X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FView.pm;h=9b62ed3c3d954d76ab173c4c32bd561b85c62b30;hb=d0147582e289fdd4cdd84e91d8b0f8ae9c230124;hp=980f265a06ddb99d9736242473e28a1db7ee894c;hpb=6b2a40269d00d014c2c94af4bb7dfecff6dff57e;p=public-inbox.git

diff --git a/lib/PublicInbox/View.pm b/lib/PublicInbox/View.pm
index 980f265a..9b62ed3c 100644
--- a/lib/PublicInbox/View.pm
+++ b/lib/PublicInbox/View.pm
@@ -10,7 +10,8 @@ use bytes (); # only for bytes::length
 use PublicInbox::MsgTime qw(msg_datestamp);
 use PublicInbox::Hval qw(ascii_html obfuscate_addrs prurl mid_href);
 use PublicInbox::Linkify;
-use PublicInbox::MID qw/id_compress mids mids_for_index references/;
+use PublicInbox::MID qw(id_compress mids mids_for_index references
+			$MID_EXTRACT);
 use PublicInbox::MsgIter;
 use PublicInbox::Address;
 use PublicInbox::WwwStream;
@@ -18,7 +19,7 @@ use PublicInbox::Reply;
 use PublicInbox::ViewDiff qw(flush_diff);
 use POSIX qw(strftime);
 use Time::Local qw(timegm);
-use PublicInbox::SearchMsg qw(subject_normalized);
+use PublicInbox::Smsg qw(subject_normalized);
 use constant COLS => 72;
 use constant INDENT => '  ';
 use constant TCHILD => '` ';
@@ -26,22 +27,14 @@ sub th_pfx ($) { $_[0] == 0 ? '' : TCHILD };
 
 sub msg_page_i {
 	my ($nr, $ctx) = @_;
-	my $more = $ctx->{more};
-	if ($nr == 1) {
-		# $more cannot be true w/o $smsg being defined:
-		$ctx->{mhref} = $more ? '../'.mid_href($ctx->{smsg}->{mid}).'/'
-				      : '';
-		multipart_text_as_html(delete $ctx->{mime}, $ctx);
-		${delete $ctx->{obuf}} .= '</pre><hr>';
-	} elsif ($more) {
-		++$ctx->{end_nr};
-		# fake an EOF if {more} retrieval fails fails;
-		eval { msg_page_more($ctx, $nr) };
-	} elsif ($nr == $ctx->{end_nr}) {
+	if (my $more = delete $ctx->{more}) { # unlikely
+		# fake an EOF if $more retrieval fails;
+		eval { msg_page_more($ctx, $nr, @$more) };
+	} elsif (my $hdr = delete $ctx->{hdr}) {
 		# fake an EOF if generating the footer fails;
 		# we want to at least show the message if something
 		# here crashes:
-		eval { html_footer($ctx) };
+		eval { html_footer($ctx, $hdr) };
 	} else {
 		undef
 	}
@@ -53,40 +46,37 @@ sub msg_page {
 	my ($ctx) = @_;
 	my $mid = $ctx->{mid};
 	my $ibx = $ctx->{-inbox};
-	my ($first);
-	my $smsg;
+	my ($smsg, $first, $next);
 	if (my $over = $ibx->over) {
 		my ($id, $prev);
-		$smsg = $over->next_by_mid($mid, \$id, \$prev);
-		$first = $ibx->msg_by_smsg($smsg) if $smsg;
-		if ($first) {
-			my $next = $over->next_by_mid($mid, \$id, \$prev);
-			$ctx->{more} = [ $id, $prev, $next ] if $next;
-		}
-		return unless $first;
+		$smsg = $over->next_by_mid($mid, \$id, \$prev) or return;
+		$first = $ibx->msg_by_smsg($smsg) or return;
+		$next = $over->next_by_mid($mid, \$id, \$prev);
+		$ctx->{more} = [ $id, $prev, $next ] if $next;
 	} else {
 		$first = $ibx->msg_by_mid($mid) or return;
 	}
-	my $mime = $ctx->{mime} = PublicInbox::MIME->new($first);
+	my $mime = PublicInbox::MIME->new($first);
 	$ctx->{-obfs_ibx} = $ibx->{obfuscate} ? $ibx : undef;
 	my $hdr = $ctx->{hdr} = $mime->header_obj;
-	_msg_page_prepare_obuf($hdr, $ctx, 0);
-	$ctx->{end_nr} = 2;
+	$ctx->{obuf} = _msg_page_prepare_obuf($hdr, $ctx, 0);
 	$ctx->{smsg} = $smsg;
+	# $next cannot be true w/o $smsg being defined:
+	$ctx->{mhref} = $next ? '../'.mid_href($smsg->{mid}).'/' : '';
+	multipart_text_as_html($mime, $ctx);
+	$ctx->{-html_tip} = (${delete $ctx->{obuf}} .= '</pre><hr>');
 	PublicInbox::WwwStream->response($ctx, 200, \&msg_page_i);
 }
 
-sub msg_page_more {
-	my ($ctx, $nr) = @_;
-	my ($id, $prev, $smsg) = @{$ctx->{more}};
+sub msg_page_more { # cold
+	my ($ctx, $nr, $id, $prev, $smsg) = @_;
 	my $ibx = $ctx->{-inbox};
-	$smsg = $ibx->smsg_mime($smsg);
 	my $next = $ibx->over->next_by_mid($ctx->{mid}, \$id, \$prev);
-	$ctx->{more} = $next ? [ $id, $prev, $next ] : undef;
-	return '' unless $smsg;
+	$ctx->{more} = [ $id, $prev, $next ] if $next;
+	$smsg = $ibx->smsg_mime($smsg) or return '';
 	$ctx->{mhref} = '../' . mid_href($smsg->{mid}) . '/';
 	my $mime = delete $smsg->{mime};
-	_msg_page_prepare_obuf($mime->header_obj, $ctx, $nr);
+	$ctx->{obuf} = _msg_page_prepare_obuf($mime->header_obj, $ctx, $nr);
 	multipart_text_as_html($mime, $ctx);
 	${delete $ctx->{obuf}} .= '</pre><hr>';
 }
@@ -160,7 +150,7 @@ sub in_reply_to {
 sub fold_addresses ($) {
 	return $_[0] if length($_[0]) <= COLS;
 	# try to fold on commas after non-word chars before $lim chars,
-	# Try to get the "," preceeded by ">" or ")", but avoid folding
+	# Try to get the "," preceded by ">" or ")", but avoid folding
 	# on the comma where somebody uses "Lastname, Firstname".
 	# We also try to keep the last and penultimate addresses in
 	# the list on the same line if possible, hence the extra \z
@@ -307,11 +297,9 @@ sub _th_index_lite {
 	my $nr_c = scalar @$children;
 	my $nr_s = 0;
 	my $siblings;
-	if (my $smsg = $node->{smsg}) {
-		# delete saves about 200KB on a 1K message thread
-		if (my $refs = delete $smsg->{references}) {
-			($$irt) = ($refs =~ m/<([^>]+)>\z/);
-		}
+	# delete saves about 200KB on a 1K message thread
+	if (my $refs = delete $node->{references}) {
+		($$irt) = ($refs =~ m/$MID_EXTRACT\z/o);
 	}
 	my $irt_map = $mapping->{$$irt} if defined $$irt;
 	if (defined $irt_map) {
@@ -320,12 +308,12 @@ sub _th_index_lite {
 		$rv .= $pad . $irt_map->[0];
 		if ($idx > 0) {
 			my $prev = $siblings->[$idx - 1];
-			my $pmid = $prev->{id};
+			my $pmid = $prev->{mid};
 			if ($idx > 2) {
 				my $s = ($idx - 1). ' preceding siblings ...';
 				$rv .= pad_link($pmid, $level, $s);
 			} elsif ($idx == 2) {
-				my $ppmid = $siblings->[0]->{id};
+				my $ppmid = $siblings->[0]->{mid};
 				$rv .= $pad . $mapping->{$ppmid}->[0];
 			}
 			$rv .= $pad . $mapping->{$pmid}->[0];
@@ -338,26 +326,26 @@ sub _th_index_lite {
 	$attr =~ s!<a\nhref=[^>]+>([^<]+)</a>!$1!s; # no point linking to self
 	$rv .= "<b>@ $attr";
 	if ($nr_c) {
-		my $cmid = $children->[0]->{id};
+		my $cmid = $children->[0]->{mid};
 		$rv .= $pad . $mapping->{$cmid}->[0];
 		if ($nr_c > 2) {
 			my $s = ($nr_c - 1). ' more replies';
 			$rv .= pad_link($cmid, $level + 1, $s);
 		} elsif (my $cn = $children->[1]) {
-			$rv .= $pad . $mapping->{$cn->{id}}->[0];
+			$rv .= $pad . $mapping->{$cn->{mid}}->[0];
 		}
 	}
 
 	my $next = $siblings->[$idx+1] if $siblings && $idx >= 0;
 	if ($next) {
-		my $nmid = $next->{id};
+		my $nmid = $next->{mid};
 		$rv .= $pad . $mapping->{$nmid}->[0];
 		my $nnext = $nr_s - $idx;
 		if ($nnext > 2) {
 			my $s = ($nnext - 1).' subsequent siblings';
 			$rv .= pad_link($nmid, $level, $s);
 		} elsif (my $nn = $siblings->[$idx + 2]) {
-			$rv .= $pad . $mapping->{$nn->{id}}->[0];
+			$rv .= $pad . $mapping->{$nn->{mid}}->[0];
 		}
 	}
 	$rv .= $pad ."<a\nhref=#r$id>$s_s, $s_c; $ctx->{s_nr}</a>\n";
@@ -379,7 +367,7 @@ sub walk_thread ($$$) {
 
 sub pre_thread  { # walk_thread callback
 	my ($ctx, $level, $node, $idx) = @_;
-	$ctx->{mapping}->{$node->{id}} = [ '', $node, $idx, $level ];
+	$ctx->{mapping}->{$node->{mid}} = [ '', $node, $idx, $level ];
 	skel_dump($ctx, $level, $node);
 }
 
@@ -398,8 +386,8 @@ sub stream_thread_i { # PublicInbox::WwwStream::getline callback
 		my $node = shift @$q or next;
 		my $cl = $level + 1;
 		unshift @$q, map { ($cl, $_) } @{$node->{children}};
-		if (my $smsg = $ctx->{-inbox}->smsg_mime($node->{smsg})) {
-			return thread_index_entry($ctx, $level, $smsg);
+		if ($ctx->{-inbox}->smsg_mime($node)) {
+			return thread_index_entry($ctx, $level, $node);
 		} else {
 			return ghost_index_entry($ctx, $level, $node);
 		}
@@ -417,7 +405,7 @@ sub stream_thread ($$) {
 		my $node = shift @q or next;
 		my $cl = $level + 1;
 		unshift @q, map { ($cl, $_) } @{$node->{children}};
-		$smsg = $ibx->smsg_mime($node->{smsg}) and last;
+		$smsg = $ibx->smsg_mime($node) and last;
 	}
 	return missing_thread($ctx) unless $smsg;
 
@@ -492,17 +480,6 @@ sub multipart_text_as_html {
 	msg_iter($_[0], \&add_text_body, $_[1], 1);
 }
 
-sub flush_quote {
-	my ($s, $l, $quot) = @_;
-
-	my $rv = $l->to_html($$quot);
-
-	# we use a <span> here to allow users to specify their own
-	# color for quoted text
-	$$quot = undef;
-	$$s .= qq(<span\nclass="q">) . $rv . '</span>'
-}
-
 sub attach_link ($$$$;$) {
 	my ($ctx, $ct, $p, $fn, $err) = @_;
 	my ($part, $depth, @idx) = @$p;
@@ -514,9 +491,6 @@ sub attach_link ($$$$;$) {
 	# spotting MUA problems:
 	$ct =~ s/;.*// unless $err;
 	$ct = ascii_html($ct);
-	my $desc = $part->header('Content-Description');
-	$desc = $fn unless defined $desc;
-	$desc = '' unless defined $desc;
 	my $sfn;
 	if (defined $fn && $fn =~ /\A$PublicInbox::Hval::FN\z/o) {
 		$sfn = $fn;
@@ -528,10 +502,13 @@ sub attach_link ($$$$;$) {
 	my $rv = $ctx->{obuf};
 	$$rv .= qq($nl<a\nhref="$ctx->{mhref}$idx-$sfn">);
 	if ($err) {
-		$$rv .= "[-- Warning: decoded text below may be mangled --]\n";
+		$$rv .= <<EOF;
+[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
+EOF
 	}
 	$$rv .= "[-- Attachment #$idx: ";
 	my $ts = "Type: $ct, Size: $size bytes";
+	my $desc = $part->header('Content-Description') // $fn // '';
 	$desc = ascii_html($desc);
 	$$rv .= ($desc eq '') ? "$ts --]" : "$desc --]\n[-- $ts --]";
 	$$rv .= "</a>\n";
@@ -553,6 +530,9 @@ sub add_text_body { # callback for msg_iter
 	# link generation in diffs with the extra '%0D'
 	$s =~ s/\r\n/\n/sg;
 
+	# will be escaped to `&#8226;' in HTML
+	obfuscate_addrs($ibx, $s, "\x{2022}") if $ibx->{obfuscate};
+
 	# always support diff-highlighting, but we can't linkify hunk
 	# headers for solver unless some coderepo are configured:
 	my $diff;
@@ -586,8 +566,8 @@ sub add_text_body { # callback for msg_iter
 	$s .= "\n" unless $s =~ /\n\z/s;
 
 	# split off quoted and unquoted blocks:
-	my @sections = split(/((?:^>[^\n]*\n)+)/sm, $s);
-	$s = '';
+	my @sections = PublicInbox::MsgIter::split_quotes($s);
+	undef $s; # free memory
 	my $rv = $ctx->{obuf};
 	if (defined($fn) || $depth > 0 || $err) {
 		# badly-encoded message with $err? tell the world about it!
@@ -597,17 +577,19 @@ sub add_text_body { # callback for msg_iter
 	my $l = $ctx->{-linkify} //= PublicInbox::Linkify->new;
 	foreach my $cur (@sections) {
 		if ($cur =~ /\A>/) {
-			flush_quote($rv, $l, \$cur);
+			# we use a <span> here to allow users to specify
+			# their own color for quoted text
+			$$rv .= qq(<span\nclass="q">);
+			$$rv .= $l->to_html($cur);
+			$$rv .= '</span>';
 		} elsif ($diff) {
-			flush_diff($rv, $ctx, \$cur);
+			flush_diff($ctx, \$cur);
 		} else {
 			# regular lines, OK
 			$$rv .= $l->to_html($cur);
-			$cur = undef;
 		}
+		undef $cur; # free memory
 	}
-
-	obfuscate_addrs($ibx, $$rv) if $ibx->{obfuscate};
 }
 
 sub _msg_page_prepare_obuf {
@@ -686,7 +668,7 @@ sub _msg_page_prepare_obuf {
 	}
 	$rv .= _parent_headers($hdr, $over);
 	$rv .= "\n";
-	$ctx->{obuf} = \$rv;
+	\$rv;
 }
 
 sub SKEL_EXPAND () {
@@ -769,9 +751,8 @@ sub _parent_headers {
 
 # returns a string buffer via ->getline
 sub html_footer {
-	my ($ctx) = @_;
+	my ($ctx, $hdr) = @_;
 	my $ibx = $ctx->{-inbox};
-	my $hdr = delete $ctx->{hdr};
 	my $upfx = '../';
 	my $skel = " <a\nhref=\"$upfx\">index</a>";
 	my $rv = '<pre>';
@@ -842,7 +823,7 @@ sub indent_for {
 sub find_mid_root {
 	my ($ctx, $level, $node, $idx) = @_;
 	++$ctx->{root_idx} if $level == 0;
-	if ($node->{id} eq $ctx->{mid}) {
+	if ($node->{mid} eq $ctx->{mid}) {
 		$ctx->{found_mid_at} = $ctx->{root_idx};
 		return 0;
 	}
@@ -899,25 +880,25 @@ sub missing_thread {
 sub dedupe_subject {
 	my ($prev_subj, $subj, $val) = @_;
 
-	my $omit = ''; # '"' denotes identical text omitted
+	my $omit; # '"' denotes identical text omitted
 	my (@prev_pop, @curr_pop);
 	while (@$prev_subj && @$subj && $subj->[-1] eq $prev_subj->[-1]) {
 		push(@prev_pop, pop(@$prev_subj));
 		push(@curr_pop, pop(@$subj));
-		$omit ||= $val;
+		$omit //= $val;
 	}
 	pop @$subj if @$subj && $subj->[-1] =~ /^re:\s*/i;
 	if (scalar(@curr_pop) == 1) {
-		$omit = '';
+		$omit = undef;
 		push @$prev_subj, @prev_pop;
 		push @$subj, @curr_pop;
 	}
-	$omit;
+	$omit // '';
 }
 
 sub skel_dump { # walk_thread callback
-	my ($ctx, $level, $node) = @_;
-	my $smsg = $node->{smsg} or return _skel_ghost($ctx, $level, $node);
+	my ($ctx, $level, $smsg) = @_;
+	$smsg->{blob} or return _skel_ghost($ctx, $level, $smsg);
 
 	my $skel = $ctx->{skel};
 	my $cur = $ctx->{cur};
@@ -1000,7 +981,7 @@ sub skel_dump { # walk_thread callback
 sub _skel_ghost {
 	my ($ctx, $level, $node) = @_;
 
-	my $mid = $node->{id};
+	my $mid = $node->{mid};
 	my $d = '     [not found] ';
 	$d .= '    '  if exists $ctx->{searchview};
 	$d .= indent_for($level) . th_pfx($level);
@@ -1023,18 +1004,23 @@ sub _skel_ghost {
 
 sub sort_ds {
 	[ sort {
-		(eval { $a->topmost->{smsg}->{ds} } || 0) <=>
-		(eval { $b->topmost->{smsg}->{ds} } || 0)
+		(eval { $a->topmost->{ds} } || 0) <=>
+		(eval { $b->topmost->{ds} } || 0)
 	} @{$_[0]} ];
 }
 
 # accumulate recent topics if search is supported
 # returns 200 if done, 404 if not
 sub acc_topic { # walk_thread callback
-	my ($ctx, $level, $node) = @_;
-	my $mid = $node->{id};
-	my $smsg = $node->{smsg} // $ctx->{-inbox}->smsg_by_mid($mid);
-	if ($smsg) {
+	my ($ctx, $level, $smsg) = @_;
+	my $mid = $smsg->{mid};
+	my $has_blob = $smsg->{blob} // do {
+		if (my $by_mid = $ctx->{-inbox}->smsg_by_mid($mid)) {
+			%$smsg = (%$smsg, %$by_mid);
+			1;
+		}
+	};
+	if ($has_blob) {
 		my $subj = subject_normalized($smsg->{subject});
 		$subj = '(no subject)' if $subj eq '';
 		my $ds = $smsg->{ds};
@@ -1067,7 +1053,7 @@ sub acc_topic { # walk_thread callback
 sub dump_topics {
 	my ($ctx) = @_;
 	my $order = delete $ctx->{order}; # [ ds, subj1, subj2, subj3, ... ]
-	if (!@$order) {
+	unless ($order) {
 		$ctx->{-html_tip} = '<pre>[No topics in range]</pre>';
 		return 404;
 	}
@@ -1225,7 +1211,7 @@ sub thread_adj_level {
 sub ghost_index_entry {
 	my ($ctx, $level, $node) = @_;
 	my ($beg, $end) = thread_adj_level($ctx,  $level);
-	$beg . '<pre>'. ghost_parent($ctx->{-upfx}, $node->{id})
+	$beg . '<pre>'. ghost_parent($ctx->{-upfx}, $node->{mid})
 		. '</pre>' . $end;
 }