-# Copyright (C) 2014-2018 all contributors <meta@public-inbox.org>
+# Copyright (C) all contributors <meta@public-inbox.org>
# License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
#
# Used for displaying the HTML web interface.
# See Documentation/design_www.txt for this.
package PublicInbox::View;
use strict;
-use warnings;
+use v5.10.1;
+use List::Util qw(max);
+use Text::Wrap qw(wrap); # stdlib, we need Perl 5.6+ for $huge
use PublicInbox::MsgTime qw(msg_datestamp);
-use PublicInbox::Hval qw/ascii_html obfuscate_addrs/;
+use PublicInbox::Hval qw(ascii_html obfuscate_addrs prurl mid_href
+ ts2str fmt_ts);
use PublicInbox::Linkify;
-use PublicInbox::MID qw/id_compress mid_escape mids references/;
+use PublicInbox::MID qw(id_compress mids mids_for_index references
+ $MID_EXTRACT);
use PublicInbox::MsgIter;
use PublicInbox::Address;
-use PublicInbox::WwwStream;
+use PublicInbox::WwwStream qw(html_oneshot);
use PublicInbox::Reply;
-require POSIX;
-
+use PublicInbox::ViewDiff qw(flush_diff);
+use PublicInbox::Eml;
+use Time::Local qw(timegm);
+use PublicInbox::Smsg qw(subject_normalized);
+use PublicInbox::ContentHash qw(content_hash);
+use constant COLS => 72;
use constant INDENT => ' ';
use constant TCHILD => '` ';
sub th_pfx ($) { $_[0] == 0 ? '' : TCHILD };
-# public functions: (unstable)
-
-sub msg_html {
- my ($ctx, $mime, $more, $smsg) = @_;
- my $hdr = $mime->header_obj;
- 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) {
- # $more cannot be true w/o $smsg being defined:
- my $upfx = $more ? '../'.mid_escape($smsg->mid).'/' : '';
- $tip . multipart_text_as_html($mime, $upfx, $obfs_ibx) .
- '</pre><hr>'
- } elsif ($more && @$more) {
- ++$end;
- msg_html_more($ctx, $more, $nr);
- } elsif ($nr == $end) {
- # fake an EOF if generating the footer fails;
- # we want to at least show the message if something
- # here crashes:
- eval {
- '<pre>' . html_footer($hdr, 1, $ctx) .
- '</pre>' . msg_reply($ctx, $hdr)
- };
- } else {
- undef
+sub msg_page_i {
+ my ($ctx, $eml) = @_;
+ if ($eml) { # called by WwwStream::async_eml or getline
+ my $smsg = $ctx->{smsg};
+ my $over = $ctx->{ibx}->over;
+ $ctx->{smsg} = $over ? $over->next_by_mid(@{$ctx->{next_arg}})
+ : $ctx->gone('over');
+ $ctx->{mhref} = ($ctx->{nr} || $ctx->{smsg}) ?
+ "../${\mid_href($smsg->{mid})}/" : '';
+ my $obuf = $ctx->{obuf} = _msg_page_prepare_obuf($eml, $ctx);
+ if (length($$obuf)) {
+ multipart_text_as_html($eml, $ctx);
+ $$obuf .= '</pre><hr>';
}
- });
+ delete $ctx->{obuf};
+ $$obuf .= html_footer($ctx, $ctx->{first_hdr}) if !$ctx->{smsg};
+ $$obuf;
+ } else { # called by WwwStream::async_next or getline
+ $ctx->{smsg}; # may be undef
+ }
}
-sub msg_page {
+# /$INBOX/$MSGID/ for unindexed v1 inboxes
+sub no_over_html ($) {
my ($ctx) = @_;
- my $mid = $ctx->{mid};
- my $ibx = $ctx->{-inbox};
- my ($first, $more, $head, $tail, $db);
- my $smsg;
- if (my $srch = $ibx->search) {
- $srch->retry_reopen(sub {
- ($head, $tail, $db) = $srch->each_smsg_by_mid($mid);
- for (; !defined($first) && $head != $tail; $head++) {
- my @args = ($head, $db, $mid);
- $smsg = PublicInbox::SearchMsg->get(@args);
- $first = $ibx->msg_by_smsg($smsg);
- }
- if ($head != $tail) {
- $more = [ $head, $tail, $db ];
- }
- });
- return unless $first;
- } else {
- $first = $ibx->msg_by_mid($mid) or return;
+ my $bref = $ctx->{ibx}->msg_by_mid($ctx->{mid}) or return; # 404
+ my $eml = PublicInbox::Eml->new($bref);
+ $ctx->{mhref} = '';
+ PublicInbox::WwwStream::init($ctx);
+ my $obuf = $ctx->{obuf} = _msg_page_prepare_obuf($eml, $ctx);
+ if (length($$obuf)) {
+ multipart_text_as_html($eml, $ctx);
+ $$obuf .= '</pre><hr>';
}
- msg_html($ctx, PublicInbox::MIME->new($first), $more, $smsg);
+ delete $ctx->{obuf};
+ eval { $$obuf .= html_footer($ctx, $eml) };
+ html_oneshot($ctx, 200, $$obuf);
}
-sub msg_html_more {
- my ($ctx, $more, $nr) = @_;
- my $str = eval {
- my $smsg;
- my ($head, $tail, $db) = @$more;
- my $mid = $ctx->{mid};
- for (; !defined($smsg) && $head != $tail; $head++) {
- my $m = PublicInbox::SearchMsg->get($head, $db, $mid);
- $smsg = $ctx->{-inbox}->smsg_mime($m);
- }
- if ($head == $tail) { # done
- @$more = ();
- } else {
- $more->[0] = $head;
- }
- if ($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,
- $ctx->{-obfs_ibx}) .
- '</pre><hr>'
- } else {
- '';
- }
- };
- if ($@) {
- warn "Error lookup up additional messages: $@\n";
- $str = '<pre>Error looking up additional messages</pre>';
- }
- $str;
+# public functions: (unstable)
+
+sub msg_page {
+ my ($ctx) = @_;
+ my $ibx = $ctx->{ibx};
+ $ctx->{-obfs_ibx} = $ibx->{obfuscate} ? $ibx : undef;
+ my $over = $ibx->over or return no_over_html($ctx);
+ my ($id, $prev);
+ my $next_arg = $ctx->{next_arg} = [ $ctx->{mid}, \$id, \$prev ];
+
+ my $smsg = $ctx->{smsg} = $over->next_by_mid(@$next_arg) or
+ return; # undef == 404
+
+ # allow user to easily browse the range around this message if
+ # they have ->over
+ $ctx->{-t_max} = $smsg->{ts};
+ PublicInbox::WwwStream::aresponse($ctx, 200, \&msg_page_i);
}
# /$INBOX/$MESSAGE_ID/#R
-sub msg_reply {
+sub msg_reply ($$) {
my ($ctx, $hdr) = @_;
my $se_url =
'https://kernel.org/pub/software/scm/git/docs/git-send-email.html';
'https://en.wikipedia.org/wiki/Posting_style#Interleaved_style';
my $info = '';
- my $ibx = $ctx->{-inbox};
+ my $ibx = $ctx->{ibx};
if (my $url = $ibx->{infourl}) {
- $url = PublicInbox::Hval::prurl($ctx->{env}, $url);
+ $url = prurl($ctx->{env}, $url);
$info = qq(\n List information: <a\nhref="$url">$url</a>\n);
}
my ($arg, $link, $reply_to_all) =
PublicInbox::Reply::mailto_arg_link($ibx, $hdr);
+ if (ref($arg) eq 'SCALAR') {
+ return '<pre id=R>'.ascii_html($$arg).'</pre>';
+ }
# mailto: link only works if address obfuscation is disabled
if ($link) {
<hr><pre
id=R><b>Reply instructions:</b>
-You may reply publically to <a
+You may reply publicly to <a
href=#t>this message</a> via plain-text email
using any one of the following methods:
<a
href="$se_url">$se_url</a>
$link</pre>
+
+ Be sure your reply has a <b>Subject:</b> header at the top and a blank line
+ before the message body.
EOF
}
$refs->[-1];
}
+sub fold_addresses ($) {
+ return $_[0] if length($_[0]) <= COLS;
+ # try to fold on commas after non-word chars before $lim chars,
+ # Try to get the "," 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
+ # 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 ($$) {
my ($hdr, $field) = @_;
- my $val = $hdr->header($field) or return '';
- ascii_html(join(', ', PublicInbox::Address::names($val)));
+ my @vals = $hdr->header($field) or return '';
+ ascii_html(join(', ', PublicInbox::Address::names(join(',', @vals))));
}
sub nr_to_s ($$$) {
$nr == 1 ? "$nr $singular" : "$nr $plural";
}
+# Displays the text of of the message for /$INBOX/$MSGID/[Tt]/ endpoint
# this is already inside a <pre>
-sub index_entry {
- my ($smsg, $ctx, $more) = @_;
- my $srch = $ctx->{srch};
- my $subj = $smsg->subject;
- my $mid_raw = $smsg->mid;
+sub eml_entry {
+ my ($ctx, $eml) = @_;
+ my $smsg = delete $ctx->{smsg};
+ my $subj = delete $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};
+ $subj = '(no subject)' if $subj eq '';
my $rv = "<a\nhref=#e$id\nid=m$id>*</a> ";
$subj = '<b>'.ascii_html($subj).'</b>';
obfuscate_addrs($obfs_ibx, $subj) if $obfs_ibx;
$rv .= $subj . "\n";
$rv .= _th_index_lite($mid_raw, \$irt, $id, $ctx);
my @tocc;
- my $mime = $smsg->{mime};
- my $hdr = $mime->header_obj;
- foreach my $f (qw(To Cc)) {
- my $dst = _hdr_names_html($hdr, $f);
- if ($dst ne '') {
- obfuscate_addrs($obfs_ibx, $dst) if $obfs_ibx;
- push @tocc, "$f: $dst";
- }
- }
- my $from = _hdr_names_html($hdr, 'From');
+ my $ds = delete $smsg->{ds}; # for v1 non-Xapian/SQLite users
+
+ # Deleting these fields saves about 400K as we iterate across 1K msgs
+ delete @$smsg{qw(ts blob)};
+
+ my $from = _hdr_names_html($eml, 'From');
obfuscate_addrs($obfs_ibx, $from) if $obfs_ibx;
- $rv .= "From: $from @ ".fmt_ts($smsg->ds)." UTC";
+ $rv .= "From: $from @ ".fmt_ts($ds)." UTC";
my $upfx = $ctx->{-upfx};
- my $mhref = $upfx . mid_escape($mid_raw) . '/';
+ my $mhref = $upfx . mid_href($mid_raw) . '/';
$rv .= qq{ (<a\nhref="$mhref">permalink</a> / };
$rv .= qq{<a\nhref="${mhref}raw">raw</a>)\n};
- $rv .= ' '.join('; +', @tocc) . "\n" if @tocc;
+ my $to = fold_addresses(_hdr_names_html($eml, 'To'));
+ my $cc = fold_addresses(_hdr_names_html($eml, '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;
+ $to_cc .= '; <b>+Cc:</b> '.$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)))) {
- my $mirt = PublicInbox::Hval->new_msgid($irt);
- my $href = $upfx . $mirt->{href}. '/';
- my $html = $mirt->as_html;
+ if (!$mapping && (defined($irt) || defined($irt = in_reply_to($eml)))) {
+ my $href = $upfx . mid_href($irt) . '/';
+ my $html = ascii_html($irt);
$rv .= qq(In-Reply-To: <<a\nhref="$href">$html</a>>\n)
}
$rv .= "\n";
# scan through all parts, looking for displayable text
- msg_iter($mime, sub { $rv .= add_text_body($mhref, $obfs_ibx, $_[0]) });
+ $ctx->{mhref} = $mhref;
+ $ctx->{end_id} = "e$id";
+ $ctx->{obuf} = \$rv;
+ $eml->each_part(\&add_text_body, $ctx, 1);
+ delete $ctx->{obuf};
# add the footer
$rv .= "\n<a\nhref=#$id_m\nid=e$id>^</a> ".
" <a\nhref=\"${mhref}raw\">raw</a>" .
" <a\nhref=\"${mhref}#R\">reply</a>";
+ delete($ctx->{-qry}) and
+ $rv .= qq[ <a\nhref="${mhref}#related">related</a>];
+
my $hr;
- if (my $pct = $ctx->{pct}) { # used by SearchView.pm
- $rv .= "\t[relevance $pct->{$mid_raw}%]";
+ if (defined(my $pct = $smsg->{pct})) { # used by SearchView.pm
+ $rv .= "\t[relevance $pct%]";
$hr = 1;
} elsif ($mapping) {
my $nested = 'nested';
my $flat = 'flat';
- my $end = '';
if ($ctx->{flat}) {
$hr = 1;
$flat = "<b>$flat</b>";
$hr = $ctx->{-hr};
}
- $rv .= $more ? '</pre><hr><pre>' : '</pre>' if $hr;
+ # do we have more messages? start a new <pre> if so
+ $rv .= scalar(@{$ctx->{msgs}}) ? '</pre><hr><pre>' : '</pre>' if $hr;
$rv;
}
sub pad_link ($$;$) {
my ($mid, $level, $s) = @_;
$s ||= '...';
- my $id = id_compress($mid, 1);
- (' 'x19).indent_for($level).th_pfx($level)."<a\nhref=#r$id>($s)</a>\n";
+ my $href = defined($mid) ?
+ ("<a\nhref=#r".id_compress($mid, 1).">($s)</a>\n") :
+ "($s)\n";
+ (' 'x19).indent_for($level).th_pfx($level).$href;
+}
+
+sub _skel_hdr {
+ # my ($mapping, $mid) = @_;
+ ($_[0]->{$_[1] // \'bogus'} // [ "(?)\n" ])->[0];
}
sub _th_index_lite {
my $nr_c = scalar @$children;
my $nr_s = 0;
my $siblings;
- if (my $smsg = $node->{smsg}) {
- ($$irt) = (($smsg->{references} || '') =~ m/<([^>]+)>\z/);
+ # delete saves about 200KB on a 1K message thread
+ if (my $refs = delete $node->{references}) {
+ ($$irt) = ($refs =~ m/$MID_EXTRACT\z/o);
}
my $irt_map = $mapping->{$$irt} if defined $$irt;
if (defined $irt_map) {
$rv .= $pad . $irt_map->[0];
if ($idx > 0) {
my $prev = $siblings->[$idx - 1];
- my $pmid = $prev->{id};
+ my $pmid = $prev->{mid};
if ($idx > 2) {
my $s = ($idx - 1). ' preceding siblings ...';
$rv .= pad_link($pmid, $level, $s);
} elsif ($idx == 2) {
- my $ppmid = $siblings->[0]->{id};
- $rv .= $pad . $mapping->{$ppmid}->[0];
+ $rv .= $pad . _skel_hdr($mapping,
+ $siblings->[0] ?
+ $siblings->[0]->{mid} : undef);
}
- $rv .= $pad . $mapping->{$pmid}->[0];
+ $rv .= $pad . _skel_hdr($mapping, $pmid);
}
}
my $s_s = nr_to_s($nr_s, 'sibling', 'siblings');
my $s_c = nr_to_s($nr_c, 'reply', 'replies');
$attr =~ s!\n\z!</b>\n!s;
- $attr =~ s!<a\nhref.*</a> !!s; # no point in duplicating subject
+ $attr =~ s!<a\nhref.*</a> (?:" )?!!s; # no point in dup subject
$attr =~ s!<a\nhref=[^>]+>([^<]+)</a>!$1!s; # no point linking to self
$rv .= "<b>@ $attr";
if ($nr_c) {
- my $cmid = $children->[0]->{id};
- $rv .= $pad . $mapping->{$cmid}->[0];
+ my $cmid = $children->[0] ? $children->[0]->{mid} : undef;
+ $rv .= $pad . _skel_hdr($mapping, $cmid);
if ($nr_c > 2) {
my $s = ($nr_c - 1). ' more replies';
$rv .= pad_link($cmid, $level + 1, $s);
} elsif (my $cn = $children->[1]) {
- $rv .= $pad . $mapping->{$cn->{id}}->[0];
+ $rv .= $pad . _skel_hdr($mapping, $cn->{mid});
}
}
my $next = $siblings->[$idx+1] if $siblings && $idx >= 0;
if ($next) {
- my $nmid = $next->{id};
- $rv .= $pad . $mapping->{$nmid}->[0];
+ my $nmid = $next->{mid};
+ $rv .= $pad . _skel_hdr($mapping, $nmid);
my $nnext = $nr_s - $idx;
if ($nnext > 2) {
my $s = ($nnext - 1).' subsequent siblings';
$rv .= pad_link($nmid, $level, $s);
} elsif (my $nn = $siblings->[$idx + 2]) {
- $rv .= $pad . $mapping->{$nn->{id}}->[0];
+ $rv .= $pad . _skel_hdr($mapping, $nn->{mid});
}
}
$rv .= $pad ."<a\nhref=#r$id>$s_s, $s_c; $ctx->{s_nr}</a>\n";
}
-sub walk_thread {
+# non-recursive thread walker
+sub walk_thread ($$$) {
my ($rootset, $ctx, $cb) = @_;
my @q = map { (0, $_, -1) } @$rootset;
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}};
}
}
-sub pre_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, $node);
}
-sub thread_index_entry {
- my ($ctx, $level, $smsg) = @_;
- my ($beg, $end) = thread_adj_level($ctx, $level);
- $beg . '<pre>' . index_entry($smsg, $ctx, 0) . '</pre>' . $end;
+sub thread_eml_entry {
+ my ($ctx, $eml) = @_;
+ my ($beg, $end) = thread_adj_level($ctx, $ctx->{level});
+ $beg . '<pre>' . eml_entry($ctx, $eml) . '</pre>' . $end;
}
-sub stream_thread ($$) {
- my ($rootset, $ctx) = @_;
- my $inbox = $ctx->{-inbox};
- my @q = map { (0, $_) } @$rootset;
- my $level;
- my $smsg;
- while (@q) {
- $level = shift @q;
- my $node = shift @q or next;
+sub next_in_queue ($$) {
+ my ($q, $ghost_ok) = @_;
+ while (@$q) {
+ my ($level, $smsg) = splice(@$q, 0, 2);
my $cl = $level + 1;
- unshift @q, map { ($cl, $_) } @{$node->{children}};
- $smsg = $inbox->smsg_mime($node->{smsg}) and last;
+ unshift @$q, map { ($cl, $_) } @{$smsg->{children}};
+ return ($level, $smsg) if $ghost_ok || exists($smsg->{blob});
}
- return missing_thread($ctx) unless $smsg;
-
- $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) {
- $level = shift @q;
- my $node = shift @q or next;
- my $cl = $level + 1;
- unshift @q, map { ($cl, $_) } @{$node->{children}};
- if ($smsg = $inbox->smsg_mime($node->{smsg})) {
- return thread_index_entry($ctx, $level, $smsg);
- } else {
- return ghost_index_entry($ctx, $level, $node);
+ undef;
+}
+
+sub stream_thread_i { # PublicInbox::WwwStream::getline callback
+ my ($ctx, $eml) = @_;
+ return thread_eml_entry($ctx, $eml) if $eml;
+ return unless exists($ctx->{skel});
+ my $ghost_ok = $ctx->{nr}++;
+ while (1) {
+ my ($lvl, $smsg) = next_in_queue($ctx->{-queue}, $ghost_ok);
+ if ($smsg) {
+ if (exists $smsg->{blob}) { # next message for cat-file
+ $ctx->{level} = $lvl;
+ if (!$ghost_ok) { # first non-ghost
+ $ctx->{-title_html} =
+ ascii_html($smsg->{subject});
+ $ctx->zmore($ctx->html_top);
+ }
+ return $smsg;
}
+ # buffer the ghost entry and loop
+ $ctx->zmore(ghost_index_entry($ctx, $lvl, $smsg));
+ } else { # all done
+ $ctx->zmore(join('', thread_adj_level($ctx, 0)));
+ $ctx->zmore(${delete($ctx->{skel})});
+ return;
}
- my $ret = join('', thread_adj_level($ctx, 0));
- $ret .= ${$ctx->{dst}}; # skel
- $ctx = undef;
- $ret;
- });
+ }
}
+sub stream_thread ($$) {
+ my ($rootset, $ctx) = @_;
+ @{$ctx->{-queue}} = map { (0, $_) } @$rootset;
+ PublicInbox::WwwStream::aresponse($ctx, 200, \&stream_thread_i);
+}
+
+# /$INBOX/$MSGID/t/ and /$INBOX/$MSGID/T/
sub thread_html {
my ($ctx) = @_;
+ $ctx->{-upfx} = '../../';
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 $ibx = $ctx->{ibx};
+ my ($nr, $msgs) = $ibx->over->get_thread($mid);
return missing_thread($ctx) if $nr == 0;
+
+ # link $INBOX_DIR/description text to "index_topics" view around
+ # the newest message in this thread
+ my $t = ts2str($ctx->{-t_max} = max(map { delete $_->{ts} } @$msgs));
+ my $t_fmt = fmt_ts($ctx->{-t_max});
+
my $skel = '<hr><pre>';
$skel .= $nr == 1 ? 'only message in thread' : 'end of thread';
- $skel .= ", back to <a\nhref=\"../../\">index</a>\n\n";
+ $skel .= <<EOF;
+, other threads:[<a
+href="../../?t=$t">~$t_fmt UTC</a> | <a
+href="../../">newest</a>]
+
+EOF
$skel .= "<b\nid=t>Thread overview:</b> ";
$skel .= $nr == 1 ? '(only message)' : "$nr+ messages";
$skel .= " (download: <a\nhref=\"../t.mbox.gz\">mbox.gz</a>";
$skel .= " / follow: <a\nhref=\"../t.atom\">Atom feed</a>)\n";
$skel .= "-- links below jump to the message on this page --\n";
- $ctx->{-upfx} = '../../';
$ctx->{cur_level} = 0;
- $ctx->{dst} = \$skel;
+ $ctx->{skel} = \$skel;
$ctx->{prev_attr} = '';
$ctx->{prev_level} = 0;
- $ctx->{root_anchor} = anchor_for($mid);
- $ctx->{mapping} = {};
- $ctx->{s_nr} = "$nr+ messages in thread";
+ $ctx->{root_anchor} = 'm' . id_compress($mid, 1);
+ $ctx->{mapping} = {}; # mid -> [ header_summary, node, idx, level ]
+ $ctx->{s_nr} = ($nr > 1 ? "$nr+ messages" : 'only message')
+ .' in thread';
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);
+ $ctx->{-obfs_ibx} = $ibx->{obfuscate} ? $ibx : undef;
+ walk_thread($rootset, $ctx, \&pre_thread);
$skel .= '</pre>';
return stream_thread($rootset, $ctx) unless $ctx->{flat};
# flat display: lazy load the full message from smsg
- 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} = '<pre>'.index_entry($smsg, $ctx, scalar @$msgs);
- $smsg = undef;
- PublicInbox::WwwStream->response($ctx, 200, sub {
- return unless $msgs;
- $smsg = undef;
- while (my $m = shift @$msgs) {
- $smsg = $inbox->smsg_mime($m) and last;
+ $ctx->{msgs} = $msgs;
+ $ctx->{-html_tip} = '<pre>';
+ PublicInbox::WwwStream::aresponse($ctx, 200, \&thread_html_i);
+}
+
+sub thread_html_i { # PublicInbox::WwwStream::getline callback
+ my ($ctx, $eml) = @_;
+ if ($eml) {
+ my $smsg = $ctx->{smsg};
+ if (exists $ctx->{-html_tip}) {
+ $ctx->{-title_html} = ascii_html($smsg->{subject});
+ $ctx->zmore($ctx->html_top);
+ }
+ return eml_entry($ctx, $eml);
+ } else {
+ while (my $smsg = shift @{$ctx->{msgs}}) {
+ return $smsg if exists($smsg->{blob});
}
- return index_entry($smsg, $ctx, scalar @$msgs) if $smsg;
- $msgs = undef;
- $skel;
- });
+ my $skel = delete($ctx->{skel}) or return; # all done
+ $ctx->zmore($$skel);
+ undef;
+ }
}
sub multipart_text_as_html {
- my ($mime, $upfx, $obfs_ibx) = @_;
- my $rv = "";
+ # ($mime, $ctx) = @_; # each_part may do "$_[0] = undef"
# scan through all parts, looking for displayable text
- msg_iter($mime, sub { $rv .= add_text_body($upfx, $obfs_ibx, $_[0]) });
- $rv;
+ $_[0]->each_part(\&add_text_body, $_[1], 1);
}
-sub flush_quote {
- my ($s, $l, $quot) = @_;
-
- # show everything in the full version with anchor from
- # short version (see above)
- my $rv = $l->linkify_1(join('', @$quot));
- @$quot = ();
-
- # we use a <span> here to allow users to specify their own
- # color for quoted text
- $rv = $l->linkify_2(ascii_html($rv));
- $$s .= qq(<span\nclass="q">) . $rv . '</span>'
+sub submsg_hdr ($$) {
+ my ($ctx, $eml) = @_;
+ my $obfs_ibx = $ctx->{-obfs_ibx};
+ my $rv = $ctx->{obuf};
+ $$rv .= "\n";
+ for my $h (qw(From To Cc Subject Date Message-ID X-Alt-Message-ID)) {
+ my @v = $eml->header($h);
+ for my $v (@v) {
+ obfuscate_addrs($obfs_ibx, $v) if $obfs_ibx;
+ $v = ascii_html($v);
+ $$rv .= "$h: $v\n";
+ }
+ }
}
sub attach_link ($$$$;$) {
- my ($upfx, $ct, $p, $fn, $err) = @_;
- my ($part, $depth, @idx) = @$p;
- my $nl = $idx[-1] > 1 ? "\n" : '';
- my $idx = join('.', @idx);
- my $size = bytes::length($part->body);
+ my ($ctx, $ct, $p, $fn, $err) = @_;
+ my ($part, $depth, $idx) = @$p;
+
+ # Eml iteration clobbers multipart ->{bdy}, so do not offer
+ # downloads for 0-byte multipart attachments
+ return unless $part->{bdy};
+
+ my $nl = $idx eq '1' ? '' : "\n"; # like join("\n", ...)
+ my $size = length($part->body);
+ delete $part->{bdy}; # save memory
# hide attributes normally, unless we want to aid users in
# spotting MUA problems:
$ct =~ s/;.*// unless $err;
$ct = ascii_html($ct);
- my $desc = $part->header('Content-Description');
- $desc = $fn unless defined $desc;
- $desc = '' unless defined $desc;
my $sfn;
- if (defined $fn && $fn =~ /\A[[:alnum:]][\w\.-]+[[:alnum:]]\z/) {
+ if (defined $fn && $fn =~ /\A$PublicInbox::Hval::FN\z/o) {
$sfn = $fn;
} elsif ($ct eq 'text/plain') {
$sfn = 'a.txt';
} else {
$sfn = 'a.bin';
}
- my $ret = qq($nl<a\nhref="$upfx$idx-$sfn">);
+ my $rv = $ctx->{obuf};
+ $$rv .= qq($nl<a\nhref="$ctx->{mhref}$idx-$sfn">);
if ($err) {
- $ret .=
-"[-- Warning: decoded text below may be mangled --]\n";
+ $$rv .= <<EOF;
+[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
+EOF
}
- $ret .= "[-- Attachment #$idx: ";
+ $$rv .= "[-- Attachment #$idx: ";
my $ts = "Type: $ct, Size: $size bytes";
+ my $desc = $part->header('Content-Description') // $fn // '';
$desc = ascii_html($desc);
- $ret .= ($desc eq '') ? "$ts --]" : "$desc --]\n[-- $ts --]";
- $ret .= "</a>\n";
+ $$rv .= ($desc eq '') ? "$ts --]" : "$desc --]\n[-- $ts --]";
+ $$rv .= "</a>\n";
+
+ submsg_hdr($ctx, $part) if $part->{is_submsg};
+
+ undef;
}
-sub add_text_body {
- my ($upfx, $obfs_ibx, $p) = @_;
- # $p - from msg_iter: [ Email::MIME, depth, @idx ]
- my ($part, $depth) = @$p; # attachment @idx is unused
+sub add_text_body { # callback for each_part
+ my ($p, $ctx) = @_;
+ my $upfx = $ctx->{mhref};
+ my $ibx = $ctx->{ibx};
+ my $l = $ctx->{-linkify} //= PublicInbox::Linkify->new;
+ # $p - from each_part: [ Email::MIME-like, depth, $idx ]
+ my ($part, $depth, $idx) = @$p;
my $ct = $part->content_type || 'text/plain';
my $fn = $part->filename;
+ my ($s, $err) = msg_part_text($part, $ct);
+ return attach_link($ctx, $ct, $p, $fn) unless defined $s;
- if ($ct =~ m!\btext/x?html\b!i) {
- return attach_link($upfx, $ct, $p, $fn);
+ my $rv = $ctx->{obuf};
+ if ($part->{is_submsg}) {
+ submsg_hdr($ctx, $part);
+ $$rv .= "\n";
}
- 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);
+ # 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;
+
+ # will be escaped to `•' 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;
+ if ($s =~ /^--- [^\n]+\n\+{3} [^\n]+\n@@ /ms) {
+ # diffstat anchors do not link across attachments or messages,
+ # -apfx is just a stable prefix for making diffstat anchors
+ # linkable to the first diff hunk w/o crossing attachments
+ $idx =~ tr!.!/!; # compatibility with previous versions
+ $ctx->{-apfx} = $upfx . $idx;
+
+ # do attr => filename mappings for diffstats in git diffs:
+ $ctx->{-anchors} = {} if $s =~ /^diff --git /sm;
+ $diff = 1;
+ delete $ctx->{-long_path};
+ my $spfx;
+ # absolute URL (Atom feeds)
+ if ($ibx->{coderepo}) {
+ if (index($upfx, '//') >= 0) {
+ $spfx = $upfx;
+ $spfx =~ s!/([^/]*)/\z!/!;
+ } else {
+ my $n_slash = $upfx =~ tr!/!/!;
+ if ($n_slash == 0) {
+ $spfx = '../';
+ } elsif ($n_slash == 1) {
+ $spfx = '';
+ } else { # nslash == 2
+ $spfx = '../../';
+ }
+ }
}
- }
+ $ctx->{-spfx} = $spfx;
+ };
- my @lines = split(/^/m, $s);
- $s = '';
- if (defined($fn) || $depth > 0 || $err) {
- $s .= attach_link($upfx, $ct, $p, $fn, $err);
- $s .= "\n";
+ # split off quoted and unquoted blocks:
+ my @sections = PublicInbox::MsgIter::split_quotes($s);
+ undef $s; # free memory
+ if (defined($fn) || ($depth > 0 && !$part->{is_submsg}) || $err) {
+ # badly-encoded message with $err? tell the world about it!
+ attach_link($ctx, $ct, $p, $fn, $err);
+ $$rv .= "\n";
}
- my @quot;
- my $l = PublicInbox::Linkify->new;
- foreach my $cur (@lines) {
- if ($cur !~ /^>/) {
- # 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));
+ delete $part->{bdy}; # save memory
+ foreach my $cur (@sections) {
+ if ($cur =~ /\A>/) {
+ # 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);
+ $$rv .= '</span>';
+ } elsif ($diff) {
+ flush_diff($ctx, \$cur);
} else {
- push @quot, $cur;
- }
- }
-
- if (@quot) { # ugh, top posted
- flush_quote(\$s, $l, \@quot);
- obfuscate_addrs($obfs_ibx, $s) if $obfs_ibx;
- $s;
- } else {
- obfuscate_addrs($obfs_ibx, $s) if $obfs_ibx;
- if ($s =~ /\n\z/s) { # common, last line ends with a newline
- $s;
- } else { # some editors don't do newlines...
- $s .= "\n";
+ # regular lines, OK
+ $$rv .= $l->to_html($cur);
}
+ undef $cur; # free memory
}
}
-sub _msg_html_prepare {
- my ($hdr, $ctx, $more, $nr) = @_;
- my $srch = $ctx->{srch} if $ctx;
- my $atom = '';
+sub _msg_page_prepare_obuf {
+ my ($eml, $ctx) = @_;
+ my $over = $ctx->{ibx}->over;
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 .=
-"<pre>WARNING: multiple messages refer to this Message-ID\n</pre>";
+ my $mids = mids_for_index($eml);
+ my $nr = $ctx->{nr}++;
+ if ($nr) { # unlikely
+ if ($ctx->{chash} eq content_hash($eml)) {
+ warn "W: BUG? @$mids not deduplicated properly\n";
+ return \$rv;
}
- $rv .= "<pre\nid=b>"; # anchor for body start
- } else {
+ $rv .=
+"<pre>WARNING: multiple messages have this Message-ID\n</pre>";
$rv .= '<pre>';
+ } else {
+ $ctx->{first_hdr} = $eml->header_obj;
+ $ctx->{chash} = content_hash($eml) if $ctx->{smsg}; # reused MID
+ $rv .= "<pre\nid=b>"; # anchor for body start
}
- if ($srch) {
- $ctx->{-upfx} = '../';
+ $ctx->{-upfx} = '../' if $over;
+ my @title; # (Subject[0], From[0])
+ for my $v ($eml->header('From')) {
+ my @n = PublicInbox::Address::names($v);
+ $v = ascii_html($v);
+ $title[1] //= ascii_html(join(', ', @n));
+ if ($obfs_ibx) {
+ obfuscate_addrs($obfs_ibx, $v);
+ obfuscate_addrs($obfs_ibx, $title[1]);
+ }
+ $rv .= "From: $v\n" if $v ne '';
}
- my @title;
- foreach my $h (qw(From To Cc Subject Date)) {
- my $v = $hdr->header($h);
- defined($v) && ($v ne '') or next;
- $v = PublicInbox::Hval->new($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: <a\nhref="#r"\nid=t>);
- $rv .= $v->as_html . "</a>\n";
- next;
- }
+ foreach my $h (qw(To Cc)) {
+ for my $v ($eml->header($h)) {
+ fold_addresses($v);
+ $v = ascii_html($v);
+ obfuscate_addrs($obfs_ibx, $v) if $obfs_ibx;
+ $rv .= "$h: $v\n" if $v ne '';
}
- $v = $v->as_html;
+ }
+ my @subj = $eml->header('Subject');
+ if (@subj) {
+ my $v = ascii_html(shift @subj);
obfuscate_addrs($obfs_ibx, $v) if $obfs_ibx;
- $rv .= "$h: $v\n";
-
+ $rv .= 'Subject: ';
+ $rv .= $over ? qq(<a\nhref="#r"\nid=t>$v</a>\n) : "$v\n";
+ $title[0] = $v;
+ for $v (@subj) { # multi-Subject message :<
+ $v = ascii_html($v);
+ obfuscate_addrs($obfs_ibx, $v) if $obfs_ibx;
+ $rv .= "Subject: $v\n";
+ }
+ } else { # dummy anchor for thread skeleton at bottom of page
+ $rv .= qq(<a\nhref="#r"\nid=t></a>) if $over;
+ $title[0] = '(no subject)';
}
- $title[0] ||= '(no subject)';
- $ctx->{-title_html} = join(' - ', @title);
- foreach (@$mids) {
- my $mid = PublicInbox::Hval->new_msgid($_) ;
- my $mhtml = $mid->as_html;
- if ($multiple) {
- my $href = $mid->{href};
- $rv .= "Message-ID: ";
- $rv .= "<a\nhref=\"../$href/\">";
- $rv .= "<$mhtml></a> ";
- $rv .= "(<a\nhref=\"../$href/raw\">raw</a>)\n";
- } else {
- $rv .= "Message-ID: <$mhtml> ";
- $rv .= "(<a\nhref=\"raw\">raw</a>)\n";
+ for my $v ($eml->header('Date')) {
+ $v = ascii_html($v);
+ obfuscate_addrs($obfs_ibx, $v) if $obfs_ibx; # possible :P
+ $rv .= qq{Date: $v\t<a\nhref="#r">[thread overview]</a>\n};
+ }
+ if (!$nr) { # first (and only) message, common case
+ $ctx->{-title_html} = join(' - ', @title);
+ $rv = $ctx->html_top . $rv;
+ }
+ if (scalar(@$mids) == 1) { # common case
+ my $mhtml = ascii_html($mids->[0]);
+ $rv .= "Message-ID: <$mhtml> ";
+ $rv .= "(<a\nhref=\"raw\">raw</a>)\n";
+ } else {
+ # X-Alt-Message-ID can happen if a message is injected from
+ # public-inbox-nntpd because of multiple Message-ID headers.
+ my $lnk = PublicInbox::Linkify->new;
+ my $s = '';
+ for my $h (qw(Message-ID X-Alt-Message-ID)) {
+ $s .= "$h: $_\n" for ($eml->header_raw($h));
}
+ $lnk->linkify_mids('..', \$s, 1);
+ $rv .= $s;
}
- $rv .= _parent_headers($hdr, $srch);
+ $rv .= _parent_headers($eml, $over);
$rv .= "\n";
+ \$rv;
}
-sub thread_skel {
- my ($dst, $ctx, $hdr, $tpfx) = @_;
- my $srch = $ctx->{srch};
- my $mid = mids($hdr)->[0];
- my $sres = $srch->get_thread($mid);
- my $nr = $sres->{total};
- my $expand = qq(expand[<a\nhref="${tpfx}T/#u">flat</a>) .
- qq(|<a\nhref="${tpfx}t/#u">nested</a>] ) .
- qq(<a\nhref="${tpfx}t.mbox.gz">mbox.gz</a> ) .
- qq(<a\nhref="${tpfx}t.atom">Atom feed</a>);
+sub SKEL_EXPAND () {
+ qq(expand[<a\nhref="T/#u">flat</a>) .
+ qq(|<a\nhref="t/#u">nested</a>] ) .
+ qq(<a\nhref="t.mbox.gz">mbox.gz</a> ) .
+ qq(<a\nhref="t.atom">Atom feed</a>);
+}
+sub thread_skel ($$$) {
+ my ($skel, $ctx, $hdr) = @_;
+ my $mid = mids($hdr)->[0];
+ my $ibx = $ctx->{ibx};
+ my ($nr, $msgs) = $ibx->over->get_thread($mid);
my $parent = in_reply_to($hdr);
- $$dst .= "\n<b>Thread overview: </b>";
+ $$skel .= "\n<b>Thread overview: </b>";
if ($nr <= 1) {
if (defined $parent) {
- $$dst .= "$expand\n ";
- $$dst .= ghost_parent("$tpfx../", $parent) . "\n";
+ $$skel .= SKEL_EXPAND."\n ";
+ $$skel .= ghost_parent('../', $parent) . "\n";
} else {
- $$dst .= "[no followups] $expand\n";
+ $$skel .= "<a\nid=r>[no followups]</a> ".
+ SKEL_EXPAND."\n";
}
$ctx->{next_msg} = undef;
$ctx->{parent_msg} = $parent;
return;
}
- $$dst .= "$nr+ messages / $expand";
- $$dst .= qq! <a\nhref="#b">top</a>\n!;
+ $$skel .= $nr;
+ $$skel .= '+ messages / '.SKEL_EXPAND.qq! <a\nhref="#b">top</a>\n!;
- my $subj = $hdr->header('Subject');
- defined $subj or $subj = '';
- $ctx->{prev_subj} = [ split(/ /, $srch->subject_normalized($subj)) ];
+ # nb: mutt only shows the first Subject in the index pane
+ # when multiple Subject: headers are present, so we follow suit:
+ my $subj = $hdr->header('Subject') // '';
+ $subj = '(no subject)' if $subj eq '';
+ $ctx->{prev_subj} = [ split(/ /, subject_normalized($subj)) ];
$ctx->{cur} = $mid;
$ctx->{prev_attr} = '';
$ctx->{prev_level} = 0;
- $ctx->{dst} = $dst;
- $sres = load_results($srch, $sres);
+ $ctx->{skel} = $skel;
# 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;
}
sub _parent_headers {
- my ($hdr, $srch) = @_;
+ my ($hdr, $over) = @_;
my $rv = '';
-
- my $irt = in_reply_to($hdr);
- if (defined $irt) {
- my $v = PublicInbox::Hval->new_msgid($irt);
- my $html = $v->as_html;
- my $href = $v->{href};
- $rv .= "In-Reply-To: <";
- $rv .= "<a\nhref=\"../$href/\">$html</a>>\n";
+ my @irt = $hdr->header_raw('In-Reply-To');
+ my $refs;
+ if (@irt) {
+ my $lnk = PublicInbox::Linkify->new;
+ $rv .= "In-Reply-To: $_\n" for @irt;
+ $lnk->linkify_mids('..', \$rv);
+ } else {
+ $refs = references($hdr);
+ my $irt = pop @$refs;
+ if (defined $irt) {
+ my $html = ascii_html($irt);
+ my $href = mid_href($irt);
+ $rv .= "In-Reply-To: <";
+ $rv .= "<a\nhref=\"../$href/\">$html</a>>\n";
+ }
}
# do not display References: if search is present,
# 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);
- }
+ return $rv if $over;
- if (@refs) {
- $rv .= 'References: '. join("\n\t", @refs) . "\n";
- }
+ $refs //= references($hdr);
+ if (@$refs) {
+ @$refs = map { linkify_ref_no_over($_) } @$refs;
+ $rv .= 'References: '. join("\n\t", @$refs) . "\n";
}
$rv;
}
+# returns a string buffer
sub html_footer {
- my ($hdr, $standalone, $ctx, $rhref) = @_;
-
- my $srch = $ctx->{srch} if $ctx;
+ my ($ctx, $hdr) = @_;
my $upfx = '../';
- my $tpfx = '';
- my $idx = $standalone ? " <a\nhref=\"$upfx\">index</a>" : '';
- my $irt = '';
- if ($idx && $srch) {
- $idx .= "\n";
- thread_skel(\$idx, $ctx, $hdr, $tpfx);
+ my $skel;
+ my $rv = '<pre>';
+ my $related;
+ my $qry = delete $ctx->{-qry};
+ if ($qry && $ctx->{ibx}->isrch) {
+ my $q = ''; # search for either ancestor or descendent patches
+ for (@{$qry->{dfpre}}, @{$qry->{dfpost}}) {
+ chop if length > 7; # include 1 abbrev "older" patches
+ $q .= "dfblob:$_ ";
+ }
+ chop $q; # omit trailing SP
+ local $Text::Wrap::columns = COLS;
+ local $Text::Wrap::huge = 'overflow';
+ $q = wrap('', '', $q);
+ my $rows = ($q =~ tr/\n/\n/) + 1;
+ $q = ascii_html($q);
+ $related = <<EOM;
+<form id=related
+action=$upfx
+><pre>find likely ancestor, descendant, or conflicting patches for <a
+href=#t>this message</a>:
+<textarea name=q cols=${\COLS} rows=$rows>$q</textarea>
+<input type=submit value=search
+/>\t(<a href=${upfx}_/text/help/>help</a>)</pre></form>
+EOM
+ }
+ if ($ctx->{ibx}->over) {
+ my $t = ts2str($ctx->{-t_max});
+ my $t_fmt = fmt_ts($ctx->{-t_max});
+ my $fallback = $related ? "\t" : "<a id=related>\t</a>";
+ $skel = <<EOF;
+${fallback}other threads:[<a
+href="$upfx?t=$t">~$t_fmt UTC</a>|<a
+href="$upfx">newest</a>]
+EOF
+
+ thread_skel(\$skel, $ctx, $hdr);
my ($next, $prev);
my $parent = ' ';
$next = $prev = ' ';
if (my $n = $ctx->{next_msg}) {
- $n = PublicInbox::Hval->new_msgid($n)->{href};
+ $n = mid_href($n);
$next = "<a\nhref=\"$upfx$n/\"\nrel=next>next</a>";
}
my $u;
my $par = $ctx->{parent_msg};
if ($par) {
- $u = PublicInbox::Hval->new_msgid($par)->{href};
+ $u = mid_href($par);
$u = "$upfx$u/";
}
if (my $p = $ctx->{prev_msg}) {
- $prev = PublicInbox::Hval->new_msgid($p)->{href};
+ $prev = mid_href($p);
if ($p && $par && $p eq $par) {
$prev = "<a\nhref=\"$upfx$prev/\"\n" .
'rel=prev>prev parent</a>';
} elsif ($u) { # unlikely
$parent = " <a\nhref=\"$u\"\nrel=prev>parent</a>";
}
- $irt = "$next $prev$parent ";
- } else {
- $irt = '';
+ $rv .= "$next $prev$parent ";
+ } else { # unindexed inboxes w/o over
+ $skel = qq( <a\nhref="$upfx">latest</a>);
}
- $rhref ||= '#R';
- $irt .= qq(<a\nhref="$rhref">reply</a>);
- $irt .= $idx;
+ $rv .= qq(<a\nhref="#R">reply</a>);
+ $rv .= $skel;
+ $rv .= '</pre>';
+ $rv .= $related // '';
+ $rv .= msg_reply($ctx, $hdr);
}
-sub linkify_ref_nosrch {
- my $v = PublicInbox::Hval->new_msgid($_[0]);
- my $html = $v->as_html;
- my $href = $v->{href};
+sub linkify_ref_no_over {
+ my ($mid) = @_;
+ my $href = mid_href($mid);
+ my $html = ascii_html($mid);
"<<a\nhref=\"../$href/\">$html</a>>";
}
-sub anchor_for {
- my ($msgid) = @_;
- 'm' . id_compress($msgid, 1);
-}
-
sub ghost_parent {
my ($upfx, $mid) = @_;
- $mid = PublicInbox::Hval->new_msgid($mid);
- my $href = $mid->{href};
- my $html = $mid->as_html;
+ my $href = mid_href($mid);
+ my $html = ascii_html($mid);
qq{[parent not found: <<a\nhref="$upfx$href/">$html</a>>]};
}
$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->{mid} eq $ctx->{mid}) {
+ $ctx->{found_mid_at} = $ctx->{root_idx};
+ return 0; # stop iterating
+ }
+ 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 $rootset = PublicInbox::SearchThread::thread($msgs, \&sort_ds, $ctx);
+
+ # 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 {
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 {
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 {
- my ($ctx, $level, $node) = @_;
- my $smsg = $node->{smsg} or return _skel_ghost($ctx, $level, $node);
+sub skel_dump { # walk_thread callback
+ my ($ctx, $level, $smsg) = @_;
+ $smsg->{blob} or return _skel_ghost($ctx, $level, $smsg);
- my $dst = $ctx->{dst};
+ my $skel = $ctx->{skel};
my $cur = $ctx->{cur};
my $mid = $smsg->{mid};
- my $f = ascii_html($smsg->from_name);
+ if ($level == 0 && $ctx->{skel_dump_roots}++) {
+ $$skel .= delete($ctx->{sl_note}) || '';
+ }
+
+ my $f = ascii_html(delete $smsg->{from_name});
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 $d = fmt_ts($smsg->{ds});
+ my $unmatched; # if lazy-loaded by SearchThread::Msg::visible()
+ if (exists $ctx->{searchview}) {
+ if (defined(my $pct = $smsg->{pct})) {
+ $d .= (sprintf(' % 2u', $pct) . '%');
+ } else {
+ $unmatched = 1;
+ $d .= ' ';
+ }
+ }
+ $d .= ' ' . indent_for($level) . th_pfx($level);
my $attr = $f;
$ctx->{first_level} ||= $level;
if ($cur) {
if ($cur eq $mid) {
delete $ctx->{cur};
- $$dst .= "<b>$d<a\nid=r\nhref=\"#t\">".
+ $$skel .= "<b>$d<a\nid=r\nhref=\"#t\">".
"$attr [this message]</a></b>\n";
- return;
+ return 1;
} else {
$ctx->{prev_msg} = $mid;
}
# 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 @subj = split(/ /, $ctx->{srch}->subject_normalized($smsg->subject));
-
+ my @subj = split(/ /, 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} || [];
}
my $m;
my $id = '';
- my $mapping = $ctx->{mapping};
+ my $mapping = $unmatched ? undef : $ctx->{mapping};
if ($mapping) {
my $map = $mapping->{$mid};
$id = id_compress($mid, 1);
$map->[0] = "$d<a\nhref=\"$m\">$end";
$id = "\nid=r".$id;
} else {
- $m = $ctx->{-upfx}.mid_escape($mid).'/';
+ $m = $ctx->{-upfx}.mid_href($mid).'/';
}
- $$dst .= $d . "<a\nhref=\"$m\"$id>" . $end;
+ $$skel .= $d . "<a\nhref=\"$m\"$id>" . $end;
+ 1;
}
sub _skel_ghost {
my ($ctx, $level, $node) = @_;
- my $mid = $node->{id};
- my $d = $ctx->{pct} ? ' [irrelevant] ' # search result
- : ' [not found] ';
+ my $mid = $node->{mid};
+ my $d = ' [not found] ';
+ $d .= ' ' if exists $ctx->{searchview};
$d .= indent_for($level) . th_pfx($level);
my $upfx = $ctx->{-upfx};
- my $m = PublicInbox::Hval->new_msgid($mid);
- my $href = $upfx . $m->{href} . '/';
- my $html = $m->as_html;
+ my $href = $upfx . mid_href($mid) . '/';
+ my $html = ascii_html($mid);
my $mapping = $ctx->{mapping};
my $map = $mapping->{$mid} if $mapping;
} else {
$d .= qq{<<a\nhref="$href">$html</a>>\n};
}
- my $dst = $ctx->{dst};
- $$dst .= $d;
+ ${$ctx->{skel}} .= $d;
+ 1;
}
sub sort_ds {
- [ sort {
- (eval { $a->topmost->{smsg}->ds } || 0) <=>
- (eval { $b->topmost->{smsg}->ds } || 0)
- } @{$_[0]} ];
+ @{$_[0]} = sort {
+ (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 {
- my ($ctx, $level, $node) = @_;
- my $srch = $ctx->{srch};
- my $mid = $node->{id};
- 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);
- $ds = $x->ds;
- if ($level == 0) {
- $topic = [ $ds, 1, { $subj => $mid }, $subj ];
+sub acc_topic { # walk_thread callback
+ my ($ctx, $level, $smsg) = @_;
+ my $mid = $smsg->{mid};
+ my $has_blob = $smsg->{blob} // do {
+ if (my $by_mid = $ctx->{ibx}->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};
+ if ($level == 0) { # new, top-level topic
+ my $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
+ # continue existing topic
+ my $topic = $ctx->{-cur_topic}; # should never be undef
$topic->[0] = $ds if $ds > $topic->[0];
- $topic->[1]++;
+ $topic->[1]++; # bump N+ message counter
my $seen = $topic->[2];
if (scalar(@$topic) == 3) { # parent was a ghost
push @$topic, $subj;
- } elsif (!$seen->{$subj}) {
- push @$topic, $level, $subj;
+ } elsif (!defined($seen->{$subj})) {
+ push @$topic, $level, $subj; # @extra messages
}
$seen->{$subj} = $mid; # latest for subject
} else { # ghost message
- return if $level != 0; # ignore child ghosts
- $topic = [ -666, 0, {} ];
- $ctx->{-cur_topic} = $topic;
+ return 1 if $level != 0; # ignore child ghosts
+ my $topic = $ctx->{-cur_topic} = [ -666, 0, {} ];
push @{$ctx->{order}}, $topic;
}
+ 1;
}
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;
}
my @out;
- my $ibx = $ctx->{-inbox};
+ my $ibx = $ctx->{ibx};
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 ($ds, $n, $seen, $top, @ex) = @$topic;
+ my ($ds, $n, $seen, $top_subj, @extra) = @$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;
+ next unless defined $top_subj; # ghost topic
+ my $mid = delete $seen->{$top_subj};
+ my $href = mid_href($mid);
+ my $prev_subj = [ split(/ /, $top_subj) ];
+ $top_subj = ascii_html($top_subj);
$ds = fmt_ts($ds);
# $n isn't the total number of posts on the topic,
$anchor = '#t'; # thread skeleton
}
- my $mbox = qq(<a\nhref="$href/t.mbox.gz">mbox.gz</a>);
- my $atom = qq(<a\nhref="$href/t.atom">Atom</a>);
- my $s = "<a\nhref=\"$href/T/$anchor\"><b>$top</b></a>\n" .
- " $ds UTC $n - $mbox / $atom\n";
- for (my $i = 0; $i < scalar(@ex); $i += 2) {
- my $level = $ex[$i];
- my $subj = $ex[$i + 1];
+ my $s = "<a\nhref=\"$href/T/$anchor\">$top_subj</a>\n" .
+ " $ds UTC $n\n";
+ while (@extra) {
+ my $level = shift @extra;
+ my $subj = shift @extra; # already normalized
$mid = delete $seen->{$subj};
- my @subj = split(/ /, $srch->subject_normalized($subj));
+ my @subj = split(/ /, $subj);
my @next_prev = @subj; # full copy
my $omit = dedupe_subject($prev_subj, \@subj, ' "');
$prev_subj = \@next_prev;
- $subj = ascii_html(join(' ', @subj));
+ $subj = join(' ', @subj);
+ $subj = ascii_html($subj);
obfuscate_addrs($obfs_ibx, $subj) if $obfs_ibx;
- $href = mid_escape($mid);
+ $href = mid_href($mid);
$s .= indent_for($level) . TCHILD;
- $s .= "<a\nhref=\"$href/T/#u\">$subj</a>$omit\n";
+ $s .= qq(<a\nhref="$href/T/#u">$subj</a>$omit\n);
}
push @out, $s;
}
200;
}
-sub index_nav { # callback for WwwStream
- my (undef, $ctx) = @_;
- delete $ctx->{qp} or return;
- my ($next, $prev);
- $next = $prev = ' ';
- my $latest = '';
+sub str2ts ($) {
+ my ($yyyy, $mon, $dd, $hh, $mm, $ss) = unpack('A4A2A2A2A2A2', $_[0]);
+ timegm($ss || 0, $mm || 0, $hh || 0, $dd, $mon - 1, $yyyy);
+}
+
+sub pagination_footer ($$) {
+ my ($ctx, $latest) = @_;
+ my $next = $ctx->{next_page} || '';
+ my $prev = $ctx->{prev_page} || '';
+ if ($prev) { # aligned padding for: 'next (older) | '
+ $next = $next ? "$next | " : ' | ';
+ $prev .= qq[ | <a\nhref="$latest">latest</a>];
+ }
+ ($next || $prev) ? "<hr><pre>page: $next$prev</pre>" : '';
+}
- my $next_o = $ctx->{-next_o};
- if ($next_o) {
- $next = qq!<a\nhref="?o=$next_o"\nrel=next>next</a>!;
+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([0-9]{8,14})-// and $after = str2ts($1);
+ $t =~ /\A([0-9]{8,14})\z/ and $before = str2ts($1);
+
+ my $msgs = $ctx->{ibx}->over->recent($opts, $after, $before);
+ my $nr = scalar @$msgs;
+ if ($nr < $lim && defined($after)) {
+ $after = $before = undef;
+ $msgs = $ctx->{ibx}->over->recent($opts);
+ $nr = scalar @$msgs;
}
- if (my $cur_o = $ctx->{-cur_o}) {
- $latest = qq! <a\nhref=.>latest</a>!;
-
- my $o = $cur_o - ($next_o - $cur_o);
- if ($o > 0) {
- $prev = qq!<a\nhref="?o=$o"\nrel=prev>prev</a>!;
- } elsif ($o == 0) {
- $prev = qq!<a\nhref=.\nrel=prev>prev</a>!;
+ 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;
}
}
- "<hr><pre>page: $next $prev$latest</pre>";
+ if (defined($oldest) && $more) {
+ my $s = ts2str($oldest);
+ $ctx->{next_page} = qq[<a\nhref="?t=$s"\nrel=next>] .
+ 'next (older)</a>';
+ }
+ if (defined($newest) && (defined($before) || defined($after))) {
+ my $s = ts2str($newest);
+ $ctx->{prev_page} = qq[<a\nhref="?t=$s-"\nrel=prev>] .
+ 'prev (newer)</a>';
+ }
+ $msgs;
}
+# GET /$INBOX - top-level inbox view for indexed inboxes
sub index_topics {
my ($ctx) = @_;
- my ($off) = (($ctx->{qp}->{o} || '0') =~ /(\d+)/);
- my $lim = 200;
- my $opts = { offset => $off, limit => $lim };
-
- $ctx->{order} = [];
- my $srch = $ctx->{srch};
-
- my $qs = '';
- # this complicated bit cuts loading time by over 400ms on my system:
- if ($off == 0) {
- my ($min, $max) = $ctx->{-inbox}->mm->minmax;
- my $n = $max - $lim;
- $n = $min if $n < $min;
- for (; $qs eq '' && $n >= $min; --$n) {
- my $smsg = $srch->lookup_article($n) or next;
- $qs = POSIX::strftime('d:%Y%m%d..', gmtime($smsg->ts));
- }
- }
-
- my $sres = $srch->query($qs, $opts);
- my $nr = scalar @{$sres->{msgs}};
- if ($nr) {
- $sres = load_results($srch, $sres);
- walk_thread(thread_results($ctx, $sres), $ctx, *acc_topic);
- }
- $ctx->{-next_o} = $off + $nr;
- $ctx->{-cur_o} = $off;
- PublicInbox::WwwStream->response($ctx, dump_topics($ctx), *index_nav);
+ my $msgs = paginate_recent($ctx, 200); # 200 is our window
+ walk_thread(thread_results($ctx, $msgs), $ctx, \&acc_topic) if @$msgs;
+ html_oneshot($ctx, dump_topics($ctx), pagination_footer($ctx, '.'));
}
sub thread_adj_level {
sub ghost_index_entry {
my ($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;
}