-# Copyright (C) 2014, Eric Wong <normalperson@yhbt.net> and all contributors
-# License: AGPLv3 or later (https://www.gnu.org/licenses/agpl-3.0.txt)
+# Copyright (C) 2014-2018 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 mid_compressed mid2path/;
-use Digest::SHA;
+use PublicInbox::MsgTime qw(msg_datestamp);
+use PublicInbox::Hval qw/ascii_html obfuscate_addrs/;
+use PublicInbox::Linkify;
+use PublicInbox::MID qw/id_compress mid_escape mids references/;
+use PublicInbox::MsgIter;
+use PublicInbox::Address;
+use PublicInbox::WwwStream;
+use PublicInbox::Reply;
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 PRE_WRAP => "<pre\nstyle=\"white-space:pre-wrap\">";
-use constant T_ANCHOR => '#u';
+use constant INDENT => ' ';
+use constant TCHILD => '` ';
+sub th_pfx ($) { $_[0] == 0 ? '' : TCHILD };
-*ascii_html = *PublicInbox::Hval::ascii_html;
+# public functions: (unstable)
-my $enc_utf8 = find_encoding('UTF-8');
-
-# public functions:
sub msg_html {
- my ($class, $mime, $full_pfx, $footer, $srch) = @_;
- if (defined $footer) {
- $footer = "\n" . $footer;
+ my ($ctx, $mime, $more, $smsg) = @_;
+ my $hdr = $mime->header_obj;
+ my $ibx = $ctx->{-inbox};
+ my $obfs_ibx = $ctx->{-obfs_ibx} = $ibx->{obfuscate} ? $ibx : undef;
+ my $tip = _msg_html_prepare($hdr, $ctx, $more, 0);
+ my $end = 2;
+ PublicInbox::WwwStream->response($ctx, 200, sub {
+ my ($nr, undef) = @_;
+ if ($nr == 1) {
+ # $more cannot be true w/o $smsg being defined:
+ my $upfx = $more ? '../'.mid_escape($smsg->mid).'/' : '';
+ $tip . multipart_text_as_html($mime, $upfx, $obfs_ibx) .
+ '</pre><hr>'
+ } elsif ($more && @$more) {
+ ++$end;
+ msg_html_more($ctx, $more, $nr);
+ } elsif ($nr == $end) {
+ # fake an EOF if generating the footer fails;
+ # we want to at least show the message if something
+ # here crashes:
+ eval {
+ '<pre>' . html_footer($hdr, 1, $ctx) .
+ '</pre>' . msg_reply($ctx, $hdr)
+ };
+ } else {
+ undef
+ }
+ });
+}
+
+sub msg_page {
+ my ($ctx) = @_;
+ my $mid = $ctx->{mid};
+ my $ibx = $ctx->{-inbox};
+ my ($first, $more, $head, $tail, $db);
+ my $smsg;
+ if (my $srch = $ibx->search) {
+ $srch->retry_reopen(sub {
+ ($head, $tail, $db) = $srch->each_smsg_by_mid($mid);
+ for (; !defined($first) && $head != $tail; $head++) {
+ my @args = ($head, $db, $mid);
+ $smsg = PublicInbox::SearchMsg->get(@args);
+ $first = $ibx->msg_by_smsg($smsg);
+ }
+ if ($head != $tail) {
+ $more = [ $head, $tail, $db ];
+ }
+ });
+ return unless $first;
} else {
- $footer = '';
+ $first = $ibx->msg_by_mid($mid) or return;
}
- headers_to_html_header($mime, $full_pfx, $srch) .
- multipart_text_as_html($mime, $full_pfx) .
- '</pre><hr /><pre>' .
- html_footer($mime, 1, $full_pfx, $srch) .
- $footer .
- '</pre></body></html>';
+ msg_html($ctx, PublicInbox::MIME->new($first), $more, $smsg);
}
-sub feed_entry {
- my ($class, $mime, $full_pfx) = @_;
-
- PRE_WRAP . multipart_text_as_html($mime, $full_pfx) . '</pre>';
+sub msg_html_more {
+ my ($ctx, $more, $nr) = @_;
+ my $str = eval {
+ my $smsg;
+ my ($head, $tail, $db) = @$more;
+ my $mid = $ctx->{mid};
+ for (; !defined($smsg) && $head != $tail; $head++) {
+ my $m = PublicInbox::SearchMsg->get($head, $db, $mid);
+ $smsg = $ctx->{-inbox}->smsg_mime($m);
+ }
+ if ($head == $tail) { # done
+ @$more = ();
+ } else {
+ $more->[0] = $head;
+ }
+ if ($smsg) {
+ my $mime = $smsg->{mime};
+ my $upfx = '../' . mid_escape($smsg->mid) . '/';
+ _msg_html_prepare($mime->header_obj, $ctx, $more, $nr) .
+ multipart_text_as_html($mime, $upfx,
+ $ctx->{-obfs_ibx}) .
+ '</pre><hr>'
+ } else {
+ '';
+ }
+ };
+ if ($@) {
+ warn "Error lookup up additional messages: $@\n";
+ $str = '<pre>Error looking up additional messages</pre>';
+ }
+ $str;
}
-# this is already inside a <pre>
-# state = [ time, seen = {}, first_commit, page_nr = 0 ]
-sub index_entry {
- my (undef, $mime, $level, $state) = @_;
- my ($srch, $seen, $first_commit) = @$state;
- my $midx = $state->[3]++;
- my ($prev, $next) = ($midx - 1, $midx + 1);
- my $part_nr = 0;
- my $enc = enc_for($mime->header("Content-Type"));
- my $subj = $mime->header('Subject');
- my $header_obj = $mime->header_obj;
-
- my $mid_raw = $header_obj->header('Message-ID');
- my $id = anchor_for($mid_raw);
- $seen->{$id} = "#$id"; # save the anchor for later
-
- my $mid = PublicInbox::Hval->new_msgid($mid_raw);
- my $from = PublicInbox::Hval->new_oneline($mime->header('From'))->raw;
- my @from = Email::Address->parse($from);
- $from = $from[0]->name;
- (defined($from) && length($from)) or $from = $from[0]->address;
-
- $from = PublicInbox::Hval->new_oneline($from)->as_html;
- $subj = PublicInbox::Hval->new_oneline($subj)->as_html;
- my $root_anchor = $seen->{root_anchor};
- my $more = 'permalink';
- my $path = $root_anchor ? '../' : '';
- my $href = $mid->as_href;
- my $irt = $header_obj->header('In-Reply-To');
- my ($anchor_idx, $anchor, $t_anchor);
- if (defined $irt) {
- $anchor_idx = anchor_for($irt);
- $anchor = $seen->{$anchor_idx};
- $t_anchor = T_ANCHOR;
- } else {
- $t_anchor = '';
- }
- if (defined $srch) {
- $subj = "<a\nhref=\"${path}t/$href.html#u\">$subj</a>";
- }
- if ($root_anchor && $root_anchor eq $id) {
- $subj = "<u\nid=\"u\">$subj</u>";
+# /$INBOX/$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->{-inbox};
+ if (my $url = $ibx->{infourl}) {
+ $url = PublicInbox::Hval::prurl($ctx->{env}, $url);
+ $info = qq(\n List information: <a\nhref="$url">$url</a>\n);
}
- my $ts = $mime->header('X-PI-TS');
- unless (defined $ts) {
- $ts = msg_timestamp($mime);
- }
- my $fmt = '%Y-%m-%d %H:%M';
- $ts = POSIX::strftime($fmt, gmtime($ts));
+ my ($arg, $link, $reply_to_all) =
+ PublicInbox::Reply::mailto_arg_link($ibx, $hdr);
- my $rv = "<table\nsummary=l$level><tr>";
- if ($level) {
- $rv .= '<td><pre>' . (' ' x $level) . '</pre></td>';
- }
- $rv .= "<td\nid=s$midx>" . PRE_WRAP;
- $rv .= "<b\nid=\"$id\">$subj</b>\n";
- $rv .= "- by $from @ $ts UTC - ";
- $rv .= "<a\nhref=\"#s$next\">next</a>";
- if ($prev >= 0) {
- $rv .= "/<a\nhref=\"#s$prev\">prev</a>";
- }
- $rv .= "\n\n";
+ # mailto: link only works if address obfuscation is disabled
+ if ($link) {
+ $link = <<EOF;
- my ($fhref, $more_ref);
- my $mhref = "${path}m/$href.html";
- if ($level > 0) {
- $fhref = "${path}f/$href.html";
- $more_ref = \$more;
+* If your mail client supports setting the <b>In-Reply-To</b> header
+ via mailto: links, try the <a
+href="$link">mailto: link</a>
+EOF
}
- # scan through all parts, looking for displayable text
- $mime->walk_parts(sub {
- $rv .= index_walk($_[0], $enc, \$part_nr, $fhref, $more_ref);
- });
- $mime->body_set('');
- $rv .= "\n<a\nhref=\"$mhref\">$more</a> ";
- my $txt = "${path}m/$href.txt";
- $rv .= "<a\nhref=\"$txt\">raw</a> ";
- $rv .= html_footer($mime, 0);
+ push @$arg, '/path/to/YOUR_REPLY';
+ $arg = ascii_html(join(" \\\n ", '', @$arg));
+ <<EOF
+<hr><pre
+id=R><b>Reply instructions:</b>
+
+You may reply publically 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
+}
- if (defined $irt) {
- unless (defined $anchor) {
- my $v = PublicInbox::Hval->new_msgid($irt);
- $v = $v->as_href;
- $anchor = "${path}m/$v.html";
- $seen->{$anchor_idx} = $anchor;
- }
- $rv .= " <a\nhref=\"$anchor\">parent</a>";
- }
+sub in_reply_to {
+ my ($hdr) = @_;
+ my $refs = references($hdr);
+ $refs->[-1];
+}
- if ($srch) {
- $rv .= " <a\nhref=\"${path}t/$href.html$t_anchor\">" .
- "threadlink</a>";
- }
+sub _hdr_names_html ($$) {
+ my ($hdr, $field) = @_;
+ my $val = $hdr->header($field) or return '';
+ ascii_html(join(', ', PublicInbox::Address::names($val)));
+}
- $rv .= '</pre></td></tr></table>';
+sub nr_to_s ($$$) {
+ my ($nr, $singular, $plural) = @_;
+ return "0 $plural" if $nr == 0;
+ $nr == 1 ? "$nr $singular" : "$nr $plural";
}
-sub thread_html {
- my (undef, $ctx, $foot, $srch) = @_;
- my $mid = mid_compressed($ctx->{mid});
- my $res = $srch->get_thread($mid);
- my $rv = '';
- my $msgs = load_results($res);
- my $nr = scalar @$msgs;
- return $rv if $nr == 0;
- my $th = thread_results($msgs);
- my $state = [ $srch, { root_anchor => anchor_for($mid) }, undef, 0 ];
- {
- require PublicInbox::GitCatFile;
- my $git = PublicInbox::GitCatFile->new($ctx->{git_dir});
- thread_entry(\$rv, $git, $state, $_, 0) for $th->rootset;
- }
- my $final_anchor = $state->[3];
- my $next = "<a\nid=\"s$final_anchor\">";
-
- if ($final_anchor == 1) {
- $next .= 'only message in thread';
+# this is already inside a <pre>
+sub index_entry {
+ my ($smsg, $ctx, $more) = @_;
+ my $srch = $ctx->{srch};
+ my $subj = $smsg->subject;
+ my $mid_raw = $smsg->mid;
+ my $id = id_compress($mid_raw, 1);
+ my $id_m = 'm'.$id;
+
+ my $root_anchor = $ctx->{root_anchor} || '';
+ my $irt;
+ my $obfs_ibx = $ctx->{-obfs_ibx};
+
+ my $rv = "<a\nhref=#e$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 $mime = $smsg->{mime};
+ my $hdr = $mime->header_obj;
+ foreach my $f (qw(To Cc)) {
+ my $dst = _hdr_names_html($hdr, $f);
+ if ($dst ne '') {
+ obfuscate_addrs($obfs_ibx, $dst) if $obfs_ibx;
+ push @tocc, "$f: $dst";
+ }
+ }
+ my $from = _hdr_names_html($hdr, 'From');
+ obfuscate_addrs($obfs_ibx, $from) if $obfs_ibx;
+ $rv .= "From: $from @ ".fmt_ts($smsg->ds)." UTC";
+ my $upfx = $ctx->{-upfx};
+ my $mhref = $upfx . mid_escape($mid_raw) . '/';
+ $rv .= qq{ (<a\nhref="$mhref">permalink</a> / };
+ $rv .= qq{<a\nhref="${mhref}raw">raw</a>)\n};
+ $rv .= ' '.join('; +', @tocc) . "\n" if @tocc;
+
+ my $mapping = $ctx->{mapping};
+ if (!$mapping && (defined($irt) || defined($irt = in_reply_to($hdr)))) {
+ my $mirt = PublicInbox::Hval->new_msgid($irt);
+ my $href = $upfx . $mirt->{href}. '/';
+ my $html = $mirt->as_html;
+ $rv .= qq(In-Reply-To: <<a\nhref="$href">$html</a>>\n)
+ }
+ $rv .= "\n";
+
+ # scan through all parts, looking for displayable text
+ msg_iter($mime, sub { $rv .= add_text_body($mhref, $obfs_ibx, $_[0]) });
+
+ # 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 (my $pct = $ctx->{pct}) { # used by SearchView.pm
+ $rv .= "\t[relevance $pct->{$mid_raw}%]";
+ $hr = 1;
+ } elsif ($mapping) {
+ my $nested = 'nested';
+ my $flat = 'flat';
+ my $end = '';
+ if ($ctx->{flat}) {
+ $hr = 1;
+ $flat = "<b>$flat</b>";
+ } else {
+ $nested = "<b>$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 {
- $next .= 'end of thread';
+ $hr = $ctx->{-hr};
}
- $next .= "</a>, back to <a\nhref=\"../\">index</a>\n";
- $rv .= "<hr />" . PRE_WRAP . $next . $foot . "</pre>";
+ $rv .= $more ? '</pre><hr><pre>' : '</pre>' if $hr;
+ $rv;
}
-# only private functions below.
+sub pad_link ($$;$) {
+ my ($mid, $level, $s) = @_;
+ $s ||= '...';
+ my $id = id_compress($mid, 1);
+ (' 'x19).indent_for($level).th_pfx($level)."<a\nhref=#r$id>($s)</a>\n";
+}
-sub index_walk {
- my ($part, $enc, $part_nr, $fhref, $more) = @_;
- my $s = add_text_body($enc, $part, $part_nr, $fhref);
+sub _th_index_lite {
+ my ($mid_raw, $irt, $id, $ctx) = @_;
+ my $rv = '';
+ my $mapping = $ctx->{mapping} or return $rv;
+ my $pad = ' ';
+ my $mid_map = $mapping->{$mid_raw};
+ defined $mid_map or
+ return 'public-inbox BUG: '.ascii_html($mid_raw).' not mapped';
+ my ($attr, $node, $idx, $level) = @$mid_map;
+ my $children = $node->{children};
+ my $nr_c = scalar @$children;
+ my $nr_s = 0;
+ my $siblings;
+ if (my $smsg = $node->{smsg}) {
+ ($$irt) = (($smsg->{references} || '') =~ m/<([^>]+)>\z/);
+ }
+ 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 = $prev->{id};
+ if ($idx > 2) {
+ my $s = ($idx - 1). ' preceding siblings ...';
+ $rv .= pad_link($pmid, $level, $s);
+ } elsif ($idx == 2) {
+ my $ppmid = $siblings->[0]->{id};
+ $rv .= $pad . $mapping->{$ppmid}->[0];
+ }
+ $rv .= $pad . $mapping->{$pmid}->[0];
+ }
+ }
+ my $s_s = nr_to_s($nr_s, 'sibling', 'siblings');
+ my $s_c = nr_to_s($nr_c, 'reply', 'replies');
+ $attr =~ s!\n\z!</b>\n!s;
+ $attr =~ s!<a\nhref.*</a> !!s; # no point in duplicating subject
+ $attr =~ s!<a\nhref=[^>]+>([^<]+)</a>!$1!s; # no point linking to self
+ $rv .= "<b>@ $attr";
+ if ($nr_c) {
+ my $cmid = $children->[0]->{id};
+ $rv .= $pad . $mapping->{$cmid}->[0];
+ if ($nr_c > 2) {
+ my $s = ($nr_c - 1). ' more replies';
+ $rv .= pad_link($cmid, $level + 1, $s);
+ } elsif (my $cn = $children->[1]) {
+ $rv .= $pad . $mapping->{$cn->{id}}->[0];
+ }
+ }
- if ($more) {
- # drop the remainder of git patches, they're usually better
- # to review when the full message is viewed
- $s =~ s!^---+\n.*\z!!ms and $$more = 'more...';
+ my $next = $siblings->[$idx+1] if $siblings && $idx >= 0;
+ if ($next) {
+ my $nmid = $next->{id};
+ $rv .= $pad . $mapping->{$nmid}->[0];
+ my $nnext = $nr_s - $idx;
+ if ($nnext > 2) {
+ my $s = ($nnext - 1).' subsequent siblings';
+ $rv .= pad_link($nmid, $level, $s);
+ } elsif (my $nn = $siblings->[$idx + 2]) {
+ $rv .= $pad . $mapping->{$nn->{id}}->[0];
+ }
+ }
+ $rv .= $pad ."<a\nhref=#r$id>$s_s, $s_c; $ctx->{s_nr}</a>\n";
+}
- # Drop signatures
- $s =~ s/^-- \n.*\z//ms and $$more = 'more...';
+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, $i);
+ ++$level;
+ $i = 0;
+ unshift @q, map { ($level, $_, $i++) } @{$node->{children}};
}
+}
- # kill any leading or trailing whitespace lines
- $s =~ s/^\s*$//sgm;
- $s =~ s/\s+\z//s;
+sub pre_thread {
+ my ($ctx, $level, $node, $idx) = @_;
+ $ctx->{mapping}->{$node->{id}} = [ '', $node, $idx, $level ];
+ skel_dump($ctx, $level, $node);
+}
- if (length $s) {
- # kill per-line trailing whitespace
- $s =~ s/[ \t]+$//sgm;
- $s .= "\n" unless $s =~ /\n\z/s;
- }
- $s;
+sub thread_index_entry {
+ my ($ctx, $level, $smsg) = @_;
+ my ($beg, $end) = thread_adj_level($ctx, $level);
+ $beg . '<pre>' . index_entry($smsg, $ctx, 0) . '</pre>' . $end;
}
-sub enc_for {
- my ($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;
+sub stream_thread ($$) {
+ my ($rootset, $ctx) = @_;
+ my $inbox = $ctx->{-inbox};
+ my @q = map { (0, $_) } @$rootset;
+ my $level;
+ my $smsg;
+ while (@q) {
+ $level = shift @q;
+ my $node = shift @q or next;
+ my $cl = $level + 1;
+ unshift @q, map { ($cl, $_) } @{$node->{children}};
+ $smsg = $inbox->smsg_mime($node->{smsg}) and last;
+ }
+ return missing_thread($ctx) unless $smsg;
+
+ $ctx->{-obfs_ibx} = $inbox->{obfuscate} ? $inbox : undef;
+ $ctx->{-title_html} = ascii_html($smsg->subject);
+ $ctx->{-html_tip} = thread_index_entry($ctx, $level, $smsg);
+ $smsg = undef;
+ PublicInbox::WwwStream->response($ctx, 200, sub {
+ return unless $ctx;
+ while (@q) {
+ $level = shift @q;
+ my $node = shift @q or next;
+ my $cl = $level + 1;
+ unshift @q, map { ($cl, $_) } @{$node->{children}};
+ if ($smsg = $inbox->smsg_mime($node->{smsg})) {
+ return thread_index_entry($ctx, $level, $smsg);
+ } else {
+ return ghost_index_entry($ctx, $level, $node);
+ }
}
+ my $ret = join('', thread_adj_level($ctx, 0));
+ $ret .= ${$ctx->{dst}}; # skel
+ $ctx = undef;
+ $ret;
+ });
+}
+
+sub thread_html {
+ my ($ctx) = @_;
+ my $mid = $ctx->{mid};
+ my $srch = $ctx->{srch};
+ my $sres = $srch->get_thread($mid);
+ my $msgs = load_results($srch, $sres);
+ my $nr = $sres->{total};
+ return missing_thread($ctx) if $nr == 0;
+ my $skel = '<hr><pre>';
+ $skel .= $nr == 1 ? 'only message in thread' : 'end of thread';
+ $skel .= ", back to <a\nhref=\"../../\">index</a>\n\n";
+ $skel .= "<b\nid=t>Thread overview:</b> ";
+ $skel .= $nr == 1 ? '(only message)' : "$nr+ messages";
+ $skel .= " (download: <a\nhref=\"../t.mbox.gz\">mbox.gz</a>";
+ $skel .= " / follow: <a\nhref=\"../t.atom\">Atom feed</a>)\n";
+ $skel .= "-- links below jump to the message on this page --\n";
+ $ctx->{-upfx} = '../../';
+ $ctx->{cur_level} = 0;
+ $ctx->{dst} = \$skel;
+ $ctx->{prev_attr} = '';
+ $ctx->{prev_level} = 0;
+ $ctx->{root_anchor} = anchor_for($mid);
+ $ctx->{mapping} = {};
+ $ctx->{s_nr} = "$nr+ messages in thread";
+
+ my $rootset = thread_results($ctx, $msgs);
+
+ # reduce hash lookups in pre_thread->skel_dump
+ my $inbox = $ctx->{-inbox};
+ $ctx->{-obfs_ibx} = $inbox->{obfuscate} ? $inbox : undef;
+ walk_thread($rootset, $ctx, *pre_thread);
+
+ $skel .= '</pre>';
+ return stream_thread($rootset, $ctx) unless $ctx->{flat};
+
+ # flat display: lazy load the full message from smsg
+ my $smsg;
+ while (my $m = shift @$msgs) {
+ $smsg = $inbox->smsg_mime($m) and last;
}
- $default;
+ return missing_thread($ctx) unless $smsg;
+ $ctx->{-title_html} = ascii_html($smsg->subject);
+ $ctx->{-html_tip} = '<pre>'.index_entry($smsg, $ctx, scalar @$msgs);
+ $smsg = undef;
+ PublicInbox::WwwStream->response($ctx, 200, sub {
+ return unless $msgs;
+ $smsg = undef;
+ while (my $m = shift @$msgs) {
+ $smsg = $inbox->smsg_mime($m) and last;
+ }
+ return index_entry($smsg, $ctx, scalar @$msgs) if $smsg;
+ $msgs = undef;
+ $skel;
+ });
}
sub multipart_text_as_html {
- my ($mime, $full_pfx, $srch) = @_;
+ my ($mime, $upfx, $obfs_ibx) = @_;
my $rv = "";
- my $part_nr = 0;
- my $enc = enc_for($mime->header("Content-Type"));
# scan through all parts, looking for displayable text
- $mime->walk_parts(sub {
- my ($part) = @_;
- $rv .= add_text_body($enc, $part, \$part_nr, $full_pfx);
- });
- $mime->body_set('');
+ msg_iter($mime, sub { $rv .= add_text_body($upfx, $obfs_ibx, $_[0]) });
$rv;
}
-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";
-}
+sub flush_quote {
+ my ($s, $l, $quot) = @_;
-my $LINK_RE = qr!\b((?:ftp|https?|nntp)://[@\w\+\&\?\.\%\;/#=-]+)!;
+ # show everything in the full version with anchor from
+ # short version (see above)
+ my $rv = $l->linkify_1(join('', @$quot));
+ @$quot = ();
-sub linkify {
- # no newlines added here since it'd break the splitting we do
- # to fold quotes
- $_[0] =~ s!$LINK_RE!<a\nhref="$1">$1</a>!g;
+ # we use a <span> here to allow users to specify their own
+ # color for quoted text
+ $rv = $l->linkify_2(ascii_html($rv));
+ $$s .= qq(<span\nclass="q">) . $rv . '</span>'
}
-sub flush_quote {
- my ($quot, $n, $part_nr, $full_pfx, $final) = @_;
-
- if ($full_pfx) {
- if (!$final && scalar(@$quot) <= MAX_INLINE_QUOTED) {
- # show quote inline
- my $rv = join("\n", map { linkify($_); $_ } @$quot);
- @$quot = ();
- return $rv . "\n";
- }
-
- # show a short snippet of quoted text and link to full version:
- @$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;
- my $nr = ++$$n;
- "> [<a\nhref=\"$full_pfx#q${part_nr}_$nr\">$cur</a>]\n";
+sub attach_link ($$$$;$) {
+ my ($upfx, $ct, $p, $fn, $err) = @_;
+ my ($part, $depth, @idx) = @$p;
+ my $nl = $idx[-1] > 1 ? "\n" : '';
+ my $idx = join('.', @idx);
+ my $size = bytes::length($part->body);
+
+ # hide attributes normally, unless we want to aid users in
+ # spotting MUA problems:
+ $ct =~ s/;.*// unless $err;
+ $ct = ascii_html($ct);
+ my $desc = $part->header('Content-Description');
+ $desc = $fn unless defined $desc;
+ $desc = '' unless defined $desc;
+ my $sfn;
+ if (defined $fn && $fn =~ /\A[[:alnum:]][\w\.-]+[[:alnum:]]\z/) {
+ $sfn = $fn;
+ } elsif ($ct eq 'text/plain') {
+ $sfn = 'a.txt';
} else {
- # show everything in the full version with anchor from
- # short version (see above)
- my $nr = ++$$n;
- my $rv = "<a\nid=q${part_nr}_$nr></a>";
- $rv .= join("\n", map { linkify($_); $_ } @$quot) . "\n";
- @$quot = ();
- $rv;
+ $sfn = 'a.bin';
}
+ my $ret = qq($nl<a\nhref="$upfx$idx-$sfn">);
+ if ($err) {
+ $ret .=
+"[-- Warning: decoded text below may be mangled --]\n";
+ }
+ $ret .= "[-- Attachment #$idx: ";
+ my $ts = "Type: $ct, Size: $size bytes";
+ $desc = ascii_html($desc);
+ $ret .= ($desc eq '') ? "$ts --]" : "$desc --]\n[-- $ts --]";
+ $ret .= "</a>\n";
}
sub add_text_body {
- my ($enc_msg, $part, $part_nr, $full_pfx) = @_;
- return '' if $part->subparts;
-
- my $ct = $part->content_type;
- # account for filter bugs...
- if (defined $ct && $ct =~ m!\btext/[xh]+tml\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);
- $s = ascii_html($s);
- my @lines = split(/\n/, $s);
- $s = '';
+ my ($upfx, $obfs_ibx, $p) = @_;
+ # $p - from msg_iter: [ Email::MIME, depth, @idx ]
+ my ($part, $depth) = @$p; # attachment @idx is unused
+ my $ct = $part->content_type || 'text/plain';
+ my $fn = $part->filename;
+
+ if ($ct =~ m!\btext/x?html\b!i) {
+ return attach_link($upfx, $ct, $p, $fn);
+ }
+
+ my $s = eval { $part->body_str };
- if ($$part_nr > 0) {
- my $fn = $part->filename;
- defined($fn) or $fn = "part #" . ($$part_nr + 1);
- $s .= add_filename_line($enc, $fn);
+ # badly-encoded message? tell the world about it!
+ my $err = $@;
+ if ($err) {
+ if ($ct =~ m!\btext/plain\b!i) {
+ # Try to assume UTF-8 because Alpine seems to
+ # do wacky things and set charset=X-UNKNOWN
+ $part->charset_set('UTF-8');
+ $s = eval { $part->body_str };
+
+ # If forcing charset=UTF-8 failed,
+ # attach_link will warn further down...
+ $s = $part->body if $@;
+ } else {
+ return attach_link($upfx, $ct, $p, $fn);
+ }
}
+ my @lines = split(/^/m, $s);
+ $s = '';
+ if (defined($fn) || $depth > 0 || $err) {
+ $s .= attach_link($upfx, $ct, $p, $fn, $err);
+ $s .= "\n";
+ }
my @quot;
- while (defined(my $cur = shift @lines)) {
- if ($cur !~ /^>/) {
+ my $l = PublicInbox::Linkify->new;
+ foreach my $cur (@lines) {
+ if ($cur !~ /^>/) {
# show the previously buffered quote inline
- if (scalar @quot) {
- $s .= flush_quote(\@quot, \$n, $$part_nr,
- $full_pfx, 0);
- }
+ flush_quote(\$s, $l, \@quot) if @quot;
# regular line, OK
- linkify($cur);
- $s .= $cur;
- $s .= "\n";
+ $l->linkify_1($cur);
+ $s .= $l->linkify_2(ascii_html($cur));
} else {
push @quot, $cur;
}
}
- $s .= flush_quote(\@quot, \$n, $$part_nr, $full_pfx, 1) if scalar @quot;
- $s .= "\n" unless $s =~ /\n\z/s;
- ++$$part_nr;
- $s;
-}
-sub headers_to_html_header {
- my ($mime, $full_pfx, $srch) = @_;
+ if (@quot) { # ugh, top posted
+ flush_quote(\$s, $l, \@quot);
+ obfuscate_addrs($obfs_ibx, $s) if $obfs_ibx;
+ $s;
+ } else {
+ obfuscate_addrs($obfs_ibx, $s) if $obfs_ibx;
+ if ($s =~ /\n\z/s) { # common, last line ends with a newline
+ $s;
+ } else { # some editors don't do newlines...
+ $s .= "\n";
+ }
+ }
+}
- my $rv = "";
+sub _msg_html_prepare {
+ my ($hdr, $ctx, $more, $nr) = @_;
+ my $srch = $ctx->{srch} if $ctx;
+ my $atom = '';
+ my $obfs_ibx = $ctx->{-obfs_ibx};
+ my $rv = '';
+ my $mids = mids($hdr);
+ my $multiple = scalar(@$mids) > 1; # zero, one, infinity
+ if ($nr == 0) {
+ if ($more) {
+ $rv .=
+"<pre>WARNING: multiple messages refer to this Message-ID\n</pre>";
+ }
+ $rv .= "<pre\nid=b>"; # anchor for body start
+ } else {
+ $rv .= '<pre>';
+ }
+ if ($srch) {
+ $ctx->{-upfx} = '../';
+ }
my @title;
- my $header_obj = $mime->header_obj;
- my $mid = $header_obj->header('Message-ID');
- $mid = PublicInbox::Hval->new_msgid($mid);
- my $mid_href = $mid->as_href;
foreach my $h (qw(From To Cc Subject Date)) {
- my $v = $mime->header($h);
- defined($v) && length($v) or next;
- $v = PublicInbox::Hval->new_oneline($v);
+ my $v = $hdr->header($h);
+ defined($v) && ($v ne '') or next;
+ $v = PublicInbox::Hval->new($v);
if ($h eq 'From') {
- my @from = Email::Address->parse($v->raw);
- $title[1] = ascii_html($from[0]->name);
+ my @n = PublicInbox::Address::names($v->raw);
+ $title[1] = ascii_html(join(', ', @n));
+ obfuscate_addrs($obfs_ibx, $title[1]) if $obfs_ibx;
} elsif ($h eq 'Subject') {
$title[0] = $v->as_html;
if ($srch) {
- $rv .= "$h: <a\nhref=\"../t/$mid_href.html\">";
+ $rv .= qq($h: <a\nhref="#r"\nid=t>);
$rv .= $v->as_html . "</a>\n";
next;
}
}
- $rv .= "$h: " . $v->as_html . "\n";
+ $v = $v->as_html;
+ obfuscate_addrs($obfs_ibx, $v) if $obfs_ibx;
+ $rv .= "$h: $v\n";
}
+ $title[0] ||= '(no subject)';
+ $ctx->{-title_html} = join(' - ', @title);
+ foreach (@$mids) {
+ my $mid = PublicInbox::Hval->new_msgid($_) ;
+ my $mhtml = $mid->as_html;
+ if ($multiple) {
+ my $href = $mid->{href};
+ $rv .= "Message-ID: ";
+ $rv .= "<a\nhref=\"../$href/\">";
+ $rv .= "<$mhtml></a> ";
+ $rv .= "(<a\nhref=\"../$href/raw\">raw</a>)\n";
+ } else {
+ $rv .= "Message-ID: <$mhtml> ";
+ $rv .= "(<a\nhref=\"raw\">raw</a>)\n";
+ }
+ }
+ $rv .= _parent_headers($hdr, $srch);
+ $rv .= "\n";
+}
+
+sub thread_skel {
+ my ($dst, $ctx, $hdr, $tpfx) = @_;
+ my $srch = $ctx->{srch};
+ my $mid = mids($hdr)->[0];
+ my $sres = $srch->get_thread($mid);
+ my $nr = $sres->{total};
+ my $expand = qq(expand[<a\nhref="${tpfx}T/#u">flat</a>) .
+ qq(|<a\nhref="${tpfx}t/#u">nested</a>] ) .
+ qq(<a\nhref="${tpfx}t.mbox.gz">mbox.gz</a> ) .
+ qq(<a\nhref="${tpfx}t.atom">Atom feed</a>);
+
+ my $parent = in_reply_to($hdr);
+ $$dst .= "\n<b>Thread overview: </b>";
+ if ($nr <= 1) {
+ if (defined $parent) {
+ $$dst .= "$expand\n ";
+ $$dst .= ghost_parent("$tpfx../", $parent) . "\n";
+ } else {
+ $$dst .= "[no followups] $expand\n";
+ }
+ $ctx->{next_msg} = undef;
+ $ctx->{parent_msg} = $parent;
+ return;
+ }
+
+ $$dst .= "$nr+ messages / $expand";
+ $$dst .= qq! <a\nhref="#b">top</a>\n!;
+
+ my $subj = $hdr->header('Subject');
+ defined $subj or $subj = '';
+ $ctx->{prev_subj} = [ split(/ /, $srch->subject_normalized($subj)) ];
+ $ctx->{cur} = $mid;
+ $ctx->{prev_attr} = '';
+ $ctx->{prev_level} = 0;
+ $ctx->{dst} = $dst;
+ $sres = load_results($srch, $sres);
+
+ # reduce hash lookups in skel_dump
+ my $ibx = $ctx->{-inbox};
+ $ctx->{-obfs_ibx} = $ibx->{obfuscate} ? $ibx : undef;
+ walk_thread(thread_results($ctx, $sres), $ctx, *skel_dump);
+
+ $ctx->{parent_msg} = $parent;
+}
- $rv .= 'Message-ID: <' . $mid->as_html . '> ';
- $mid_href = "../m/$mid_href" unless $full_pfx;
- $rv .= "(<a\nhref=\"$mid_href.txt\">raw</a>)\n";
+sub _parent_headers {
+ my ($hdr, $srch) = @_;
+ my $rv = '';
- my $irt = $header_obj->header('In-Reply-To');
+ 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;
+ my $href = $v->{href};
$rv .= "In-Reply-To: <";
- $rv .= "<a\nhref=\"$href.html\">$html</a>>\n";
+ $rv .= "<a\nhref=\"../$href/\">$html</a>>\n";
}
- my $refs = $header_obj->header('References');
+ # 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{mid_clean($irt)} = 1 if defined $irt;
+ $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($ref);
+ push @refs, linkify_ref_nosrch($ref);
}
if (@refs) {
- $rv .= 'References: '. join(' ', @refs) . "\n";
+ $rv .= 'References: '. join("\n\t", @refs) . "\n";
}
}
-
- $rv .= "\n";
-
- ("<html><head><title>". join(' - ', @title) .
- '</title></head><body>' . PRE_WRAP . $rv);
+ $rv;
}
sub html_footer {
- my ($mime, $standalone, $full_pfx, $srch) = @_;
- my %cc; # everyone else
- my $to; # this is the From address
-
- foreach my $h (qw(From To Cc)) {
- my $v = $mime->header($h);
- defined($v) && length($v) 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;
- }
- }
- Email::Address->purge_cache if $standalone;
-
- my $subj = $mime->header('Subject') || '';
- $subj = "Re: $subj" unless $subj =~ /\bRe:/;
- my $mid = $mime->header_obj->header('Message-ID');
- my $irt = uri_escape_utf8($mid);
- delete $cc{$to};
- $to = uri_escape_utf8($to);
- $subj = uri_escape_utf8($subj);
+ my ($hdr, $standalone, $ctx, $rhref) = @_;
- my $cc = uri_escape_utf8(join(',', sort values %cc));
- my $href = "mailto:$to?In-Reply-To=$irt&Cc=${cc}&Subject=$subj";
-
- my $idx = $standalone ? " <a\nhref=\"../\">index</a>" : '';
+ my $srch = $ctx->{srch} if $ctx;
+ my $upfx = '../';
+ my $tpfx = '';
+ my $idx = $standalone ? " <a\nhref=\"$upfx\">index</a>" : '';
+ my $irt = '';
if ($idx && $srch) {
- $irt = $mime->header_obj->header('In-Reply-To') || '';
- $mid = mid_compressed(mid_clean($mid));
- my $t_anchor = length $irt ? T_ANCHOR : '';
- $idx = " <a\nhref=\"../t/$mid.html$t_anchor\">".
- "threadlink</a>$idx";
- my $res = $srch->get_followups($mid);
- if (my $c = $res->{count}) {
- $c = $c == 1 ? '1 followup' : "$c followups";
- $idx .= "\n$c:\n";
- $res->{srch} = $srch;
- thread_followups(\$idx, $mime, $res);
- } else {
- $idx .= "\n(no followups, yet)\n";
+ $idx .= "\n";
+ thread_skel(\$idx, $ctx, $hdr, $tpfx);
+ my ($next, $prev);
+ my $parent = ' ';
+ $next = $prev = ' ';
+
+ if (my $n = $ctx->{next_msg}) {
+ $n = PublicInbox::Hval->new_msgid($n)->{href};
+ $next = "<a\nhref=\"$upfx$n/\"\nrel=next>next</a>";
}
- if ($irt) {
- $irt = PublicInbox::Hval->new_msgid($irt);
- $irt = $irt->as_href;
- $irt = "<a\nhref=\"$irt\">parent</a> ";
- } else {
- $irt = ' ' x length('parent ');
+ my $u;
+ my $par = $ctx->{parent_msg};
+ if ($par) {
+ $u = PublicInbox::Hval->new_msgid($par)->{href};
+ $u = "$upfx$u/";
}
+ if (my $p = $ctx->{prev_msg}) {
+ $prev = PublicInbox::Hval->new_msgid($p)->{href};
+ 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>";
+ }
+ $irt = "$next $prev$parent ";
} else {
$irt = '';
}
-
- "$irt<a\nhref=\"" . ascii_html($href) . '">reply</a>' . $idx;
+ $rhref ||= '#R';
+ $irt .= qq(<a\nhref="$rhref">reply</a>);
+ $irt .= $idx;
}
-sub linkify_ref {
+sub linkify_ref_nosrch {
my $v = PublicInbox::Hval->new_msgid($_[0]);
my $html = $v->as_html;
- my $href = $v->as_href;
- "<<a\nhref=\"$href.html\">$html</a>>";
+ my $href = $v->{href};
+ "<<a\nhref=\"../$href/\">$html</a>>";
}
sub anchor_for {
my ($msgid) = @_;
- my $id = $msgid;
- if ($id !~ /\A[a-f0-9]{40}\z/) {
- $id = mid_compressed(mid_clean($id), 1);
- }
- 'm' . $id;
-}
-
-sub simple_dump {
- my ($dst, $root, $node, $level) = @_;
- return unless $node;
- # $root = [ Root Message-ID, \%seen, $srch ];
- if (my $x = $node->message) {
- my $mid = $x->header('Message-ID');
- if ($root->[0] ne $mid) {
- my $pfx = ' ' x $level;
- $$dst .= $pfx;
- my $s = $x->header('Subject');
- my $h = $root->[2]->subject_path($s);
- if ($root->[1]->{$h}) {
- $s = '';
- } else {
- $root->[1]->{$h} = 1;
- $s = PublicInbox::Hval->new($s);
- $s = $s->as_html;
- }
- my $m = PublicInbox::Hval->new_msgid($mid);
- my $f = PublicInbox::Hval->new($x->header('X-PI-From'));
- my $d = PublicInbox::Hval->new($x->header('X-PI-Date'));
- $m = $m->as_href . '.html';
- $f = $f->as_html;
- $d = $d->as_html . ' UTC';
- if (length($s) == 0) {
- $$dst .= "` <a\nhref=\"$m\">$f @ $d</a>\n";
- } else {
- $$dst .= "` <a\nhref=\"$m\">$s</a>\n" .
- "$pfx by $f @ $d\n";
- }
+ 'm' . id_compress($msgid, 1);
+}
+
+sub ghost_parent {
+ my ($upfx, $mid) = @_;
+
+ $mid = PublicInbox::Hval->new_msgid($mid);
+ my $href = $mid->{href};
+ my $html = $mid->as_html;
+ qq{[parent not found: <<a\nhref="$upfx$href/">$html</a>>]};
+}
+
+sub indent_for {
+ my ($level) = @_;
+ $level ? INDENT x ($level - 1) : '';
+}
+
+sub load_results {
+ my ($srch, $sres) = @_;
+ my $msgs = delete $sres->{msgs};
+ $srch->retry_reopen(sub { [ map { $_->mid; $_ } @$msgs ] });
+}
+
+sub thread_results {
+ my ($ctx, $msgs) = @_;
+ require PublicInbox::SearchThread;
+ PublicInbox::SearchThread::thread($msgs, *sort_ds, $ctx->{-inbox});
+}
+
+sub missing_thread {
+ my ($ctx) = @_;
+ require PublicInbox::ExtMsg;
+ PublicInbox::ExtMsg::ext_msg($ctx);
+}
+
+sub _msg_date {
+ my ($hdr) = @_;
+ fmt_ts(msg_datestamp($hdr));
+}
+
+sub fmt_ts { POSIX::strftime('%Y-%m-%d %k:%M', gmtime($_[0])) }
+
+sub dedupe_subject {
+ my ($prev_subj, $subj, $val) = @_;
+
+ my $omit = ''; # '"' denotes identical text omitted
+ my (@prev_pop, @curr_pop);
+ while (@$prev_subj && @$subj && $subj->[-1] eq $prev_subj->[-1]) {
+ push(@prev_pop, pop(@$prev_subj));
+ push(@curr_pop, pop(@$subj));
+ $omit ||= $val;
+ }
+ pop @$subj if @$subj && $subj->[-1] =~ /^re:\s*/i;
+ if (scalar(@curr_pop) == 1) {
+ $omit = '';
+ push @$prev_subj, @prev_pop;
+ push @$subj, @curr_pop;
+ }
+ $omit;
+}
+
+sub skel_dump {
+ my ($ctx, $level, $node) = @_;
+ my $smsg = $node->{smsg} or return _skel_ghost($ctx, $level, $node);
+
+ my $dst = $ctx->{dst};
+ my $cur = $ctx->{cur};
+ my $mid = $smsg->{mid};
+
+ my $f = ascii_html($smsg->from_name);
+ my $obfs_ibx = $ctx->{-obfs_ibx};
+ obfuscate_addrs($obfs_ibx, $f) if $obfs_ibx;
+
+ my $d = fmt_ts($smsg->{ds}) . ' ' . indent_for($level) . th_pfx($level);
+ my $attr = $f;
+ $ctx->{first_level} ||= $level;
+
+ if ($attr ne $ctx->{prev_attr} || $ctx->{prev_level} > $level) {
+ $ctx->{prev_attr} = $attr;
+ }
+ $ctx->{prev_level} = $level;
+
+ if ($cur) {
+ if ($cur eq $mid) {
+ delete $ctx->{cur};
+ $$dst .= "<b>$d<a\nid=r\nhref=\"#t\">".
+ "$attr [this message]</a></b>\n";
+ return;
+ } else {
+ $ctx->{prev_msg} = $mid;
}
+ } else {
+ $ctx->{next_msg} ||= $mid;
}
- simple_dump($dst, $root, $node->child, $level+1);
- simple_dump($dst, $root, $node->next, $level);
-}
-
-sub thread_followups {
- my ($dst, $root, $res) = @_;
- $root->header_set('X-PI-TS', '0');
- my $msgs = load_results($res);
- push @$msgs, $root;
- my $th = thread_results($msgs);
- my $srch = $res->{srch};
- my $subj = $srch->subject_path($root->header('Subject'));
- my %seen = ($subj => 1);
- $root = [ $root->header('Message-ID'), \%seen, $srch ];
- simple_dump($dst, $root, $_, 0) for $th->rootset;
-}
-
-sub thread_html_head {
- my ($mime) = @_;
- my $s = PublicInbox::Hval->new_oneline($mime->header('Subject'));
- $s = $s->as_html;
- "<html><head><title>$s</title></head><body>";
-}
-
-sub thread_entry {
- my ($dst, $git, $state, $node, $level) = @_;
- return unless $node;
- # $state = [ $search_res, $seen, undef, 0 (msg_nr) ];
- # $seen is overloaded with 3 types of fields:
- # 1) "root_anchor" => anchor_for(Message-ID),
- # 2) seen subject hashes: sha1(subject) => 1
- # 3) anchors hashes: "#$sha1_hex" (same as $seen in index_entry)
- if (my $mime = $node->message) {
-
- # lazy load the full message from mini_mime:
- my $path = mid2path(mid_clean($mime->header('Message-ID')));
- $mime = eval { Email::MIME->new($git->cat_file("HEAD:$path")) };
- if ($mime) {
- if (length($$dst) == 0) {
- $$dst .= thread_html_head($mime);
- }
- $$dst .= index_entry(undef, $mime, $level, $state);
+
+ # Subject is never undef, this mail was loaded from
+ # our Xapian which would've resulted in '' if it were
+ # really missing (and Filter rejects empty subjects)
+ my @subj = split(/ /, $ctx->{srch}->subject_normalized($smsg->subject));
+
+ # 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 {
+ $end = "$f</a>\n";
+ }
+ my $m;
+ my $id = '';
+ my $mapping = $ctx->{mapping};
+ if ($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 {
+ $m = $ctx->{-upfx}.mid_escape($mid).'/';
+ }
+ $$dst .= $d . "<a\nhref=\"$m\"$id>" . $end;
+}
+
+sub _skel_ghost {
+ my ($ctx, $level, $node) = @_;
+
+ my $mid = $node->{id};
+ my $d = $ctx->{pct} ? ' [irrelevant] ' # search result
+ : ' [not found] ';
+ $d .= indent_for($level) . th_pfx($level);
+ my $upfx = $ctx->{-upfx};
+ my $m = PublicInbox::Hval->new_msgid($mid);
+ my $href = $upfx . $m->{href} . '/';
+ my $html = $m->as_html;
+
+ my $mapping = $ctx->{mapping};
+ my $map = $mapping->{$mid} if $mapping;
+ if ($map) {
+ my $id = id_compress($mid, 1);
+ $map->[0] = $d . qq{<<a\nhref=#r$id>$html</a>>\n};
+ $d .= qq{<<a\nhref="$href"\nid=r$id>$html</a>>\n};
+ } else {
+ $d .= qq{<<a\nhref="$href">$html</a>>\n};
+ }
+ my $dst = $ctx->{dst};
+ $$dst .= $d;
+}
+
+sub sort_ds {
+ [ sort {
+ (eval { $a->topmost->{smsg}->ds } || 0) <=>
+ (eval { $b->topmost->{smsg}->ds } || 0)
+ } @{$_[0]} ];
+}
+
+# accumulate recent topics if search is supported
+# returns 200 if done, 404 if not
+sub acc_topic {
+ my ($ctx, $level, $node) = @_;
+ my $srch = $ctx->{srch};
+ my $mid = $node->{id};
+ my $x = $node->{smsg} || $ctx->{-inbox}->smsg_by_mid($mid);
+ my ($subj, $ds);
+ my $topic;
+ if ($x) {
+ $subj = $x->subject;
+ $subj = $srch->subject_normalized($subj);
+ $ds = $x->ds;
+ if ($level == 0) {
+ $topic = [ $ds, 1, { $subj => $mid }, $subj ];
+ $ctx->{-cur_topic} = $topic;
+ push @{$ctx->{order}}, $topic;
+ return;
+ }
+
+ $topic = $ctx->{-cur_topic}; # should never be undef
+ $topic->[0] = $ds if $ds > $topic->[0];
+ $topic->[1]++;
+ my $seen = $topic->[2];
+ if (scalar(@$topic) == 3) { # parent was a ghost
+ push @$topic, $subj;
+ } elsif (!$seen->{$subj}) {
+ push @$topic, $level, $subj;
}
+ $seen->{$subj} = $mid; # latest for subject
+ } else { # ghost message
+ return if $level != 0; # ignore child ghosts
+ $topic = [ -666, 0, {} ];
+ $ctx->{-cur_topic} = $topic;
+ push @{$ctx->{order}}, $topic;
}
- thread_entry($dst, $git, $state, $node->child, $level + 1);
- thread_entry($dst, $git, $state, $node->next, $level);
}
-sub load_results {
- my ($res) = @_;
+sub dump_topics {
+ my ($ctx) = @_;
+ my $order = delete $ctx->{order}; # [ ds, subj1, subj2, subj3, ... ]
+ if (!@$order) {
+ $ctx->{-html_tip} = '<pre>[No topics in range]</pre>';
+ return 404;
+ }
- [ map { $_->mini_mime } @{delete $res->{msgs}} ];
+ my @out;
+ my $ibx = $ctx->{-inbox};
+ my $obfs_ibx = $ibx->{obfuscate} ? $ibx : undef;
+ my $srch = $ctx->{srch};
+
+ # sort by recency, this allows new posts to "bump" old topics...
+ foreach my $topic (sort { $b->[0] <=> $a->[0] } @$order) {
+ my ($ds, $n, $seen, $top, @ex) = @$topic;
+ @$topic = ();
+ next unless defined $top; # ghost topic
+ my $mid = delete $seen->{$top};
+ my $href = mid_escape($mid);
+ my $prev_subj = [ split(/ /, $top) ];
+ $top = PublicInbox::Hval->new($top)->as_html;
+ $ds = fmt_ts($ds);
+
+ # $n isn't the total number of posts on the topic,
+ # just the number of posts in the current results window
+ my $anchor;
+ if ($n == 1) {
+ $n = '';
+ $anchor = '#u'; # top of only message
+ } else {
+ $n = " ($n+ messages)";
+ $anchor = '#t'; # thread skeleton
+ }
+
+ my $mbox = qq(<a\nhref="$href/t.mbox.gz">mbox.gz</a>);
+ my $atom = qq(<a\nhref="$href/t.atom">Atom</a>);
+ my $s = "<a\nhref=\"$href/T/$anchor\"><b>$top</b></a>\n" .
+ " $ds UTC $n - $mbox / $atom\n";
+ for (my $i = 0; $i < scalar(@ex); $i += 2) {
+ my $level = $ex[$i];
+ my $subj = $ex[$i + 1];
+ $mid = delete $seen->{$subj};
+ my @subj = split(/ /, $srch->subject_normalized($subj));
+ my @next_prev = @subj; # full copy
+ my $omit = dedupe_subject($prev_subj, \@subj, ' "');
+ $prev_subj = \@next_prev;
+ $subj = ascii_html(join(' ', @subj));
+ obfuscate_addrs($obfs_ibx, $subj) if $obfs_ibx;
+ $href = mid_escape($mid);
+ $s .= indent_for($level) . TCHILD;
+ $s .= "<a\nhref=\"$href/T/#u\">$subj</a>$omit\n";
+ }
+ push @out, $s;
+ }
+ $ctx->{-html_tip} = '<pre>' . join("\n", @out) . '</pre>';
+ 200;
}
-sub msg_timestamp {
- my ($mime) = @_;
- my $ts = eval { str2time($mime->header('Date')) };
- defined($ts) ? $ts : 0;
+sub index_nav { # callback for WwwStream
+ my (undef, $ctx) = @_;
+ delete $ctx->{qp} or return;
+ my ($next, $prev);
+ $next = $prev = ' ';
+ my $latest = '';
+
+ my $next_o = $ctx->{-next_o};
+ if ($next_o) {
+ $next = qq!<a\nhref="?o=$next_o"\nrel=next>next</a>!;
+ }
+ if (my $cur_o = $ctx->{-cur_o}) {
+ $latest = qq! <a\nhref=.>latest</a>!;
+
+ my $o = $cur_o - ($next_o - $cur_o);
+ if ($o > 0) {
+ $prev = qq!<a\nhref="?o=$o"\nrel=prev>prev</a>!;
+ } elsif ($o == 0) {
+ $prev = qq!<a\nhref=.\nrel=prev>prev</a>!;
+ }
+ }
+ "<hr><pre>page: $next $prev$latest</pre>";
}
-sub thread_results {
- my ($msgs) = @_;
- require PublicInbox::Thread;
- my $th = PublicInbox::Thread->new(@$msgs);
- $th->thread;
- $th->order(*PublicInbox::Thread::sort_ts);
- $th
+sub index_topics {
+ my ($ctx) = @_;
+ my ($off) = (($ctx->{qp}->{o} || '0') =~ /(\d+)/);
+ my $lim = 200;
+ my $opts = { offset => $off, limit => $lim };
+
+ $ctx->{order} = [];
+ my $srch = $ctx->{srch};
+
+ my $qs = '';
+ # this complicated bit cuts loading time by over 400ms on my system:
+ if ($off == 0) {
+ my ($min, $max) = $ctx->{-inbox}->mm->minmax;
+ my $n = $max - $lim;
+ $n = $min if $n < $min;
+ for (; $qs eq '' && $n >= $min; --$n) {
+ my $smsg = $srch->lookup_article($n) or next;
+ $qs = POSIX::strftime('d:%Y%m%d..', gmtime($smsg->ts));
+ }
+ }
+
+ my $sres = $srch->query($qs, $opts);
+ my $nr = scalar @{$sres->{msgs}};
+ if ($nr) {
+ $sres = load_results($srch, $sres);
+ walk_thread(thread_results($ctx, $sres), $ctx, *acc_topic);
+ }
+ $ctx->{-next_o} = $off + $nr;
+ $ctx->{-cur_o} = $off;
+ PublicInbox::WwwStream->response($ctx, dump_topics($ctx), *index_nav);
+}
+
+sub thread_adj_level {
+ my ($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>';
+ ($beg, '</li>');
+ }
+}
+
+sub ghost_index_entry {
+ my ($ctx, $level, $node) = @_;
+ my ($beg, $end) = thread_adj_level($ctx, $level);
+ $beg . '<pre>'. ghost_parent($ctx->{-upfx}, $node->{id})
+ . '</pre>' . $end;
}
1;