X-Git-Url: http://www.git.stargrave.org/?p=public-inbox.git;a=blobdiff_plain;f=lib%2FPublicInbox%2FView.pm;h=58851edc59f5226bc47a4c6ba3f7627a19ba3e00;hp=9ef4712f313cc0bf59f21cbbf38fcbb8beeb0bac;hb=55b22e6414b78336633ca243c113c712af2c3bcd;hpb=92f27ed0be327ab6acb61aeedf7a77702cc6c25f diff --git a/lib/PublicInbox/View.pm b/lib/PublicInbox/View.pm index 9ef4712f..58851edc 100644 --- a/lib/PublicInbox/View.pm +++ b/lib/PublicInbox/View.pm @@ -1,35 +1,47 @@ -# Copyright (C) 2014-2015 all contributors -# License: AGPLv3 or later (https://www.gnu.org/licenses/agpl-3.0.txt) +# Copyright (C) 2014-2018 all contributors +# License: AGPL-3.0+ # # Used for displaying the HTML web interface. # See Documentation/design_www.txt for this. package PublicInbox::View; use strict; use warnings; -use URI::Escape qw/uri_escape_utf8/; -use Date::Parse qw/str2time/; -use PublicInbox::Hval qw/ascii_html/; +use PublicInbox::MsgTime qw(msg_datestamp); +use PublicInbox::Hval qw/ascii_html obfuscate_addrs/; use PublicInbox::Linkify; -use PublicInbox::MID qw/mid_clean id_compress mid_mime mid_escape/; +use PublicInbox::MID qw/id_compress mid_escape mids references/; use PublicInbox::MsgIter; use PublicInbox::Address; use PublicInbox::WwwStream; +use PublicInbox::Reply; require POSIX; +use Time::Local qw(timegm); +use constant COLS => 72; use constant INDENT => ' '; use constant TCHILD => '` '; sub th_pfx ($) { $_[0] == 0 ? '' : TCHILD }; # public functions: (unstable) + sub msg_html { - my ($ctx, $mime) = @_; + my ($ctx, $mime, $more, $smsg) = @_; my $hdr = $mime->header_obj; - my $tip = _msg_html_prepare($hdr, $ctx); + my $ibx = $ctx->{-inbox}; + my $obfs_ibx = $ctx->{-obfs_ibx} = $ibx->{obfuscate} ? $ibx : undef; + my $tip = _msg_html_prepare($hdr, $ctx, $more, 0); + my $end = 2; PublicInbox::WwwStream->response($ctx, 200, sub { my ($nr, undef) = @_; if ($nr == 1) { - $tip . multipart_text_as_html($mime, '') . '
' - } elsif ($nr == 2) { + # $more cannot be true w/o $smsg being defined: + my $upfx = $more ? '../'.mid_escape($smsg->mid).'/' : ''; + $tip . multipart_text_as_html($mime, $upfx, $obfs_ibx) . + '
' + } elsif ($more && @$more) { + ++$end; + msg_html_more($ctx, $more, $nr); + } elsif ($nr == $end) { # fake an EOF if generating the footer fails; # we want to at least show the message if something # here crashes: @@ -43,6 +55,53 @@ sub msg_html { }); } +sub msg_page { + my ($ctx) = @_; + my $mid = $ctx->{mid}; + my $ibx = $ctx->{-inbox}; + my ($first, $more); + my $smsg; + if (my $srch = $ibx->search) { + my ($id, $prev); + $smsg = $srch->next_by_mid($mid, \$id, \$prev); + $first = $ibx->msg_by_smsg($smsg) if $smsg; + if ($first) { + my $next = $srch->next_by_mid($mid, \$id, \$prev); + $more = [ $id, $prev, $next ] if $next; + } + return unless $first; + } else { + $first = $ibx->msg_by_mid($mid) or return; + } + msg_html($ctx, PublicInbox::MIME->new($first), $more, $smsg); +} + +sub msg_html_more { + my ($ctx, $more, $nr) = @_; + my $str = eval { + my ($id, $prev, $smsg) = @$more; + my $mid = $ctx->{mid}; + $smsg = $ctx->{-inbox}->smsg_mime($smsg); + my $next = $ctx->{srch}->next_by_mid($mid, \$id, \$prev); + @$more = $next ? ($id, $prev, $next) : (); + if ($smsg) { + my $mime = $smsg->{mime}; + my $upfx = '../' . mid_escape($smsg->mid) . '/'; + _msg_html_prepare($mime->header_obj, $ctx, $more, $nr) . + multipart_text_as_html($mime, $upfx, + $ctx->{-obfs_ibx}) . + '
' + } else { + ''; + } + }; + if ($@) { + warn "Error lookup up additional messages: $@\n"; + $str = '
Error looking up additional messages
'; + } + $str; +} + # /$INBOX/$MESSAGE_ID/#R sub msg_reply { my ($ctx, $hdr) = @_; @@ -52,12 +111,25 @@ sub msg_reply { 'https://en.wikipedia.org/wiki/Posting_style#Interleaved_style'; my $info = ''; - if (my $url = $ctx->{-inbox}->{infourl}) { + my $ibx = $ctx->{-inbox}; + if (my $url = $ibx->{infourl}) { $url = PublicInbox::Hval::prurl($ctx->{env}, $url); $info = qq(\n List information: $url\n); } - my ($arg, $link) = mailto_arg_link($hdr); + my ($arg, $link, $reply_to_all) = + PublicInbox::Reply::mailto_arg_link($ibx, $hdr); + + # mailto: link only works if address obfuscation is disabled + if ($link) { + $link = <In-Reply-To header + via mailto: links, try the mailto: link +EOF + } + push @$arg, '/path/to/YOUR_REPLY'; $arg = ascii_html(join(" \\\n ", '', @$arg)); <this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, - and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: $p_url $info -* Reply to all the recipients using the --to, --cc, - and --in-reply-to switches of git-send-email(1): +* Reply using the --to, --cc, and --in-reply-to + switches of git-send-email(1): git send-email$arg $se_url - -* If your mail client supports setting the In-Reply-To header - via mailto: links, try the mailto: link +$link EOF } sub in_reply_to { my ($hdr) = @_; - my %mid = map { $_ => 1 } $hdr->header_raw('Message-ID'); - my @refs = ($hdr->header_raw('References'), - $hdr->header_raw('In-Reply-To')); - @refs = ((join(' ', @refs)) =~ /<([^>]+)>/g); - while (defined(my $irt = pop @refs)) { - next if $mid{"<$irt>"}; - return $irt; - } - undef; + my $refs = references($hdr); + $refs->[-1]; } -sub _hdr_names ($$) { +sub fold_addresses ($) { + return $_[0] if length($_[0]) <= COLS; + # try to fold on commas after non-word chars before $lim chars, + # Try to get the "," preceeded by ">" or ")", but avoid folding + # on the comma where somebody uses "Lastname, Firstname". + # We also try to keep the last and penultimate addresses in + # the list on the same line if possible, hence the extra \z + # Fall back to folding on spaces at $lim + 1 chars + my $lim = COLS - 8; # 8 = "\t" display width + my $too_long = $lim + 1; + $_[0] =~ s/\s*\z//s; # Email::Simple doesn't strip trailing spaces + $_[0] = join("\n\t", + ($_[0] =~ /(.{0,$lim}\W(?:,|\z)| + .{1,$lim}(?:,|\z)| + .{1,$lim}| + .{$too_long,}?)(?:\s|\z)/xgo)); +} + +sub _hdr_names_html ($$) { my ($hdr, $field) = @_; my $val = $hdr->header($field) or return ''; ascii_html(join(', ', PublicInbox::Address::names($val))); @@ -117,34 +197,51 @@ sub nr_to_s ($$$) { # this is already inside a
 sub index_entry {
-	my ($mime, $ctx, $more) = @_;
+	my ($smsg, $ctx, $more) = @_;
 	my $srch = $ctx->{srch};
-	my $hdr = $mime->header_obj;
-	my $subj = $hdr->header('Subject');
-
-	my $mid_raw = mid_clean(mid_mime($mime));
+	my $subj = $smsg->subject;
+	my $mid_raw = $smsg->mid;
 	my $id = id_compress($mid_raw, 1);
 	my $id_m = 'm'.$id;
 
 	my $root_anchor = $ctx->{root_anchor} || '';
 	my $irt;
+	my $obfs_ibx = $ctx->{-obfs_ibx};
 
 	my $rv = "* ";
 	$subj = ''.ascii_html($subj).'';
+	obfuscate_addrs($obfs_ibx, $subj) if $obfs_ibx;
 	$subj = "$subj" if $root_anchor eq $id_m;
 	$rv .= $subj . "\n";
 	$rv .= _th_index_lite($mid_raw, \$irt, $id, $ctx);
 	my @tocc;
-	foreach my $f (qw(To Cc)) {
-		my $dst = _hdr_names($hdr, $f);
-		push @tocc, "$f: $dst" if $dst ne '';
-	}
-	$rv .= "From: "._hdr_names($hdr, 'From').' @ '._msg_date($hdr)." UTC";
+	my $mime = $smsg->{mime};
+	my $hdr = $mime->header_obj;
+	my $from = _hdr_names_html($hdr, 'From');
+	obfuscate_addrs($obfs_ibx, $from) if $obfs_ibx;
+	$rv .= "From: $from @ ".fmt_ts($smsg->ds)." UTC";
 	my $upfx = $ctx->{-upfx};
 	my $mhref = $upfx . mid_escape($mid_raw) . '/';
 	$rv .= qq{ (permalink / };
 	$rv .= qq{raw)\n};
-	$rv .= '  '.join('; +', @tocc) . "\n" if @tocc;
+	my $to = fold_addresses(_hdr_names_html($hdr, 'To'));
+	my $cc = fold_addresses(_hdr_names_html($hdr, 'Cc'));
+	my ($tlen, $clen) = (length($to), length($cc));
+	my $to_cc = '';
+	if (($tlen + $clen) > COLS) {
+		$to_cc .= '  To: '.$to."\n" if $tlen;
+		$to_cc .= '  Cc: '.$cc."\n" if $clen;
+	} else {
+		if ($tlen) {
+			$to_cc .= '  To: '.$to;
+			$to_cc .= '; +Cc: '.$cc if $clen;
+		} else {
+			$to_cc .= '  Cc: '.$cc if $clen;
+		}
+		$to_cc .= "\n";
+	}
+	obfuscate_addrs($obfs_ibx, $to_cc) if $obfs_ibx;
+	$rv .= $to_cc;
 
 	my $mapping = $ctx->{mapping};
 	if (!$mapping && (defined($irt) || defined($irt = in_reply_to($hdr)))) {
@@ -156,7 +253,7 @@ sub index_entry {
 	$rv .= "\n";
 
 	# scan through all parts, looking for displayable text
-	msg_iter($mime, sub { $rv .= add_text_body($mhref, $_[0]) });
+	msg_iter($mime, sub { $rv .= add_text_body($mhref, $obfs_ibx, $_[0]) });
 
 	# add the footer
 	$rv .= "\n^ ".
@@ -169,17 +266,17 @@ sub index_entry {
 		$rv .= "\t[relevance $pct->{$mid_raw}%]";
 		$hr = 1;
 	} elsif ($mapping) {
-		my $threaded = 'threaded';
+		my $nested = 'nested';
 		my $flat = 'flat';
 		my $end = '';
 		if ($ctx->{flat}) {
 			$hr = 1;
 			$flat = "$flat";
 		} else {
-			$threaded = "$threaded";
+			$nested = "$nested";
 		}
 		$rv .= "\t[$flat";
-		$rv .= "|$threaded]";
+		$rv .= "|$nested]";
 		$rv .= " $ctx->{s_nr}";
 	} else {
 		$hr = $ctx->{-hr};
@@ -282,29 +379,30 @@ sub pre_thread  {
 }
 
 sub thread_index_entry {
-	my ($ctx, $level, $mime) = @_;
+	my ($ctx, $level, $smsg) = @_;
 	my ($beg, $end) = thread_adj_level($ctx, $level);
-	$beg . '
' . index_entry($mime, $ctx, 0) . '
' . $end; + $beg . '
' . index_entry($smsg, $ctx, 0) . '
' . $end; } sub stream_thread ($$) { my ($rootset, $ctx) = @_; my $inbox = $ctx->{-inbox}; - my $mime; my @q = map { (0, $_) } @$rootset; my $level; + my $smsg; while (@q) { $level = shift @q; my $node = shift @q or next; my $cl = $level + 1; unshift @q, map { ($cl, $_) } @{$node->{children}}; - $mime = $inbox->msg_by_smsg($node->{smsg}) and last; + $smsg = $inbox->smsg_mime($node->{smsg}) and last; } - return missing_thread($ctx) unless $mime; + return missing_thread($ctx) unless $smsg; - $mime = PublicInbox::MIME->new($mime); - $ctx->{-title_html} = ascii_html($mime->header('Subject')); - $ctx->{-html_tip} = thread_index_entry($ctx, $level, $mime); + $ctx->{-obfs_ibx} = $inbox->{obfuscate} ? $inbox : undef; + $ctx->{-title_html} = ascii_html($smsg->subject); + $ctx->{-html_tip} = thread_index_entry($ctx, $level, $smsg); + $smsg = undef; PublicInbox::WwwStream->response($ctx, 200, sub { return unless $ctx; while (@q) { @@ -312,10 +410,8 @@ sub stream_thread ($$) { my $node = shift @q or next; my $cl = $level + 1; unshift @q, map { ($cl, $_) } @{$node->{children}}; - my $mid = $node->{id}; - if ($mime = $inbox->msg_by_smsg($node->{smsg})) { - $mime = PublicInbox::MIME->new($mime); - return thread_index_entry($ctx, $level, $mime); + if ($smsg = $inbox->smsg_mime($node->{smsg})) { + return thread_index_entry($ctx, $level, $smsg); } else { return ghost_index_entry($ctx, $level, $node); } @@ -331,9 +427,7 @@ sub thread_html { my ($ctx) = @_; my $mid = $ctx->{mid}; my $srch = $ctx->{srch}; - my $sres = $srch->get_thread($mid); - my $msgs = load_results($srch, $sres); - my $nr = $sres->{total}; + my ($nr, $msgs) = $srch->get_thread($mid); return missing_thread($ctx) if $nr == 0; my $skel = '
';
 	$skel .= $nr == 1 ? 'only message in thread' : 'end of thread';
@@ -349,49 +443,46 @@ sub thread_html {
 	$ctx->{prev_attr} = '';
 	$ctx->{prev_level} = 0;
 	$ctx->{root_anchor} = anchor_for($mid);
-	$ctx->{seen} = {};
 	$ctx->{mapping} = {};
 	$ctx->{s_nr} = "$nr+ messages in thread";
 
-	my $rootset = thread_results($msgs);
+	my $rootset = thread_results($ctx, $msgs);
+
+	# reduce hash lookups in pre_thread->skel_dump
+	my $inbox = $ctx->{-inbox};
+	$ctx->{-obfs_ibx} = $inbox->{obfuscate} ? $inbox : undef;
 	walk_thread($rootset, $ctx, *pre_thread);
+
 	$skel .= '
'; return stream_thread($rootset, $ctx) unless $ctx->{flat}; # flat display: lazy load the full message from smsg - my $inbox = $ctx->{-inbox}; - my $mime; - while ($mime = shift @$msgs) { - $mime = $inbox->msg_by_smsg($mime) and last; - } - return missing_thread($ctx) unless $mime; - $mime = PublicInbox::MIME->new($mime); - $ctx->{-title_html} = ascii_html($mime->header('Subject')); - $ctx->{-html_tip} = '
'.index_entry($mime, $ctx, scalar @$msgs);
-	$mime = undef;
+	my $smsg;
+	while (my $m = shift @$msgs) {
+		$smsg = $inbox->smsg_mime($m) and last;
+	}
+	return missing_thread($ctx) unless $smsg;
+	$ctx->{-title_html} = ascii_html($smsg->subject);
+	$ctx->{-html_tip} = '
'.index_entry($smsg, $ctx, scalar @$msgs);
+	$smsg = undef;
 	PublicInbox::WwwStream->response($ctx, 200, sub {
 		return unless $msgs;
-		while ($mime = shift @$msgs) {
-			$mime = $inbox->msg_by_smsg($mime) and last;
-		}
-		if ($mime) {
-			$mime = PublicInbox::MIME->new($mime);
-			return index_entry($mime, $ctx, scalar @$msgs);
+		$smsg = undef;
+		while (my $m = shift @$msgs) {
+			$smsg = $inbox->smsg_mime($m) and last;
 		}
+		return index_entry($smsg, $ctx, scalar @$msgs) if $smsg;
 		$msgs = undef;
 		$skel;
 	});
 }
 
 sub multipart_text_as_html {
-	my ($mime, $upfx) = @_;
+	my ($mime, $upfx, $obfs_ibx) = @_;
 	my $rv = "";
 
 	# scan through all parts, looking for displayable text
-	msg_iter($mime, sub {
-		my ($p) = @_;
-		$rv .= add_text_body($upfx, $p);
-	});
+	msg_iter($mime, sub { $rv .= add_text_body($upfx, $obfs_ibx, $_[0]) });
 	$rv;
 }
 
@@ -444,7 +535,8 @@ sub attach_link ($$$$;$) {
 }
 
 sub add_text_body {
-	my ($upfx, $p) = @_; # from msg_iter: [ Email::MIME, depth, @idx ]
+	my ($upfx, $obfs_ibx, $p) = @_;
+	# $p - from msg_iter: [ Email::MIME, depth, @idx ]
 	my ($part, $depth) = @$p; # attachment @idx is unused
 	my $ct = $part->content_type || 'text/plain';
 	my $fn = $part->filename;
@@ -495,50 +587,91 @@ sub add_text_body {
 
 	if (@quot) { # ugh, top posted
 		flush_quote(\$s, $l, \@quot);
-	} elsif ($s =~ /\n\z/s) { # common, last line ends with a newline
+		obfuscate_addrs($obfs_ibx, $s) if $obfs_ibx;
 		$s;
-	} else { # some editors don't do newlines...
-		$s .= "\n";
+	} else {
+		obfuscate_addrs($obfs_ibx, $s) if $obfs_ibx;
+		if ($s =~ /\n\z/s) { # common, last line ends with a newline
+			$s;
+		} else { # some editors don't do newlines...
+			$s .= "\n";
+		}
 	}
 }
 
 sub _msg_html_prepare {
-	my ($hdr, $ctx) = @_;
+	my ($hdr, $ctx, $more, $nr) = @_;
 	my $srch = $ctx->{srch} if $ctx;
 	my $atom = '';
-	my $rv = ""; # anchor for body start
-
+	my $obfs_ibx = $ctx->{-obfs_ibx};
+	my $rv = '';
+	my $mids = mids($hdr);
+	my $multiple = scalar(@$mids) > 1; # zero, one, infinity
+	if ($nr == 0) {
+		if ($more) {
+			$rv .=
+"
WARNING: multiple messages refer to this Message-ID\n
"; + } + $rv .= ""; # anchor for body start + } else { + $rv .= '
';
+	}
 	if ($srch) {
 		$ctx->{-upfx} = '../';
 	}
 	my @title;
-	my $mid = mid_clean($hdr->header_raw('Message-ID'));
-	$mid = PublicInbox::Hval->new_msgid($mid);
-	foreach my $h (qw(From To Cc Subject Date)) {
-		my $v = $hdr->header($h);
-		defined($v) && ($v ne '') or next;
+	my $v;
+	if (defined($v = $hdr->header('From'))) {
 		$v = PublicInbox::Hval->new($v);
-
-		if ($h eq 'From') {
-			my @n = PublicInbox::Address::names($v->raw);
-			$title[1] = ascii_html(join(', ', @n));
-		} elsif ($h eq 'Subject') {
-			$title[0] = $v->as_html;
-			if ($srch) {
-				$rv .= qq($h: );
-				$rv .= $v->as_html . "\n";
-				next;
-			}
-		}
+		my @n = PublicInbox::Address::names($v->raw);
+		$title[1] = ascii_html(join(', ', @n));
 		$v = $v->as_html;
-		$v =~ s/(\@[^,]+,) /$1\n\t/g if ($h eq 'Cc' || $h eq 'To');
-		$rv .= "$h: $v\n";
-
+		if ($obfs_ibx) {
+			obfuscate_addrs($obfs_ibx, $v);
+			obfuscate_addrs($obfs_ibx, $title[1]);
+		}
+		$rv .= "From: $v\n" if $v ne '';
+	}
+	foreach my $h (qw(To Cc)) {
+		defined($v = $hdr->header($h)) or next;
+		fold_addresses($v);
+		$v = ascii_html($v);
+		obfuscate_addrs($obfs_ibx, $v) if $obfs_ibx;
+		$rv .= "$h: $v\n" if $v ne '';
+	}
+	if (defined($v = $hdr->header('Subject')) && ($v ne '')) {
+		$v = ascii_html($v);
+		obfuscate_addrs($obfs_ibx, $v) if $obfs_ibx;
+		if ($srch) {
+			$rv .= qq(Subject: $v\n);
+		} else {
+			$rv .= "Subject: $v\n";
+		}
+		$title[0] = $v;
+	} else { # dummy anchor for thread skeleton at bottom of page
+		$rv .= qq() if $srch;
+		$title[0] = '(no subject)';
+	}
+	if (defined($v = $hdr->header('Date'))) {
+		$v = ascii_html($v);
+		obfuscate_addrs($obfs_ibx, $v) if $obfs_ibx; # possible :P
+		$rv .= "Date: $v\n";
 	}
-	$title[0] ||= '(no subject)';
 	$ctx->{-title_html} = join(' - ', @title);
-	$rv .= 'Message-ID: <' . $mid->as_html . '> ';
-	$rv .= "(raw)\n";
+	foreach (@$mids) {
+		my $mid = PublicInbox::Hval->new_msgid($_) ;
+		my $mhtml = $mid->as_html;
+		if ($multiple) {
+			my $href = $mid->{href};
+			$rv .= "Message-ID: ";
+			$rv .= "";
+			$rv .= "<$mhtml> ";
+			$rv .= "(raw)\n";
+		} else {
+			$rv .= "Message-ID: <$mhtml> ";
+			$rv .= "(raw)\n";
+		}
+	}
 	$rv .= _parent_headers($hdr, $srch);
 	$rv .= "\n";
 }
@@ -546,38 +679,43 @@ sub _msg_html_prepare {
 sub thread_skel {
 	my ($dst, $ctx, $hdr, $tpfx) = @_;
 	my $srch = $ctx->{srch};
-	my $mid = mid_clean($hdr->header_raw('Message-ID'));
-	my $sres = $srch->get_thread($mid);
-	my $nr = $sres->{total};
-	my $expand = qq(expand ) .
-			qq(/ mbox.gz ) .
-			qq(/ Atom feed);
+	my $mid = mids($hdr)->[0];
+	my ($nr, $msgs) = $srch->get_thread($mid);
+	my $expand = qq(expand[flat) .
+	                qq(|nested]  ) .
+			qq(mbox.gz  ) .
+			qq(Atom feed);
 
 	my $parent = in_reply_to($hdr);
 	$$dst .= "\nThread overview: ";
 	if ($nr <= 1) {
 		if (defined $parent) {
-			$$dst .= "($expand)\n ";
+			$$dst .= "$expand\n ";
 			$$dst .= ghost_parent("$tpfx../", $parent) . "\n";
 		} else {
-			$$dst .= "[no followups, yet] ($expand)\n";
+			$$dst .= "[no followups] $expand\n";
 		}
 		$ctx->{next_msg} = undef;
 		$ctx->{parent_msg} = $parent;
 		return;
 	}
 
-	$$dst .= "$nr+ messages in thread ($expand";
-	$$dst .= qq! / [top])\n!;
+	$$dst .= "$nr+ messages / $expand";
+	$$dst .= qq!  top\n!;
 
-	my $subj = $srch->subject_path($hdr->header('Subject'));
-	$ctx->{seen} = { $subj => 1 };
+	my $subj = $hdr->header('Subject');
+	defined $subj or $subj = '';
+	$ctx->{prev_subj} = [ split(/ /, $srch->subject_normalized($subj)) ];
 	$ctx->{cur} = $mid;
 	$ctx->{prev_attr} = '';
 	$ctx->{prev_level} = 0;
 	$ctx->{dst} = $dst;
-	$sres = load_results($srch, $sres);
-	walk_thread(thread_results($sres), $ctx, *skel_dump);
+
+	# reduce hash lookups in skel_dump
+	my $ibx = $ctx->{-inbox};
+	$ctx->{-obfs_ibx} = $ibx->{obfuscate} ? $ibx : undef;
+	walk_thread(thread_results($ctx, $msgs), $ctx, *skel_dump);
+
 	$ctx->{parent_msg} = $parent;
 }
 
@@ -585,7 +723,8 @@ sub _parent_headers {
 	my ($hdr, $srch) = @_;
 	my $rv = '';
 
-	my $irt = in_reply_to($hdr);
+	my $refs = references($hdr);
+	my $irt = pop @$refs;
 	if (defined $irt) {
 		my $v = PublicInbox::Hval->new_msgid($irt);
 		my $html = $v->as_html;
@@ -598,69 +737,13 @@ sub _parent_headers {
 	# we show the thread skeleton at the bottom, instead.
 	return $rv if $srch;
 
-	my $refs = $hdr->header_raw('References');
-	if ($refs) {
-		# avoid redundant URLs wasting bandwidth
-		my %seen;
-		$seen{$irt} = 1 if defined $irt;
-		my @refs;
-		my @raw_refs = ($refs =~ /<([^>]+)>/g);
-		foreach my $ref (@raw_refs) {
-			next if $seen{$ref};
-			$seen{$ref} = 1;
-			push @refs, linkify_ref_nosrch($ref);
-		}
-
-		if (@refs) {
-			$rv .= 'References: '. join("\n\t", @refs) . "\n";
-		}
+	if (@$refs) {
+		@$refs = map { linkify_ref_nosrch($_) } @$refs;
+		$rv .= 'References: '. join("\n\t", @$refs) . "\n";
 	}
 	$rv;
 }
 
-sub squote_maybe ($) {
-	my ($val) = @_;
-	if ($val =~ m{([^\w@\./,\%\+\-])}) {
-		$val =~ s/(['!])/'\\$1'/g; # '!' for csh
-		return "'$val'";
-	}
-	$val;
-}
-
-sub mailto_arg_link {
-	my ($hdr) = @_;
-	my %cc; # everyone else
-	my $to; # this is the From address
-
-	foreach my $h (qw(From To Cc)) {
-		my $v = $hdr->header($h);
-		defined($v) && ($v ne '') or next;
-		my @addrs = PublicInbox::Address::emails($v);
-		foreach my $address (@addrs) {
-			my $dst = lc($address);
-			$cc{$dst} ||= $address;
-			$to ||= $dst;
-		}
-	}
-	my @arg;
-
-	my $subj = $hdr->header('Subject') || '';
-	$subj = "Re: $subj" unless $subj =~ /\bRe:/i;
-	my $mid = $hdr->header_raw('Message-ID');
-	push @arg, '--in-reply-to='.squote_maybe(mid_clean($mid));
-	my $irt = mid_escape($mid);
-	delete $cc{$to};
-	push @arg, "--to=$to";
-	$to = uri_escape_utf8($to);
-	$subj = uri_escape_utf8($subj);
-	my @cc = sort values %cc;
-	push(@arg, map { "--cc=$_" } @cc);
-	my $cc = uri_escape_utf8(join(',', @cc));
-	my $href = "mailto:$to?In-Reply-To=$irt&Cc=${cc}&Subject=$subj";
-
-	(\@arg, ascii_html($href));
-}
-
 sub html_footer {
 	my ($hdr, $standalone, $ctx, $rhref) = @_;
 
@@ -732,25 +815,13 @@ sub ghost_parent {
 
 sub indent_for {
 	my ($level) = @_;
-	INDENT x ($level - 1);
-}
-
-sub load_results {
-	my ($srch, $sres) = @_;
-	my $msgs = delete $sres->{msgs};
-	$srch->retry_reopen(sub { [ map { $_->mid; $_ } @$msgs ] });
-}
-
-sub msg_timestamp {
-	my ($hdr) = @_;
-	my $ts = eval { str2time($hdr->header('Date')) };
-	defined($ts) ? $ts : 0;
+	$level ? INDENT x ($level - 1) : '';
 }
 
 sub thread_results {
-	my ($msgs) = @_;
+	my ($ctx, $msgs) = @_;
 	require PublicInbox::SearchThread;
-	PublicInbox::SearchThread::thread($msgs, *sort_ts);
+	PublicInbox::SearchThread::thread($msgs, *sort_ds, $ctx->{-inbox});
 }
 
 sub missing_thread {
@@ -761,12 +832,30 @@ sub missing_thread {
 
 sub _msg_date {
 	my ($hdr) = @_;
-	my $ts = $hdr->header('X-PI-TS') || msg_timestamp($hdr);
-	fmt_ts($ts);
+	fmt_ts(msg_datestamp($hdr));
 }
 
 sub fmt_ts { POSIX::strftime('%Y-%m-%d %k:%M', gmtime($_[0])) }
 
+sub dedupe_subject {
+	my ($prev_subj, $subj, $val) = @_;
+
+	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 ||= $val;
+	}
+	pop @$subj if @$subj && $subj->[-1] =~ /^re:\s*/i;
+	if (scalar(@curr_pop) == 1) {
+		$omit = '';
+		push @$prev_subj, @prev_pop;
+		push @$subj, @curr_pop;
+	}
+	$omit;
+}
+
 sub skel_dump {
 	my ($ctx, $level, $node) = @_;
 	my $smsg = $node->{smsg} or return _skel_ghost($ctx, $level, $node);
@@ -774,8 +863,12 @@ sub skel_dump {
 	my $dst = $ctx->{dst};
 	my $cur = $ctx->{cur};
 	my $mid = $smsg->{mid};
+
 	my $f = ascii_html($smsg->from_name);
-	my $d = fmt_ts($smsg->{ts}) . ' ' . indent_for($level) . th_pfx($level);
+	my $obfs_ibx = $ctx->{-obfs_ibx};
+	obfuscate_addrs($obfs_ibx, $f) if $obfs_ibx;
+
+	my $d = fmt_ts($smsg->{ds}) . ' ' . indent_for($level) . th_pfx($level);
 	my $attr = $f;
 	$ctx->{first_level} ||= $level;
 
@@ -800,19 +893,25 @@ sub skel_dump {
 	# Subject is never undef, this mail was loaded from
 	# our Xapian which would've resulted in '' if it were
 	# really missing (and Filter rejects empty subjects)
-	my $s = $smsg->subject;
-	my $h = $ctx->{srch}->subject_path($s);
-	if ($ctx->{seen}->{$h}) {
-		$s = undef;
+	my @subj = split(/ /, $ctx->{srch}->subject_normalized($smsg->subject));
+
+	# remove common suffixes from the subject if it matches the previous,
+	# so we do not show redundant text at the end.
+	my $prev_subj = $ctx->{prev_subj} || [];
+	$ctx->{prev_subj} = [ @subj ];
+	my $omit = dedupe_subject($prev_subj, \@subj, '" ');
+	my $end;
+	if (@subj) {
+		my $subj = join(' ', @subj);
+		$subj = ascii_html($subj);
+		obfuscate_addrs($obfs_ibx, $subj) if $obfs_ibx;
+		$end = "$subj $omit$f\n"
 	} else {
-		$ctx->{seen}->{$h} = 1;
-		$s = PublicInbox::Hval->new($s);
-		$s = $s->as_html;
+		$end = "$f\n";
 	}
 	my $m;
 	my $id = '';
 	my $mapping = $ctx->{mapping};
-	my $end = defined($s) ? "$s $f\n" : "$f\n";
 	if ($mapping) {
 		my $map = $mapping->{$mid};
 		$id = id_compress($mid, 1);
@@ -850,10 +949,10 @@ sub _skel_ghost {
 	$$dst .= $d;
 }
 
-sub sort_ts {
+sub sort_ds {
 	[ sort {
-		(eval { $a->topmost->{smsg}->ts } || 0) <=>
-		(eval { $b->topmost->{smsg}->ts } || 0)
+		(eval { $a->topmost->{smsg}->ds } || 0) <=>
+		(eval { $b->topmost->{smsg}->ds } || 0)
 	} @{$_[0]} ];
 }
 
@@ -863,22 +962,22 @@ sub acc_topic {
 	my ($ctx, $level, $node) = @_;
 	my $srch = $ctx->{srch};
 	my $mid = $node->{id};
-	my $x = $node->{smsg} || $srch->lookup_mail($mid);
-	my ($subj, $ts);
+	my $x = $node->{smsg} || $ctx->{-inbox}->smsg_by_mid($mid);
+	my ($subj, $ds);
 	my $topic;
 	if ($x) {
 		$subj = $x->subject;
 		$subj = $srch->subject_normalized($subj);
-		$ts = $x->ts;
+		$ds = $x->ds;
 		if ($level == 0) {
-			$topic = [ $ts, 1, { $subj => $mid }, $subj ];
+			$topic = [ $ds, 1, { $subj => $mid }, $subj ];
 			$ctx->{-cur_topic} = $topic;
 			push @{$ctx->{order}}, $topic;
 			return;
 		}
 
 		$topic = $ctx->{-cur_topic}; # should never be undef
-		$topic->[0] = $ts if $ts > $topic->[0];
+		$topic->[0] = $ds if $ds > $topic->[0];
 		$topic->[1]++;
 		my $seen = $topic->[2];
 		if (scalar(@$topic) == 3) { # parent was a ghost
@@ -897,23 +996,27 @@ sub acc_topic {
 
 sub dump_topics {
 	my ($ctx) = @_;
-	my $order = delete $ctx->{order}; # [ ts, subj1, subj2, subj3, ... ]
+	my $order = delete $ctx->{order}; # [ ds, subj1, subj2, subj3, ... ]
 	if (!@$order) {
 		$ctx->{-html_tip} = '
[No topics in range]
'; return 404; } my @out; + my $ibx = $ctx->{-inbox}; + my $obfs_ibx = $ibx->{obfuscate} ? $ibx : undef; + my $srch = $ctx->{srch}; # sort by recency, this allows new posts to "bump" old topics... foreach my $topic (sort { $b->[0] <=> $a->[0] } @$order) { - my ($ts, $n, $seen, $top, @ex) = @$topic; + my ($ds, $n, $seen, $top, @ex) = @$topic; @$topic = (); next unless defined $top; # ghost topic my $mid = delete $seen->{$top}; my $href = mid_escape($mid); + my $prev_subj = [ split(/ /, $top) ]; $top = PublicInbox::Hval->new($top)->as_html; - $ts = fmt_ts($ts); + $ds = fmt_ts($ds); # $n isn't the total number of posts on the topic, # just the number of posts in the current results window @@ -929,15 +1032,20 @@ sub dump_topics { my $mbox = qq(mbox.gz); my $atom = qq(Atom); my $s = "$top\n" . - " $ts UTC $n - $mbox / $atom\n"; + " $ds UTC $n - $mbox / $atom\n"; for (my $i = 0; $i < scalar(@ex); $i += 2) { my $level = $ex[$i]; - my $sub = $ex[$i + 1]; - $mid = delete $seen->{$sub}; - $sub = PublicInbox::Hval->new($sub)->as_html; + my $subj = $ex[$i + 1]; + $mid = delete $seen->{$subj}; + my @subj = split(/ /, $srch->subject_normalized($subj)); + my @next_prev = @subj; # full copy + my $omit = dedupe_subject($prev_subj, \@subj, ' "'); + $prev_subj = \@next_prev; + $subj = ascii_html(join(' ', @subj)); + obfuscate_addrs($obfs_ibx, $subj) if $obfs_ibx; $href = mid_escape($mid); $s .= indent_for($level) . TCHILD; - $s .= "$sub\n"; + $s .= "$subj$omit\n"; } push @out, $s; } @@ -945,45 +1053,81 @@ sub dump_topics { 200; } +sub ts2str ($) { + my ($ts) = @_; + POSIX::strftime('%Y%m%d%H%M%S', gmtime($ts)); +} + +sub str2ts ($) { + my ($yyyy, $mon, $dd, $hh, $mm, $ss) = unpack('A4A2A2A2A2A2', $_[0]); + timegm($ss, $mm, $hh, $dd, $mon - 1, $yyyy); +} + +sub pagination_footer ($$) { + my ($ctx, $latest) = @_; + delete $ctx->{qp} or return; + my $next = $ctx->{next_page} || ''; + my $prev = $ctx->{prev_page} || ''; + if ($prev) { + $next = $next ? "$next " : ' '; + $prev .= qq! latest!; + } + "
page: $next$prev
"; +} + sub index_nav { # callback for WwwStream my (undef, $ctx) = @_; - delete $ctx->{qp} or return; - my ($next, $prev); - $next = $prev = ' '; - my $latest = ''; - - my $next_o = $ctx->{-next_o}; - if ($next_o) { - $next = qq!next!; - } - if (my $cur_o = $ctx->{-cur_o}) { - $latest = qq! latest!; - - my $o = $cur_o - ($next_o - $cur_o); - if ($o > 0) { - $prev = qq!prev!; - } elsif ($o == 0) { - $prev = qq!prev!; + pagination_footer($ctx, '.') +} + +sub paginate_recent ($$) { + my ($ctx, $lim) = @_; + my $t = $ctx->{qp}->{t} || ''; + my $opts = { limit => $lim }; + my ($after, $before); + + # Xapian uses '..' but '-' is perhaps friendier to URL linkifiers + # if only $after exists "YYYYMMDD.." because "." could be skipped + # if interpreted as an end-of-sentence + $t =~ s/\A(\d{8,14})-// and $after = str2ts($1); + $t =~ /\A(\d{8,14})\z/ and $before = str2ts($1); + + my $ibx = $ctx->{-inbox}; + my $msgs = $ibx->recent($opts, $after, $before); + my $nr = scalar @$msgs; + if ($nr < $lim && defined($after)) { + $after = $before = undef; + $msgs = $ibx->recent($opts); + $nr = scalar @$msgs; + } + my $more = $nr == $lim; + my ($newest, $oldest); + if ($nr) { + $newest = $msgs->[0]->{ts}; + $oldest = $msgs->[-1]->{ts}; + # if we only had $after, our SQL query in ->recent ordered + if ($newest < $oldest) { + ($oldest, $newest) = ($newest, $oldest); + $more = 0 if defined($after) && $after < $oldest; } } - "
page: $next $prev$latest
"; + if (defined($oldest) && $more) { + my $s = ts2str($oldest); + $ctx->{next_page} = qq!next!; + } + if (defined($newest) && (defined($before) || defined($after))) { + my $s = ts2str($newest); + $ctx->{prev_page} = qq!prev!; + } + $msgs; } sub index_topics { my ($ctx) = @_; - my ($off) = (($ctx->{qp}->{o} || '0') =~ /(\d+)/); - my $opts = { offset => $off, limit => 200 }; - - $ctx->{order} = []; - my $srch = $ctx->{srch}; - my $sres = $srch->query('', $opts); - my $nr = scalar @{$sres->{msgs}}; - if ($nr) { - $sres = load_results($srch, $sres); - walk_thread(thread_results($sres), $ctx, *acc_topic); + my $msgs = paginate_recent($ctx, 200); # 200 is our window + if (@$msgs) { + walk_thread(thread_results($ctx, $msgs), $ctx, *acc_topic); } - $ctx->{-next_o} = $off+ $nr; - $ctx->{-cur_o} = $off; PublicInbox::WwwStream->response($ctx, dump_topics($ctx), *index_nav); }