-# Copyright (C) 2014-2015 all contributors <meta@public-inbox.org>
-# License: AGPLv3 or later (https://www.gnu.org/licenses/agpl-3.0.txt)
+# Copyright (C) 2014-2021 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 URI::Escape qw/uri_escape_utf8/;
-use Date::Parse qw/str2time/;
-use Encode qw/find_encoding/;
-use Encode::MIME::Header;
-use Email::MIME::ContentType qw/parse_content_type/;
-use PublicInbox::Hval;
-use PublicInbox::MID qw/mid_clean id_compress mid2path/;
-use Digest::SHA qw/sha1_hex/;
-my $SALT = rand;
-require POSIX;
-
-# TODO: make these constants tunable
-use constant MAX_INLINE_QUOTED => 12; # half an 80x24 terminal
-use constant MAX_TRUNC_LEN => 72;
-use constant T_ANCHOR => '#u';
+use v5.10.1;
+use List::Util qw(max);
+use PublicInbox::MsgTime qw(msg_datestamp);
+use PublicInbox::Hval qw(ascii_html obfuscate_addrs prurl mid_href
+ ts2str fmt_ts);
+use PublicInbox::Linkify;
+use PublicInbox::MID qw(id_compress mids mids_for_index references
+ $MID_EXTRACT);
+use PublicInbox::MsgIter;
+use PublicInbox::Address;
+use PublicInbox::WwwStream qw(html_oneshot);
+use PublicInbox::Reply;
+use PublicInbox::ViewDiff qw(flush_diff);
+use PublicInbox::Eml;
+use Time::Local qw(timegm);
+use PublicInbox::Smsg qw(subject_normalized);
+use PublicInbox::ContentHash qw(content_hash);
+use constant COLS => 72;
use constant INDENT => ' ';
+use constant TCHILD => '` ';
+sub th_pfx ($) { $_[0] == 0 ? '' : TCHILD };
+
+sub msg_page_i {
+ my ($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
+ $ctx->{smsg}; # may be undef
+ }
+}
+
+# /$INBOX/$MSGID/ for unindexed v1 inboxes
+sub no_over_html ($) {
+ my ($ctx) = @_;
+ 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>';
+ }
+ delete $ctx->{obuf};
+ eval { $$obuf .= html_footer($ctx, $eml) };
+ html_oneshot($ctx, 200, $obuf);
+}
-*ascii_html = *PublicInbox::Hval::ascii_html;
+# public functions: (unstable)
-my $enc_utf8 = find_encoding('UTF-8');
+sub msg_page {
+ my ($ctx) = @_;
+ my $ibx = $ctx->{ibx};
+ $ctx->{-obfs_ibx} = $ibx->{obfuscate} ? $ibx : undef;
+ my $over = $ibx->over or return no_over_html($ctx);
+ my ($id, $prev);
+ my $next_arg = $ctx->{next_arg} = [ $ctx->{mid}, \$id, \$prev ];
-# public functions:
-sub msg_html {
- my ($ctx, $mime, $full_pfx, $footer) = @_;
- if (defined $footer) {
- $footer = "\n" . $footer;
- } else {
- $footer = '';
- }
- my $hdr = $mime->header_obj;
- headers_to_html_header($hdr, $full_pfx, $ctx) .
- multipart_text_as_html($mime, $full_pfx) .
- '</pre><hr /><pre>' .
- html_footer($hdr, 1, $full_pfx, $ctx) .
- $footer .
- '</pre></body></html>';
-}
-
-# /$LISTNAME/$MESSAGE_ID/R/
-sub msg_reply {
- my ($ctx, $hdr, $footer) = @_;
- my $s = $hdr->header('Subject');
- $s = '(no subject)' if (!defined $s) || ($s eq '');
- my $f = $hdr->header('From');
- $f = '' unless defined $f;
- $s = PublicInbox::Hval->new_oneline($s);
- my $mid = $hdr->header('Message-ID');
- $mid = PublicInbox::Hval->new_msgid($mid);
- my $t = $s->as_html;
+ my $smsg = $ctx->{smsg} = $over->next_by_mid(@$next_arg) or
+ return; # undef == 404
+
+ # allow user to easily browse the range around this message if
+ # they have ->over
+ $ctx->{-t_max} = $smsg->{ts};
+ PublicInbox::WwwStream::aresponse($ctx, 200, \&msg_page_i);
+}
+
+# /$INBOX/$MESSAGE_ID/#R
+sub msg_reply ($$) {
+ my ($ctx, $hdr) = @_;
my $se_url =
'https://kernel.org/pub/software/scm/git/docs/git-send-email.html';
+ my $p_url =
+ 'https://en.wikipedia.org/wiki/Posting_style#Interleaved_style';
+
+ my $info = '';
+ my $ibx = $ctx->{ibx};
+ if (my $url = $ibx->{infourl}) {
+ $url = prurl($ctx->{env}, $url);
+ $info = qq(\n List information: <a\nhref="$url">$url</a>\n);
+ }
- my ($arg, $link) = mailto_arg_link($hdr);
- push @$arg, '/path/to/YOUR_REPLY';
+ my ($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) {
+ $link = <<EOF;
- "<html><head><title>replying to \"$t\"</title></head><body><pre>" .
- "replying to message:\n\n" .
- "Subject: <b>$t</b>\n" .
- "From: ". ascii_html($f) .
- "\nDate: " . ascii_html($hdr->header('Date')) .
- "\nMessage-ID: <" . $mid->as_html . ">\n\n" .
- "There are multiple ways to reply:\n\n" .
- "* Save the following mbox file, import it into your mail client,\n" .
- " and reply-to-all from there: <a\nhref=../raw>mbox</a>\n\n" .
- "* Reply to all the recipients using the <b>--to</b>, <b>--cc</b>,\n" .
- " and <b>--in-reply-to</b> switches of git-send-email(1):\n\n" .
- "\tgit send-email \\\n\t\t" .
- join(" \\ \n\t\t", @$arg ). "\n\n" .
- qq( <a\nhref="$se_url">$se_url</a>\n\n) .
- "* If your mail client supports setting the <b>In-Reply-To</b>" .
- " header\n via mailto: links, try the " .
- qq(<a\nhref="$link">mailto: link</a>\n) .
- "\nFor context, the original <a\nhref=../>message</a> or " .
- qq(<a\nhref="../t/#u">thread</a>) .
- '</pre><hr /><pre>' . $footer . '</pre></body></html>';
-}
-
-sub feed_entry {
- my ($class, $mime, $full_pfx) = @_;
-
- # no <head> here for <style>...
- PublicInbox::Hval::PRE .
- multipart_text_as_html($mime, $full_pfx) . '</pre>';
+* If your mail client supports setting the <b>In-Reply-To</b> header
+ via mailto: links, try the <a
+href="$link">mailto: link</a>
+EOF
+ }
+
+ push @$arg, '/path/to/YOUR_REPLY';
+ $arg = ascii_html(join(" \\\n ", '', @$arg));
+ <<EOF
+<hr><pre
+id=R><b>Reply instructions:</b>
+
+You may reply publicly to <a
+href=#t>this message</a> 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: <a
+href=raw>mbox</a>
+
+ Avoid top-posting and favor interleaved quoting:
+ <a
+href="$p_url">$p_url</a>
+$info
+* Reply using the <b>--to</b>, <b>--cc</b>, and <b>--in-reply-to</b>
+ switches of git-send-email(1):
+
+ git send-email$arg
+
+ <a
+href="$se_url">$se_url</a>
+$link</pre>
+EOF
}
sub in_reply_to {
my ($hdr) = @_;
- my $irt = $hdr->header('In-Reply-To');
+ my $refs = references($hdr);
+ $refs->[-1];
+}
- return mid_clean($irt) if (defined $irt);
+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));
+}
- my $refs = $hdr->header('References');
- if ($refs && $refs =~ /<([^>]+)>\s*\z/s) {
- return $1;
- }
- undef;
+sub _hdr_names_html ($$) {
+ my ($hdr, $field) = @_;
+ my @vals = $hdr->header($field) or return '';
+ ascii_html(join(', ', PublicInbox::Address::names(join(',', @vals))));
+}
+
+sub nr_to_s ($$$) {
+ my ($nr, $singular, $plural) = @_;
+ return "0 $plural" if $nr == 0;
+ $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 ($fh, $mime, $level, $state) = @_;
- my $midx = $state->{anchor_idx}++;
- my $ctx = $state->{ctx};
- my $srch = $ctx->{srch};
- my ($prev, $next) = ($midx - 1, $midx + 1);
- my $part_nr = 0;
- my $hdr = $mime->header_obj;
- my $enc = enc_for($hdr->header("Content-Type"));
- my $subj = $hdr->header('Subject');
-
- my $mid_raw = mid_clean($hdr->header('Message-ID'));
- my $id = anchor_for($mid_raw);
- my $seen = $state->{seen};
- $seen->{$id} = "#$id"; # save the anchor for children, later
-
- my $mid = PublicInbox::Hval->new_msgid($mid_raw);
- my $from = PublicInbox::Hval->new_oneline($hdr->header('From'))->raw;
- my @from = Email::Address->parse($from);
- $from = $from[0]->name;
-
- $from = PublicInbox::Hval->new_oneline($from)->as_html;
- $subj = PublicInbox::Hval->new_oneline($subj)->as_html;
- my $root_anchor = $state->{root_anchor} || '';
- my $path = $root_anchor ? '../../' : '';
- my $href = $mid->as_href;
- my $irt = in_reply_to($hdr);
- my $parent_anchor = $seen->{anchor_for($irt)} if defined $irt;
-
- if ($srch) {
- my $t = $ctx->{flat} ? 'T' : 't';
- $subj = "<a\nhref=\"${path}$href/$t/#u\">$subj</a>";
- }
- if ($root_anchor eq $id) {
- $subj = "<u\nid=u>$subj</u>";
- }
-
- my $ts = _msg_date($hdr);
- my $rv = "<pre\nid=s$midx>";
- $rv .= "<b\nid=$id>$subj</b>\n";
- $rv .= "- $from @ $ts UTC - ";
- $rv .= "<a\nhref=\"#s$next\">next</a>";
- if ($prev >= 0) {
- $rv .= "/<a\nhref=\"#s$prev\">prev</a>";
- }
- $fh->write($rv .= "\n\n");
-
- my ($fhref, $more_ref);
- my $mhref = "${path}$href/";
- my $more = 'permalink';
-
- # show full message if it's our root message
- my $neq = $root_anchor ne $id;
- if ($neq || ($neq && $level != 0 && !$ctx->{flat})) {
- $fhref = "${path}$href/f/";
- $more_ref = \$more;
- }
- # scan through all parts, looking for displayable text
- $mime->walk_parts(sub {
- index_walk($fh, $_[0], $enc, \$part_nr, $fhref, $more_ref);
- });
- $mime->body_set('');
-
- my $txt = "${path}$href/raw";
- $rv = "\n<a\nhref=\"$mhref\">$more</a> <a\nhref=\"$txt\">raw</a> ";
- $rv .= html_footer($hdr, 0, undef, $ctx, $mhref);
-
- if (defined $irt) {
- unless (defined $parent_anchor) {
- my $v = PublicInbox::Hval->new_msgid($irt, 1);
- $v = $v->as_href;
- $parent_anchor = "${path}$v/";
+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;
+ $subj = "<u\nid=u>$subj</u>" if $root_anchor eq $id_m;
+ $rv .= $subj . "\n";
+ $rv .= _th_index_lite($mid_raw, \$irt, $id, $ctx);
+ my @tocc;
+ 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_href($mid_raw) . '/';
+ $rv .= qq{ (<a\nhref="$mhref">permalink</a> / };
+ $rv .= qq{<a\nhref="${mhref}raw">raw</a>)\n};
+ 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;
}
- $rv .= " <a\nhref=\"$parent_anchor\">parent</a>";
+ $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($eml)))) {
+ my $href = $upfx . mid_href($irt) . '/';
+ my $html = ascii_html($irt);
+ $rv .= qq(In-Reply-To: <<a\nhref="$href">$html</a>>\n)
}
- if (my $pct = $state->{pct}) { # used by SearchView.pm
- $rv .= " [relevance $pct->{$mid_raw}%]";
- } elsif ($srch) {
+ $rv .= "\n";
+
+ # scan through all parts, looking for displayable text
+ $ctx->{mhref} = $mhref;
+ $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\">permalink</a>" .
+ " <a\nhref=\"${mhref}raw\">raw</a>" .
+ " <a\nhref=\"${mhref}#R\">reply</a>";
+
+ my $hr;
+ 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}) {
- $rv .= " [<a\nhref=\"${path}$href/t/#u\">threaded</a>" .
- "|<b>flat</b>]";
+ $hr = 1;
+ $flat = "<b>$flat</b>";
} else {
- $rv .= " [<b>threaded</b>|" .
- "<a\nhref=\"${path}$href/T/#u\">flat</a>]";
+ $nested = "<b>$nested</b>";
}
+ $rv .= "\t[<a\nhref=\"${mhref}T/#u\">$flat</a>";
+ $rv .= "|<a\nhref=\"${mhref}t/#u\">$nested</a>]";
+ $rv .= " <a\nhref=#r$id>$ctx->{s_nr}</a>";
+ } else {
+ $hr = $ctx->{-hr};
}
- $fh->write($rv .= '</pre>');
+
+ # do we have more messages? start a new <pre> if so
+ $rv .= scalar(@{$ctx->{msgs}}) ? '</pre><hr><pre>' : '</pre>' if $hr;
+ $rv;
}
-sub thread_html {
- my ($ctx, $foot, $srch) = @_;
- sub { emit_thread_html($_[0], $ctx, $foot, $srch) }
+sub pad_link ($$;$) {
+ my ($mid, $level, $s) = @_;
+ $s ||= '...';
+ my $href = defined($mid) ?
+ ("<a\nhref=#r".id_compress($mid, 1).">($s)</a>\n") :
+ "($s)\n";
+ (' 'x19).indent_for($level).th_pfx($level).$href;
}
-# only private functions below.
+sub _skel_hdr {
+ # my ($mapping, $mid) = @_;
+ ($_[0]->{$_[1] // \'bogus'} // [ "(?)\n" ])->[0];
+}
-sub emit_thread_html {
- my ($cb, $ctx, $foot, $srch) = @_;
- my $mid = $ctx->{mid};
- my $res = $srch->get_thread($mid);
- my $msgs = load_results($res);
- my $nr = scalar @$msgs;
- return missing_thread($cb, $ctx) if $nr == 0;
- my $flat = $ctx->{flat};
- my $orig_cb = $cb;
- my $seen = {};
- my $state = {
- ctx => $ctx,
- seen => $seen,
- root_anchor => anchor_for($mid),
- anchor_idx => 0,
- cur_level => 0,
- };
+sub _th_index_lite {
+ my ($mid_raw, $irt, $id, $ctx) = @_;
+ my $rv = '';
+ my $mapping = $ctx->{mapping} or return $rv;
+ my $pad = ' ';
+ 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 $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->[0];
+ if ($idx > 0) {
+ my $prev = $siblings->[$idx - 1];
+ my $pmid = $prev->{mid};
+ if ($idx > 2) {
+ my $s = ($idx - 1). ' preceding siblings ...';
+ $rv .= pad_link($pmid, $level, $s);
+ } elsif ($idx == 2) {
+ $rv .= $pad . _skel_hdr($mapping,
+ $siblings->[0] ?
+ $siblings->[0]->{mid} : undef);
+ }
+ $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 dup subject
+ $attr =~ s!<a\nhref=[^>]+>([^<]+)</a>!$1!s; # no point linking to self
+ $rv .= "<b>@ $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 = $children->[1]) {
+ $rv .= $pad . _skel_hdr($mapping, $cn->{mid});
+ }
+ }
- require PublicInbox::Git;
- my $git = $ctx->{git} ||= PublicInbox::Git->new($ctx->{git_dir});
- if ($flat) {
- pre_anchor_entry($seen, $_) for (@$msgs);
- __thread_entry(\$cb, $git, $state, $_, 0) for (@$msgs);
- } else {
- my $th = thread_results($msgs);
- thread_entry(\$cb, $git, $state, $_, 0) for $th->rootset;
- if (my $max = $state->{cur_level}) {
- $cb->write(('</ul></li>' x ($max - 1)) . '</ul>');
+ 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 = $siblings->[$idx + 2]) {
+ $rv .= $pad . _skel_hdr($mapping, $nn->{mid});
}
}
- $git = undef;
- Email::Address->purge_cache;
+ $rv .= $pad ."<a\nhref=#r$id>$s_s, $s_c; $ctx->{s_nr}</a>\n";
+}
- # there could be a race due to a message being deleted in git
- # but still being in the Xapian index:
- return missing_thread($cb, $ctx) if ($orig_cb eq $cb);
+# 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, $i) or return;
+ ++$level;
+ $i = 0;
+ unshift @q, map { ($level, $_, $i++) } @{$node->{children}};
+ }
+}
- my $final_anchor = $state->{anchor_idx};
- my $next = "<a\nid=s$final_anchor>";
- $next .= $final_anchor == 1 ? 'only message in' : 'end of';
- $next .= " thread</a>, back to <a\nhref=\"../../\">index</a>";
- $next .= "\ndownload thread: ";
- $next .= "<a\nhref=\"../t.mbox.gz\">mbox.gz</a>";
- $next .= " / follow: <a\nhref=\"../t.atom\">Atom feed</a>";
- $cb->write('<hr /><pre>' . $next . "\n\n".
- $foot . '</pre></body></html>');
- $cb->close;
+sub pre_thread { # walk_thread callback
+ my ($ctx, $level, $node, $idx) = @_;
+ $ctx->{mapping}->{$node->{mid}} = [ '', $node, $idx, $level ];
+ skel_dump($ctx, $level, $node);
}
-sub index_walk {
- my ($fh, $part, $enc, $part_nr, $fhref, $more) = @_;
- my $s = add_text_body($enc, $part, $part_nr, $fhref);
+sub thread_eml_entry {
+ my ($ctx, $eml) = @_;
+ my ($beg, $end) = thread_adj_level($ctx, $ctx->{level});
+ $beg . '<pre>' . eml_entry($ctx, $eml) . '</pre>' . $end;
+}
- return if $s eq '';
+sub next_in_queue ($$) {
+ my ($q, $ghost_ok) = @_;
+ while (@$q) {
+ my ($level, $smsg) = splice(@$q, 0, 2);
+ my $cl = $level + 1;
+ unshift @$q, map { ($cl, $_) } @{$smsg->{children}};
+ return ($level, $smsg) if $ghost_ok || exists($smsg->{blob});
+ }
+ undef;
+}
- $s .= "\n"; # ensure there's a trailing newline
+sub stream_thread_i { # PublicInbox::WwwStream::getline callback
+ my ($ctx, $eml) = @_;
+ return thread_eml_entry($ctx, $eml) if $eml;
+ return unless exists($ctx->{skel});
+ my $ghost_ok = $ctx->{nr}++;
+ while (1) {
+ my ($lvl, $smsg) = next_in_queue($ctx->{-queue}, $ghost_ok);
+ if ($smsg) {
+ if (exists $smsg->{blob}) { # next message for cat-file
+ $ctx->{level} = $lvl;
+ if (!$ghost_ok) { # first non-ghost
+ $ctx->{-title_html} =
+ ascii_html($smsg->{subject});
+ $ctx->zmore($ctx->html_top);
+ }
+ return $smsg;
+ }
+ # buffer the ghost entry and loop
+ $ctx->zmore(ghost_index_entry($ctx, $lvl, $smsg));
+ } else { # all done
+ $ctx->zmore(join('', thread_adj_level($ctx, 0)));
+ $ctx->zmore(${delete($ctx->{skel})});
+ return;
+ }
+ }
+}
- $fh->write($s);
+sub stream_thread ($$) {
+ my ($rootset, $ctx) = @_;
+ $ctx->{-queue} = [ map { (0, $_) } @$rootset ];
+ PublicInbox::WwwStream::aresponse($ctx, 200, \&stream_thread_i);
}
-sub enc_for {
- my ($ct, $default) = @_;
- $default ||= $enc_utf8;
- defined $ct or return $default;
- my $ct_parsed = parse_content_type($ct);
- if ($ct_parsed) {
- if (my $charset = $ct_parsed->{attributes}->{charset}) {
- my $enc = find_encoding($charset);
- return $enc if $enc;
+# /$INBOX/$MSGID/t/ and /$INBOX/$MSGID/T/
+sub thread_html {
+ my ($ctx) = @_;
+ $ctx->{-upfx} = '../../';
+ my $mid = $ctx->{mid};
+ 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';
+ $skel .= <<EOF;
+, other threads:[<a
+href="../../?t=$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->{cur_level} = 0;
+ $ctx->{skel} = \$skel;
+ $ctx->{prev_attr} = '';
+ $ctx->{prev_level} = 0;
+ $ctx->{root_anchor} = 'm' . id_compress($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);
+
+ $skel .= '</pre>';
+ return stream_thread($rootset, $ctx) unless $ctx->{flat};
+
+ # flat display: lazy load the full message from smsg
+ $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});
}
+ my $skel = delete($ctx->{skel}) or return; # all done
+ $ctx->zmore($$skel);
+ undef;
}
- $default;
}
sub multipart_text_as_html {
- my ($mime, $full_pfx, $srch) = @_;
- my $rv = "";
- my $part_nr = 0;
- my $enc = enc_for($mime->header("Content-Type"));
+ # ($mime, $ctx) = @_; # each_part may do "$_[0] = undef"
# scan through all parts, looking for displayable text
- $mime->walk_parts(sub {
- my ($part) = @_;
- $part = add_text_body($enc, $part, \$part_nr, $full_pfx, 1);
- $rv .= $part;
- $rv .= "\n" if $part ne '';
- });
- $mime->body_set('');
- $rv;
+ $_[0]->each_part(\&add_text_body, $_[1], 1);
}
-sub add_filename_line {
- my ($enc, $fn) = @_;
- my $len = 72;
- my $pad = "-";
- $fn = $enc->decode($fn);
- $len -= length($fn);
- $pad x= ($len/2) if ($len > 0);
- "$pad " . ascii_html($fn) . " $pad\n";
-}
-
-my $LINK_RE = qr!\b((?:ftp|https?|nntp)://
- [\@:\w\.-]+/
- ?[\@\w\+\&\?\.\%\;/#=-]*)!x;
-
-sub linkify_1 {
- my ($link_map, $s) = @_;
- $s =~ s!$LINK_RE!
- my $url = $1;
- # salt this, as this could be exploited to show
- # links in the HTML which don't show up in the raw mail.
- my $key = sha1_hex($url . $SALT);
- $link_map->{$key} = $url;
- 'PI-LINK-'. $key;
- !ge;
- $s;
-}
-
-sub linkify_2 {
- my ($link_map, $s) = @_;
-
- # Added "PI-LINK-" prefix to avoid false-positives on git commits
- $s =~ s!\bPI-LINK-([a-f0-9]{40})\b!
- my $key = $1;
- my $url = $link_map->{$key};
- if (defined $url) {
- $url = ascii_html($url);
- "<a\nhref=\"$url\">$url</a>";
- } else {
- # false positive or somebody tried to mess with us
- $key;
- }
- !ge;
- $s;
-}
-
-sub flush_quote {
- my ($quot, $n, $part_nr, $full_pfx, $final, $do_anchor) = @_;
-
- # n.b.: do not use <blockquote> since it screws up alignment
- # w.r.t. unquoted text. Repliers may rely on pre-formatted
- # alignment to point out a certain word in quoted text.
- if ($full_pfx) {
- if (!$final && scalar(@$quot) <= MAX_INLINE_QUOTED) {
- # show quote inline
- my %l;
- my $rv = join('', map { linkify_1(\%l, $_) } @$quot);
- @$quot = ();
- $rv = ascii_html($rv);
- return linkify_2(\%l, $rv);
+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: $v\n";
}
+ }
+}
- # show a short snippet of quoted text and link to full version:
- @$quot = map { s/^(?:>\s*)+//gm; $_ } @$quot;
- my $cur = join(' ', @$quot);
- @$quot = split(/\s+/, $cur);
- $cur = '';
- do {
- my $tmp = shift(@$quot);
- my $len = length($tmp) + length($cur);
- if ($len > MAX_TRUNC_LEN) {
- @$quot = ();
- } else {
- $cur .= $tmp . ' ';
- }
- } while (@$quot && length($cur) < MAX_TRUNC_LEN);
- @$quot = ();
- $cur =~ s/ \z/ .../s;
- $cur = ascii_html($cur);
- my $nr = ++$$n;
- "> [<a\nhref=\"$full_pfx#q${part_nr}_$nr\">$cur</a>]\n";
+sub attach_link ($$$$;$) {
+ my ($ctx, $ct, $p, $fn, $err) = @_;
+ my ($part, $depth, $idx) = @$p;
+
+ # Eml iteration clobbers multipart ->{bdy}, so do not offer
+ # downloads for 0-byte multipart attachments
+ return unless $part->{bdy};
+
+ my $nl = $idx eq '1' ? '' : "\n"; # like join("\n", ...)
+ my $size = length($part->body);
+ delete $part->{bdy}; # save memory
+
+ # hide attributes normally, unless we want to aid users in
+ # spotting MUA problems:
+ $ct =~ s/;.*// unless $err;
+ $ct = ascii_html($ct);
+ my $sfn;
+ if (defined $fn && $fn =~ /\A$PublicInbox::Hval::FN\z/o) {
+ $sfn = $fn;
+ } elsif ($ct eq 'text/plain') {
+ $sfn = 'a.txt';
} else {
- # show everything in the full version with anchor from
- # short version (see above)
- my %l;
- my $rv .= join('', map { linkify_1(\%l, $_) } @$quot);
- @$quot = ();
- $rv = ascii_html($rv);
- return linkify_2(\%l, $rv) unless $do_anchor;
- my $nr = ++$$n;
- "<a\nid=q${part_nr}_$nr></a>" . linkify_2(\%l, $rv);
- }
-}
-
-sub add_text_body {
- my ($enc_msg, $part, $part_nr, $full_pfx, $do_anchor) = @_;
- return '' if $part->subparts;
-
- my $ct = $part->content_type;
- # account for filter bugs...
- if (defined $ct && $ct =~ m!\btext/x?html\b!i) {
- $part->body_set('');
- return '';
- }
- my $enc = enc_for($ct, $enc_msg);
- my $n = 0;
- my $nr = 0;
- my $s = $part->body;
- $part->body_set('');
- $s = $enc->decode($s);
- my @lines = split(/^/m, $s);
- $s = '';
-
- if ($$part_nr > 0) {
- my $fn = $part->filename;
- defined($fn) or $fn = "part #" . ($$part_nr + 1);
- $s .= add_filename_line($enc, $fn);
- }
-
- my @quot;
- while (defined(my $cur = shift @lines)) {
- if ($cur !~ /^>/) {
- # show the previously buffered quote inline
- if (scalar @quot) {
- $s .= flush_quote(\@quot, \$n, $$part_nr,
- $full_pfx, 0, $do_anchor);
- }
+ $sfn = 'a.bin';
+ }
+ my $rv = $ctx->{obuf};
+ $$rv .= qq($nl<a\nhref="$ctx->{mhref}$idx-$sfn">);
+ if ($err) {
+ $$rv .= <<EOF;
+[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
+EOF
+ }
+ $$rv .= "[-- Attachment #$idx: ";
+ my $ts = "Type: $ct, Size: $size bytes";
+ my $desc = $part->header('Content-Description') // $fn // '';
+ $desc = ascii_html($desc);
+ $$rv .= ($desc eq '') ? "$ts --]" : "$desc --]\n[-- $ts --]";
+ $$rv .= "</a>\n";
- # regular line, OK
- my %l;
- $cur = linkify_1(\%l, $cur);
- $cur = ascii_html($cur);
- $s .= linkify_2(\%l, $cur);
- } else {
- push @quot, $cur;
- }
+ submsg_hdr($ctx, $part) if $part->{is_submsg};
+
+ undef;
+}
+
+sub add_text_body { # callback for each_part
+ my ($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;
+
+ my $rv = $ctx->{obuf};
+ if ($part->{is_submsg}) {
+ submsg_hdr($ctx, $part);
+ $$rv .= "\n";
}
- if (scalar @quot) {
- $s .= flush_quote(\@quot, \$n, $$part_nr, $full_pfx, 1,
- $do_anchor);
- }
- ++$$part_nr;
-
- $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 (final "\n" added if ne '')
- $s;
-}
-
-sub headers_to_html_header {
- my ($hdr, $full_pfx, $ctx) = @_;
- my $srch = $ctx->{srch} if $ctx;
- my $rv = "";
- my @title;
- my $mid = $hdr->header('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_oneline($v);
-
- if ($h eq 'From') {
- my @from = Email::Address->parse($v->raw);
- $title[1] = ascii_html($from[0]->name);
- } elsif ($h eq 'Subject') {
- $title[0] = $v->as_html;
- if ($srch) {
- $rv .= "$h: <b\nid=t>";
- $rv .= $v->as_html . "</b>\n";
- next;
+
+ # 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 = '../../';
+ }
}
}
- $rv .= "$h: " . $v->as_html . "\n";
+ $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";
+ }
+ 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);
+ $$rv .= '</span>';
+ } elsif ($diff) {
+ flush_diff($ctx, \$cur);
+ } else {
+ # regular lines, OK
+ $$rv .= $l->to_html($cur);
+ }
+ undef $cur; # free memory
}
- $rv .= 'Message-ID: <' . $mid->as_html . '> ';
- my $upfx = $full_pfx ? '' : '../';
- $rv .= "(<a\nhref=\"${upfx}raw\">raw</a>)\n";
- my $atom;
- if ($srch) {
- thread_inline(\$rv, $ctx, $hdr, $upfx);
+}
- $atom = qq{<link\nrel=alternate\ntitle="Atom feed"\n} .
- qq!href="${upfx}t.atom"\ntype="application/atom+xml"/>!;
+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;
+ }
+ $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
+ }
+ $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 .= $over ? qq(<a\nhref="#r"\nid=t>$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)';
+ }
+ 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 {
- $rv .= _parent_headers_nosrch($hdr);
- $atom = '';
+ # 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;
}
+ $rv .= _parent_headers($eml, $over);
$rv .= "\n";
-
- ("<html><head><title>". join(' - ', @title) . "</title>$atom".
- PublicInbox::Hval::STYLE . "</head><body><pre>" . $rv);
+ \$rv;
}
-sub thread_inline {
- my ($dst, $ctx, $hdr, $upfx) = @_;
- my $srch = $ctx->{srch};
- my $mid = mid_clean($hdr->header('Message-ID'));
- my $res = $srch->get_thread($mid);
- my $nr = $res->{total};
- my $expand = "<a\nhref=\"${upfx}t/#u\">expand</a> " .
- "/ <a\nhref=\"${upfx}t.mbox.gz\">mbox.gz</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>);
+}
- $$dst .= 'Thread: ';
+sub thread_skel ($$$) {
+ 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);
+ $$skel .= "\n<b>Thread overview: </b>";
if ($nr <= 1) {
if (defined $parent) {
- $$dst .= "($expand)\n ";
- $$dst .= ghost_parent("$upfx../", $parent) . "\n";
+ $$skel .= SKEL_EXPAND."\n ";
+ $$skel .= ghost_parent('../', $parent) . "\n";
} else {
- $$dst .= "[no followups, yet] ($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";
- if ($nr > MAX_INLINE_QUOTED) {
- $$dst .= qq! / <a\nhref="#b">[scroll down]</a>!;
- }
- $$dst .= ")\n";
-
- my $subj = $srch->subject_path($hdr->header('Subject'));
- my $state = {
- seen => { $subj => 1 },
- srch => $srch,
- cur => $mid,
- parent_cmp => defined $parent ? $parent : '',
- parent => $parent,
- prev_attr => '',
- prev_level => 0,
- };
- for (thread_results(load_results($res))->rootset) {
- inline_dump($dst, $state, $upfx, $_, 0);
- }
- $$dst .= "<a\nid=b></a>"; # anchor for body start
- $ctx->{next_msg} = $state->{next_msg};
- $ctx->{parent_msg} = $state->{parent};
-}
+ $$skel .= $nr;
+ $$skel .= '+ messages / '.SKEL_EXPAND.qq! <a\nhref="#b">top</a>\n!;
-sub _parent_headers_nosrch {
- my ($hdr) = @_;
- my $rv = '';
+ # 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->{skel} = $skel;
- my $irt = in_reply_to($hdr);
- if (defined $irt) {
- my $v = PublicInbox::Hval->new_msgid($irt, 1);
- my $html = $v->as_html;
- my $href = $v->as_href;
- $rv .= "In-Reply-To: <";
- $rv .= "<a\nhref=\"../$href/\">$html</a>>\n";
- }
-
- my $refs = $hdr->header('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);
- }
+ # reduce hash lookups in skel_dump
+ $ctx->{-obfs_ibx} = $ibx->{obfuscate} ? $ibx : undef;
+ walk_thread(thread_results($ctx, $msgs), $ctx, \&skel_dump);
- if (@refs) {
- $rv .= 'References: '. join(' ', @refs) . "\n";
- }
- }
- $rv;
+ $ctx->{parent_msg} = $parent;
}
-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 = Email::Address->parse($v);
- foreach my $recip (@addrs) {
- my $address = $recip->address;
- my $dst = lc($address);
- $cc{$dst} ||= $address;
- $to ||= $dst;
+sub _parent_headers {
+ my ($hdr, $over) = @_;
+ my $rv = '';
+ 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";
}
}
- Email::Address->purge_cache;
- my @arg;
- my $subj = $hdr->header('Subject') || '';
- $subj = "Re: $subj" unless $subj =~ /\bRe:/i;
- my $mid = $hdr->header('Message-ID');
- push @arg, "--in-reply-to='" . ascii_html($mid) . "'";
- my $irt = uri_escape_utf8($mid);
- delete $cc{$to};
- push @arg, '--to=' . ascii_html($to);
- $to = uri_escape_utf8($to);
- $subj = uri_escape_utf8($subj);
- my $cc = join(',', sort values %cc);
- push @arg, '--cc=' . ascii_html($cc);
- $cc = uri_escape_utf8($cc);
- my $href = "mailto:$to?In-Reply-To=$irt&Cc=${cc}&Subject=$subj";
- $href =~ s/%20/+/g;
+ # do not display References: if search is present,
+ # we show the thread skeleton at the bottom, instead.
+ return $rv if $over;
- (\@arg, $href);
+ $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 ($mime, $standalone, $full_pfx, $ctx, $mhref) = @_;
-
- my $srch = $ctx->{srch} if $ctx;
- my $upfx = $full_pfx ? '../' : '../../';
- my $tpfx = $full_pfx ? '' : '../';
- my $idx = $standalone ? " <a\nhref=\"$upfx\">index</a>" : '';
- my $irt = '';
-
- if ($srch && $standalone) {
- $idx .= qq{ / follow: <a\nhref="${tpfx}t.atom">Atom feed</a>\n};
- }
- if ($idx && $srch) {
- my $p = $ctx->{parent_msg};
- my $next = $ctx->{next_msg};
- if ($p) {
- $p = PublicInbox::Hval->new_oneline($p);
- $p = $p->as_href;
- $irt = "<a\nhref=\"$upfx$p/\">parent</a> ";
- } else {
- $irt = ' ' x length('parent ');
+ my ($ctx, $hdr) = @_;
+ my $ibx = $ctx->{ibx};
+ my $upfx = '../';
+ my $skel;
+ my $rv = '<pre>';
+ if ($ibx->over) {
+ my $t = ts2str($ctx->{-t_max});
+ my $t_fmt = fmt_ts($ctx->{-t_max});
+ $skel .= <<EOF;
+ 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 = mid_href($n);
+ $next = "<a\nhref=\"$upfx$n/\"\nrel=next>next</a>";
}
- if ($next) {
- $irt .= "<a\nhref=\"$upfx$next/\">next</a> ";
- } else {
- $irt .= ' ' x length('next ');
+ my $u;
+ my $par = $ctx->{parent_msg};
+ if ($par) {
+ $u = mid_href($par);
+ $u = "$upfx$u/";
}
- if ($p || $next) {
- $irt .= "<a\nhref=\"${tpfx}t/#u\">thread</a> ";
- } else {
- $irt .= ' ' x length('thread ');
+ if (my $p = $ctx->{prev_msg}) {
+ $prev = mid_href($p);
+ if ($p && $par && $p eq $par) {
+ $prev = "<a\nhref=\"$upfx$prev/\"\n" .
+ 'rel=prev>prev parent</a>';
+ $parent = '';
+ } else {
+ $prev = "<a\nhref=\"$upfx$prev/\"\n" .
+ 'rel=prev>prev</a>';
+ $parent = " <a\nhref=\"$u\">parent</a>" if $u;
+ }
+ } elsif ($u) { # unlikely
+ $parent = " <a\nhref=\"$u\"\nrel=prev>parent</a>";
}
- } else {
- $irt = '';
+ $rv .= "$next $prev$parent ";
+ } else { # unindexed inboxes w/o over
+ $skel = qq( <a\nhref="$upfx">latest</a>);
}
-
- $mhref = './' unless defined $mhref;
- $irt . qq(<a\nhref="${mhref}R/">reply</a>) . $idx;
+ $rv .= qq(<a\nhref="#R">reply</a>);
+ $rv .= $skel;
+ $rv .= '</pre>';
+ $rv .= msg_reply($ctx, $hdr);
}
-sub linkify_ref_nosrch {
- my $v = PublicInbox::Hval->new_msgid($_[0], 1);
- 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);
"<<a\nhref=\"../$href/\">$html</a>>";
}
-sub anchor_for {
- my ($msgid) = @_;
- my $id = $msgid;
- if ($id !~ /\A[a-f0-9]{40}\z/) {
- $id = id_compress(mid_clean($id), 1);
- }
- 'm' . $id;
-}
-
-sub thread_html_head {
- my ($cb, $header, $state) = @_;
- $$cb = $$cb->([200, ['Content-Type'=> 'text/html; charset=UTF-8']]);
-
- my $s = PublicInbox::Hval->new_oneline($header->header('Subject'));
- $s = $s->as_html;
- $$cb->write("<html><head><title>$s</title>".
- qq{<link\nrel=alternate\ntitle="Atom feed"\n} .
- qq!href="../t.atom"\ntype="application/atom+xml"/>! .
- PublicInbox::Hval::STYLE .
- "</head><body>");
-}
-
-sub pre_anchor_entry {
- my ($seen, $mime) = @_;
- my $id = anchor_for($mime->header('Message-ID'));
- $seen->{$id} = "#$id"; # save the anchor for children, later
-}
-
sub ghost_parent {
my ($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: <<a\nhref="$upfx$href/">$html</a>>]};
}
-sub thread_adj_level {
- my ($fh, $state, $level) = @_;
-
- my $max = $state->{cur_level};
- if ($level <= 0) {
- return '' if $max == 0; # flat output
-
- # reset existing lists
- my $x = $max > 1 ? ('</ul></li>' x ($max - 1)) : '';
- $fh->write($x . '</ul>');
- $state->{cur_level} = 0;
- return '';
- }
- if ($level == $max) { # continue existing list
- $fh->write('<li>');
- } elsif ($level < $max) {
- my $x = $max > 1 ? ('</ul></li>' x ($max - $level)) : '';
- $fh->write($x .= '<li>');
- $state->{cur_level} = $level;
- } else { # ($level > $max) # start a new level
- $state->{cur_level} = $level;
- $fh->write(($max ? '<li>' : '') . '<ul><li>');
- }
- '</li>';
-}
-
-sub ghost_flush {
- my ($fh, $state, $upfx, $mid, $level) = @_;
-
- my $end = thread_adj_level($fh, $state, $level);
- $fh->write('<pre>'. ghost_parent($upfx, $mid) . '</pre>' . $end);
+sub indent_for {
+ my ($level) = @_;
+ $level ? INDENT x ($level - 1) : '';
}
-sub __thread_entry {
- my ($cb, $git, $state, $mime, $level) = @_;
-
- # lazy load the full message from mini_mime:
- $mime = eval {
- my $path = mid2path(mid_clean($mime->header('Message-ID')));
- Email::MIME->new($git->cat_file('HEAD:'.$path));
- } or return;
-
- if ($state->{anchor_idx} == 0) {
- thread_html_head($cb, $mime, $state, $level);
- }
- my $fh = $$cb;
- if (my $ghost = delete $state->{ghost}) {
- # n.b. ghost messages may only be parents, not children
- foreach my $g (@$ghost) {
- ghost_flush($fh, $state, '../../', @$g);
- }
+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
}
- my $end = thread_adj_level($fh, $state, $level);
- index_entry($fh, $mime, $level, $state);
- $fh->write($end) if $end;
-
1;
}
-sub indent_for {
- my ($level) = @_;
- INDENT x ($level - 1);
-}
+sub strict_loose_note ($) {
+ my ($nr) = @_;
+ my $msg =
+" -- strict thread matches above, loose matches on Subject: below --\n";
-sub __ghost_prepare {
- my ($state, $node, $level) = @_;
- my $ghost = $state->{ghost} ||= [];
- push @$ghost, [ $node->messageid, $level ];
+ if ($nr > PublicInbox::Over::DEFAULT_LIMIT()) {
+ $msg .=
+" -- use mbox.gz link to download all $nr messages --\n";
+ }
+ $msg;
}
-sub thread_entry {
- my ($cb, $git, $state, $node, $level) = @_;
- return unless $node;
- if (my $mime = $node->message) {
- unless (__thread_entry($cb, $git, $state, $mime, $level)) {
- __ghost_prepare($state, $node, $level);
+sub thread_results {
+ 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);
}
- } else {
- __ghost_prepare($state, $node, $level);
}
-
- thread_entry($cb, $git, $state, $node->child, $level + 1);
- thread_entry($cb, $git, $state, $node->next, $level);
+ $rootset
}
-sub load_results {
- my ($res) = @_;
-
- [ map { $_->mini_mime } @{delete $res->{msgs}} ];
+sub missing_thread {
+ my ($ctx) = @_;
+ require PublicInbox::ExtMsg;
+ PublicInbox::ExtMsg::ext_msg($ctx);
}
-sub msg_timestamp {
- my ($hdr) = @_;
- my $ts = eval { str2time($hdr->header('Date')) };
- defined($ts) ? $ts : 0;
-}
+sub dedupe_subject {
+ my ($prev_subj, $subj, $val) = @_;
-sub thread_results {
- my ($msgs, $nosubject) = @_;
- require PublicInbox::Thread;
- my $th = PublicInbox::Thread->new(@$msgs);
- no warnings 'once';
- $Mail::Thread::nosubject = $nosubject;
- $th->thread;
- $th->order(*sort_ts);
- $th
+ 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 missing_thread {
- my ($cb, $ctx) = @_;
- require PublicInbox::ExtMsg;
+sub skel_dump { # walk_thread callback
+ my ($ctx, $level, $smsg) = @_;
+ $smsg->{blob} or return _skel_ghost($ctx, $level, $smsg);
- $cb->(PublicInbox::ExtMsg::ext_msg($ctx))
-}
+ my $skel = $ctx->{skel};
+ my $cur = $ctx->{cur};
+ my $mid = $smsg->{mid};
-sub _msg_date {
- my ($hdr) = @_;
- my $ts = $hdr->header('X-PI-TS') || msg_timestamp($hdr);
- fmt_ts($ts);
-}
-
-sub fmt_ts { POSIX::strftime('%Y-%m-%d %k:%M', gmtime($_[0])) }
+ if ($level == 0 && $ctx->{skel_dump_roots}++) {
+ $$skel .= delete($ctx->{sl_note}) || '';
+ }
-sub _inline_header {
- my ($dst, $state, $upfx, $hdr, $level) = @_;
- my $dot = $level == 0 ? '' : '` ';
+ my $f = ascii_html(delete $smsg->{from_name});
+ my $obfs_ibx = $ctx->{-obfs_ibx};
+ obfuscate_addrs($obfs_ibx, $f) if $obfs_ibx;
- my $cur = $state->{cur};
- my $mid = mid_clean($hdr->header('Message-ID'));
- my $f = $hdr->header('X-PI-From');
- my $d = _msg_date($hdr);
- $f = PublicInbox::Hval->new_oneline($f)->as_html;
- my $pfx = ' ' . $d . ' ' . indent_for($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;
- $state->{first_level} ||= $level;
+ $ctx->{first_level} ||= $level;
- if ($attr ne $state->{prev_attr} || $state->{prev_level} > $level) {
- $state->{prev_attr} = $attr;
- } else {
- $attr = '';
+ if ($attr ne $ctx->{prev_attr} || $ctx->{prev_level} > $level) {
+ $ctx->{prev_attr} = $attr;
}
- $state->{prev_level} = $level;
+ $ctx->{prev_level} = $level;
if ($cur) {
if ($cur eq $mid) {
- delete $state->{cur};
- $$dst .= "$pfx$dot<b><a\nid=r\nhref=\"#b\">".
+ delete $ctx->{cur};
+ $$skel .= "<b>$d<a\nid=r\nhref=\"#t\">".
"$attr [this message]</a></b>\n";
-
- return;
+ return 1;
+ } else {
+ $ctx->{prev_msg} = $mid;
}
} else {
- $state->{next_msg} ||= $mid;
+ $ctx->{next_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 $s = $hdr->header('Subject');
- my $h = $state->{srch}->subject_path($s);
- if ($state->{seen}->{$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</a> $omit$f\n"
} else {
- $state->{seen}->{$h} = 1;
- $s = PublicInbox::Hval->new($s);
- $s = $s->as_html;
- }
- my $m = PublicInbox::Hval->new_msgid($mid);
- $m = $upfx . '../' . $m->as_href . '/';
- if (defined $s) {
- $$dst .= "$pfx$dot<a\nhref=\"$m\">$s</a> $attr\n";
+ $end = "$f</a>\n";
+ }
+ my $m;
+ my $id = '';
+ my $mapping = $unmatched ? undef : $ctx->{mapping};
+ if ($mapping) {
+ my $map = $mapping->{$mid};
+ $id = id_compress($mid, 1);
+ $m = '#m'.$id;
+ $map->[0] = "$d<a\nhref=\"$m\">$end";
+ $id = "\nid=r".$id;
} else {
- $$dst .= "$pfx$dot<a\nhref=\"$m\">$f</a>\n";
+ $m = $ctx->{-upfx}.mid_href($mid).'/';
}
+ $$skel .= $d . "<a\nhref=\"$m\"$id>" . $end;
+ 1;
}
-sub inline_dump {
- my ($dst, $state, $upfx, $node, $level) = @_;
- return unless $node;
- if (my $mime = $node->message) {
- my $hdr = $mime->header_obj;
- my $mid = mid_clean($hdr->header('Message-ID'));
- if ($mid eq $state->{parent_cmp}) {
- $state->{parent} = $mid;
- }
- _inline_header($dst, $state, $upfx, $hdr, $level);
+sub _skel_ghost {
+ my ($ctx, $level, $node) = @_;
+
+ my $mid = $node->{mid};
+ my $d = ' [not found] ';
+ $d .= ' ' if exists $ctx->{searchview};
+ $d .= indent_for($level) . th_pfx($level);
+ my $upfx = $ctx->{-upfx};
+ my $href = $upfx . mid_href($mid) . '/';
+ my $html = ascii_html($mid);
+
+ my $mapping = $ctx->{mapping};
+ my $map = $mapping->{$mid} if $mapping;
+ if ($map) {
+ my $id = id_compress($mid, 1);
+ $map->[0] = $d . qq{<<a\nhref=#r$id>$html</a>>\n};
+ $d .= qq{<<a\nhref="$href"\nid=r$id>$html</a>>\n};
} else {
- my $dot = $level == 0 ? '' : '` ';
- my $pfx = (' ' x length(' 1970-01-01 13:37 ')).
- indent_for($level) . $dot;
- $$dst .= $pfx;
- $$dst .= ghost_parent("$upfx../", $node->messageid) . "\n";
+ $d .= qq{<<a\nhref="$href">$html</a>>\n};
}
- inline_dump($dst, $state, $upfx, $node->child, $level+1);
- inline_dump($dst, $state, $upfx, $node->next, $level);
-}
-
-sub sort_ts {
- sort {
- (eval { $a->topmost->message->header('X-PI-TS') } || 0) <=>
- (eval { $b->topmost->message->header('X-PI-TS') } || 0)
- } @_;
+ ${$ctx->{skel}} .= $d;
+ 1;
}
-sub rsort_ts {
- sort {
- (eval { $b->topmost->message->header('X-PI-TS') } || 0) <=>
- (eval { $a->topmost->message->header('X-PI-TS') } || 0)
- } @_;
+sub sort_ds {
+ @{$_[0]} = sort {
+ (eval { $a->topmost->{ds} } || 0) <=>
+ (eval { $b->topmost->{ds} } || 0)
+ } @{$_[0]};
}
# accumulate recent topics if search is supported
-# returns 1 if done, undef if not
-sub add_topic {
- my ($state, $node, $level) = @_;
- return unless $node;
- my $child_adjust = 1;
+# returns 200 if done, 404 if not
+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 1;
+ }
- if (my $x = $node->message) {
- $x = $x->header_obj;
- my $subj;
+ # 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 (!defined($seen->{$subj})) {
+ push @$topic, $level, $subj; # @extra messages
+ }
+ $seen->{$subj} = $mid; # latest for subject
+ } else { # ghost message
+ 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, ... ]
+ unless ($order) {
+ $ctx->{-html_tip} = '<pre>[No topics in range]</pre>';
+ return 404;
+ }
- $subj = $x->header('Subject');
- $subj = $state->{srch}->subject_normalized($subj);
+ 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 ($ds, $n, $seen, $top_subj, @extra) = @$topic;
+ @$topic = ();
+ 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);
- if (++$state->{subjs}->{$subj} == 1) {
- push @{$state->{order}}, [ $level, $subj ];
+ # $n isn't the total number of posts on the topic,
+ # just the number of posts in the current results window
+ my $anchor;
+ if ($n == 1) {
+ $n = '';
+ $anchor = '#u'; # top of only message
+ } else {
+ $n = " ($n+ messages)";
+ $anchor = '#t'; # thread skeleton
}
- my $mid = mid_clean($x->header('Message-ID'));
-
- my $ts = $x->header('X-PI-TS');
- my $exist = $state->{latest}->{$subj};
- if (!$exist || $exist->[1] < $ts) {
- $state->{latest}->{$subj} = [ $mid, $ts ];
+ my $s = "<a\nhref=\"$href/T/$anchor\">$top_subj</a>\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 .= qq(<a\nhref="$href/T/#u">$subj</a>$omit\n);
}
- } else {
- # ghost message, do not bump level
- $child_adjust = 0;
+ push @out, $s;
}
+ $ctx->{-html_tip} = '<pre>' . join("\n", @out) . '</pre>';
+ 200;
+}
- add_topic($state, $node->child, $level + $child_adjust);
- add_topic($state, $node->next, $level);
+sub str2ts ($) {
+ my ($yyyy, $mon, $dd, $hh, $mm, $ss) = unpack('A4A2A2A2A2A2', $_[0]);
+ timegm($ss || 0, $mm || 0, $hh || 0, $dd, $mon - 1, $yyyy);
}
-sub dump_topics {
- my ($state) = @_;
- my $order = $state->{order};
- my $subjs = $state->{subjs};
- my $latest = $state->{latest};
- return "\n[No topics in range]</pre>" unless (scalar @$order);
- my $dst = '';
- my $pfx;
- my $prev = 0;
- my $prev_attr = '';
- while (defined(my $info = shift @$order)) {
- my ($level, $subj) = @$info;
- my $n = delete $subjs->{$subj};
- my ($mid, $ts) = @{delete $latest->{$subj}};
- $mid = PublicInbox::Hval->new_msgid($mid)->as_href;
- $subj = PublicInbox::Hval->new($subj)->as_html;
- $pfx = indent_for($level);
- my $nl = $level == $prev ? "\n" : '';
- my $dot = $level == 0 ? '' : '` ';
- $dst .= "$nl$pfx$dot<a\nhref=\"$mid/t/#u\"><b>$subj</b></a>\n";
-
- my $attr;
- $ts = fmt_ts($ts);
- $attr = " $ts UTC";
+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>];
+ }
+ "<hr><pre>page: $next$prev</pre>";
+}
- # $n isn't the total number of posts on the topic,
- # just the number of posts in the current results
- # window, so leave it unlabeled
- $n = $n == 1 ? '' : " ($n+ messages)";
-
- if ($level == 0 || $attr ne $prev_attr) {
- my $mbox = qq(<a\nhref="$mid/t.mbox.gz">mbox.gz</a>);
- my $atom = qq(<a\nhref="$mid/t.atom">Atom</a>);
- $pfx .= INDENT if $level > 0;
- $dst .= $pfx . $attr . $n . " - $mbox / $atom\n";
- $prev_attr = $attr;
+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;
}
}
- $dst .= '</pre>';
+ if (defined($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>';
+ }
+ $msgs;
}
-sub emit_index_topics {
- my ($state, $fh) = @_;
- my $off = $state->{ctx}->{cgi}->param('o');
- $off = 0 unless defined $off;
- $state->{order} = [];
- $state->{subjs} = {};
- $state->{latest} = {};
- my $max = 25;
- my %opts = ( offset => int $off, limit => $max * 4 );
- while (scalar @{$state->{order}} < $max) {
- my $res = $state->{srch}->query('', \%opts);
- my $nr = scalar @{$res->{msgs}} or last;
+# GET /$INBOX - top-level inbox view for indexed inboxes
+sub index_topics {
+ my ($ctx) = @_;
+ my $msgs = paginate_recent($ctx, 200); # 200 is our window
+ if (@$msgs) {
+ walk_thread(thread_results($ctx, $msgs), $ctx, \&acc_topic);
+ }
+ html_oneshot($ctx, dump_topics($ctx), \pagination_footer($ctx, '.'));
- for (rsort_ts(thread_results(load_results($res), 1)->rootset)) {
- add_topic($state, $_, 0);
- }
- $opts{offset} += $nr;
+}
+
+sub thread_adj_level {
+ my ($ctx, $level) = @_;
+
+ my $max = $ctx->{cur_level};
+ if ($level <= 0) {
+ return ('', '') if $max == 0; # flat output
+
+ # reset existing lists
+ my $beg = $max > 1 ? ('</ul></li>' x ($max - 1)) : '';
+ $ctx->{cur_level} = 0;
+ ("$beg</ul>", '');
+ } elsif ($level == $max) { # continue existing list
+ qw(<li> </li>);
+ } elsif ($level < $max) {
+ my $beg = $max > 1 ? ('</ul></li>' x ($max - $level)) : '';
+ $ctx->{cur_level} = $level;
+ ("$beg<li>", '</li>');
+ } else { # ($level > $max) # start a new level
+ $ctx->{cur_level} = $level;
+ my $beg = ($max ? '<li>' : '') . '<ul><li>';
+ ($beg, '</li>');
}
+}
- $fh->write(dump_topics($state));
- $opts{offset};
+sub ghost_index_entry {
+ my ($ctx, $level, $node) = @_;
+ my ($beg, $end) = thread_adj_level($ctx, $level);
+ $beg . '<pre>'. ghost_parent($ctx->{-upfx}, $node->{mid} // '?')
+ . '</pre>' . $end;
}
1;