-# Copyright (C) 2014-2018 all contributors <meta@public-inbox.org>
+# Copyright (C) 2014-2019 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.
package PublicInbox::View;
use strict;
use warnings;
+use bytes (); # only for bytes::length
use PublicInbox::MsgTime qw(msg_datestamp);
-use PublicInbox::Hval qw/ascii_html obfuscate_addrs/;
+use PublicInbox::Hval qw(ascii_html obfuscate_addrs prurl);
use PublicInbox::Linkify;
-use PublicInbox::MID qw/id_compress mid_escape mids references/;
+use PublicInbox::MID qw/id_compress mid_escape mids mids_for_index references/;
use PublicInbox::MsgIter;
use PublicInbox::Address;
use PublicInbox::WwwStream;
use PublicInbox::Reply;
-require POSIX;
+use PublicInbox::ViewDiff qw(flush_diff);
+use POSIX qw(strftime);
use Time::Local qw(timegm);
-
+use PublicInbox::SearchMsg qw(subject_normalized);
use constant COLS => 72;
use constant INDENT => ' ';
use constant TCHILD => '` ';
sub th_pfx ($) { $_[0] == 0 ? '' : TCHILD };
+sub msg_html_i {
+ my ($nr, $ctx) = @_;
+ my $more = $ctx->{more};
+ if ($nr == 1) {
+ # $more cannot be true w/o $smsg being defined:
+ my $upfx = $more ? '../'.mid_escape($ctx->{smsg}->mid).'/' : '';
+ $ctx->{tip} .
+ multipart_text_as_html(delete $ctx->{mime}, $upfx,
+ $ctx) . '</pre><hr>'
+ } elsif ($more) {
+ ++$ctx->{end_nr};
+ # fake an EOF if {more} retrieval fails fails;
+ eval { msg_html_more($ctx, $nr) };
+ } elsif ($nr == $ctx->{end_nr}) {
+ # fake an EOF if generating the footer fails;
+ # we want to at least show the message if something
+ # here crashes:
+ eval { html_footer($ctx) };
+ } else {
+ undef
+ }
+}
+
# public functions: (unstable)
sub msg_html {
- my ($ctx, $mime, $more, $smsg) = @_;
- my $hdr = $mime->header_obj;
+ my ($ctx, $mime, $smsg) = @_;
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
- }
- });
+ $ctx->{-obfs_ibx} = $ibx->{obfuscate} ? $ibx : undef;
+ my $hdr = $ctx->{hdr} = $mime->header_obj;
+ $ctx->{tip} = _msg_html_prepare($hdr, $ctx, 0);
+ $ctx->{end_nr} = 2;
+ $ctx->{smsg} = $smsg;
+ $ctx->{mime} = $mime;
+ PublicInbox::WwwStream->response($ctx, 200, \&msg_html_i);
}
sub msg_page {
my ($ctx) = @_;
my $mid = $ctx->{mid};
my $ibx = $ctx->{-inbox};
- my ($first, $more);
+ my ($first);
my $smsg;
- if (my $srch = $ibx->search) {
+ if (my $over = $ibx->over) {
my ($id, $prev);
- $smsg = $srch->next_by_mid($mid, \$id, \$prev);
+ $smsg = $over->next_by_mid($mid, \$id, \$prev);
$first = $ibx->msg_by_smsg($smsg) if $smsg;
if ($first) {
- my $next = $srch->next_by_mid($mid, \$id, \$prev);
- $more = [ $id, $prev, $next ] if $next;
+ my $next = $over->next_by_mid($mid, \$id, \$prev);
+ $ctx->{more} = [ $id, $prev, $next ] if $next;
}
return unless $first;
} else {
$first = $ibx->msg_by_mid($mid) or return;
}
- msg_html($ctx, PublicInbox::MIME->new($first), $more, $smsg);
+ msg_html($ctx, PublicInbox::MIME->new($first), $smsg);
}
sub msg_html_more {
- my ($ctx, $more, $nr) = @_;
- my $str = eval {
- my ($id, $prev, $smsg) = @$more;
- my $mid = $ctx->{mid};
- $smsg = $ctx->{-inbox}->smsg_mime($smsg);
- my $next = $ctx->{srch}->next_by_mid($mid, \$id, \$prev);
- @$more = $next ? ($id, $prev, $next) : ();
- if ($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>';
- }
- $str;
+ my ($ctx, $nr) = @_;
+ my ($id, $prev, $smsg) = @{$ctx->{more}};
+ my $ibx = $ctx->{-inbox};
+ $smsg = $ibx->smsg_mime($smsg);
+ my $next = $ibx->over->next_by_mid($ctx->{mid}, \$id, \$prev);
+ $ctx->{more} = $next ? [ $id, $prev, $next ] : undef;
+ return '' unless $smsg;
+ my $upfx = '../' . mid_escape($smsg->mid) . '/';
+ my $mime = delete $smsg->{mime};
+ _msg_html_prepare($mime->header_obj, $ctx, $nr) .
+ multipart_text_as_html($mime, $upfx, $ctx) .
+ '</pre><hr>'
}
# /$INBOX/$MESSAGE_ID/#R
-sub msg_reply {
+sub msg_reply ($$) {
my ($ctx, $hdr) = @_;
my $se_url =
'https://kernel.org/pub/software/scm/git/docs/git-send-email.html';
my $info = '';
my $ibx = $ctx->{-inbox};
if (my $url = $ibx->{infourl}) {
- $url = PublicInbox::Hval::prurl($ctx->{env}, $url);
+ $url = prurl($ctx->{env}, $url);
$info = qq(\n List information: <a\nhref="$url">$url</a>\n);
}
my ($arg, $link, $reply_to_all) =
PublicInbox::Reply::mailto_arg_link($ibx, $hdr);
+ if (ref($arg) eq 'SCALAR') {
+ return '<pre id=R>'.ascii_html($$arg).'</pre>';
+ }
# mailto: link only works if address obfuscation is disabled
if ($link) {
<hr><pre
id=R><b>Reply instructions:</b>
-You may reply public
ally to <a
+You may reply publicly to <a
href=#t>this message</a> via plain-text email
using any one of the following methods:
sub _hdr_names_html ($$) {
my ($hdr, $field) = @_;
- my $val = $hdr->header($field) or return '';
- ascii_html(join(', ', PublicInbox::Address::names($val)));
+ my @vals = $hdr->header($field) or return '';
+ ascii_html(join(', ', PublicInbox::Address::names(join(',', @vals))));
}
sub nr_to_s ($$$) {
$nr == 1 ? "$nr $singular" : "$nr $plural";
}
+# human-friendly format
+sub fmt_ts ($) { strftime('%Y-%m-%d %k:%M', gmtime($_[0])) }
+
# this is already inside a <pre>
sub index_entry {
my ($smsg, $ctx, $more) = @_;
- my $srch = $ctx->{srch};
my $subj = $smsg->subject;
my $mid_raw = $smsg->mid;
my $id = id_compress($mid_raw, 1);
my $irt;
my $obfs_ibx = $ctx->{-obfs_ibx};
+ $subj = '(no subject)' if $subj eq '';
my $rv = "<a\nhref=#e$id\nid=m$id>*</a> ";
$subj = '<b>'.ascii_html($subj).'</b>';
obfuscate_addrs($obfs_ibx, $subj) if $obfs_ibx;
$rv .= $subj . "\n";
$rv .= _th_index_lite($mid_raw, \$irt, $id, $ctx);
my @tocc;
- my $mime = $smsg->{mime};
+ my $ds = $smsg->ds; # for v1 non-Xapian/SQLite users
+ # deleting {mime} is critical to memory use,
+ # the rest of the fields saves about 400K as we iterate across 1K msgs
+ my ($mime) = delete @$smsg{qw(mime ds ts blob subject)};
+
my $hdr = $mime->header_obj;
my $from = _hdr_names_html($hdr, 'From');
obfuscate_addrs($obfs_ibx, $from) if $obfs_ibx;
- $rv .= "From: $from @ ".fmt_ts($smsg->ds)." UTC";
+ $rv .= "From: $from @ ".fmt_ts($ds)." UTC";
my $upfx = $ctx->{-upfx};
my $mhref = $upfx . mid_escape($mid_raw) . '/';
$rv .= qq{ (<a\nhref="$mhref">permalink</a> / };
$rv .= "\n";
# scan through all parts, looking for displayable text
- msg_iter($mime, sub { $rv .= add_text_body($mhref, $obfs_ibx, $_[0]) });
+ $ctx->{mhref} = $mhref;
+ $ctx->{rv} = \$rv;
+ msg_iter($mime, \&add_text_body, $ctx, 1);
+ delete $ctx->{rv};
# add the footer
$rv .= "\n<a\nhref=#$id_m\nid=e$id>^</a> ".
" <a\nhref=\"${mhref}#R\">reply</a>";
my $hr;
- if (my $pct = $ctx->{pct}) { # used by SearchView.pm
- $rv .= "\t[relevance $pct->{$mid_raw}%]";
+ if (defined(my $pct = $smsg->{pct})) { # used by SearchView.pm
+ $rv .= "\t[relevance $pct%]";
$hr = 1;
} elsif ($mapping) {
my $nested = 'nested';
my $nr_s = 0;
my $siblings;
if (my $smsg = $node->{smsg}) {
- ($$irt) = (($smsg->{references} || '') =~ m/<([^>]+)>\z/);
+ # delete saves about 200KB on a 1K message thread
+ if (my $refs = delete $smsg->{references}) {
+ ($$irt) = ($refs =~ m/<([^>]+)>\z/);
+ }
}
my $irt_map = $mapping->{$$irt} if defined $$irt;
if (defined $irt_map) {
$rv .= $pad ."<a\nhref=#r$id>$s_s, $s_c; $ctx->{s_nr}</a>\n";
}
-sub walk_thread {
+# non-recursive thread walker
+sub walk_thread ($$$) {
my ($rootset, $ctx, $cb) = @_;
my @q = map { (0, $_, -1) } @$rootset;
while (@q) {
my ($level, $node, $i) = splice(@q, 0, 3);
defined $node or next;
- $cb->($ctx, $level, $node, $i);
+ $cb->($ctx, $level, $node, $i) or return;
++$level;
$i = 0;
unshift @q, map { ($level, $_, $i++) } @{$node->{children}};
}
}
-sub pre_thread {
+sub pre_thread { # walk_thread callback
my ($ctx, $level, $node, $idx) = @_;
$ctx->{mapping}->{$node->{id}} = [ '', $node, $idx, $level ];
skel_dump($ctx, $level, $node);
$beg . '<pre>' . index_entry($smsg, $ctx, 0) . '</pre>' . $end;
}
+sub stream_thread_i { # PublicInbox::WwwStream::getline callback
+ my ($nr, $ctx) = @_;
+ return unless exists($ctx->{skel});
+ my $q = $ctx->{-queue};
+ while (@$q) {
+ my $level = shift @$q;
+ my $node = shift @$q or next;
+ my $cl = $level + 1;
+ unshift @$q, map { ($cl, $_) } @{$node->{children}};
+ if (my $smsg = $ctx->{-inbox}->smsg_mime($node->{smsg})) {
+ return thread_index_entry($ctx, $level, $smsg);
+ } else {
+ return ghost_index_entry($ctx, $level, $node);
+ }
+ }
+ join('', thread_adj_level($ctx, 0)) . ${delete $ctx->{skel}};
+}
+
sub stream_thread ($$) {
my ($rootset, $ctx) = @_;
- my $inbox = $ctx->{-inbox};
+ my $ibx = $ctx->{-inbox};
my @q = map { (0, $_) } @$rootset;
- my $level;
- my $smsg;
+ my ($smsg, $level);
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;
+ $smsg = $ibx->smsg_mime($node->{smsg}) and last;
}
return missing_thread($ctx) unless $smsg;
- $ctx->{-obfs_ibx} = $inbox->{obfuscate} ? $inbox : undef;
+ $ctx->{-obfs_ibx} = $ibx->{obfuscate} ? $ibx : 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;
- });
+ $ctx->{-queue} = \@q;
+ PublicInbox::WwwStream->response($ctx, 200, \&stream_thread_i);
}
+# /$INBOX/$MESSAGE_ID/t/
sub thread_html {
my ($ctx) = @_;
my $mid = $ctx->{mid};
- my $srch = $ctx->{srch};
- my ($nr, $msgs) = $srch->get_thread($mid);
+ my $ibx = $ctx->{-inbox};
+ my ($nr, $msgs) = $ibx->over->get_thread($mid);
return missing_thread($ctx) if $nr == 0;
my $skel = '<hr><pre>';
$skel .= $nr == 1 ? 'only message in thread' : 'end of thread';
$skel .= "-- links below jump to the message on this page --\n";
$ctx->{-upfx} = '../../';
$ctx->{cur_level} = 0;
- $ctx->{dst} = \$skel;
+ $ctx->{skel} = \$skel;
$ctx->{prev_attr} = '';
$ctx->{prev_level} = 0;
$ctx->{root_anchor} = anchor_for($mid);
$ctx->{mapping} = {};
- $ctx->{s_nr} = "$nr+ messages in thread";
+ $ctx->{s_nr} = ($nr > 1 ? "$nr+ messages" : 'only message')
+ .' in thread';
my $rootset = thread_results($ctx, $msgs);
# reduce hash lookups in pre_thread->skel_dump
- my $inbox = $ctx->{-inbox};
- $ctx->{-obfs_ibx} = $inbox->{obfuscate} ? $inbox : undef;
- walk_thread($rootset, $ctx, *pre_thread);
+ $ctx->{-obfs_ibx} = $ibx->{obfuscate} ? $ibx : undef;
+ walk_thread($rootset, $ctx, \&pre_thread);
$skel .= '</pre>';
return stream_thread($rootset, $ctx) unless $ctx->{flat};
# flat display: lazy load the full message from smsg
my $smsg;
while (my $m = shift @$msgs) {
- $smsg = $inbox->smsg_mime($m) and last;
+ $smsg = $ibx->smsg_mime($m) and last;
}
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;
- });
+ $ctx->{msgs} = $msgs;
+ PublicInbox::WwwStream->response($ctx, 200, \&thread_html_i);
+}
+
+sub thread_html_i { # PublicInbox::WwwStream::getline callback
+ my ($nr, $ctx) = @_;
+ my $msgs = $ctx->{msgs} or return;
+ while (my $smsg = shift @$msgs) {
+ $ctx->{-inbox}->smsg_mime($smsg) or next;
+ return index_entry($smsg, $ctx, scalar @$msgs);
+ }
+ my ($skel) = delete @$ctx{qw(skel msgs)};
+ $$skel;
}
sub multipart_text_as_html {
- my ($mime, $upfx, $obfs_ibx) = @_;
- my $rv = "";
+ my (undef, $mhref, $ctx) = @_; # $mime = $_[0]
+ $ctx->{mhref} = $mhref;
+ $ctx->{rv} = \(my $rv = '');
# scan through all parts, looking for displayable text
- msg_iter($mime, sub { $rv .= add_text_body($upfx, $obfs_ibx, $_[0]) });
- $rv;
+ msg_iter($_[0], \&add_text_body, $ctx, 1);
+ ${delete $ctx->{rv}};
}
sub flush_quote {
# show everything in the full version with anchor from
# short version (see above)
- my $rv = $l->linkify_1(join('', @$quot));
- @$quot = ();
+ my $rv = $l->linkify_1($$quot);
# we use a <span> here to allow users to specify their own
# color for quoted text
$rv = $l->linkify_2(ascii_html($rv));
+ $$quot = undef;
$$s .= qq(<span\nclass="q">) . $rv . '</span>'
}
sub attach_link ($$$$;$) {
- my ($upfx, $ct, $p, $fn, $err) = @_;
+ my ($ctx, $ct, $p, $fn, $err) = @_;
my ($part, $depth, @idx) = @$p;
my $nl = $idx[-1] > 1 ? "\n" : '';
my $idx = join('.', @idx);
$desc = $fn unless defined $desc;
$desc = '' unless defined $desc;
my $sfn;
- if (defined $fn && $fn =~ /\A[[:alnum:]][\w\.-]+[[:alnum:]]\z/) {
+ if (defined $fn && $fn =~ /\A$PublicInbox::Hval::FN\z/o) {
$sfn = $fn;
} elsif ($ct eq 'text/plain') {
$sfn = 'a.txt';
} else {
$sfn = 'a.bin';
}
- my $ret = qq($nl<a\nhref="$upfx$idx-$sfn">);
+ my $rv = $ctx->{rv};
+ $$rv .= qq($nl<a\nhref="$ctx->{mhref}$idx-$sfn">);
if ($err) {
- $ret .=
-"[-- Warning: decoded text below may be mangled --]\n";
+ $$rv .= "[-- Warning: decoded text below may be mangled --]\n";
}
- $ret .= "[-- Attachment #$idx: ";
+ $$rv .= "[-- Attachment #$idx: ";
my $ts = "Type: $ct, Size: $size bytes";
$desc = ascii_html($desc);
- $ret .= ($desc eq '') ? "$ts --]" : "$desc --]\n[-- $ts --]";
- $ret .= "</a>\n";
+ $$rv .= ($desc eq '') ? "$ts --]" : "$desc --]\n[-- $ts --]";
+ $$rv .= "</a>\n";
+ undef;
}
-sub add_text_body {
- my ($upfx, $obfs_ibx, $p) = @_;
+sub add_text_body { # callback for msg_iter
+ my ($p, $ctx) = @_;
+ my $upfx = $ctx->{mhref};
+ my $ibx = $ctx->{-inbox};
# $p - from msg_iter: [ Email::MIME, depth, @idx ]
- my ($part, $depth) = @$p; # attachment @idx is unused
+ my ($part, $depth, @idx) = @$p;
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 forcing charset=UTF-8 failed,
- # attach_link will warn further down...
- $s = $part->body if $@;
- } else {
- return attach_link($upfx, $ct, $p, $fn);
+ my ($s, $err) = msg_part_text($part, $ct);
+ return attach_link($ctx, $ct, $p, $fn) unless defined $s;
+
+ # makes no difference to browsers, and don't screw up filename
+ # link generation in diffs with the extra '%0D'
+ $s =~ s/\r\n/\n/sg;
+
+ # always support diff-highlighting, but we can't linkify hunk
+ # headers for solver unless some coderepo are configured:
+ my $diff;
+ if ($s =~ /^(?:diff|---|\+{3}) /ms) {
+ # diffstat anchors do not link across attachments or messages:
+ $idx[0] = $upfx . $idx[0] if $upfx ne '';
+ $ctx->{-apfx} = join('/', @idx);
+ $ctx->{-anchors} = {}; # attr => filename
+ $ctx->{-diff} = $diff = [];
+ delete $ctx->{-long_path};
+ my $spfx;
+ if ($ibx->{-repo_objs}) {
+ if (index($upfx, '//') >= 0) { # absolute URL (Atom feeds)
+ $spfx = $upfx;
+ $spfx =~ s!/([^/]*)/\z!/!;
+ } else {
+ my $n_slash = $upfx =~ tr!/!/!;
+ if ($n_slash == 0) {
+ $spfx = '../';
+ } elsif ($n_slash == 1) {
+ $spfx = '';
+ } else { # nslash == 2
+ $spfx = '../../';
+ }
+ }
}
- }
+ $ctx->{-spfx} = $spfx;
+ };
- my @lines = split(/^/m, $s);
+ # some editors don't put trailing newlines at the end:
+ $s .= "\n" unless $s =~ /\n\z/s;
+
+ # split off quoted and unquoted blocks:
+ my @sections = split(/((?:^>[^\n]*\n)+)/sm, $s);
$s = '';
+ my $rv = $ctx->{rv};
if (defined($fn) || $depth > 0 || $err) {
- $s .= attach_link($upfx, $ct, $p, $fn, $err);
- $s .= "\n";
+ # badly-encoded message with $err? tell the world about it!
+ attach_link($ctx, $ct, $p, $fn, $err);
+ $$rv .= "\n";
}
- my @quot;
my $l = PublicInbox::Linkify->new;
- foreach my $cur (@lines) {
- if ($cur !~ /^>/) {
- # show the previously buffered quote inline
- flush_quote(\$s, $l, \@quot) if @quot;
-
- # regular line, OK
- $l->linkify_1($cur);
- $s .= $l->linkify_2(ascii_html($cur));
+ foreach my $cur (@sections) {
+ if ($cur =~ /\A>/) {
+ flush_quote($rv, $l, \$cur);
+ } elsif ($diff) {
+ @$diff = split(/^/m, $cur);
+ $cur = undef;
+ flush_diff($rv, $ctx, $l);
} else {
- push @quot, $cur;
+ # regular lines, OK
+ $l->linkify_1($cur);
+ $$rv .= $l->linkify_2(ascii_html($cur));
+ $cur = undef;
}
}
- 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...
- $s .= "\n";
- }
- }
+ obfuscate_addrs($ibx, $$rv) if $ibx->{obfuscate};
}
sub _msg_html_prepare {
- my ($hdr, $ctx, $more, $nr) = @_;
- my $srch = $ctx->{srch} if $ctx;
+ my ($hdr, $ctx, $nr) = @_;
my $atom = '';
+ my $over = $ctx->{-inbox}->over;
my $obfs_ibx = $ctx->{-obfs_ibx};
my $rv = '';
- my $mids = mids($hdr);
- my $multiple = scalar(@$mids) > 1; # zero, one, infinity
+ my $mids = mids_for_index($hdr);
if ($nr == 0) {
- if ($more) {
+ if ($ctx->{more}) {
$rv .=
-"<pre>WARNING: multiple messages refer to this Message-ID\n</pre>";
+"<pre>WARNING: multiple messages have this Message-ID\n</pre>";
}
$rv .= "<pre\nid=b>"; # anchor for body start
} else {
$rv .= '<pre>';
}
- if ($srch) {
+ if ($over) {
$ctx->{-upfx} = '../';
}
- my @title;
- my $v;
- if (defined($v = $hdr->header('From'))) {
+ my @title; # (Subject[0], From[0])
+ for my $v ($hdr->header('From')) {
$v = PublicInbox::Hval->new($v);
my @n = PublicInbox::Address::names($v->raw);
- $title[1] = ascii_html(join(', ', @n));
+ $title[1] //= ascii_html(join(', ', @n));
$v = $v->as_html;
if ($obfs_ibx) {
obfuscate_addrs($obfs_ibx, $v);
$rv .= "From: $v\n" if $v ne '';
}
foreach my $h (qw(To Cc)) {
- defined($v = $hdr->header($h)) or next;
- fold_addresses($v);
- $v = ascii_html($v);
- obfuscate_addrs($obfs_ibx, $v) if $obfs_ibx;
- $rv .= "$h: $v\n" if $v ne '';
+ for my $v ($hdr->header($h)) {
+ fold_addresses($v);
+ $v = ascii_html($v);
+ obfuscate_addrs($obfs_ibx, $v) if $obfs_ibx;
+ $rv .= "$h: $v\n" if $v ne '';
+ }
}
- if (defined($v = $hdr->header('Subject')) && ($v ne '')) {
- $v = ascii_html($v);
- obfuscate_addrs($obfs_ibx, $v) if $obfs_ibx;
- if ($srch) {
- $rv .= qq(Subject: <a\nhref="#r"\nid=t>$v</a>\n);
- } else {
- $rv .= "Subject: $v\n";
+ my @subj = $hdr->header('Subject');
+ if (@subj) {
+ for my $v (@subj) {
+ $v = ascii_html($v);
+ obfuscate_addrs($obfs_ibx, $v) if $obfs_ibx;
+ $rv .= 'Subject: ';
+ if ($over) {
+ $rv .= qq(<a\nhref="#r"\nid=t>$v</a>\n);
+ } else {
+ $rv .= "$v\n";
+ }
+ $title[0] //= $v;
}
- $title[0] = $v;
} else { # dummy anchor for thread skeleton at bottom of page
- $rv .= qq(<a\nhref="#r"\nid=t></a>) if $srch;
+ $rv .= qq(<a\nhref="#r"\nid=t></a>) if $over;
$title[0] = '(no subject)';
}
- if (defined($v = $hdr->header('Date'))) {
+ for my $v ($hdr->header('Date')) {
$v = ascii_html($v);
obfuscate_addrs($obfs_ibx, $v) if $obfs_ibx; # possible :P
$rv .= "Date: $v\n";
}
$ctx->{-title_html} = join(' - ', @title);
- foreach (@$mids) {
- my $mid = PublicInbox::Hval->new_msgid($_) ;
+ if (scalar(@$mids) == 1) { # common case
+ my $mid = PublicInbox::Hval->new_msgid($mids->[0]);
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 .= "Message-ID: <$mhtml> ";
+ $rv .= "(<a\nhref=\"raw\">raw</a>)\n";
+ } else {
+ # X-Alt-Message-ID can happen if a message is injected from
+ # public-inbox-nntpd because of multiple Message-ID headers.
+ my $lnk = PublicInbox::Linkify->new;
+ my $s = '';
+ for my $h (qw(Message-ID X-Alt-Message-ID)) {
+ $s .= "$h: $_\n" for ($hdr->header_raw($h));
}
+ $lnk->linkify_mids('..', \$s, 1);
+ $rv .= $s;
}
- $rv .= _parent_headers($hdr, $srch);
+ $rv .= _parent_headers($hdr, $over);
$rv .= "\n";
}
-sub thread_skel {
- my ($dst, $ctx, $hdr, $tpfx) = @_;
- my $srch = $ctx->{srch};
- my $mid = mids($hdr)->[0];
- my ($nr, $msgs) = $srch->get_thread($mid);
- 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>);
+sub SKEL_EXPAND () {
+ qq(expand[<a\nhref="T/#u">flat</a>) .
+ qq(|<a\nhref="t/#u">nested</a>] ) .
+ qq(<a\nhref="t.mbox.gz">mbox.gz</a> ) .
+ qq(<a\nhref="t.atom">Atom feed</a>);
+}
+sub thread_skel ($$$) {
+ my ($skel, $ctx, $hdr) = @_;
+ my $mid = mids($hdr)->[0];
+ my $ibx = $ctx->{-inbox};
+ my ($nr, $msgs) = $ibx->over->get_thread($mid);
my $parent = in_reply_to($hdr);
- $$dst .= "\n<b>Thread overview: </b>";
+ $$skel .= "\n<b>Thread overview: </b>";
if ($nr <= 1) {
if (defined $parent) {
- $$dst .= "$expand\n ";
- $$dst .= ghost_parent("$tpfx../", $parent) . "\n";
+ $$skel .= SKEL_EXPAND."\n ";
+ $$skel .= ghost_parent('../', $parent) . "\n";
} else {
- $$dst .= "[no followups] $expand\n";
+ $$skel .= '[no followups] '.SKEL_EXPAND."\n";
}
$ctx->{next_msg} = undef;
$ctx->{parent_msg} = $parent;
return;
}
- $$dst .= "$nr+ messages / $expand";
- $$
dst .= qq! <a\nhref="#b">top</a>\n!;
+ $$skel .= $nr;
+ $$
skel .= '+ messages / '.SKEL_EXPAND.qq! <a\nhref="#b">top</a>\n!;
- my $subj = $hdr->header('Subject');
- defined $subj or $subj = '';
- $ctx->{prev_subj} = [ split(/ /, $srch->subject_normalized($subj)) ];
+ # nb: mutt only shows the first Subject in the index pane
+ # when multiple Subject: headers are present, so we follow suit:
+ my $subj = $hdr->header('Subject') // '';
+ $subj = '(no subject)' if $subj eq '';
+ $ctx->{prev_subj} = [ split(/ /, subject_normalized($subj)) ];
$ctx->{cur} = $mid;
$ctx->{prev_attr} = '';
$ctx->{prev_level} = 0;
- $ctx->{dst} = $dst;
+ $ctx->{skel} = $skel;
# reduce hash lookups in skel_dump
- my $ibx = $ctx->{-inbox};
$ctx->{-obfs_ibx} = $ibx->{obfuscate} ? $ibx : undef;
- walk_thread(thread_results($ctx, $msgs), $ctx, *skel_dump);
+ walk_thread(thread_results($ctx, $msgs), $ctx, \&skel_dump);
$ctx->{parent_msg} = $parent;
}
sub _parent_headers {
- my ($hdr, $srch) = @_;
+ my ($hdr, $over) = @_;
my $rv = '';
-
- my $irt = in_reply_to($hdr);
- if (defined $irt) {
- my $v = PublicInbox::Hval->new_msgid($irt);
- my $html = $v->as_html;
- my $href = $v->{href};
- $rv .= "In-Reply-To: <";
- $rv .= "<a\nhref=\"../$href/\">$html</a>>\n";
+ my @irt = $hdr->header_raw('In-Reply-To');
+ my $refs;
+ if (@irt) {
+ my $lnk = PublicInbox::Linkify->new;
+ $rv .= "In-Reply-To: $_\n" for @irt;
+ $lnk->linkify_mids('..', \$rv);
+ } else {
+ $refs = references($hdr);
+ my $irt = pop @$refs;
+ if (defined $irt) {
+ my $v = PublicInbox::Hval->new_msgid($irt);
+ my $html = $v->as_html;
+ my $href = $v->{href};
+ $rv .= "In-Reply-To: <";
+ $rv .= "<a\nhref=\"../$href/\">$html</a>>\n";
+ }
}
# do not display References: if search is present,
# we show the thread skeleton at the bottom, instead.
- return $rv if $srch;
-
- my $refs = $hdr->header_raw('References');
- if ($refs) {
- # avoid redundant URLs wasting bandwidth
- my %seen;
- $seen{$irt} = 1 if defined $irt;
- my @refs;
- my @raw_refs = ($refs =~ /<([^>]+)>/g);
- foreach my $ref (@raw_refs) {
- next if $seen{$ref};
- $seen{$ref} = 1;
- push @refs, linkify_ref_nosrch($ref);
- }
+ return $rv if $over;
- if (@refs) {
- $rv .= 'References: '. join("\n\t", @refs) . "\n";
- }
+ $refs //= references($hdr);
+ if (@$refs) {
+ @$refs = map { linkify_ref_no_over($_) } @$refs;
+ $rv .= 'References: '. join("\n\t", @$refs) . "\n";
}
$rv;
}
+# returns a string buffer via ->getline
sub html_footer {
- my ($hdr, $standalone, $ctx, $rhref) = @_;
-
- my $srch = $ctx->{srch} if $ctx;
+ my ($ctx) = @_;
+ my $ibx = $ctx->{-inbox};
+ my $hdr = delete $ctx->{hdr};
my $upfx = '../';
- my $tpfx = '';
- my $idx = $standalone ? " <a\nhref=\"$upfx\">index</a>" : '';
- my $irt = '';
- if ($idx && $srch) {
- $idx .= "\n";
- thread_skel(\$idx, $ctx, $hdr, $tpfx);
+ my $skel = " <a\nhref=\"$upfx\">index</a>";
+ my $rv = '<pre>';
+ if ($ibx->over) {
+ $skel .= "\n";
+ thread_skel(\$skel, $ctx, $hdr);
my ($next, $prev);
my $parent = ' ';
$next = $prev = ' ';
} elsif ($u) { # unlikely
$parent = " <a\nhref=\"$u\"\nrel=prev>parent</a>";
}
- $irt = "$next $prev$parent ";
- } else {
- $irt = '';
+ $rv .= "$next $prev$parent ";
}
- $rhref ||= '#R';
- $irt .= qq(<a\nhref="$rhref">reply</a>);
- $irt .= $idx;
+ $rv .= qq(<a\nhref="#R">reply</a>);
+ $rv .= $skel;
+ $rv .= '</pre>';
+ $rv .= msg_reply($ctx, $hdr);
}
-sub linkify_ref_nosrch {
+sub linkify_ref_no_over {
my $v = PublicInbox::Hval->new_msgid($_[0]);
my $html = $v->as_html;
my $href = $v->{href};
$level ? INDENT x ($level - 1) : '';
}
+sub find_mid_root {
+ my ($ctx, $level, $node, $idx) = @_;
+ ++$ctx->{root_idx} if $level == 0;
+ if ($node->{id} eq $ctx->{mid}) {
+ $ctx->{found_mid_at} = $ctx->{root_idx};
+ return 0;
+ }
+ 1;
+}
+
+sub strict_loose_note ($) {
+ my ($nr) = @_;
+ my $msg =
+" -- strict thread matches above, loose matches on Subject: below --\n";
+
+ if ($nr > PublicInbox::Over::DEFAULT_LIMIT()) {
+ $msg .=
+" -- use mbox.gz link to download all $nr messages --\n";
+ }
+ $msg;
+}
+
sub thread_results {
my ($ctx, $msgs) = @_;
require PublicInbox::SearchThread;
- PublicInbox::SearchThread::thread($msgs, *sort_ds, $ctx->{-inbox});
+ my $rootset = PublicInbox::SearchThread::thread($msgs, \&sort_ds, $ctx);
+
+ # FIXME: `tid' is broken on --reindex, so that needs to be fixed
+ # and preserved in the future. This bug is hidden by `sid' matches
+ # in get_thread, so we never noticed it until now. And even when
+ # reindexing is fixed, we'll keep this code until a SCHEMA_VERSION
+ # bump since reindexing is expensive and users may not do it
+
+ # loose threading could've returned too many results,
+ # put the root the message we care about at the top:
+ my $mid = $ctx->{mid};
+ if (defined($mid) && scalar(@$rootset) > 1) {
+ $ctx->{root_idx} = -1;
+ my $nr = scalar @$msgs;
+ walk_thread($rootset, $ctx, \&find_mid_root);
+ my $idx = $ctx->{found_mid_at};
+ if (defined($idx) && $idx != 0) {
+ my $tip = splice(@$rootset, $idx, 1);
+ @$rootset = reverse @$rootset;
+ unshift @$rootset, $tip;
+ $ctx->{sl_note} = strict_loose_note($nr);
+ }
+ }
+ $rootset
}
sub missing_thread {
PublicInbox::ExtMsg::ext_msg($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) = @_;
$omit;
}
-sub skel_dump {
+sub skel_dump { # walk_thread callback
my ($ctx, $level, $node) = @_;
my $smsg = $node->{smsg} or return _skel_ghost($ctx, $level, $node);
- my $dst = $ctx->{dst};
+ my $skel = $ctx->{skel};
my $cur = $ctx->{cur};
my $mid = $smsg->{mid};
+ if ($level == 0 && $ctx->{skel_dump_roots}++) {
+ $$skel .= delete($ctx->{sl_note}) || '';
+ }
+
my $f = ascii_html($smsg->from_name);
my $obfs_ibx = $ctx->{-obfs_ibx};
obfuscate_addrs($obfs_ibx, $f) if $obfs_ibx;
- my $d = fmt_ts($smsg->{ds}) . ' ' . indent_for($level) . th_pfx($level);
+ my $d = fmt_ts($smsg->{ds});
+ my $unmatched; # if lazy-loaded by SearchThread::Msg::visible()
+ if (exists $ctx->{searchview}) {
+ if (defined(my $pct = $smsg->{pct})) {
+ $d .= (sprintf(' % 2u', $pct) . '%');
+ } else {
+ $unmatched = 1;
+ $d .= ' ';
+ }
+ }
+ $d .= ' ' . indent_for($level) . th_pfx($level);
my $attr = $f;
$ctx->{first_level} ||= $level;
if ($cur) {
if ($cur eq $mid) {
delete $ctx->{cur};
- $$
dst .= "<b>$d<a\nid=r\nhref=\"#t\">".
+ $$
skel .= "<b>$d<a\nid=r\nhref=\"#t\">".
"$attr [this message]</a></b>\n";
- return;
+ return 1;
} else {
$ctx->{prev_msg} = $mid;
}
# Subject is never undef, this mail was loaded from
# our Xapian which would've resulted in '' if it were
# really missing (and Filter rejects empty subjects)
- my @subj = split(/ /, $ctx->{srch}->subject_normalized($smsg->subject));
-
+ my @subj = split(/ /, subject_normalized($smsg->subject));
# remove common suffixes from the subject if it matches the previous,
# so we do not show redundant text at the end.
my $prev_subj = $ctx->{prev_subj} || [];
}
my $m;
my $id = '';
- my $mapping = $ctx->{mapping};
+ my $mapping = $unmatched ? undef : $ctx->{mapping};
if ($mapping) {
my $map = $mapping->{$mid};
$id = id_compress($mid, 1);
} else {
$m = $ctx->{-upfx}.mid_escape($mid).'/';
}
- $$dst .= $d . "<a\nhref=\"$m\"$id>" . $end;
+ $$skel .= $d . "<a\nhref=\"$m\"$id>" . $end;
+ 1;
}
sub _skel_ghost {
my ($ctx, $level, $node) = @_;
my $mid = $node->{id};
- my $d = $ctx->{pct} ? ' [irrelevant] ' # search result
- : ' [not found] ';
+ my $d = ' [not found] ';
+ $d .= ' ' if exists $ctx->{searchview};
$d .= indent_for($level) . th_pfx($level);
my $upfx = $ctx->{-upfx};
my $m = PublicInbox::Hval->new_msgid($mid);
} else {
$d .= qq{<<a\nhref="$href">$html</a>>\n};
}
- my $dst = $ctx->{dst};
- $$dst .= $d;
+ ${$ctx->{skel}} .= $d;
+ 1;
}
sub sort_ds {
# accumulate recent topics if search is supported
# returns 200 if done, 404 if not
-sub acc_topic {
+sub acc_topic { # walk_thread callback
my ($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);
+ $subj = subject_normalized($subj);
+ $subj = '(no subject)' if $subj eq '';
$ds = $x->ds;
if ($level == 0) {
$topic = [ $ds, 1, { $subj => $mid }, $subj ];
$ctx->{-cur_topic} = $topic;
push @{$ctx->{order}}, $topic;
- return;
+ return 1;
}
$topic = $ctx->{-cur_topic}; # should never be undef
}
$seen->{$subj} = $mid; # latest for subject
} else { # ghost message
- return if $level != 0; # ignore child ghosts
+ return 1 if $level != 0; # ignore child ghosts
$topic = [ -666, 0, {} ];
$ctx->{-cur_topic} = $topic;
push @{$ctx->{order}}, $topic;
}
+ 1;
}
sub dump_topics {
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 $mbox = qq(<a\nhref="$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" .
+ my $s = "<a\nhref=\"$href/T/$anchor\">$top</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 @subj = split(/ /, subject_normalized($subj));
my @next_prev = @subj; # full copy
my $omit = dedupe_subject($prev_subj, \@subj, ' "');
$prev_subj = \@next_prev;
200;
}
-sub ts2str ($) {
- my ($ts) = @_;
- POSIX::strftime('%Y%m%d%H%M%S', gmtime($ts));
-}
+# only for the t= query parameter passed to overview DB
+sub ts2str ($) { strftime('%Y%m%d%H%M%S', gmtime($_[0])) };
sub str2ts ($) {
my ($yyyy, $mon, $dd, $hh, $mm, $ss) = unpack('A4A2A2A2A2A2', $_[0]);
# Xapian uses '..' but '-' is perhaps friendier to URL linkifiers
# if only $after exists "YYYYMMDD.." because "." could be skipped
# if interpreted as an end-of-sentence
- $t =~ s/\A(\d{8,14})-// and $after = str2ts($1);
- $t =~ /\A(\d{8,14})\z/ and $before = str2ts($1);
+ $t =~ s/\A([0-9]{8,14})-// and $after = str2ts($1);
+ $t =~ /\A([0-9]{8,14})\z/ and $before = str2ts($1);
my $ibx = $ctx->{-inbox};
my $msgs = $ibx->recent($opts, $after, $before);
my ($ctx) = @_;
my $msgs = paginate_recent($ctx, 200); # 200 is our window
if (@$msgs) {
- walk_thread(thread_results($ctx, $msgs), $ctx, *acc_topic);
+ walk_thread(thread_results($ctx, $msgs), $ctx, \&acc_topic);
}
- PublicInbox::WwwStream->response($ctx, dump_topics($ctx), *index_nav);
+ PublicInbox::WwwStream->response($ctx, dump_topics($ctx), \&index_nav);
}
sub thread_adj_level {