^ ".
@@ -288,7 +316,10 @@ sub _th_index_lite {
my $nr_s = 0;
my $siblings;
if (my $smsg = $node->{smsg}) {
- ($$irt) = (($smsg->{references} || '') =~ m/<([^>]+)>\z/);
+ # delete saves about 200KB on a 1K message thread
+ if (my $refs = delete $smsg->{references}) {
+ ($$irt) = ($refs =~ m/<([^>]+)>\z/);
+ }
}
my $irt_map = $mapping->{$$irt} if defined $$irt;
if (defined $irt_map) {
@@ -346,7 +377,7 @@ sub walk_thread {
while (@q) {
my ($level, $node, $i) = splice(@q, 0, 3);
defined $node or next;
- $cb->($ctx, $level, $node, $i);
+ $cb->($ctx, $level, $node, $i) or return;
++$level;
$i = 0;
unshift @q, map { ($level, $_, $i++) } @{$node->{children}};
@@ -408,9 +439,7 @@ 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 .= $nr == 1 ? 'only message in thread' : 'end of thread';
@@ -461,11 +490,11 @@ sub thread_html {
}
sub multipart_text_as_html {
- my ($mime, $upfx, $obfs_ibx) = @_;
+ my ($mime, $upfx, $ibx) = @_;
my $rv = "";
# scan through all parts, looking for displayable text
- msg_iter($mime, sub { $rv .= add_text_body($upfx, $obfs_ibx, $_[0]) });
+ msg_iter($mime, sub { $rv .= add_text_body($upfx, $ibx, $_[0]) });
$rv;
}
@@ -518,38 +547,39 @@ sub attach_link ($$$$;$) {
}
sub add_text_body {
- my ($upfx, $obfs_ibx, $p) = @_;
+ my ($upfx, $ibx, $p) = @_;
+ my $obfs_ibx = $ibx->{obfuscate} ? $ibx : undef;
# $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;
-
- if ($ct =~ m!\btext/x?html\b!i) {
- return attach_link($upfx, $ct, $p, $fn);
- }
-
- my $s = eval { $part->body_str };
-
- # badly-encoded message? tell the world about it!
- my $err = $@;
- if ($err) {
- if ($ct =~ m!\btext/plain\b!i) {
- # Try to assume UTF-8 because Alpine seems to
- # do wacky things and set charset=X-UNKNOWN
- $part->charset_set('UTF-8');
- $s = eval { $part->body_str };
-
- # If forcing charset=UTF-8 failed,
- # attach_link will warn further down...
- $s = $part->body if $@;
- } else {
- return attach_link($upfx, $ct, $p, $fn);
+ my ($s, $err) = msg_part_text($part, $ct);
+
+ return attach_link($upfx, $ct, $p, $fn) unless defined $s;
+
+ # makes no difference to browsers, and don't screw up filename
+ # link generation in diffs with the extra '%0D'
+ $s =~ s/\r\n/\n/sg;
+
+ my ($diff, $spfx);
+ if ($s =~ /^(?:diff|---|\+{3}) /ms) {
+ $diff = [];
+ if ($ibx->{-repo_objs}) {
+ my $n_slash = $upfx =~ tr!/!/!;
+ if ($n_slash == 0) {
+ $spfx = '../';
+ } elsif ($n_slash == 1) {
+ $spfx = '';
+ } else { # nslash == 2
+ $spfx = '../../';
+ }
}
- }
+ };
my @lines = split(/^/m, $s);
$s = '';
if (defined($fn) || $depth > 0 || $err) {
+ # badly-encoded message with $err? tell the world about it!
$s .= attach_link($upfx, $ct, $p, $fn, $err);
$s .= "\n";
}
@@ -560,19 +590,26 @@ sub add_text_body {
# show the previously buffered quote inline
flush_quote(\$s, $l, \@quot) if @quot;
- # regular line, OK
- $l->linkify_1($cur);
- $s .= $l->linkify_2(ascii_html($cur));
+ if ($diff) {
+ push @$diff, $cur;
+ } else {
+ # regular line, OK
+ $l->linkify_1($cur);
+ $s .= $l->linkify_2(ascii_html($cur));
+ }
} else {
+ flush_diff(\$s, $spfx, $l, $diff) if $diff && @$diff;
push @quot, $cur;
}
}
if (@quot) { # ugh, top posted
flush_quote(\$s, $l, \@quot);
+ flush_diff(\$s, $spfx, $l, $diff) if $diff && @$diff;
obfuscate_addrs($obfs_ibx, $s) if $obfs_ibx;
$s;
} else {
+ flush_diff(\$s, $spfx, $l, $diff) if $diff && @$diff;
obfuscate_addrs($obfs_ibx, $s) if $obfs_ibx;
if ($s =~ /\n\z/s) { # common, last line ends with a newline
$s;
@@ -603,29 +640,43 @@ sub _msg_html_prepare {
$ctx->{-upfx} = '../';
}
my @title;
- 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));
- obfuscate_addrs($obfs_ibx, $title[1]) if $obfs_ibx;
- } 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;
+ 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";
-
+ $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);
foreach (@$mids) {
my $mid = PublicInbox::Hval->new_msgid($_) ;
@@ -649,8 +700,7 @@ sub thread_skel {
my ($dst, $ctx, $hdr, $tpfx) = @_;
my $srch = $ctx->{srch};
my $mid = mids($hdr)->[0];
- my $sres = $srch->get_thread($mid);
- my $nr = $sres->{total};
+ my ($nr, $msgs) = $srch->get_thread($mid);
my $expand = qq(expand[flat) .
qq(|nested] ) .
qq(mbox.gz ) .
@@ -680,12 +730,11 @@ sub thread_skel {
$ctx->{prev_attr} = '';
$ctx->{prev_level} = 0;
$ctx->{dst} = $dst;
- $sres = load_results($srch, $sres);
# reduce hash lookups in skel_dump
my $ibx = $ctx->{-inbox};
$ctx->{-obfs_ibx} = $ibx->{obfuscate} ? $ibx : undef;
- walk_thread(thread_results($ctx, $sres), $ctx, *skel_dump);
+ walk_thread(thread_results($ctx, $msgs), $ctx, *skel_dump);
$ctx->{parent_msg} = $parent;
}
@@ -694,7 +743,8 @@ 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;
@@ -707,22 +757,9 @@ 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;
}
@@ -801,16 +838,56 @@ sub indent_for {
$level ? INDENT x ($level - 1) : '';
}
-sub load_results {
- my ($srch, $sres) = @_;
- my $msgs = delete $sres->{msgs};
- $srch->retry_reopen(sub { [ map { $_->mid; $_ } @$msgs ] });
+sub find_mid_root {
+ my ($ctx, $level, $node, $idx) = @_;
+ ++$ctx->{root_idx} if $level == 0;
+ if ($node->{id} eq $ctx->{mid}) {
+ $ctx->{found_mid_at} = $ctx->{root_idx};
+ return 0;
+ }
+ 1;
+}
+
+sub strict_loose_note ($) {
+ my ($nr) = @_;
+ my $msg =
+" -- strict thread matches above, loose matches on Subject: below --\n";
+
+ if ($nr > PublicInbox::Over::DEFAULT_LIMIT()) {
+ $msg .=
+" -- use mbox.gz link to download all $nr messages --\n";
+ }
+ $msg;
}
sub thread_results {
my ($ctx, $msgs) = @_;
require PublicInbox::SearchThread;
- PublicInbox::SearchThread::thread($msgs, *sort_ds, $ctx->{-inbox});
+ my $ibx = $ctx->{-inbox};
+ my $rootset = PublicInbox::SearchThread::thread($msgs, *sort_ds, $ibx);
+
+ # FIXME: `tid' is broken on --reindex, so that needs to be fixed
+ # and preserved in the future. This bug is hidden by `sid' matches
+ # in get_thread, so we never noticed it until now. And even when
+ # reindexing is fixed, we'll keep this code until a SCHEMA_VERSION
+ # bump since reindexing is expensive and users may not do it
+
+ # loose threading could've returned too many results,
+ # put the root the message we care about at the top:
+ my $mid = $ctx->{mid};
+ if (defined($mid) && scalar(@$rootset) > 1) {
+ $ctx->{root_idx} = -1;
+ my $nr = scalar @$msgs;
+ walk_thread($rootset, $ctx, *find_mid_root);
+ my $idx = $ctx->{found_mid_at};
+ if (defined($idx) && $idx != 0) {
+ my $tip = splice(@$rootset, $idx, 1);
+ @$rootset = reverse @$rootset;
+ unshift @$rootset, $tip;
+ $ctx->{sl_note} = strict_loose_note($nr);
+ }
+ }
+ $rootset
}
sub missing_thread {
@@ -819,11 +896,6 @@ sub missing_thread {
PublicInbox::ExtMsg::ext_msg($ctx);
}
-sub _msg_date {
- my ($hdr) = @_;
- fmt_ts(msg_datestamp($hdr));
-}
-
sub fmt_ts { POSIX::strftime('%Y-%m-%d %k:%M', gmtime($_[0])) }
sub dedupe_subject {
@@ -853,6 +925,10 @@ sub skel_dump {
my $cur = $ctx->{cur};
my $mid = $smsg->{mid};
+ if ($level == 0 && $ctx->{skel_dump_roots}++) {
+ $$dst .= delete $ctx->{sl_note} || '';
+ }
+
my $f = ascii_html($smsg->from_name);
my $obfs_ibx = $ctx->{-obfs_ibx};
obfuscate_addrs($obfs_ibx, $f) if $obfs_ibx;
@@ -871,7 +947,7 @@ sub skel_dump {
delete $ctx->{cur};
$$dst .= "$d".
"$attr [this message]\n";
- return;
+ return 1;
} else {
$ctx->{prev_msg} = $mid;
}
@@ -911,6 +987,7 @@ sub skel_dump {
$m = $ctx->{-upfx}.mid_escape($mid).'/';
}
$$dst .= $d . "" . $end;
+ 1;
}
sub _skel_ghost {
@@ -936,6 +1013,7 @@ sub _skel_ghost {
}
my $dst = $ctx->{dst};
$$dst .= $d;
+ 1;
}
sub sort_ds {
@@ -962,7 +1040,7 @@ sub acc_topic {
$topic = [ $ds, 1, { $subj => $mid }, $subj ];
$ctx->{-cur_topic} = $topic;
push @{$ctx->{order}}, $topic;
- return;
+ return 1;
}
$topic = $ctx->{-cur_topic}; # should never be undef
@@ -976,11 +1054,12 @@ sub acc_topic {
}
$seen->{$subj} = $mid; # latest for subject
} else { # ghost message
- return if $level != 0; # ignore child ghosts
+ return 1 if $level != 0; # ignore child ghosts
$topic = [ -666, 0, {} ];
$ctx->{-cur_topic} = $topic;
push @{$ctx->{order}}, $topic;
}
+ 1;
}
sub dump_topics {
@@ -1020,7 +1099,7 @@ sub dump_topics {
my $mbox = qq(mbox.gz);
my $atom = qq(Atom);
- my $s = "$top\n" .
+ my $s = "$top\n" .
" $ds UTC $n - $mbox / $atom\n";
for (my $i = 0; $i < scalar(@ex); $i += 2) {
my $level = $ex[$i];
@@ -1042,45 +1121,81 @@ sub dump_topics {
200;
}
+sub ts2str ($) {
+ my ($ts) = @_;
+ POSIX::strftime('%Y%m%d%H%M%S', gmtime($ts));
+}
+
+sub str2ts ($) {
+ my ($yyyy, $mon, $dd, $hh, $mm, $ss) = unpack('A4A2A2A2A2A2', $_[0]);
+ timegm($ss, $mm, $hh, $dd, $mon - 1, $yyyy);
+}
+
+sub pagination_footer ($$) {
+ 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 $latest = '';
+ pagination_footer($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 $next_o = $ctx->{-next_o};
- if ($next_o) {
- $next = qq!next!;
+ 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;
}
- 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!;
+ 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, $sres);
- walk_thread(thread_results($ctx, $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);
}