X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FSearchView.pm;h=94a55b8d8a3b1a3cbac5b2b569c2f90160c7de88;hb=c44a47c6d3463e35c68ab570cf286068f6b6657d;hp=219006a0d7070dbfc58530255eba7aeba1919ff4;hpb=3f3d9cf7d88a851721f1f8468e1311a4f0c02ff6;p=public-inbox.git diff --git a/lib/PublicInbox/SearchView.pm b/lib/PublicInbox/SearchView.pm index 219006a0..94a55b8d 100644 --- a/lib/PublicInbox/SearchView.pm +++ b/lib/PublicInbox/SearchView.pm @@ -1,4 +1,4 @@ -# Copyright (C) 2015-2018 all contributors +# Copyright (C) 2015-2019 all contributors # License: AGPL-3.0+ # # Displays search results for the web interface @@ -10,11 +10,9 @@ use PublicInbox::SearchMsg; use PublicInbox::Hval qw/ascii_html obfuscate_addrs/; use PublicInbox::View; use PublicInbox::WwwAtomStream; -use PublicInbox::MID qw(mid2path mid_mime mid_clean mid_escape MID_ESC); -use PublicInbox::MIME; -require PublicInbox::Git; -require PublicInbox::SearchThread; +use PublicInbox::SearchThread; our $LIM = 200; +my %rmap_inc; sub noop {} @@ -22,28 +20,38 @@ sub mbox_results { my ($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); } 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}, + asc => $asc, }; my ($mset, $total, $err, $cb); retry: eval { - $mset = $ctx->{srch}->query($query, $opts); + $mset = $srch->query($query, $opts); $total = $mset->get_matches_estimated; }; $err = $@; @@ -77,17 +85,6 @@ retry: PublicInbox::WwwStream->response($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) - }); - } -} - # display non-nested search results similar to what users expect from # regular WWW search engines: sub mset_summary { @@ -97,13 +94,13 @@ sub mset_summary { my $pad = length("$total"); my $pfx = ' ' x $pad; my $res = \($ctx->{-html_tip}); - my $srch = $ctx->{srch}; 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 = $m->get_percent; - my $smsg = load_doc_retry($srch, $m); + my $pct = get_pct($m); + my $smsg = PublicInbox::SearchMsg::from_mitem($m, $srch); unless ($smsg) { eval { $m = "$m ".$m->get_docid . " expired\n"; @@ -117,20 +114,38 @@ sub mset_summary { obfuscate_addrs($obfs_ibx, $s); obfuscate_addrs($obfs_ibx, $f); } - my $ts = PublicInbox::View::fmt_ts($smsg->ts); + 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 @ $ts UTC [$pct%]\n\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 $u for help on using search}; @@ -180,89 +195,116 @@ sub search_nav_top { sub search_nav_bot { my ($mset, $q) = @_; my $total = $mset->get_matches_estimated; - my $nr = scalar $mset->items; - my $o = $q->{o}; my $l = $q->{l}; - my $end = $o + $nr; - my $beg = $o + 1; 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 $n = $o + $l;
+	my ($next, $join, $prev);
 
-	if ($n < $total) {
-		my $qs = $q->qs_html(o => $n, l => $l);
-		$rv .= qq{  next}
-	}
-	if ($o > 0) {
-		$rv .= $n < $total ? '/' : '       ';
-		my $p = $o - $l;
-		my $qs = $q->qs_html(o => ($p > 0 ? $p : 0));
-		$rv .= qq{prev};
+	if ($o >= 0) { # sort descending
+		my $n = $o + $l;
+		if ($n < $total) {
+			$next = $q->qs_html(o => $n, l => $l);
+		}
+		if ($o > 0) {
+			$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));
+		}
 	}
-	$rv .= '
'; + + $rv .= qq{ next} if $next; + $rv .= $join if $join; + $rv .= qq{prev} if $prev; + + my $rev = $q->qs_html(o => $o < 0 ? 0 : -1); + $rv .= qq{ | reverse results}; } sub sort_relevance { - my ($pct) = @_; - sub { - [ sort { (eval { $pct->{$b->topmost->{id}} } || 0) - <=> - (eval { $pct->{$a->topmost->{id}} } || 0) - } @{$_[0]} ] }; + [ sort { + (eval { $b->topmost->{smsg}->{pct} } // 0) <=> + (eval { $a->topmost->{smsg}->{pct} } // 0) + } @{$_[0]} ] +} + +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 %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 $msgs = $ctx->{-inbox}->search->retry_reopen(\&load_msgs, $mset); my $r = $q->{r}; my $rootset = PublicInbox::SearchThread::thread($msgs, - $r ? sort_relevance(\%pct) : *PublicInbox::View::sort_ts, - $srch); + $r ? \&sort_relevance : \&PublicInbox::View::sort_ds, + $ctx); my $skel = search_nav_bot($mset, $q). "
";
-	my $inbox = $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->{seen} = {};
 	$ctx->{s_nr} = scalar(@$msgs).'+ results';
 
 	# reduce hash lookups in skel_dump
 	$ctx->{-obfuscate} = $ctx->{-inbox}->{obfuscate};
 	PublicInbox::View::walk_thread($rootset, $ctx,
-		*PublicInbox::View::pre_thread);
+		\&PublicInbox::View::pre_thread);
 
 	@$msgs = reverse @$msgs if $r;
-	my $mime;
-	sub {
-		return unless $msgs;
-		while ($mime = pop @$msgs) {
-			$mime = $inbox->msg_by_smsg($mime) and last;
-		}
-		if ($mime) {
-			$mime = PublicInbox::MIME->new($mime);
-			return PublicInbox::View::index_entry($mime, $ctx,
-				scalar @$msgs);
-		}
-		$msgs = undef;
-		$skel .= "\n
"; - }; + $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(skel msgs)}; + $$skel .= "\n"; } sub ctx_prepare { @@ -284,37 +326,39 @@ 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 = $ibx->msg_by_smsg($x) and - return PublicInbox::MIME->new($x); - } - return undef; - }); + $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 $smsg; + } } package PublicInbox::SearchQuery; use strict; use warnings; use URI::Escape qw(uri_escape); -use PublicInbox::Hval; use PublicInbox::MID qw(MID_ESC); sub new { my ($class, $qp) = @_; my $r = $qp->{r}; - my ($l) = (($qp->{l} || '') =~ /(\d+)/); + my ($l) = (($qp->{l} || '') =~ /([0-9]+)/); $l = $LIM if !$l || $l > $LIM; bless { q => $qp->{'q'}, x => $qp->{x} || '', - o => (($qp->{o} || '0') =~ /(\d+)/), + o => (($qp->{o} || '0') =~ /(-?[0-9]+)/), l => $l, r => (defined $r && $r ne '0'), }, $class; @@ -339,7 +383,7 @@ sub qs_html { $qs .= "&o=$o"; } if (my $l = $self->{l}) { - $qs .= "&l=$l"; + $qs .= "&l=$l" unless $l == $LIM; } if (my $r = $self->{r}) { $qs .= "&r";