X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FView.pm;h=b6d7acaf64642d31c240febcf431891298e778c3;hb=4da62f284003d75abe7cb35594414eb2224f42bc;hp=5baaffaf5deda70395a5078d47c09f9adff4eac7;hpb=87678710135973f72722258e171fc00f85c86ec8;p=public-inbox.git

diff --git a/lib/PublicInbox/View.pm b/lib/PublicInbox/View.pm
index 5baaffaf..b6d7acaf 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
+			$MID_EXTRACT);
 use PublicInbox::MsgIter;
 use PublicInbox::Address;
 use PublicInbox::WwwStream;
@@ -149,7 +150,7 @@ sub in_reply_to {
 sub fold_addresses ($) {
 	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
@@ -299,7 +300,7 @@ sub _th_index_lite {
 	if (my $smsg = $node->{smsg}) {
 		# delete saves about 200KB on a 1K message thread
 		if (my $refs = delete $smsg->{references}) {
-			($$irt) = ($refs =~ m/<([^>]+)>\z/);
+			($$irt) = ($refs =~ m/$MID_EXTRACT\z/o);
 		}
 	}
 	my $irt_map = $mapping->{$$irt} if defined $$irt;
@@ -481,17 +482,6 @@ 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;
-	$$s .= qq(<span\nclass="q">) . $rv . '</span>'
-}
-
 sub attach_link ($$$$;$) {
 	my ($ctx, $ct, $p, $fn, $err) = @_;
 	my ($part, $depth, @idx) = @$p;
@@ -503,9 +493,6 @@ sub attach_link ($$$$;$) {
 	# spotting MUA problems:
 	$ct =~ s/;.*// unless $err;
 	$ct = ascii_html($ct);
-	my $desc = $part->header('Content-Description');
-	$desc = $fn unless defined $desc;
-	$desc = '' unless defined $desc;
 	my $sfn;
 	if (defined $fn && $fn =~ /\A$PublicInbox::Hval::FN\z/o) {
 		$sfn = $fn;
@@ -517,10 +504,13 @@ sub attach_link ($$$$;$) {
 	my $rv = $ctx->{obuf};
 	$$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";
@@ -542,6 +532,9 @@ 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;
@@ -575,8 +568,8 @@ 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!
@@ -586,17 +579,19 @@ 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 {
@@ -887,20 +882,20 @@ 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
@@ -1055,7 +1050,7 @@ 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;
 	}