-# Copyright (C) 2014, Eric Wong <normalperson@yhbt.net> and all contributors
+# Copyright (C) 2014-2015 all contributors <meta@public-inbox.org>
# License: AGPLv3 or later (https://www.gnu.org/licenses/agpl-3.0.txt)
package PublicInbox::View;
use strict;
use warnings;
-use PublicInbox::Hval;
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_compress mid2path/;
+use Digest::SHA qw/sha1_hex/;
+my $SALT = rand;
require POSIX;
# TODO: make these constants tunable
use constant MAX_INLINE_QUOTED => 12; # half an 80x24 terminal
use constant MAX_TRUNC_LEN => 72;
use constant PRE_WRAP => "<pre\nstyle=\"white-space:pre-wrap\">";
+use constant T_ANCHOR => '#u';
+use constant INDENT => ' ';
*ascii_html = *PublicInbox::Hval::ascii_html;
# public functions:
sub msg_html {
- my ($class, $mime, $full_pfx, $footer) = @_;
+ my ($ctx, $mime, $full_pfx, $footer) = @_;
if (defined $footer) {
$footer = "\n" . $footer;
} else {
$footer = '';
}
- headers_to_html_header($mime, $full_pfx) .
+ headers_to_html_header($mime, $full_pfx, $ctx) .
multipart_text_as_html($mime, $full_pfx) .
'</pre><hr />' . PRE_WRAP .
- html_footer($mime, 1) . $footer .
+ html_footer($mime, 1, $full_pfx, $ctx) .
+ $footer .
'</pre></body></html>';
}
PRE_WRAP . multipart_text_as_html($mime, $full_pfx) . '</pre>';
}
+sub in_reply_to {
+ my ($header_obj) = @_;
+ my $irt = $header_obj->header('In-Reply-To');
+
+ return mid_clean($irt) if (defined $irt);
+
+ my $refs = $header_obj->header('References');
+ if ($refs && $refs =~ /<([^>]+)>\s*\z/s) {
+ return $1;
+ }
+ undef;
+}
+
# this is already inside a <pre>
sub index_entry {
- my ($class, $mime, $level, $state) = @_;
- my ($now, $seen, $first) = @$state;
- my $midx = $state->[3]++;
+ my ($fh, $mime, $level, $state) = @_;
+ my $midx = $state->{anchor_idx}++;
+ my $ctx = $state->{ctx};
+ my $srch = $ctx->{srch};
my ($prev, $next) = ($midx - 1, $midx + 1);
- my $rv = '';
my $part_nr = 0;
- my $enc_msg = enc_for($mime->header("Content-Type"));
+ 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_raw('Message-ID');
+ my $mid_raw = mid_clean($header_obj->header('Message-ID'));
my $id = anchor_for($mid_raw);
- $seen->{$id} = "#$id"; # save the anchor for later
+ my $seen = $state->{seen};
+ $seen->{$id} = "#$id"; # save the anchor for children, later
my $mid = PublicInbox::Hval->new_msgid($mid_raw);
my $from = PublicInbox::Hval->new_oneline($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 $pfx = (' ' x $level);
+ my $more = 'permalink';
+ my $root_anchor = $state->{root_anchor} || '';
+ my $path = $root_anchor ? '../../' : '';
+ my $href = $mid->as_href;
+ my $irt = in_reply_to($header_obj);
+ my $parent_anchor = $seen->{anchor_for($irt)} if defined $irt;
- my $ts = $mime->header('X-PI-Date');
- my $fmt = '%Y-%m-%d %H:%M UTC';
- $ts = POSIX::strftime($fmt, gmtime($ts));
+ if ($srch) {
+ my $t = $ctx->{flat} ? 'T' : 't';
+ $subj = "<a\nhref=\"${path}$href/$t/#u\">$subj</a>";
+ }
+ if ($root_anchor eq $id) {
+ $subj = "<u\nid=\"u\">$subj</u>";
+ }
- $rv .= "$pfx<b\nid=\"$id\">$subj</b>\n$pfx";
- $rv .= "- by $from @ $ts - ";
- $rv .= "<a\nid=\"s$midx\"\nhref=\"#s$next\">next</a>";
+ my $ts = _msg_date($mime);
+ my $rv = "<table\nsummary=l$level><tr>";
+ if ($level) {
+ $rv .= '<td><pre>' . (INDENT 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";
+ $fh->write($rv .= "\n\n");
+
+ my ($fhref, $more_ref);
+ my $mhref = "${path}$href/";
- my $irp = $header_obj->header_raw('In-Reply-To');
- my ($anchor_idx, $anchor);
- if (defined $irp) {
- $anchor_idx = anchor_for($irp);
- $anchor = $seen->{$anchor_idx};
+ # show full message if it's our root message
+ if ($root_anchor ne $id || ($level != 0 && !$ctx->{flat})) {
+ $fhref = "${path}$href/f/";
+ $more_ref = \$more;
}
- my $href = $mid->as_href;
- my $mhref = "m/$href.html";
- my $fhref = "f/$href.html";
- my $more = 'message';
# scan through all parts, looking for displayable text
$mime->walk_parts(sub {
- $rv .= index_walk($_[0], $pfx, $enc_msg, $part_nr, $fhref,
- \$more);
- $part_nr++;
+ index_walk($fh, $_[0], $enc, \$part_nr, $fhref, $more_ref);
});
+ $mime->body_set('');
+
+ my $txt = "${path}$href/raw";
+ $rv = "\n<a\nhref=\"$mhref\">$more</a> <a\nhref=\"$txt\">raw</a> ";
+ $rv .= html_footer($mime, 0, undef, $ctx);
- $rv .= "\n$pfx<a\nhref=\"$mhref\">$more</a> ";
- my $txt = "m/$href.txt";
- $rv .= "<a\nhref=\"$txt\">raw</a> ";
- $rv .= html_footer($mime, 0);
-
- if (defined $irp) {
- unless (defined $anchor) {
- my $v = PublicInbox::Hval->new_msgid($irp);
- my $html = $v->as_html;
- $anchor = 'm/' . $v->as_href . '.html';
- $seen->{$anchor_idx} = $anchor;
+ if (defined $irt) {
+ unless (defined $parent_anchor) {
+ my $v = PublicInbox::Hval->new_msgid($irt, 1);
+ $v = $v->as_href;
+ $parent_anchor = "${path}$v/";
+ }
+ $rv .= " <a\nhref=\"$parent_anchor\">parent</a>";
+ }
+ if (my $pct = $state->{pct}) {
+ $rv .= " [$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>]";
}
- $rv .= " <a\nhref=\"$anchor\">parent</a>";
}
- $rv .= " <a\nhref=\"?r=$first#$id\">threadlink</a>";
- $rv . "\n\n";
+ $fh->write($rv .= '</pre></td></tr></table>');
}
-# only private functions below.
-
-sub index_walk {
- my ($part, $pfx, $enc_msg, $part_nr, $fhref, $more) = @_;
- my $rv = '';
- return $rv if $part->subparts; # walk_parts already recurses
- my $ct = $part->content_type;
-
- # account for filter bugs...
- return if defined $ct && $ct =~ m!\btext/[xh]+tml\b!i;
+sub thread_html {
+ my ($ctx, $foot, $srch) = @_;
+ sub { emit_thread_html($_[0], $ctx, $foot, $srch) }
+}
- my $enc = enc_for($ct, $enc_msg);
+# only private functions below.
- if ($part_nr > 0) {
- my $fn = $part->filename;
- defined($fn) or $fn = "part #" . ($part_nr + 1);
- $rv .= $pfx . add_filename_line($enc->decode($fn));
+sub emit_thread_html {
+ my ($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,
+ };
+
+ require PublicInbox::GitCatFile;
+ my $git = PublicInbox::GitCatFile->new($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;
}
+ $git = undef;
+ Email::Address->purge_cache;
+
+ # there could be a race due to a message being deleted in git
+ # but still being in the Xapian index:
+ return missing_thread($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: <a\nhref=\"../t.mbox.gz\">mbox.gz</a>";
+ $next .= " / follow: <a\nhref=\"../t.atom\">Atom feed</a>";
+ $cb->write("<hr />" . PRE_WRAP . $next . "\n\n". $foot .
+ "</pre></body></html>");
+ $cb->close;
+}
- my $s = add_text_body_short($enc, $part, $part_nr, $fhref);
+sub index_walk {
+ my ($fh, $part, $enc, $part_nr, $fhref, $more) = @_;
+ my $s = add_text_body($enc, $part, $part_nr, $fhref);
- # drop the remainder of git patches, they're usually better
- # to review when the full message is viewed
- $s =~ s!^---+\n.*\z!!ms and $$more = 'more...';
+ if ($more) {
+ # drop the remainder of git patches, they're usually better
+ # to review when the full message is viewed
+ $s =~ s!^---+\n.*\z!!ms and $$more = 'more...';
- # Drop signatures
- $s =~ s/^-- \n.*\z//ms and $$more = 'more...';
+ # Drop signatures
+ $s =~ s/^-- \n.*\z//ms and $$more = 'more...';
+ }
- # kill any leading or trailing whitespace
- $s =~ s/\A\s+//s;
+ # kill any leading or trailing whitespace lines
+ $s =~ s/^\s*$//sgm;
$s =~ s/\s+\z//s;
- if (length $s) {
- # add prefix:
- $s =~ s/^/$pfx/sgm;
-
- $rv .= $s . "\n";
+ if ($s ne '') {
+ # kill per-line trailing whitespace
+ $s =~ s/[ \t]+$//sgm;
+ $s .= "\n" unless $s =~ /\n\z/s;
}
- $rv;
+ $fh->write($s);
}
sub enc_for {
}
sub multipart_text_as_html {
- my ($mime, $full_pfx) = @_;
+ my ($mime, $full_pfx, $srch) = @_;
my $rv = "";
my $part_nr = 0;
- my $enc_msg = enc_for($mime->header("Content-Type"));
+ my $enc = enc_for($mime->header("Content-Type"));
# scan through all parts, looking for displayable text
$mime->walk_parts(sub {
my ($part) = @_;
- return if $part->subparts; # walk_parts already recurses
- my $ct = $part->content_type;
-
- # account for filter bugs...
- return if defined $ct && $ct =~ m!\btext/[xh]+tml\b!i;
-
- my $enc = enc_for($ct, $enc_msg);
-
- if ($part_nr > 0) {
- my $fn = $part->filename;
- defined($fn) or $fn = "part #" . ($part_nr + 1);
- $rv .= add_filename_line($enc->decode($fn));
- }
-
- if (defined $full_pfx) {
- $rv .= add_text_body_short($enc, $part, $part_nr,
- $full_pfx);
- } else {
- $rv .= add_text_body_full($enc, $part, $part_nr);
- }
- $rv .= "\n" unless $rv =~ /\n\z/s;
- ++$part_nr;
+ $rv .= add_text_body($enc, $part, \$part_nr, $full_pfx, 1);
});
+ $mime->body_set('');
$rv;
}
sub add_filename_line {
- my ($fn) = @_;
+ 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\+\&\?\.\%\;/#=-]+)!;
-
-sub linkify {
- # no newlines added here since it'd break the splitting we do
- # to fold quotes
- $_[0] =~ s!$LINK_RE!<a href="$1">$1</a>!g;
+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;
+ $s;
}
-sub add_text_body_short {
- my ($enc, $part, $part_nr, $full_pfx) = @_;
- my $n = 0;
- my $s = ascii_html($enc->decode($part->body));
- linkify($s);
- $s =~ s!^((?:(?:>[^\n]*)\n)+)!
- my $cur = $1;
- my @lines = split(/\n/, $cur);
- if (@lines > MAX_INLINE_QUOTED) {
- # show a short snippet of quoted text
- $cur = join(' ', @lines);
- $cur =~ s/^>\s*//;
-
- my @sum = split(/\s+/, $cur);
- $cur = '';
- do {
- my $tmp = shift(@sum);
- my $len = length($tmp) + length($cur);
- if ($len > MAX_TRUNC_LEN) {
- @sum = ();
- } else {
- $cur .= $tmp . ' ';
- }
- } while (@sum && length($cur) < MAX_TRUNC_LEN);
- $cur =~ s/ \z/ .../;
- "> <<a\nhref=\"${full_pfx}#q${part_nr}_" . $n++ .
- "\">$cur<\/a>>\n";
+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 {
- $cur;
+ # false positive or somebody tried to mess with us
+ $key;
}
- !emg;
+ !ge;
$s;
}
-sub add_text_body_full {
- my ($enc, $part, $part_nr) = @_;
+sub flush_quote {
+ my ($quot, $n, $part_nr, $full_pfx, $final, $do_anchor) = @_;
+
+ if ($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, $rv);
+ }
+
+ # show a short snippet of quoted text and link to full version:
+ @$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 = ++$$n;
+ "> [<a\nhref=\"$full_pfx#q${part_nr}_$nr\">$cur</a>]\n";
+ } else {
+ # show everything in the full version with anchor from
+ # short version (see above)
+ my %l;
+ my $rv .= join('', map { linkify_1(\%l, $_) } @$quot);
+ @$quot = ();
+ $rv = ascii_html($rv);
+ return linkify_2(\%l, $rv) unless $do_anchor;
+ my $nr = ++$$n;
+ "<a\nid=q${part_nr}_$nr></a>" . linkify_2(\%l, $rv);
+ }
+}
+
+sub add_text_body {
+ my ($enc_msg, $part, $part_nr, $full_pfx, $do_anchor) = @_;
+ return '' if $part->subparts;
+
+ my $ct = $part->content_type;
+ # account for filter bugs...
+ if (defined $ct && $ct =~ m!\btext/x?html\b!i) {
+ $part->body_set('');
+ return '';
+ }
+ my $enc = enc_for($ct, $enc_msg);
my $n = 0;
- my $s = ascii_html($enc->decode($part->body));
- linkify($s);
- $s =~ s!^((?:(?:>[^\n]*)\n)+)!
- my $cur = $1;
- my @lines = split(/\n/, $cur);
- if (@lines > MAX_INLINE_QUOTED) {
- "<a\nid=q${part_nr}_" . $n++ . ">$cur</a>";
+ my $nr = 0;
+ my $s = $part->body;
+ $part->body_set('');
+ $s = $enc->decode($s);
+ my @lines = split(/^/m, $s);
+ $s = '';
+
+ if ($$part_nr > 0) {
+ my $fn = $part->filename;
+ defined($fn) or $fn = "part #" . ($$part_nr + 1);
+ $s .= add_filename_line($enc, $fn);
+ }
+
+ my @quot;
+ while (defined(my $cur = shift @lines)) {
+ if ($cur !~ /^>/) {
+ # show the previously buffered quote inline
+ if (scalar @quot) {
+ $s .= flush_quote(\@quot, \$n, $$part_nr,
+ $full_pfx, 0, $do_anchor);
+ }
+
+ # regular line, OK
+ my %l;
+ $cur = linkify_1(\%l, $cur);
+ $cur = ascii_html($cur);
+ $s .= linkify_2(\%l, $cur);
} else {
- $cur;
+ push @quot, $cur;
}
- !emg;
+ }
+ if (scalar @quot) {
+ $s .= flush_quote(\@quot, \$n, $$part_nr, $full_pfx, 1,
+ $do_anchor);
+ }
+ $s .= "\n" unless $s =~ /\n\z/s;
+ ++$$part_nr;
$s;
}
sub headers_to_html_header {
- my ($mime, $full_pfx) = @_;
-
+ my ($mime, $full_pfx, $ctx) = @_;
+ my $srch = $ctx->{srch} if $ctx;
my $rv = "";
my @title;
+ my $header_obj = $mime->header_obj;
+ my $mid = $header_obj->header('Message-ID');
+ $mid = PublicInbox::Hval->new_msgid($mid);
foreach my $h (qw(From To Cc Subject Date)) {
- my $v = $mime->header($h);
- defined($v) && length($v) or next;
+ my $v = $header_obj->header($h);
+ defined($v) && ($v ne '') or next;
$v = PublicInbox::Hval->new_oneline($v);
- $rv .= "$h: " . $v->as_html . "\n";
if ($h eq 'From') {
my @from = Email::Address->parse($v->raw);
- $v = $from[0]->name;
- unless (defined($v) && length($v)) {
- $v = '<' . $from[0]->address . '>';
- }
- $title[1] = ascii_html($v);
+ $title[1] = ascii_html($from[0]->name);
} elsif ($h eq 'Subject') {
$title[0] = $v->as_html;
+ if ($srch) {
+ my $p = $full_pfx ? '' : '../';
+ $rv .= "$h: <a\nid=\"t\"\nhref=\"${p}t/#u\">";
+ $rv .= $v->as_html . "</a>\n";
+ next;
+ }
+ }
+ $rv .= "$h: " . $v->as_html . "\n";
+
+ }
+ $rv .= 'Message-ID: <' . $mid->as_html . '> ';
+ my $upfx = $full_pfx ? '' : '../';
+ $rv .= "(<a\nhref=\"${upfx}raw\">raw</a>)\n";
+ my $atom;
+ if ($srch) {
+ if ($header_obj->header('In-Reply-To') ||
+ $header_obj->header('References')) {
+ $rv .= "<a\nhref=\"${upfx}t/#u\">" .
+ "References: [expand]</a>\n";
}
+
+ $atom = qq{<link\nrel=alternate\ntitle="Atom feed"\n} .
+ qq!href="${upfx}t.atom"\ntype="application/atom+xml"/>!;
+ } else {
+ $rv .= _parent_headers_nosrch($header_obj);
+ $atom = '';
}
+ $rv .= "\n";
- my $header_obj = $mime->header_obj;
- my $mid = $header_obj->header_raw('Message-ID');
- if (defined $mid) {
- $mid = PublicInbox::Hval->new_msgid($mid);
- $rv .= 'Message-ID: <' . $mid->as_html . '> ';
- my $href = $mid->as_href;
- $href = "../m/$href" unless $full_pfx;
- $rv .= "(<a\nhref=\"$href.txt\">raw</a>)\n";
- }
-
- my $irp = $header_obj->header_raw('In-Reply-To');
- if (defined $irp) {
- my $v = PublicInbox::Hval->new_msgid($irp);
+ ("<html><head><title>". join(' - ', @title) .
+ "</title>$atom</head><body>" . PRE_WRAP . $rv);
+}
+
+sub thread_inline {
+ my ($dst, $ctx, $cur, $full_pfx) = @_;
+ my $srch = $ctx->{srch};
+ my $mid = mid_clean($cur->header('Message-ID'));
+ my $res = $srch->get_thread($mid);
+ my $nr = $res->{total};
+
+ if ($nr <= 1) {
+ $$dst .= "\n[no followups, yet]\n";
+ return (undef, in_reply_to($cur));
+ }
+ my $upfx = $full_pfx ? '' : '../';
+
+ $$dst .= "\n\n~$nr messages in thread: ".
+ "(<a\nhref=\"${upfx}t/#u\">expand</a>)\n";
+ my $subj = $srch->subject_path($cur->header('Subject'));
+ my $parent = in_reply_to($cur);
+ 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);
+ }
+ ($state->{next_msg}, $state->{parent});
+}
+
+sub _parent_headers_nosrch {
+ my ($header_obj) = @_;
+ my $rv = '';
+
+ my $irt = in_reply_to($header_obj);
+ if (defined $irt) {
+ my $v = PublicInbox::Hval->new_msgid($irt, 1);
my $html = $v->as_html;
my $href = $v->as_href;
$rv .= "In-Reply-To: <";
- $rv .= "<a\nhref=\"$href.html\">$html</a>>\n";
+ $rv .= "<a\nhref=\"../$href/\">$html</a>>\n";
}
- my $refs = $header_obj->header_raw('References');
+ my $refs = $header_obj->header('References');
if ($refs) {
- $refs =~ s/\s*\Q$irp\E\s*// if (defined $irp);
- my @refs = ($refs =~ /<([^>]+)>/g);
+ # avoid redundant URLs wasting bandwidth
+ my %seen;
+ $seen{$irt} = 1 if defined $irt;
+ my @refs;
+ my @raw_refs = ($refs =~ /<([^>]+)>/g);
+ foreach my $ref (@raw_refs) {
+ next if $seen{$ref};
+ $seen{$ref} = 1;
+ push @refs, linkify_ref_nosrch($ref);
+ }
+
if (@refs) {
- $rv .= 'References: '. linkify_refs(@refs) . "\n";
+ $rv .= 'References: '. join(' ', @refs) . "\n";
}
}
-
- $rv .= "\n";
-
- ("<html><head><title>". join(' - ', @title) .
- '</title></head><body>' . PRE_WRAP . $rv);
+ $rv;
}
sub html_footer {
- my ($mime, $standalone) = @_;
+ my ($mime, $standalone, $full_pfx, $ctx) = @_;
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;
+ defined($v) && ($v ne '') or next;
my @addrs = Email::Address->parse($v);
foreach my $recip (@addrs) {
my $address = $recip->address;
Email::Address->purge_cache if $standalone;
my $subj = $mime->header('Subject') || '';
- $subj = "Re: $subj" unless $subj =~ /\bRe:/;
- my $irp = uri_escape_utf8(
- $mime->header_obj->header_raw('Message-ID') || '');
+ $subj = "Re: $subj" unless $subj =~ /\bRe:/i;
+ my $mid = $mime->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=$irp&Cc=${cc}&Subject=$subj";
+ 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 = $full_pfx ? '../' : '../../';
+ my $idx = $standalone ? " <a\nhref=\"$upfx\">index</a>" : '';
- "<a\nhref=\"" . ascii_html($href) . '">reply</a>' . $idx;
+ if ($srch && $standalone) {
+ $idx .= qq{ / follow: <a\nhref="t.atom">Atom feed</a>};
+ }
+ if ($idx && $srch) {
+ my ($next, $p) = thread_inline(\$idx, $ctx, $mime, $full_pfx);
+ if (defined $p) {
+ $p = PublicInbox::Hval->new_oneline($p);
+ $p = $p->as_href;
+ $irt = "<a\nhref=\"$upfx$p/\">parent</a> ";
+ } else {
+ $irt = ' ' x length('parent ');
+ }
+ if ($next) {
+ $irt .= "<a\nhref=\"$upfx$next/\">next</a> ";
+ } else {
+ $irt .= ' ';
+ }
+ } else {
+ $irt = '';
+ }
+
+ "$irt<a\nhref=\"" . ascii_html($href) . '">reply</a>' . $idx;
}
-sub linkify_refs {
- join(' ', map {
- my $v = PublicInbox::Hval->new_msgid($_);
- my $html = $v->as_html;
- my $href = $v->as_href;
- "<<a\nhref=\"$href.html\">$html</a>>";
- } @_);
+sub linkify_ref_nosrch {
+ my $v = PublicInbox::Hval->new_msgid($_[0], 1);
+ my $html = $v->as_html;
+ my $href = $v->as_href;
+ "<<a\nhref=\"../$href/\">$html</a>>";
}
-require Digest::SHA;
sub anchor_for {
my ($msgid) = @_;
- $msgid =~ s/\A\s*<?//;
- $msgid =~ s/>?\s*\z//;
- 'm' . Digest::SHA::sha1_hex($msgid);
+ my $id = $msgid;
+ if ($id !~ /\A[a-f0-9]{40}\z/) {
+ $id = mid_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>$s</title>".
+ qq{<link\nrel=alternate\ntitle="Atom feed"\n} .
+ qq!href="../t.atom"\ntype="application/atom+xml"/>! .
+ "</head><body>");
+}
+
+sub pre_anchor_entry {
+ my ($seen, $mime) = @_;
+ my $id = anchor_for($mime->header('Message-ID'));
+ $seen->{$id} = "#$id"; # save the anchor for children, later
+}
+
+sub ghost_parent {
+ my ($upfx, $mid) = @_;
+ # 'subject dummy' is used internally by Mail::Thread
+ return '[no common parent]' if ($mid eq 'subject dummy');
+
+ $mid = PublicInbox::Hval->new_msgid($mid);
+ my $href = $mid->as_href;
+ my $html = $mid->as_html;
+ qq{[parent not found: <<a\nhref="$upfx$href/">$html</a>>]};
+}
+
+sub ghost_table {
+ my ($upfx, $mid, $level) = @_;
+ "<table\nsummary=ghost><tr><td>" .
+ (INDENT x $level) . "</td><td>" .
+ PRE_WRAP . ghost_parent($upfx, $mid) .
+ '</pre></td></table>';
+}
+
+sub __thread_entry {
+ my ($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);
+ }
+
+ if (my $ghost = delete $state->{ghost}) {
+ # n.b. ghost messages may only be parents, not children
+ foreach my $g (@$ghost) {
+ $$cb->write(ghost_table('../../', @$g));
+ }
+ }
+ index_entry($$cb, $mime, $level, $state);
+ 1;
+}
+
+sub __ghost_entry {
+ my ($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, $level)) {
+ __ghost_entry($state, $node, $level);
+ }
+ } else {
+ __ghost_entry($state, $node, $level);
+ }
+
+ thread_entry($cb, $git, $state, $node->child, $level + 1);
+ thread_entry($cb, $git, $state, $node->next, $level);
+}
+
+sub load_results {
+ my ($res) = @_;
+
+ [ map { $_->mini_mime } @{delete $res->{msgs}} ];
+}
+
+sub msg_timestamp {
+ my ($mime) = @_;
+ my $ts = eval { str2time($mime->header('Date')) };
+ defined($ts) ? $ts : 0;
+}
+
+sub thread_results {
+ my ($msgs, $nosubject) = @_;
+ require PublicInbox::Thread;
+ my $th = PublicInbox::Thread->new(@$msgs);
+ no warnings 'once';
+ $Mail::Thread::nosubject = $nosubject;
+ $th->thread;
+ $th->order(*sort_ts);
+ $th
+}
+
+sub missing_thread {
+ my ($cb, $ctx) = @_;
+ require PublicInbox::ExtMsg;
+
+ $cb->(PublicInbox::ExtMsg::ext_msg($ctx))
+}
+
+sub _msg_date {
+ my ($mime) = @_;
+ my $ts = $mime->header('X-PI-TS') || msg_timestamp($mime);
+ POSIX::strftime('%Y-%m-%d %H:%M', gmtime($ts));
+}
+
+sub _inline_header {
+ my ($dst, $state, $upfx, $mime, $level) = @_;
+ my $pfx = INDENT x ($level - 1);
+ my $dot = $level == 0 ? '' : '` ';
+
+ my $cur = $state->{cur};
+ my $mid = mid_clean($mime->header('Message-ID'));
+ my $f = $mime->header('X-PI-From');
+ my $d = _msg_date($mime);
+ $f = PublicInbox::Hval->new($f)->as_html;
+ $d = PublicInbox::Hval->new($d)->as_html;
+ my $attr = "$f @ $d";
+ $state->{first_level} ||= $level;
+ if ($attr ne $state->{prev_attr} || $state->{prev_level} > $level) {
+ $state->{prev_attr} = $attr;
+ $attr = ' - ' . $attr;
+ $attr .= ' UTC' if $level >= $state->{first_level};
+ } else {
+ $attr = '';
+ }
+ $state->{prev_level} = $level;
+
+ if ($cur) {
+ if ($cur eq $mid) {
+ delete $state->{cur};
+ $$dst .= "$pfx$dot<b><a\nid=\"r\"\nhref=\"#t\">".
+ "[this message]</a></b>$attr\n";
+
+ return;
+ }
+ } else {
+ $state->{next_msg} ||= $mid;
+ }
+
+ # Subject is never undef, this mail was loaded from
+ # our Xapian which would've resulted in '' if it were
+ # really missing (and Filter rejects empty subjects)
+ my $s = $mime->header('Subject');
+ my $h = $state->{srch}->subject_path($s);
+ if ($state->{seen}->{$h}) {
+ $s = undef;
+ } else {
+ $state->{seen}->{$h} = 1;
+ $s = PublicInbox::Hval->new($s);
+ $s = $s->as_html;
+ }
+ my $m = PublicInbox::Hval->new_msgid($mid);
+ $m = $upfx . '../' . $m->as_href . '/';
+ if (defined $s) {
+ $$dst .= "$pfx$dot<a\nhref=\"$m\">$s</a>$attr\n";
+ } else {
+ $$dst .= "$pfx$dot<a\nhref=\"$m\">$f @ $d</a>\n";
+ }
+}
+
+sub inline_dump {
+ my ($dst, $state, $upfx, $node, $level) = @_;
+ return unless $node;
+ if (my $mime = $node->message) {
+ my $mid = mid_clean($mime->header('Message-ID'));
+ if ($mid eq $state->{parent_cmp}) {
+ $state->{parent} = $mid;
+ }
+ _inline_header($dst, $state, $upfx, $mime, $level);
+ } else {
+ my $dot = $level == 0 ? '' : '` ';
+ my $pfx = (INDENT x $level) . $dot;
+ $$dst .= $pfx;
+ $$dst .= ghost_parent("$upfx../", $node->messageid) . "\n";
+ }
+ inline_dump($dst, $state, $upfx, $node->child, $level+1);
+ inline_dump($dst, $state, $upfx, $node->next, $level);
+}
+
+sub sort_ts {
+ sort {
+ (eval { $a->topmost->message->header('X-PI-TS') } || 0) <=>
+ (eval { $b->topmost->message->header('X-PI-TS') } || 0)
+ } @_;
+}
+
+sub rsort_ts {
+ sort {
+ (eval { $b->topmost->message->header('X-PI-TS') } || 0) <=>
+ (eval { $a->topmost->message->header('X-PI-TS') } || 0)
+ } @_;
+}
+
+# accumulate recent topics if search is supported
+# returns 1 if done, undef if not
+sub add_topic {
+ my ($state, $node, $level) = @_;
+ return unless $node;
+ my $child_adjust = 1;
+
+ if (my $x = $node->message) {
+ $x = $x->header_obj;
+ my ($topic, $subj);
+
+ $subj = $x->header('Subject');
+ $subj = $state->{srch}->subject_normalized($subj);
+ $topic = $subj;
+
+ # kill "[PATCH v2]" etc. for summarization
+ $topic =~ s/\A\s*\[[^\]]+\]\s*//g;
+ $topic = substr($topic, 0, 30);
+
+ if (++$state->{subjs}->{$topic} == 1) {
+ push @{$state->{order}}, [ $level, $subj, $topic ];
+ }
+
+ my $mid = mid_clean($x->header('Message-ID'));
+
+ my $u = $x->header('X-PI-From');
+ my $ts = $x->header('X-PI-TS');
+ $state->{latest}->{$topic} = [ $mid, $u, $ts ];
+ } else {
+ # ghost message, do not bump level
+ $child_adjust = 0;
+ }
+
+ add_topic($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 = $state->{latest};
+ return "\n[No recent topics]</pre>" unless (scalar @$order);
+ my $dst = '';
+ my $pfx;
+ my $prev = 0;
+ my $prev_attr = '';
+ while (defined(my $info = shift @$order)) {
+ my ($level, $subj, $topic) = @$info;
+ my $n = delete $subjs->{$topic};
+ my ($mid, $u, $ts) = @{delete $latest->{$topic}};
+ $mid = PublicInbox::Hval->new($mid)->as_href;
+ $subj = PublicInbox::Hval->new($subj)->as_html;
+ $u = PublicInbox::Hval->new($u)->as_html;
+ $pfx = INDENT x ($level - 1);
+ my $nl = $level == $prev ? "\n" : '';
+ my $dot = $level == 0 ? '' : '` ';
+ $dst .= "$nl$pfx$dot<a\nhref=\"$mid/t/#u\"><b>$subj</b></a>\n";
+
+ my $attr;
+ $ts = POSIX::strftime('%Y-%m-%d %H:%M', gmtime($ts));
+ if ($n == 1) {
+ $attr = "created by $u @ $ts UTC";
+ $n = "\n";
+ } else {
+ # $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
+ $attr = "updated by $u @ $ts UTC";
+ $n = " ($n)\n";
+ }
+ if ($level == 0 || $attr ne $prev_attr) {
+ $pfx .= INDENT if $level > 0;
+ $dst .= "$pfx- ". $attr . $n;
+ $prev_attr = $attr;
+ }
+ }
+ $dst .= '</pre>';
+}
+
+sub emit_index_topics {
+ my ($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;
+
+ for (rsort_ts(thread_results(load_results($res), 1)->rootset)) {
+ add_topic($state, $_, 0);
+ }
+ $opts{offset} += $nr;
+ }
+
+ $fh->write(dump_topics($state));
+ $opts{offset};
}
1;