};
+	$ctx->{searchview} = 1;
+	$ctx->{prev_attr} = '';
+	$ctx->{prev_level} = 0;
+	$ctx->{s_nr} = scalar(@$msgs).'+ results';
+
+	# reduce hash lookups in skel_dump
+	$ctx->{-obfuscate} = $ctx->{-inbox}->{obfuscate};
+	PublicInbox::View::walk_thread($rootset, $ctx,
+		*PublicInbox::View::pre_thread);
+
+	@$msgs = reverse @$msgs if $r;
+	$ctx->{msgs} = $msgs;
+	\&mset_thread_i;
+}
+
+# callback for PublicInbox::WwwStream::getline
+sub mset_thread_i {
+	my ($nr, $ctx) = @_;
+	my $msgs = $ctx->{msgs} or return;
+	while (my $smsg = pop @$msgs) {
+		$ctx->{-inbox}->smsg_mime($smsg) or next;
+		return PublicInbox::View::index_entry($smsg, $ctx,
+							scalar @$msgs);
+	}
+	my ($skel) = delete @$ctx{qw(dst msgs)};
+	$$skel .= "\n";
+}
+
+sub ctx_prepare {
+	my ($q, $ctx) = @_;
+	my $qh = $q->{'q'};
+	utf8::decode($qh);
+	$qh = ascii_html($qh);
+	$ctx->{-q_value_html} = $qh;
+	$ctx->{-atom} = '?'.$q->qs_html(x => 'A', r => undef);
+	$ctx->{-title_html} = "$qh - search results";
+	my $extra = '';
+	$extra .= qq{} if $q->{r};
+	if (my $x = $q->{x}) {
+		$x = ascii_html($x);
+		$extra .= qq{};
+	}
+	$ctx->{-extra_form_html} = $extra;
+}
+
+sub adump {
+	my ($cb, $mset, $q, $ctx) = @_;
+	$ctx->{items} = [ $mset->items ];
+	$ctx->{search_query} = $q; # used by WwwAtomStream::atom_header
+	$ctx->{srch} = $ctx->{-inbox}->search;
+	PublicInbox::WwwAtomStream->response($ctx, 200, \&adump_i);
+}
+
+# callback for PublicInbox::WwwAtomStream::getline
+sub adump_i {
+	my ($ctx) = @_;
+	while (my $mi = shift @{$ctx->{items}}) {
+		my $smsg = eval {
+			PublicInbox::SearchMsg::from_mitem($mi, $ctx->{srch});
+		} or next;
+		$ctx->{-inbox}->smsg_mime($smsg) and return $smsg;
+	}
+}
+
+package PublicInbox::SearchQuery;
+use strict;
+use warnings;
+use URI::Escape qw(uri_escape);
+use PublicInbox::MID qw(MID_ESC);
+
+sub new {
+	my ($class, $qp) = @_;
+
+	my $r = $qp->{r};
+	my ($l) = (($qp->{l} || '') =~ /([0-9]+)/);
+	$l = $LIM if !$l || $l > $LIM;
+	bless {
+		q => $qp->{'q'},
+		x => $qp->{x} || '',
+		o => (($qp->{o} || '0') =~ /(-?[0-9]+)/),
+		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');
 	}
+	$qs;
 }
 
 1;