Reply instructions:
-You may reply publically to 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 $irt = $hdr->header_raw('In-Reply-To');
-
- return mid_clean($irt) if defined $irt && $irt ne '';
+ my $refs = references($hdr);
+ $refs->[-1];
+}
- my $refs = $hdr->header_raw('References');
- if ($refs && $refs =~ /<([^>]+)>\s*\z/s) {
- return $1;
- }
- undef;
+sub fold_addresses ($) {
+ 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 ($$) {
+sub _hdr_names_html ($$) {
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 ($$$) {
@@ -115,79 +181,116 @@ sub nr_to_s ($$$) {
$nr == 1 ? "$nr $singular" : "$nr $plural";
}
+# Displays the text of of the message for /$INBOX/$MSGID/[Tt]/ endpoint
# this is already inside a
-sub index_entry {
- my ($mime, $ctx, $more) = @_;
- my $srch = $ctx->{srch};
- my $hdr = $mime->header_obj;
- my $subj = $hdr->header('Subject');
-
- my $mid_raw = mid_clean(mid_mime($mime));
+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 $mid = PublicInbox::Hval->new_msgid($mid_raw);
-
my $root_anchor = $ctx->{root_anchor} || '';
- my $irt = in_reply_to($hdr);
+ my $irt;
+ my $obfs_ibx = $ctx->{-obfs_ibx};
+ $subj = '(no subject)' if $subj eq '';
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);
+ $rv .= _th_index_lite($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 $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: $from @ ".fmt_ts($ds)." UTC";
my $upfx = $ctx->{-upfx};
- my $mhref = $upfx . $mid->as_href . '/';
+ my $mhref = $upfx . mid_href($mid_raw) . '/';
$rv .= qq{ (permalink / };
$rv .= qq{raw)\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;
+ $to_cc .= '; +Cc: '.$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 && $irt) {
- my $mirt = PublicInbox::Hval->new_msgid($irt);
- my $href = $upfx . $mirt->as_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);
$rv .= qq(In-Reply-To: <$html>\n)
}
$rv .= "\n";
# scan through all parts, looking for displayable text
- msg_iter($mime, sub { $rv .= add_text_body($mhref, $_[0]) });
+ $ctx->{mhref} = $mhref;
+ $ctx->{obuf} = \$rv;
+ $eml->each_part(\&add_text_body, $ctx, 1);
+ delete $ctx->{obuf};
# add the footer
$rv .= "\n^ ".
"permalink" .
" raw" .
" reply";
- if (my $pct = $ctx->{pct}) { # used by SearchView.pm
- $rv .= "\t[relevance $pct->{$mid_raw}%]";
+
+ my $hr;
+ if (defined(my $pct = $smsg->{pct})) { # used by SearchView.pm
+ $rv .= "\t[relevance $pct%]";
+ $hr = 1;
} elsif ($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};
}
- $rv .= $more ? "\n\n" : "\n";
+ # do we have more messages? start a new if so
+ $rv .= scalar(@{$ctx->{msgs}}) ? '
' : '
' if $hr;
+ $rv;
}
sub pad_link ($$;$) {
my ($mid, $level, $s) = @_;
$s ||= '...';
- my $id = id_compress($mid, 1);
- (' 'x19).indent_for($level).th_pfx($level)."($s)\n";
+ my $href = defined($mid) ?
+ ("($s)\n") :
+ "($s)\n";
+ (' 'x19).indent_for($level).th_pfx($level).$href;
+}
+
+sub _skel_hdr {
+ # my ($mapping, $mid) = @_;
+ ($_[0]->{$_[1] // \'bogus'} // [ "(?)\n" ])->[0];
}
sub _th_index_lite {
@@ -195,472 +298,556 @@ sub _th_index_lite {
my $rv = '';
my $mapping = $ctx->{mapping} or return $rv;
my $pad = ' ';
- # map = [children, attr, node, idx, level]
- my $map = $mapping->{$mid_raw};
- my $nr_c = scalar @{$map->[0]};
+ my $mid_map = $mapping->{$mid_raw};
+ defined $mid_map or
+ return 'public-inbox BUG: '.ascii_html($mid_raw).' not mapped';
+ my ($attr, $node, $idx, $level) = @$mid_map;
+ my $children = $node->{children};
+ my $nr_c = scalar @$children;
my $nr_s = 0;
- my $level = $map->[4];
- my $idx = $map->[3];
- if (defined $irt) {
- my $irt_map = $mapping->{$irt};
- my $siblings = $irt_map->[0];
+ my $siblings;
+ # delete saves about 200KB on a 1K message thread
+ if (my $refs = delete $node->{references}) {
+ ($$irt) = ($refs =~ m/$MID_EXTRACT\z/o);
+ }
+ my $irt_map = $mapping->{$$irt} if defined $$irt;
+ if (defined $irt_map) {
+ $siblings = $irt_map->[1]->{children};
$nr_s = scalar(@$siblings) - 1;
- $rv .= $pad . $irt_map->[1];
+ $rv .= $pad . $irt_map->[0];
if ($idx > 0) {
my $prev = $siblings->[$idx - 1];
- my $pmid = $prev->messageid;
+ 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]->messageid;
- $rv .= $pad . $mapping->{$ppmid}->[1];
+ $rv .= $pad . _skel_hdr($mapping,
+ $siblings->[0] ?
+ $siblings->[0]->{mid} : undef);
}
- $rv .= $pad . $mapping->{$pmid}->[1];
+ $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');
- my $this = $map->[1];
- $this =~ s!\n\z!\n!s;
- $this =~ s! !!s; # no point in duplicating subject
- $this =~ s!]+>([^<]+)!$1!s; # no point linking to self
- $rv .= "@ $this";
- my $node = $map->[2];
- if (my $child = $node->child) {
- my $cmid = $child->messageid;
- $rv .= $pad . $mapping->{$cmid}->[1];
+ $attr =~ s!\n\z!\n!s;
+ $attr =~ s! (?:" )?!!s; # no point in dup subject
+ $attr =~ s!]+>([^<]+)!$1!s; # no point linking to self
+ $rv .= "@ $attr";
+ if ($nr_c) {
+ 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 = $child->next) {
- $rv .= $pad . $mapping->{$cn->messageid}->[1];
+ } elsif (my $cn = $children->[1]) {
+ $rv .= $pad . _skel_hdr($mapping, $cn->{mid});
}
}
- if (my $next = $node->next) {
- my $nmid = $next->messageid;
- $rv .= $pad . $mapping->{$nmid}->[1];
+
+ my $next = $siblings->[$idx+1] if $siblings && $idx >= 0;
+ if ($next) {
+ 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 = $next->next) {
- $rv .= $pad . $mapping->{$nn->messageid}->[1];
+ } elsif (my $nn = $siblings->[$idx + 2]) {
+ $rv .= $pad . _skel_hdr($mapping, $nn->{mid});
}
}
$rv .= $pad ."$s_s, $s_c; $ctx->{s_nr}\n";
}
-sub walk_thread {
- my ($th, $ctx, $cb) = @_;
- my @q = map { (0, $_) } $th->rootset;
+# non-recursive thread walker
+sub walk_thread ($$$) {
+ my ($rootset, $ctx, $cb) = @_;
+ my @q = map { (0, $_, -1) } @$rootset;
while (@q) {
- my $level = shift @q;
- my $node = shift @q or next;
- $cb->($ctx, $level, $node);
- unshift @q, $level+1, $node->child, $level, $node->next;
+ my ($level, $node, $i) = splice(@q, 0, 3);
+ defined $node or next;
+ $cb->($ctx, $level, $node, $i) or return;
+ ++$level;
+ $i = 0;
+ unshift @q, map { ($level, $_, $i++) } @{$node->{children}};
}
}
-sub pre_thread {
- my ($ctx, $level, $node) = @_;
- my $mapping = $ctx->{mapping};
- my $idx = -1;
- if (my $parent = $node->parent) {
- my $m = $mapping->{$parent->messageid}->[0];
- $idx = scalar @$m;
- push @$m, $node;
- }
- $mapping->{$node->messageid} = [ [], '', $node, $idx, $level ];
+sub pre_thread { # walk_thread callback
+ my ($ctx, $level, $node, $idx) = @_;
+ $ctx->{mapping}->{$node->{mid}} = [ '', $node, $idx, $level ];
skel_dump($ctx, $level, $node);
}
-sub thread_index_entry {
- my ($ctx, $level, $mime) = @_;
- my ($beg, $end) = thread_adj_level($ctx, $level);
- $beg . '' . index_entry($mime, $ctx, 0) . '
' . $end;
+sub thread_eml_entry {
+ my ($ctx, $eml) = @_;
+ my ($beg, $end) = thread_adj_level($ctx, $ctx->{level});
+ $beg . '' . eml_entry($ctx, $eml) . '
' . $end;
}
-sub stream_thread ($$) {
- my ($th, $ctx) = @_;
- my $inbox = $ctx->{-inbox};
- my $mime;
- my @q = map { (0, $_) } $th->rootset;
- my $level;
- while (@q) {
- $level = shift @q;
- my $node = shift @q or next;
- unshift @q, $level+1, $node->child, $level, $node->next;
- $mime = $inbox->msg_by_mid($node->messageid) and last;
- }
- return missing_thread($ctx) unless $mime;
-
- $mime = Email::MIME->new($mime);
- $ctx->{-title_html} = ascii_html($mime->header('Subject'));
- $ctx->{-html_tip} = thread_index_entry($ctx, $level, $mime);
- PublicInbox::WwwStream->response($ctx, 200, sub {
- return unless $ctx;
- while (@q) {
- $level = shift @q;
- my $node = shift @q or next;
- unshift @q, $level+1, $node->child, $level, $node->next;
- my $mid = $node->messageid;
- if ($mime = $inbox->msg_by_mid($mid)) {
- $mime = Email::MIME->new($mime);
- return thread_index_entry($ctx, $level, $mime);
- } else {
- return ghost_index_entry($ctx, $level, $mid);
+sub next_in_queue ($$) {
+ my ($q, $ghost_ok) = @_;
+ while (@$q) {
+ my ($level, $smsg) = splice(@$q, 0, 2);
+ my $cl = $level + 1;
+ unshift @$q, map { ($cl, $_) } @{$smsg->{children}};
+ return ($level, $smsg) if $ghost_ok || exists($smsg->{blob});
+ }
+ undef;
+}
+
+sub stream_thread_i { # PublicInbox::WwwStream::getline callback
+ my ($ctx, $eml) = @_;
+ return thread_eml_entry($ctx, $eml) if $eml;
+ return unless exists($ctx->{skel});
+ my $ghost_ok = $ctx->{nr}++;
+ while (1) {
+ my ($lvl, $smsg) = next_in_queue($ctx->{-queue}, $ghost_ok);
+ if ($smsg) {
+ if (exists $smsg->{blob}) { # next message for cat-file
+ $ctx->{level} = $lvl;
+ if (!$ghost_ok) { # first non-ghost
+ $ctx->{-title_html} =
+ ascii_html($smsg->{subject});
+ $ctx->zmore($ctx->html_top);
+ }
+ return $smsg;
}
+ # buffer the ghost entry and loop
+ $ctx->zmore(ghost_index_entry($ctx, $lvl, $smsg));
+ } else { # all done
+ $ctx->zmore(join('', thread_adj_level($ctx, 0)));
+ $ctx->zmore(${delete($ctx->{skel})});
+ return;
}
- my $ret = join('', thread_adj_level($ctx, 0));
- $ret .= ${$ctx->{dst}}; # skel
- $ctx = undef;
- $ret;
- });
+ }
+}
+
+sub stream_thread ($$) {
+ 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 $sres = $ctx->{srch}->get_thread($mid, { asc => 1 });
- my $msgs = load_results($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 = '
';
$skel .= $nr == 1 ? 'only message in thread' : 'end of thread';
- $skel .= ", back to index\n\n";
+ $skel .= <~$t_fmt UTC | newest]
+
+EOF
$skel .= "Thread overview: ";
$skel .= $nr == 1 ? '(only message)' : "$nr+ messages";
$skel .= " (download: mbox.gz";
$skel .= " / follow: Atom feed)\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->{seen} = {};
- $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
+ $ctx->{-obfs_ibx} = $ibx->{obfuscate} ? $ibx : undef;
+ walk_thread($rootset, $ctx, \&pre_thread);
- my $th = thread_results($msgs);
- walk_thread($th, $ctx, *pre_thread);
$skel .= '
';
- return stream_thread($th, $ctx) unless $ctx->{flat};
-
- # flat display: lazy load the full message from mini_mime:
- my $inbox = $ctx->{-inbox};
- my $mime;
- while ($mime = shift @$msgs) {
- $mime = $inbox->msg_by_mid(mid_clean(mid_mime($mime))) and last;
- }
- return missing_thread($ctx) unless $mime;
- $mime = Email::MIME->new($mime);
- $ctx->{-title_html} = ascii_html($mime->header('Subject'));
- $ctx->{-html_tip} = ''.index_entry($mime, $ctx, scalar @$msgs);
- $mime = undef;
- PublicInbox::WwwStream->response($ctx, 200, sub {
- return unless $msgs;
- while ($mime = shift @$msgs) {
- $mid = mid_clean(mid_mime($mime));
- $mime = $inbox->msg_by_mid($mid) and last;
+ return stream_thread($rootset, $ctx) unless $ctx->{flat};
+
+ # flat display: lazy load the full message from smsg
+ $ctx->{msgs} = $msgs;
+ $ctx->{-html_tip} = '';
+ PublicInbox::WwwStream::aresponse($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);
}
- if ($mime) {
- $mime = Email::MIME->new($mime);
- return index_entry($mime, $ctx, scalar @$msgs);
+ return eml_entry($ctx, $eml);
+ } else {
+ while (my $smsg = shift @{$ctx->{msgs}}) {
+ return $smsg if exists($smsg->{blob});
}
- $msgs = undef;
- '
'.$skel;
- });
+ my $skel = delete($ctx->{skel}) or return; # all done
+ $ctx->zmore($$skel);
+ undef;
+ }
}
sub multipart_text_as_html {
- my ($mime, $upfx) = @_;
- my $rv = "";
+ # ($mime, $ctx) = @_; # each_part may do "$_[0] = undef"
# scan through all parts, looking for displayable text
- msg_iter($mime, sub {
- my ($p) = @_;
- $rv .= add_text_body($upfx, $p);
- });
- $rv;
+ $_[0]->each_part(\&add_text_body, $_[1], 1);
}
-sub flush_quote {
- my ($s, $l, $quot) = @_;
+sub submsg_hdr ($$) {
+ 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";
+ }
+ }
+}
- # show everything in the full version with anchor from
- # short version (see above)
- my $rv = $l->linkify_1(join('', @$quot));
- @$quot = ();
+sub attach_link ($$$$;$) {
+ my ($ctx, $ct, $p, $fn, $err) = @_;
+ my ($part, $depth, $idx) = @$p;
- # we use a here to allow users to specify their own
- # color for quoted text
- $rv = $l->linkify_2(ascii_html($rv));
- $$s .= qq() . $rv . ''
-}
+ # Eml iteration clobbers multipart ->{bdy}, so do not offer
+ # downloads for 0-byte multipart attachments
+ return unless $part->{bdy};
-sub attach_link ($$$$) {
- my ($upfx, $ct, $p, $fn) = @_;
- my ($part, $depth, @idx) = @$p;
- my $nl = $idx[-1] > 1 ? "\n" : '';
- my $idx = join('.', @idx);
- my $size = bytes::length($part->body);
- $ct ||= 'text/plain';
- $ct =~ s/;.*//; # no attributes
+ my $nl = $idx eq '1' ? '' : "\n"; # like join("\n", ...)
+ my $size = length($part->body);
+ delete $part->{bdy}; # save memory
+
+ # hide attributes normally, unless we want to aid users in
+ # spotting MUA problems:
+ $ct =~ s/;.*// unless $err;
$ct = ascii_html($ct);
- my $desc = $part->header('Content-Description');
- $desc = $fn unless defined $desc;
- $desc = '' unless defined $desc;
my $sfn;
- if (defined $fn && $fn =~ /\A[[:alnum:]][\w\.-]+[[:alnum:]]\z/) {
+ if (defined $fn && $fn =~ /\A$PublicInbox::Hval::FN\z/o) {
$sfn = $fn;
} elsif ($ct eq 'text/plain') {
$sfn = 'a.txt';
} else {
$sfn = 'a.bin';
}
- my @ret = qq($nl[-- Attachment #$idx: );
+ my $rv = $ctx->{obuf};
+ $$rv .= qq($nl{mhref}$idx-$sfn">);
+ if ($err) {
+ $$rv .= <\n");
+ my $desc = $part->header('Content-Description') // $fn // '';
+ $desc = ascii_html($desc);
+ $$rv .= ($desc eq '') ? "$ts --]" : "$desc --]\n[-- $ts --]";
+ $$rv .= "\n";
+
+ submsg_hdr($ctx, $part) if $part->{is_submsg};
+
+ undef;
}
-sub add_text_body {
- my ($upfx, $p) = @_; # from msg_iter: [ Email::MIME, depth, @idx ]
- my ($part, $depth, @idx) = @$p;
- my $ct = $part->content_type;
+sub add_text_body { # callback for each_part
+ my ($p, $ctx) = @_;
+ my $upfx = $ctx->{mhref};
+ my $ibx = $ctx->{ibx};
+ my $l = $ctx->{-linkify} //= PublicInbox::Linkify->new;
+ # $p - from each_part: [ Email::MIME-like, depth, $idx ]
+ my ($part, $depth, $idx) = @$p;
+ my $ct = $part->content_type || 'text/plain';
my $fn = $part->filename;
+ my ($s, $err) = msg_part_text($part, $ct);
+ return attach_link($ctx, $ct, $p, $fn) unless defined $s;
- if (defined $ct && $ct =~ m!\btext/x?html\b!i) {
- return attach_link($upfx, $ct, $p, $fn);
+ my $rv = $ctx->{obuf};
+ if ($part->{is_submsg}) {
+ submsg_hdr($ctx, $part);
+ $$rv .= "\n";
}
- my $s = eval { $part->body_str };
-
- # badly-encoded message? tell the world about it!
- return attach_link($upfx, $ct, $p, $fn) if $@;
-
- my @lines = split(/^/m, $s);
- $s = '';
- if (defined($fn) || $depth > 0) {
- $s .= attach_link($upfx, $ct, $p, $fn);
- $s .= "\n";
+ # 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} = $spfx;
+ };
+
+ # split off quoted and unquoted blocks:
+ my @sections = PublicInbox::MsgIter::split_quotes($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);
+ $$rv .= "\n";
}
- my @quot;
- my $l = PublicInbox::Linkify->new;
- while (defined(my $cur = shift @lines)) {
- if ($cur !~ /^>/) {
- # show the previously buffered quote inline
- flush_quote(\$s, $l, \@quot) if @quot;
-
- # regular line, OK
- $cur = $l->linkify_1($cur);
- $cur = ascii_html($cur);
- $s .= $l->linkify_2($cur);
+ delete $part->{bdy}; # save memory
+ foreach my $cur (@sections) {
+ if ($cur =~ /\A>/) {
+ # we use a here to allow users to specify
+ # their own color for quoted text
+ $$rv .= qq();
+ $$rv .= $l->to_html($cur);
+ $$rv .= '';
+ } elsif ($diff) {
+ flush_diff($ctx, \$cur);
} else {
- push @quot, $cur;
+ # regular lines, OK
+ $$rv .= $l->to_html($cur);
}
+ undef $cur; # free memory
}
+}
- my $end = "\n";
- if (@quot) {
- $end = '';
- flush_quote(\$s, $l, \@quot);
- }
- $s =~ s/[ \t]+$//sgm; # kill per-line trailing whitespace
- $s =~ s/\A\n+//s; # kill leading blank lines
- $s =~ s/\s+\z//s; # kill all trailing spaces
- $s .= $end;
-}
-
-sub _msg_html_prepare {
- my ($hdr, $ctx) = @_;
- my $srch = $ctx->{srch} if $ctx;
- my $atom = '';
- my $rv = ""; # anchor for body start
-
- if ($srch) {
- $ctx->{-upfx} = '../';
- }
- my @title;
- my $mid = $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;
- $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;
- }
+sub _msg_page_prepare_obuf {
+ my ($eml, $ctx) = @_;
+ my $over = $ctx->{ibx}->over;
+ my $obfs_ibx = $ctx->{-obfs_ibx};
+ my $rv = '';
+ 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;
}
- $v = $v->as_html;
- $v =~ s/(\@[^,]+,) /$1\n\t/g if ($h eq 'Cc' || $h eq 'To');
- $rv .= "$h: $v\n";
-
+ $rv .=
+"WARNING: multiple messages have this Message-ID\n
";
+ $rv .= '';
+ } else {
+ $ctx->{first_hdr} = $eml->header_obj;
+ $ctx->{chash} = content_hash($eml) if $ctx->{smsg}; # reused MID
+ $rv .= ""; # anchor for body start
+ }
+ $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 '';
+ }
+ 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 '';
+ }
+ }
+ my @subj = $eml->header('Subject');
+ if (@subj) {
+ my $v = ascii_html(shift @subj);
+ obfuscate_addrs($obfs_ibx, $v) if $obfs_ibx;
+ $rv .= 'Subject: ';
+ $rv .= $over ? qq($v\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() if $over;
+ $title[0] = '(no subject)';
+ }
+ 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[thread overview]\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 .= "(raw)\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;
}
- $title[0] ||= '(no subject)';
- $ctx->{-title_html} = join(' - ', @title);
- $rv .= 'Message-ID: <' . $mid->as_html . '> ';
- $rv .= "(raw)\n";
- $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 = 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);
+sub SKEL_EXPAND () {
+ qq(expand[flat) .
+ qq(|nested] ) .
+ qq(mbox.gz ) .
+ qq(Atom feed);
+}
+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);
- $$dst .= "\nThread overview: ";
+ $$skel .= "\nThread overview: ";
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 {
- $$dst .= "[no followups, yet] ($expand)\n";
+ $$skel .= "[no followups] ".
+ SKEL_EXPAND."\n";
}
$ctx->{next_msg} = undef;
$ctx->{parent_msg} = $parent;
return;
}
- $$dst .= "$nr+ messages in thread ($expand";
- $$dst .= qq! / [top])\n!;
+ $$skel .= $nr;
+ $$skel .= '+ messages / '.SKEL_EXPAND.qq! top\n!;
- my $subj = $srch->subject_path($hdr->header('Subject'));
- $ctx->{seen} = { $subj => 1 };
+ # 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} = $dst;
- walk_thread(thread_results(load_results($sres)), $ctx, *skel_dump);
+ $ctx->{skel} = $skel;
+
+ # reduce hash lookups in skel_dump
+ $ctx->{-obfs_ibx} = $ibx->{obfuscate} ? $ibx : undef;
+ walk_thread(thread_results($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->as_href;
- $rv .= "In-Reply-To: <";
- $rv .= "$html>\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: <";
+ $rv .= "$html>\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";
}
$rv;
}
-sub squote_maybe ($) {
- 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 $irt = uri_escape_utf8($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";
- $href =~ s/%20/+/g;
-
- (\@arg, ascii_html($href));
-}
-
+# returns a string buffer
sub html_footer {
- my ($hdr, $standalone, $ctx, $rhref) = @_;
-
- my $srch = $ctx->{srch} if $ctx;
+ my ($ctx, $hdr) = @_;
+ my $ibx = $ctx->{ibx};
my $upfx = '../';
- my $tpfx = '';
- my $idx = $standalone ? " index" : '';
- my $irt = '';
- if ($idx && $srch) {
- $idx .= "\n";
- thread_skel(\$idx, $ctx, $hdr, $tpfx);
+ my $skel;
+ my $rv = '';
+ if ($ibx->over) {
+ my $t = ts2str($ctx->{-t_max});
+ my $t_fmt = fmt_ts($ctx->{-t_max});
+ $skel .= <~$t_fmt UTC|newest]
+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)->as_href;
+ $n = mid_href($n);
$next = "next";
}
my $u;
my $par = $ctx->{parent_msg};
if ($par) {
- $u = PublicInbox::Hval->new_msgid($par)->as_href;
+ $u = mid_href($par);
$u = "$upfx$u/";
}
if (my $p = $ctx->{prev_msg}) {
- $prev = PublicInbox::Hval->new_msgid($p)->as_href;
+ $prev = mid_href($p);
if ($p && $par && $p eq $par) {
$prev = "prev parent';
@@ -673,62 +860,85 @@ sub html_footer {
} elsif ($u) { # unlikely
$parent = " parent";
}
- $irt = "$next $prev$parent ";
- } else {
- $irt = '';
+ $rv .= "$next $prev$parent ";
+ } else { # unindexed inboxes w/o over
+ $skel = qq( latest);
}
- $rhref ||= '#R';
- $irt .= qq(reply);
- $irt .= $idx;
+ $rv .= qq(reply);
+ $rv .= $skel;
+ $rv .= '
';
+ $rv .= msg_reply($ctx, $hdr);
}
-sub linkify_ref_nosrch {
- my $v = PublicInbox::Hval->new_msgid($_[0]);
- my $html = $v->as_html;
- my $href = $v->as_href;
+sub linkify_ref_no_over {
+ my ($mid) = @_;
+ my $href = mid_href($mid);
+ my $html = ascii_html($mid);
"<$html>";
}
-sub anchor_for {
- my ($msgid) = @_;
- 'm' . id_compress($msgid, 1);
-}
-
sub ghost_parent {
my ($upfx, $mid) = @_;
- # 'subject dummy' is used internally by Mail::Thread
- return '[no common parent]' if ($mid eq 'subject dummy');
- $mid = PublicInbox::Hval->new_msgid($mid);
- my $href = $mid->as_href;
- my $html = $mid->as_html;
+ my $href = mid_href($mid);
+ my $html = ascii_html($mid);
qq{[parent not found: <$html>]};
}
sub indent_for {
my ($level) = @_;
- INDENT x ($level - 1);
+ $level ? INDENT x ($level - 1) : '';
}
-sub load_results {
- my ($sres) = @_;
-
- [ map { $_->mini_mime } @{delete $sres->{msgs}} ];
+sub find_mid_root {
+ my ($ctx, $level, $node, $idx) = @_;
+ ++$ctx->{root_idx} if $level == 0;
+ if ($node->{mid} eq $ctx->{mid}) {
+ $ctx->{found_mid_at} = $ctx->{root_idx};
+ return 0; # stop iterating
+ }
+ 1;
}
-sub msg_timestamp {
- my ($hdr) = @_;
- my $ts = eval { str2time($hdr->header('Date')) };
- defined($ts) ? $ts : 0;
+sub strict_loose_note ($) {
+ 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";
+ }
+ $msg;
}
sub thread_results {
- my ($msgs) = @_;
- require PublicInbox::Thread;
- my $th = PublicInbox::Thread->new(@$msgs);
- $th->thread;
- $th->order(*sort_ts);
- $th
+ my ($ctx, $msgs) = @_;
+ require PublicInbox::SearchThread;
+ 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);
+ }
+ }
+ $rootset
}
sub missing_thread {
@@ -737,22 +947,52 @@ sub missing_thread {
PublicInbox::ExtMsg::ext_msg($ctx);
}
-sub _msg_date {
- my ($hdr) = @_;
- my $ts = $hdr->header('X-PI-TS') || msg_timestamp($hdr);
- fmt_ts($ts);
-}
+sub dedupe_subject {
+ my ($prev_subj, $subj, $val) = @_;
-sub fmt_ts { POSIX::strftime('%Y-%m-%d %k:%M', gmtime($_[0])) }
+ 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) {
+ $omit = undef;
+ push @$prev_subj, @prev_pop;
+ push @$subj, @curr_pop;
+ }
+ $omit // '';
+}
-sub _skel_header {
- my ($ctx, $hdr, $level) = @_;
+sub skel_dump { # walk_thread callback
+ my ($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 $mid = mid_clean($hdr->header_raw('Message-ID'));
- my $f = ascii_html($hdr->header('X-PI-From'));
- my $d = _msg_date($hdr) . ' ' . indent_for($level) . th_pfx($level);
+ my $mid = $smsg->{mid};
+
+ if ($level == 0 && $ctx->{skel_dump_roots}++) {
+ $$skel .= delete($ctx->{sl_note}) || '';
+ }
+
+ my $f = ascii_html(delete $smsg->{from_name});
+ my $obfs_ibx = $ctx->{-obfs_ibx};
+ obfuscate_addrs($obfs_ibx, $f) if $obfs_ibx;
+
+ my $d = fmt_ts($smsg->{ds});
+ my $unmatched; # if lazy-loaded by SearchThread::Msg::visible()
+ if (exists $ctx->{searchview}) {
+ if (defined(my $pct = $smsg->{pct})) {
+ $d .= (sprintf(' % 2u', $pct) . '%');
+ } else {
+ $unmatched = 1;
+ $d .= ' ';
+ }
+ }
+ $d .= ' ' . indent_for($level) . th_pfx($level);
my $attr = $f;
$ctx->{first_level} ||= $level;
@@ -764,9 +1004,9 @@ sub _skel_header {
if ($cur) {
if ($cur eq $mid) {
delete $ctx->{cur};
- $$dst .= "$d".
+ $$skel .= "$d".
"$attr [this message]\n";
- return;
+ return 1;
} else {
$ctx->{prev_msg} = $mid;
}
@@ -777,136 +1017,131 @@ sub _skel_header {
# Subject is never undef, this mail was loaded from
# our Xapian which would've resulted in '' if it were
# really missing (and Filter rejects empty subjects)
- my $s = $hdr->header('Subject');
- my $h = $ctx->{srch}->subject_path($s);
- if ($ctx->{seen}->{$h}) {
- $s = undef;
+ my @subj = split(/ /, subject_normalized($smsg->{subject}));
+ # remove common suffixes from the subject if it matches the previous,
+ # so we do not show redundant text at the end.
+ my $prev_subj = $ctx->{prev_subj} || [];
+ $ctx->{prev_subj} = [ @subj ];
+ my $omit = dedupe_subject($prev_subj, \@subj, '" ');
+ my $end;
+ if (@subj) {
+ my $subj = join(' ', @subj);
+ $subj = ascii_html($subj);
+ obfuscate_addrs($obfs_ibx, $subj) if $obfs_ibx;
+ $end = "$subj $omit$f\n"
} else {
- $ctx->{seen}->{$h} = 1;
- $s = PublicInbox::Hval->new($s);
- $s = $s->as_html;
+ $end = "$f\n";
}
- my $m = PublicInbox::Hval->new_msgid($mid);
+ my $m;
my $id = '';
- my $mapping = $ctx->{mapping};
- my $end = defined($s) ? "$s $f\n" : "$f\n";
+ my $mapping = $unmatched ? undef : $ctx->{mapping};
if ($mapping) {
my $map = $mapping->{$mid};
$id = id_compress($mid, 1);
$m = '#m'.$id;
- $map->[1] = "$d$end";
+ $map->[0] = "$d$end";
$id = "\nid=r".$id;
} else {
- $m = $ctx->{-upfx}.$m->as_href.'/';
+ $m = $ctx->{-upfx}.mid_href($mid).'/';
}
- $$dst .= $d . "" . $end;
+ $$skel .= $d . "" . $end;
+ 1;
}
-sub skel_dump {
+sub _skel_ghost {
my ($ctx, $level, $node) = @_;
- if (my $mime = $node->message) {
- _skel_header($ctx, $mime->header_obj, $level);
+
+ my $mid = $node->{mid};
+ my $d = ' [not found] ';
+ $d .= ' ' if exists $ctx->{searchview};
+ $d .= indent_for($level) . th_pfx($level);
+ my $upfx = $ctx->{-upfx};
+ my $href = $upfx . mid_href($mid) . '/';
+ my $html = ascii_html($mid);
+
+ my $mapping = $ctx->{mapping};
+ my $map = $mapping->{$mid} if $mapping;
+ if ($map) {
+ my $id = id_compress($mid, 1);
+ $map->[0] = $d . qq{<$html>\n};
+ $d .= qq{<$html>\n};
} else {
- my $mid = $node->messageid;
- my $dst = $ctx->{dst};
- my $mapping = $ctx->{mapping};
- my $map = $mapping->{$mid} if $mapping;
- if ($mid eq 'subject dummy') {
- my $ncp = "\t[no common parent]\n";
- $map->[1] = $ncp if $map;
- $$dst .= $ncp;
- return;
- }
- my $d = $ctx->{pct} ? ' [irrelevant] ' # search result
- : ' [not found] ';
- $d .= indent_for($level) . th_pfx($level);
- my $upfx = $ctx->{-upfx};
- my $m = PublicInbox::Hval->new_msgid($mid);
- my $href = $upfx . $m->as_href . '/';
- my $html = $m->as_html;
-
- if ($map) {
- my $id = id_compress($mid, 1);
- $map->[1] = $d . qq{<$html>\n};
- $d .= qq{<$html>\n};
- } else {
- $d .= qq{<$html>\n};
- }
- $$dst .= $d;
+ $d .= qq{<$html>\n};
}
+ ${$ctx->{skel}} .= $d;
+ 1;
}
-sub sort_ts {
- sort {
- (eval { $a->topmost->message->header('X-PI-TS') } || 0) <=>
- (eval { $b->topmost->message->header('X-PI-TS') } || 0)
- } @_;
-}
-
-sub _tryload_ghost ($$) {
- my ($srch, $mid) = @_;
- my $smsg = $srch->lookup_mail($mid) or return;
- $smsg->mini_mime;
+sub sort_ds {
+ @{$_[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->messageid;
- my $x = $node->message || _tryload_ghost($srch, $mid);
- my ($subj, $ts);
- my $topic;
- if ($x) {
- $x = $x->header_obj;
- $subj = $x->header('Subject') || '';
- $subj = $srch->subject_normalized($subj);
- $ts = $x->header('X-PI-TS');
- if ($level == 0) {
- $topic = [ $ts, 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
- $topic->[0] = $ts if $ts > $topic->[0];
- $topic->[1]++;
+ # continue existing topic
+ my $topic = $ctx->{-cur_topic}; # should never be undef
+ $topic->[0] = $ds if $ds > $topic->[0];
+ $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 $ctx->{order}; # [ ts, subj1, subj2, subj3, ... ]
- if (!@$order) {
+ my $order = delete $ctx->{order}; # [ ds, subj1, subj2, subj3, ... ]
+ unless ($order) {
$ctx->{-html_tip} = '[No topics in range]
';
return 404;
}
my @out;
+ my $ibx = $ctx->{ibx};
+ my $obfs_ibx = $ibx->{obfuscate} ? $ibx : undef;
# 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, $top_subj, @extra) = @$topic;
@$topic = ();
- next unless defined $top; # ghost topic
- my $mid = delete $seen->{$top};
- my $href = PublicInbox::Hval->new_msgid($mid)->as_href;
- $top = PublicInbox::Hval->new($top)->as_html;
- $ts = fmt_ts($ts);
+ 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,
# just the number of posts in the current results window
@@ -919,18 +1154,22 @@ sub dump_topics {
$anchor = '#t'; # thread skeleton
}
- my $mbox = qq(mbox.gz);
- my $atom = qq(Atom);
- my $s = "$top\n" .
- " $ts 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;
- $href = PublicInbox::Hval->new_msgid($mid)->as_href;
+ my $s = "$top_subj\n" .
+ " $ds UTC $n\n";
+ for (my $i = 0; $i < scalar(@extra); $i += 2) {
+ my $level = $extra[$i];
+ my $subj = $extra[$i + 1]; # already normalized
+ $mid = delete $seen->{$subj};
+ my @subj = split(/ /, $subj);
+ my @next_prev = @subj; # full copy
+ my $omit = dedupe_subject($prev_subj, \@subj, ' "');
+ $prev_subj = \@next_prev;
+ $subj = join(' ', @subj);
+ $subj = ascii_html($subj);
+ obfuscate_addrs($obfs_ibx, $subj) if $obfs_ibx;
+ $href = mid_href($mid);
$s .= indent_for($level) . TCHILD;
- $s .= "$sub\n";
+ $s .= qq($subj$omit\n);
}
push @out, $s;
}
@@ -938,45 +1177,75 @@ sub dump_topics {
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);
+}
- my $next_o = $ctx->{-next_o};
- if ($next_o) {
- $next = qq!next!;
+sub pagination_footer ($$) {
+ my ($ctx, $latest) = @_;
+ my $next = $ctx->{next_page} || '';
+ my $prev = $ctx->{prev_page} || '';
+ if ($prev) { # aligned padding for: 'next (older) | '
+ $next = $next ? "$next | " : ' | ';
+ $prev .= qq[ | latest];
}
- if (my $cur_o = $ctx->{-cur_o}) {
- $latest = qq! latest!;
+ "
page: $next$prev
";
+}
- my $o = $cur_o - ($next_o - $cur_o);
- if ($o > 0) {
- $prev = qq!prev!;
- } elsif ($o == 0) {
- $prev = qq!prev!;
+sub paginate_recent ($$) {
+ 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 $ibx = $ctx->{ibx};
+ 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);
+ $ctx->{next_page} = qq[] .
+ 'next (older)';
+ }
+ if (defined($newest) && (defined($before) || defined($after))) {
+ my $s = ts2str($newest);
+ $ctx->{prev_page} = qq[] .
+ 'prev (newer)';
+ }
+ $msgs;
}
+# GET /$INBOX - top-level inbox view for indexed inboxes
sub index_topics {
my ($ctx) = @_;
- my ($off) = (($ctx->{qp}->{o} || '0') =~ /(\d+)/);
- my $opts = { offset => $off, limit => 200 };
-
- $ctx->{order} = [];
- my $sres = $ctx->{srch}->query('', $opts);
- my $nr = scalar @{$sres->{msgs}};
- if ($nr) {
- $sres = load_results($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($ctx), *index_nav);
+ html_oneshot($ctx, dump_topics($ctx), \pagination_footer($ctx, '.'));
+
}
sub thread_adj_level {
@@ -1004,9 +1273,10 @@ sub thread_adj_level {
}
sub ghost_index_entry {
- my ($ctx, $level, $mid) = @_;
+ my ($ctx, $level, $node) = @_;
my ($beg, $end) = thread_adj_level($ctx, $level);
- $beg . ''. ghost_parent($ctx->{-upfx}, $mid) . '
' . $end;
+ $beg . ''. ghost_parent($ctx->{-upfx}, $node->{mid} // '?')
+ . '
' . $end;
}
1;