X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FSearchView.pm;h=7521a66ddb82c51bc2a6b9b7b1b77772403d2531;hb=6c853f5256f3a324c018a37a9e8a6b9fc8fdc063;hp=a8b66dda9004ad6cca983bff668f9ab06b7ef181;hpb=84d8920b92686e975929aebe845b6d4ea0a9ef0d;p=public-inbox.git

diff --git a/lib/PublicInbox/SearchView.pm b/lib/PublicInbox/SearchView.pm
index a8b66dda..7521a66d 100644
--- a/lib/PublicInbox/SearchView.pm
+++ b/lib/PublicInbox/SearchView.pm
@@ -1,31 +1,31 @@
-# 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(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;
+use PublicInbox::WwwStream qw(html_oneshot);
+use PublicInbox::SearchThread;
+use PublicInbox::SearchQuery;
+use PublicInbox::Search qw(mdocid);
 my %rmap_inc;
 
-sub noop {}
-
 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 {
@@ -49,9 +49,10 @@ sub sres_top_html {
 		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 = $srch->query($query, $opts);
@@ -61,8 +62,7 @@ retry:
 	ctx_prepare($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:
@@ -73,30 +73,16 @@ retry:
 			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
@@ -109,34 +95,38 @@ sub mset_summary {
 	my $pfx = ' ' x $pad;
 	my $res = \($ctx->{-html_tip});
 	my $ibx = $ctx->{-inbox};
-	my $srch = $ibx->search;
 	my $obfs_ibx = $ibx->{obfuscate} ? $ibx : undef;
+	my @nums = @{$ibx->search->mset_to_artnums($mset)};
+	my %num2msg = map { $_->{num} => $_ } @{$ibx->over->get_all(@nums)};
+
 	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};
+
+		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";
 	}
 	$$res .= search_nav_bot($mset, $q);
-	*noop;
+	undef;
 }
 
 # shorten "/full/path/to/Foo/Bar.pm" to "Foo/Bar.pm" so error
@@ -167,7 +157,7 @@ sub err_txt {
 
 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) {
@@ -202,8 +192,19 @@ sub search_nav_top {
 	}
 	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 {
@@ -256,58 +257,64 @@ sub search_nav_bot {
 }
 
 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 $ibx = $ctx->{-inbox};
-	my $msgs = $ibx->search->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 $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};
 	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>";
 	$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 {
@@ -329,69 +336,18 @@ sub ctx_prepare {
 
 sub adump {
 	my ($cb, $mset, $q, $ctx) = @_;
-	my $ibx = $ctx->{-inbox};
-	my @items = $mset->items;
-	$ctx->{search_query} = $q;
-	my $srch = $ibx->search;
-	PublicInbox::WwwAtomStream->response($ctx, 200, sub {
-		while (my $x = shift @items) {
-			$x = load_doc_retry($srch, $x);
-			$x = $ibx->smsg_mime($x) and return $x;
-		}
-		return undef;
-	});
-}
-
-package PublicInbox::SearchQuery;
-use strict;
-use warnings;
-use URI::Escape qw(uri_escape);
-use PublicInbox::Hval;
-use PublicInbox::MID qw(MID_ESC);
-
-sub new {
-	my ($class, $qp) = @_;
-
-	my $r = $qp->{r};
-	my ($l) = (($qp->{l} || '') =~ /([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;
+	$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);
 }
 
-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
-		$qs .= "&amp;o=$o";
-	}
-	if (my $l = $self->{l}) {
-		$qs .= "&amp;l=$l" unless $l == $LIM;
-	}
-	if (my $r = $self->{r}) {
-		$qs .= "&amp;r";
-	}
-	if (my $x = $self->{x}) {
-		$qs .= "&amp;x=$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;