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 $SALT = rand;
+use PublicInbox::Hval qw/ascii_html obfuscate_addrs/;
+use PublicInbox::Linkify;
+use PublicInbox::MID qw/mid_clean id_compress mid_mime mid_escape/;
+use PublicInbox::MsgIter;
+use PublicInbox::Address;
+use PublicInbox::WwwStream;
+use PublicInbox::Reply;
require POSIX;
-# TODO: make these constants tunable
-use constant MAX_INLINE_QUOTED => 12; # half an 80x24 terminal
-use constant MAX_TRUNC_LEN => 72;
-use constant T_ANCHOR => '#u';
use constant INDENT => ' ';
+use constant TCHILD => '` ';
+sub th_pfx ($) { $_[0] == 0 ? '' : TCHILD };
-*ascii_html = *PublicInbox::Hval::ascii_html;
-
-my $enc_utf8 = find_encoding('UTF-8');
-
-# public functions:
+# public functions: (unstable)
sub msg_html {
- my ($ctx, $mime, $full_pfx, $footer) = @_;
- if (defined $footer) {
- $footer = "\n" . $footer;
- } else {
- $footer = '';
- }
+ my ($ctx, $mime) = @_;
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>';
+ my $ibx = $ctx->{-inbox};
+ my $obfs_ibx = $ibx->{obfuscate} ? $ibx : undef;
+ my $tip = _msg_html_prepare($hdr, $ctx, $obfs_ibx);
+ PublicInbox::WwwStream->response($ctx, 200, sub {
+ my ($nr, undef) = @_;
+ if ($nr == 1) {
+ $tip . multipart_text_as_html($mime, '', $obfs_ibx) .
+ '</pre><hr>'
+ } elsif ($nr == 2) {
+ # 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, $hdr)
+ };
+ } else {
+ undef
+ }
+ });
}
-sub feed_entry {
- my ($class, $mime, $full_pfx) = @_;
+# /$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">$url</a>\n);
+ }
+
+ my ($arg, $link) = PublicInbox::Reply::mailto_arg_link($ibx, $hdr);
+
+ # mailto: link only works if address obfuscation is disabled
+ if ($link) {
+ $link = <<EOF;
- # no <head> here for <style>...
- PublicInbox::Hval::PRE .
- multipart_text_as_html($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></pre>
+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 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>
+$info
+* Reply to all the recipients using the <b>--to</b>, <b>--cc</b>,
+ and <b>--in-reply-to</b> switches of git-send-email(1):
+
+ git send-email$arg
+
+ <a
+href="$se_url">$se_url</a>
+$link
+EOF
}
sub in_reply_to {
my ($hdr) = @_;
- my $irt = $hdr->header('In-Reply-To');
-
- return mid_clean($irt) if (defined $irt);
-
- my $refs = $hdr->header('References');
- if ($refs && $refs =~ /<([^>]+)>\s*\z/s) {
- return $1;
+ my %mid = map { $_ => 1 } $hdr->header_raw('Message-ID');
+ my @refs = (($hdr->header_raw('References') || '') =~ /<([^>]+)>/g);
+ push(@refs, (($hdr->header_raw('In-Reply-To') || '') =~ /<([^>]+)>/g));
+ while (defined(my $irt = pop @refs)) {
+ next if $mid{"<$irt>"};
+ return $irt;
}
undef;
}
+sub _hdr_names_html ($$) {
+ my ($hdr, $field) = @_;
+ my $val = $hdr->header($field) or return '';
+ ascii_html(join(', ', PublicInbox::Address::names($val)));
+}
+
+sub nr_to_s ($$$) {
+ my ($nr, $singular, $plural) = @_;
+ return "0 $plural" if $nr == 0;
+ $nr == 1 ? "$nr $singular" : "$nr $plural";
+}
+
# this is already inside a <pre>
sub index_entry {
- my ($fh, $mime, $level, $state) = @_;
- my $midx = $state->{anchor_idx}++;
- my $ctx = $state->{ctx};
+ my ($mime, $ctx, $more) = @_;
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>";
+ my $mid_raw = mid_clean(mid_mime($mime));
+ 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;
+ 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: $dst";
+ }
}
- if ($root_anchor eq $id) {
- $subj = "<u\nid=u>$subj</u>";
+ my $from = _hdr_names_html($hdr, 'From');
+ obfuscate_addrs($obfs_ibx, $from) if $obfs_ibx;
+ $rv .= "From: $from @ "._msg_date($hdr)." 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";
- 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>";
+ # 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 {
+ $hr = $ctx->{-hr};
}
- $fh->write($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('');
+ $rv .= $more ? '</pre><hr><pre>' : '</pre>' if $hr;
+ $rv;
+}
- my $txt = "${path}$href/raw";
- $rv = "\n<a\nhref=\"$mhref\">$more</a> <a\nhref=\"$txt\">raw</a> ";
- $rv .= html_footer($hdr, 0, undef, $ctx);
+sub pad_link ($$;$) {
+ my ($mid, $level, $s) = @_;
+ $s ||= '...';
+ my $id = id_compress($mid, 1);
+ (' 'x19).indent_for($level).th_pfx($level)."<a\nhref=#r$id>($s)</a>\n";
+}
- if (defined $irt) {
- unless (defined $parent_anchor) {
- my $v = PublicInbox::Hval->new_msgid($irt, 1);
- $v = $v->as_href;
- $parent_anchor = "${path}$v/";
+sub _th_index_lite {
+ my ($mid_raw, $irt, $id, $ctx) = @_;
+ my $rv = '';
+ my $mapping = $ctx->{mapping} or return $rv;
+ my $pad = ' ';
+ my $mid_map = $mapping->{$mid_raw};
+ defined $mid_map or
+ return 'public-inbox BUG: '.ascii_html($mid_raw).' not mapped';
+ my ($attr, $node, $idx, $level) = @$mid_map;
+ my $children = $node->{children};
+ my $nr_c = scalar @$children;
+ my $nr_s = 0;
+ my $siblings;
+ if (my $smsg = $node->{smsg}) {
+ ($$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];
}
- $rv .= " <a\nhref=\"$parent_anchor\">parent</a>";
}
- if (my $pct = $state->{pct}) { # used by SearchView.pm
- $rv .= " [relevance $pct->{$mid_raw}%]";
- } elsif ($srch) {
- if ($ctx->{flat}) {
- $rv .= " [<a\nhref=\"${path}$href/t/#u\">threaded</a>" .
- "|<b>flat</b>]";
- } else {
- $rv .= " [<b>threaded</b>|" .
- "<a\nhref=\"${path}$href/T/#u\">flat</a>]";
+ 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];
+ }
+ }
+
+ 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];
}
}
- $fh->write($rv .= '</pre>');
+ $rv .= $pad ."<a\nhref=#r$id>$s_s, $s_c; $ctx->{s_nr}</a>\n";
}
-sub thread_html {
- my ($ctx, $foot, $srch) = @_;
- sub { emit_thread_html($_[0], $ctx, $foot, $srch) }
+sub walk_thread {
+ my ($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}};
+ }
}
-# only private functions below.
+sub pre_thread {
+ my ($ctx, $level, $node, $idx) = @_;
+ $ctx->{mapping}->{$node->{id}} = [ '', $node, $idx, $level ];
+ skel_dump($ctx, $level, $node);
+}
-sub emit_thread_html {
- my ($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,
- };
-
- 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>');
- }
+sub thread_index_entry {
+ my ($ctx, $level, $mime) = @_;
+ my ($beg, $end) = thread_adj_level($ctx, $level);
+ $beg . '<pre>' . index_entry($mime, $ctx, 0) . '</pre>' . $end;
+}
+
+sub stream_thread ($$) {
+ my ($rootset, $ctx) = @_;
+ my $inbox = $ctx->{-inbox};
+ my $mime;
+ my @q = map { (0, $_) } @$rootset;
+ my $level;
+ while (@q) {
+ $level = shift @q;
+ my $node = shift @q or next;
+ my $cl = $level + 1;
+ unshift @q, map { ($cl, $_) } @{$node->{children}};
+ $mime = $inbox->msg_by_smsg($node->{smsg}) and last;
}
- $git = undef;
- Email::Address->purge_cache;
-
- # there could be a race due to a message being deleted in git
- # but still being in the Xapian index:
- return missing_thread($cb, $ctx) if ($orig_cb eq $cb);
-
- my $final_anchor = $state->{anchor_idx};
- my $next = "<a\nid=s$final_anchor>";
- $next .= $final_anchor == 1 ? 'only message in' : 'end of';
- $next .= " thread</a>, back to <a\nhref=\"../../\">index</a>";
- $next .= "\ndownload thread: ";
- $next .= "<a\nhref=\"../t.mbox.gz\">mbox.gz</a>";
- $next .= " / follow: <a\nhref=\"../t.atom\">Atom feed</a>";
- $cb->write('<hr /><pre>' . $next . "\n\n".
- $foot . '</pre></body></html>');
- $cb->close;
-}
-
-sub index_walk {
- my ($fh, $part, $enc, $part_nr, $fhref, $more) = @_;
- my $s = add_text_body($enc, $part, $part_nr, $fhref);
-
- if ($more) {
- my $m = 0;
- # drop the remainder of git patches, they're usually better
- # to review when the full message is viewed
- $s =~ s!^---+\n.*\z!!ms and $m = 1;
-
- # Drop signatures
- $s =~ s/^-- \n.*\z//ms and $m = 1;
- $$more = "<b>More...</b>\n\n$$more" if $m;
- }
-
- # kill any leading or trailing whitespace lines
- $s =~ s/^\s*$//sgm;
- $s =~ s/\s+\z//s;
-
- if ($s ne '') {
- # kill per-line trailing whitespace
- $s =~ s/[ \t]+$//sgm;
- $s .= "\n" unless $s =~ /\n\z/s;
- }
- $fh->write($s);
-}
-
-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;
+ return missing_thread($ctx) unless $mime;
+
+ $ctx->{-obfs_ibx} = $inbox->{obfuscate} ? $inbox : undef;
+ $mime = PublicInbox::MIME->new($mime);
+ $ctx->{-title_html} = ascii_html($mime->header('Subject'));
+ $ctx->{-html_tip} = thread_index_entry($ctx, $level, $mime);
+ PublicInbox::WwwStream->response($ctx, 200, sub {
+ return unless $ctx;
+ while (@q) {
+ $level = shift @q;
+ my $node = shift @q or next;
+ my $cl = $level + 1;
+ unshift @q, map { ($cl, $_) } @{$node->{children}};
+ my $mid = $node->{id};
+ if ($mime = $inbox->msg_by_smsg($node->{smsg})) {
+ $mime = PublicInbox::MIME->new($mime);
+ return thread_index_entry($ctx, $level, $mime);
+ } 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($msgs, $srch);
+
+ # 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 $mime;
+ while ($mime = shift @$msgs) {
+ $mime = $inbox->msg_by_smsg($mime) and last;
}
- $default;
+ return missing_thread($ctx) unless $mime;
+ $mime = PublicInbox::MIME->new($mime);
+ $ctx->{-title_html} = ascii_html($mime->header('Subject'));
+ $ctx->{-html_tip} = '<pre>'.index_entry($mime, $ctx, scalar @$msgs);
+ $mime = undef;
+ PublicInbox::WwwStream->response($ctx, 200, sub {
+ return unless $msgs;
+ while ($mime = shift @$msgs) {
+ $mime = $inbox->msg_by_smsg($mime) and last;
+ }
+ if ($mime) {
+ $mime = PublicInbox::MIME->new($mime);
+ return index_entry($mime, $ctx, scalar @$msgs);
+ }
+ $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, 1);
- });
- $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";
-}
-
-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);
- }
+ my ($s, $l, $quot) = @_;
- # 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";
+ # show everything in the full version with anchor from
+ # short version (see above)
+ my $rv = $l->linkify_1(join('', @$quot));
+ @$quot = ();
+
+ # we use a <span> here to allow users to specify their own
+ # color for quoted text
+ $rv = $l->linkify_2(ascii_html($rv));
+ $$s .= qq(<span\nclass="q">) . $rv . '</span>'
+}
+
+sub attach_link ($$$$;$) {
+ my ($upfx, $ct, $p, $fn, $err) = @_;
+ my ($part, $depth, @idx) = @$p;
+ my $nl = $idx[-1] > 1 ? "\n" : '';
+ my $idx = join('.', @idx);
+ my $size = bytes::length($part->body);
+
+ # hide attributes normally, unless we want to aid users in
+ # spotting MUA problems:
+ $ct =~ s/;.*// unless $err;
+ $ct = ascii_html($ct);
+ my $desc = $part->header('Content-Description');
+ $desc = $fn unless defined $desc;
+ $desc = '' unless defined $desc;
+ my $sfn;
+ if (defined $fn && $fn =~ /\A[[:alnum:]][\w\.-]+[[:alnum:]]\z/) {
+ $sfn = $fn;
+ } elsif ($ct eq 'text/plain') {
+ $sfn = 'a.txt';
} else {
- # show everything in the full version with anchor from
- # short version (see above)
- my %l;
- my $rv .= join('', map { linkify_1(\%l, $_) } @$quot);
- @$quot = ();
- $rv = ascii_html($rv);
- return linkify_2(\%l, $rv) unless $do_anchor;
- my $nr = ++$$n;
- "<a\nid=q${part_nr}_$nr></a>" . linkify_2(\%l, $rv);
+ $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, $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);
- $s = '';
+ my ($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 };
+
+ # 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 ($$part_nr > 0) {
- my $fn = $part->filename;
- defined($fn) or $fn = "part #" . ($$part_nr + 1);
- $s .= add_filename_line($enc, $fn);
+ # 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)) {
+ 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, $$part_nr,
- $full_pfx, 0, $do_anchor);
- }
+ flush_quote(\$s, $l, \@quot) if @quot;
# regular line, OK
- my %l;
- $cur = linkify_1(\%l, $cur);
- $cur = ascii_html($cur);
- $s .= linkify_2(\%l, $cur);
+ $l->linkify_1($cur);
+ $s .= $l->linkify_2(ascii_html($cur));
} else {
push @quot, $cur;
}
}
- if (scalar @quot) {
- $s .= flush_quote(\@quot, \$n, $$part_nr, $full_pfx, 1,
- $do_anchor);
+
+ 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";
+ }
}
- $s .= "\n" unless $s =~ /\n\z/s;
- ++$$part_nr;
- $s;
}
-sub headers_to_html_header {
- my ($hdr, $full_pfx, $ctx) = @_;
+sub _msg_html_prepare {
+ my ($hdr, $ctx, $obfs_ibx) = @_;
my $srch = $ctx->{srch} if $ctx;
- my $rv = "";
+ my $atom = '';
+ my $rv = "<pre\nid=b>"; # anchor for body start
+
+ if ($srch) {
+ $ctx->{-upfx} = '../';
+ }
my @title;
- my $mid = $hdr->header('Message-ID');
+ my $mid = mid_clean($hdr->header_raw('Message-ID'));
$mid = PublicInbox::Hval->new_msgid($mid);
foreach my $h (qw(From To Cc Subject Date)) {
my $v = $hdr->header($h);
defined($v) && ($v ne '') or next;
- $v = PublicInbox::Hval->new_oneline($v);
+ $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: <b\nid=t>";
- $rv .= $v->as_html . "</b>\n";
+ $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);
$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"/>!;
- } else {
- $rv .= _parent_headers_nosrch($hdr);
- $atom = '';
- }
+ $rv .= "(<a\nhref=\"raw\">raw</a>)\n";
+ $rv .= _parent_headers($hdr, $srch);
$rv .= "\n";
-
- ("<html><head><title>". join(' - ', @title) . "</title>$atom".
- PublicInbox::Hval::STYLE . "</head><body><pre>" . $rv);
}
-sub thread_inline {
- my ($dst, $ctx, $hdr, $upfx) = @_;
+sub thread_skel {
+ my ($dst, $ctx, $hdr, $tpfx) = @_;
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>";
+ my $mid = mid_clean($hdr->header_raw('Message-ID'));
+ my $sres = $srch->get_thread($mid);
+ my $nr = $sres->{total};
+ my $expand = qq(<a\nhref="${tpfx}T/#u">expand</a> ) .
+ qq(/ <a\nhref="${tpfx}t.mbox.gz">mbox.gz</a> ) .
+ qq(/ <a\nhref="${tpfx}t.atom">Atom feed</a>);
- $$dst .= 'Thread: ';
my $parent = in_reply_to($hdr);
+ $$dst .= "\n<b>Thread overview: </b>";
if ($nr <= 1) {
if (defined $parent) {
$$dst .= "($expand)\n ";
- $$dst .= ghost_parent("$upfx../", $parent) . "\n";
+ $$dst .= ghost_parent("$tpfx../", $parent) . "\n";
} else {
$$dst .= "[no followups, yet] ($expand)\n";
}
return;
}
- $$dst .= "~$nr messages ($expand";
- if ($nr > MAX_INLINE_QUOTED) {
- $$dst .= qq! / <a\nhref="#b">[scroll down]</a>!;
- }
- $$dst .= ")\n";
+ $$dst .= "$nr+ messages in thread ($expand";
+ $$dst .= qq! / <a\nhref="#b">[top]</a>)\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};
+ 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($sres, $srch), $ctx, *skel_dump);
+
+ $ctx->{parent_msg} = $parent;
}
-sub _parent_headers_nosrch {
- my ($hdr) = @_;
+sub _parent_headers {
+ my ($hdr, $srch) = @_;
my $rv = '';
my $irt = in_reply_to($hdr);
if (defined $irt) {
- my $v = PublicInbox::Hval->new_msgid($irt, 1);
+ 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</a>>\n";
}
- my $refs = $hdr->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;
}
if (@refs) {
- $rv .= 'References: '. join(' ', @refs) . "\n";
+ $rv .= 'References: '. join("\n\t", @refs) . "\n";
}
}
$rv;
}
sub html_footer {
- my ($hdr, $standalone, $full_pfx, $ctx) = @_;
- 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;
- }
- }
- Email::Address->purge_cache if $standalone;
-
- my $subj = $hdr->header('Subject') || '';
- $subj = "Re: $subj" unless $subj =~ /\bRe:/i;
- my $mid = $hdr->header('Message-ID');
- my $irt = uri_escape_utf8($mid);
- delete $cc{$to};
- $to = uri_escape_utf8($to);
- $subj = uri_escape_utf8($subj);
-
- my $cc = uri_escape_utf8(join(',', sort values %cc));
- my $href = "mailto:$to?In-Reply-To=$irt&Cc=${cc}&Subject=$subj";
- $href =~ s/%20/+/g;
+ my ($hdr, $standalone, $ctx, $rhref) = @_;
my $srch = $ctx->{srch} if $ctx;
- my $upfx = $full_pfx ? '../' : '../../';
- my $tpfx = $full_pfx ? '' : '../';
+ my $upfx = '../';
+ my $tpfx = '';
my $idx = $standalone ? " <a\nhref=\"$upfx\">index</a>" : '';
-
- if ($srch && $standalone) {
- $idx .= qq{ / follow: <a\nhref="${tpfx}t.atom">Atom feed</a>\n};
- }
+ my $irt = '';
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 ');
+ $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 ($next) {
- $irt .= "<a\nhref=\"$upfx$next/\">next</a> ";
- } else {
- $irt .= ' ' x length('next ');
+ my $u;
+ my $par = $ctx->{parent_msg};
+ if ($par) {
+ $u = PublicInbox::Hval->new_msgid($par)->{href};
+ $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 = 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_nosrch {
- my $v = PublicInbox::Hval->new_msgid($_[0], 1);
+ my $v = PublicInbox::Hval->new_msgid($_[0]);
my $html = $v->as_html;
- my $href = $v->as_href;
+ 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 = 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} = "#$id"; # save the anchor for children, later
+ 'm' . id_compress($msgid, 1);
}
sub ghost_parent {
my ($upfx, $mid) = @_;
- # 'subject dummy' is used internally by Mail::Thread
- return '[no common parent]' if ($mid eq 'subject dummy');
$mid = PublicInbox::Hval->new_msgid($mid);
- my $href = $mid->as_href;
+ my $href = $mid->{href};
my $html = $mid->as_html;
qq{[parent not found: <<a\nhref="$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>' . $end);
-}
-
-sub __thread_entry {
- my ($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);
- }
- }
- 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 __ghost_prepare {
- my ($state, $node, $level) = @_;
- my $ghost = $state->{ghost} ||= [];
- push @$ghost, [ $node->messageid, $level ];
-}
-
-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);
- }
- } else {
- __ghost_prepare($state, $node, $level);
- }
-
- thread_entry($cb, $git, $state, $node->child, $level + 1);
- thread_entry($cb, $git, $state, $node->next, $level);
+ $level ? INDENT x ($level - 1) : '';
}
sub load_results {
- my ($res) = @_;
-
- [ map { $_->mini_mime } @{delete $res->{msgs}} ];
+ my ($srch, $sres) = @_;
+ my $msgs = delete $sres->{msgs};
+ $srch->retry_reopen(sub { [ map { $_->mid; $_ } @$msgs ] });
}
sub msg_timestamp {
}
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 ($msgs, $srch) = @_;
+ require PublicInbox::SearchThread;
+ PublicInbox::SearchThread::thread($msgs, *sort_ts, $srch);
}
sub missing_thread {
- my ($cb, $ctx) = @_;
+ my ($ctx) = @_;
require PublicInbox::ExtMsg;
-
- $cb->(PublicInbox::ExtMsg::ext_msg($ctx))
+ PublicInbox::ExtMsg::ext_msg($ctx);
}
sub _msg_date {
sub fmt_ts { POSIX::strftime('%Y-%m-%d %k:%M', gmtime($_[0])) }
-sub _inline_header {
- my ($dst, $state, $upfx, $hdr, $level) = @_;
- my $dot = $level == 0 ? '' : '` ';
+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 $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->{ts}) . ' ' . 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};
+ $$dst .= "<b>$d<a\nid=r\nhref=\"#t\">".
"$attr [this message]</a></b>\n";
-
return;
+ } else {
+ $ctx->{prev_msg} = $mid;
}
} else {
- $state->{next_msg} ||= $mid;
+ $ctx->{next_msg} ||= $mid;
}
# Subject is never undef, this mail was loaded from
# our Xapian which would've resulted in '' if it were
# really missing (and Filter rejects empty subjects)
- my $s = $hdr->header('Subject');
- my $h = $state->{srch}->subject_path($s);
- if ($state->{seen}->{$h}) {
- $s = undef;
+ my @subj = split(/ /, $ctx->{srch}->subject_normalized($smsg->subject));
+
+ # remove common suffixes from the subject if it matches the previous,
+ # so we do not show redundant text at the end.
+ my $prev_subj = $ctx->{prev_subj} || [];
+ $ctx->{prev_subj} = [ @subj ];
+ my $omit = ''; # '"' denotes identical text omitted
+ while (@$prev_subj && @subj && $subj[-1] eq $prev_subj->[-1]) {
+ pop @$prev_subj;
+ pop @subj;
+ $omit ||= '" ';
+ }
+ pop @subj if @subj && $subj[-1] =~ /^re:\s*/i;
+ my $end;
+ if (@subj) {
+ my $subj = join(' ', @subj);
+ $subj = ascii_html($subj);
+ obfuscate_addrs($obfs_ibx, $subj) if $obfs_ibx;
+ $end = "$subj</a> $omit$f\n"
} else {
- $state->{seen}->{$h} = 1;
- $s = PublicInbox::Hval->new($s);
- $s = $s->as_html;
+ $end = "$f</a>\n";
}
- my $m = PublicInbox::Hval->new_msgid($mid);
- $m = $upfx . '../' . $m->as_href . '/';
- if (defined $s) {
- $$dst .= "$pfx$dot<a\nhref=\"$m\">$s</a> $attr\n";
+ 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 {
- $$dst .= "$pfx$dot<a\nhref=\"$m\">$f</a>\n";
+ $m = $ctx->{-upfx}.mid_escape($mid).'/';
}
+ $$dst .= $d . "<a\nhref=\"$m\"$id>" . $end;
}
-sub inline_dump {
- my ($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->{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 {
- 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);
+ my $dst = $ctx->{dst};
+ $$dst .= $d;
}
sub sort_ts {
- sort {
- (eval { $a->topmost->message->header('X-PI-TS') } || 0) <=>
- (eval { $b->topmost->message->header('X-PI-TS') } || 0)
- } @_;
-}
-
-sub rsort_ts {
- sort {
- (eval { $b->topmost->message->header('X-PI-TS') } || 0) <=>
- (eval { $a->topmost->message->header('X-PI-TS') } || 0)
- } @_;
+ [ sort {
+ (eval { $a->topmost->{smsg}->ts } || 0) <=>
+ (eval { $b->topmost->{smsg}->ts } || 0)
+ } @{$_[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;
-
- if (my $x = $node->message) {
- $x = $x->header_obj;
- my $subj;
-
- $subj = $x->header('Subject');
- $subj = $state->{srch}->subject_normalized($subj);
-
- if (++$state->{subjs}->{$subj} == 1) {
- push @{$state->{order}}, [ $level, $subj ];
+# returns 200 if done, 404 if not
+sub acc_topic {
+ my ($ctx, $level, $node) = @_;
+ my $srch = $ctx->{srch};
+ my $mid = $node->{id};
+ my $x = $node->{smsg} || $srch->lookup_mail($mid);
+ my ($subj, $ts);
+ my $topic;
+ if ($x) {
+ $subj = $x->subject;
+ $subj = $srch->subject_normalized($subj);
+ $ts = $x->ts;
+ if ($level == 0) {
+ $topic = [ $ts, 1, { $subj => $mid }, $subj ];
+ $ctx->{-cur_topic} = $topic;
+ push @{$ctx->{order}}, $topic;
+ return;
}
- 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 ];
+ $topic = $ctx->{-cur_topic}; # should never be undef
+ $topic->[0] = $ts if $ts > $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;
}
- } else {
- # ghost message, do not bump level
- $child_adjust = 0;
+ $seen->{$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}}, $topic;
}
-
- add_topic($state, $node->child, $level + $child_adjust);
- add_topic($state, $node->next, $level);
}
sub dump_topics {
- my ($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 $attr;
+ my ($ctx) = @_;
+ my $order = delete $ctx->{order}; # [ ts, subj1, subj2, subj3, ... ]
+ if (!@$order) {
+ $ctx->{-html_tip} = '<pre>[No topics in range]</pre>';
+ return 404;
+ }
+
+ 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 ($ts, $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;
$ts = fmt_ts($ts);
- $attr = " $ts UTC";
# $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;
+ # 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" .
+ " $ts 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 $omit = ''; # '"' denotes identical text omitted
+ while (@$prev_subj && @subj &&
+ $subj[-1] eq $prev_subj->[-1]) {
+ pop @$prev_subj;
+ pop @subj;
+ $omit ||= ' "';
+ }
+ pop @subj if @subj && $subj[-1] =~ /^re:\s*/i;
+ $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;
}
- $dst .= '</pre>';
+ $ctx->{-html_tip} = '<pre>' . join("\n", @out) . '</pre>';
+ 200;
}
-sub emit_index_topics {
- my ($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;
+sub index_nav { # callback for WwwStream
+ my (undef, $ctx) = @_;
+ delete $ctx->{qp} or return;
+ my ($next, $prev);
+ $next = $prev = ' ';
+ my $latest = '';
- for (rsort_ts(thread_results(load_results($res), 1)->rootset)) {
- add_topic($state, $_, 0);
+ 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>!;
}
- $opts{offset} += $nr;
}
+ "<hr><pre>page: $next $prev$latest</pre>";
+}
+
+sub index_topics {
+ my ($ctx) = @_;
+ my ($off) = (($ctx->{qp}->{o} || '0') =~ /(\d+)/);
+ my $opts = { offset => $off, limit => 200 };
+
+ $ctx->{order} = [];
+ my $srch = $ctx->{srch};
+ my $sres = $srch->query('', $opts);
+ my $nr = scalar @{$sres->{msgs}};
+ if ($nr) {
+ $sres = load_results($srch, $sres);
+ walk_thread(thread_results($sres, $srch), $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>');
+ }
+}
- $fh->write(dump_topics($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->{id})
+ . '</pre>' . $end;
}
1;