X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;ds=sidebyside;f=lib%2FPublicInbox%2FView.pm;h=81e83d97738256511a530bc4093bfd2793267eb8;hb=956ede734d7c2e8d0a3003c6e2d554114586643e;hp=748e39106a58d5a73450df89b7b26159044aad0b;hpb=712b8bb3a61cc261a4b8c1bfdb6e39f40cee3188;p=public-inbox.git diff --git a/lib/PublicInbox/View.pm b/lib/PublicInbox/View.pm index 748e3910..81e83d97 100644 --- a/lib/PublicInbox/View.pm +++ b/lib/PublicInbox/View.pm @@ -6,14 +6,14 @@ package PublicInbox::View; use strict; use warnings; -use URI::Escape qw/uri_escape_utf8/; use Date::Parse qw/str2time/; -use PublicInbox::Hval qw/ascii_html/; +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; use constant INDENT => ' '; @@ -24,11 +24,14 @@ sub th_pfx ($) { $_[0] == 0 ? '' : TCHILD }; sub msg_html { my ($ctx, $mime) = @_; my $hdr = $mime->header_obj; - my $tip = _msg_html_prepare($hdr, $ctx); + 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, '') . '
'; @@ -347,23 +367,26 @@ sub thread_html { $ctx->{prev_attr} = ''; $ctx->{prev_level} = 0; $ctx->{root_anchor} = anchor_for($mid); - $ctx->{seen} = {}; $ctx->{mapping} = {}; $ctx->{s_nr} = "$nr+ messages in thread"; - my $th = thread_results($msgs); - walk_thread($th, $ctx, *pre_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); + $skel .= ''; - return stream_thread($th, $ctx) unless $ctx->{flat}; + return stream_thread($rootset, $ctx) unless $ctx->{flat}; # flat display: lazy load the full message from smsg - my $inbox = $ctx->{-inbox}; my $mime; while ($mime = shift @$msgs) { $mime = $inbox->msg_by_smsg($mime) and last; } return missing_thread($ctx) unless $mime; - $mime = Email::MIME->new($mime); + $mime = PublicInbox::MIME->new($mime); $ctx->{-title_html} = ascii_html($mime->header('Subject')); $ctx->{-html_tip} = '
'.index_entry($mime, $ctx, scalar @$msgs); $mime = undef; @@ -373,7 +396,7 @@ sub thread_html { $mime = $inbox->msg_by_smsg($mime) and last; } if ($mime) { - $mime = Email::MIME->new($mime); + $mime = PublicInbox::MIME->new($mime); return index_entry($mime, $ctx, scalar @$msgs); } $msgs = undef; @@ -382,14 +405,11 @@ sub thread_html { } sub multipart_text_as_html { - my ($mime, $upfx) = @_; + my ($mime, $upfx, $obfs_ibx) = @_; my $rv = ""; # scan through all parts, looking for displayable text - msg_iter($mime, sub { - my ($p) = @_; - $rv .= add_text_body($upfx, $p); - }); + msg_iter($mime, sub { $rv .= add_text_body($upfx, $obfs_ibx, $_[0]) }); $rv; } @@ -436,13 +456,15 @@ sub attach_link ($$$$;$) { } $ret .= "[-- Attachment #$idx: "; my $ts = "Type: $ct, Size: $size bytes"; + $desc = ascii_html($desc); $ret .= ($desc eq '') ? "$ts --]" : "$desc --]\n[-- $ts --]"; $ret .= "\n"; } sub add_text_body { - my ($upfx, $p) = @_; # from msg_iter: [ Email::MIME, depth, @idx ] - my ($part, $depth, @idx) = @$p; + 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; @@ -477,33 +499,35 @@ sub add_text_body { } my @quot; my $l = PublicInbox::Linkify->new; - while (defined(my $cur = shift @lines)) { + foreach my $cur (@lines) { if ($cur !~ /^>/) { # show the previously buffered quote inline flush_quote(\$s, $l, \@quot) if @quot; # regular line, OK - $cur = $l->linkify_1($cur); - $cur = ascii_html($cur); - $s .= $l->linkify_2($cur); + $l->linkify_1($cur); + $s .= $l->linkify_2(ascii_html($cur)); } else { push @quot, $cur; } } - my $end = "\n"; - if (@quot) { - $end = ''; + 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 =~ s/[ \t]+$//sgm; # kill per-line trailing whitespace - $s =~ s/\A\n+//s; # kill leading blank lines - $s =~ s/\s+\z//s; # kill all trailing spaces - $s .= $end; } sub _msg_html_prepare { - my ($hdr, $ctx) = @_; + my ($hdr, $ctx, $obfs_ibx) = @_; my $srch = $ctx->{srch} if $ctx; my $atom = ''; my $rv = ""; # anchor for body start @@ -522,6 +546,7 @@ sub _msg_html_prepare { if ($h eq 'From') { 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) { @@ -531,7 +556,7 @@ sub _msg_html_prepare { } } $v = $v->as_html; - $v =~ s/(\@[^,]+,) /$1\n\t/g if ($h eq 'Cc' || $h eq 'To'); + obfuscate_addrs($obfs_ibx, $v) if $obfs_ibx; $rv .= "$h: $v\n"; } @@ -570,13 +595,20 @@ sub thread_skel { $$dst .= "$nr+ messages in thread ($expand"; $$dst .= qq! / [top])\n!; - my $subj = $srch->subject_path($hdr->header('Subject')); - $ctx->{seen} = { $subj => 1 }; + 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; - walk_thread(thread_results(load_results($sres)), $ctx, *skel_dump); + $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; } @@ -617,50 +649,6 @@ sub _parent_headers { $rv; } -sub squote_maybe ($) { - my ($val) = @_; - if ($val =~ m{([^\w@\./,\%\+\-])}) { - $val =~ s/(['!])/'\\$1'/g; # '!' for csh - return "'$val'"; - } - $val; -} - -sub mailto_arg_link { - my ($hdr) = @_; - my %cc; # everyone else - my $to; # this is the From address - - foreach my $h (qw(From To Cc)) { - my $v = $hdr->header($h); - defined($v) && ($v ne '') or next; - my @addrs = PublicInbox::Address::emails($v); - foreach my $address (@addrs) { - my $dst = lc($address); - $cc{$dst} ||= $address; - $to ||= $dst; - } - } - my @arg; - - my $subj = $hdr->header('Subject') || ''; - $subj = "Re: $subj" unless $subj =~ /\bRe:/i; - my $mid = $hdr->header_raw('Message-ID'); - push @arg, '--in-reply-to='.squote_maybe(mid_clean($mid)); - my $irt = mid_escape($mid); - delete $cc{$to}; - push @arg, "--to=$to"; - $to = uri_escape_utf8($to); - $subj = uri_escape_utf8($subj); - my @cc = sort values %cc; - push(@arg, map { "--cc=$_" } @cc); - my $cc = uri_escape_utf8(join(',', @cc)); - my $href = "mailto:$to?In-Reply-To=$irt&Cc=${cc}&Subject=$subj"; - $href =~ s/%20/+/g; - - (\@arg, ascii_html($href)); -} - sub html_footer { my ($hdr, $standalone, $ctx, $rhref) = @_; @@ -732,13 +720,13 @@ sub ghost_parent { sub indent_for { my ($level) = @_; - INDENT x ($level - 1); + $level ? INDENT x ($level - 1) : ''; } sub load_results { - my ($sres) = @_; - - [ map { $_->ensure_metadata; $_ } @{delete $sres->{msgs}} ]; + my ($srch, $sres) = @_; + my $msgs = delete $sres->{msgs}; + $srch->retry_reopen(sub { [ map { $_->mid; $_ } @$msgs ] }); } sub msg_timestamp { @@ -748,12 +736,9 @@ sub msg_timestamp { } sub thread_results { - my ($msgs) = @_; + my ($msgs, $srch) = @_; require PublicInbox::SearchThread; - my $th = PublicInbox::SearchThread->new($msgs); - $th->thread; - $th->order(*sort_ts); - $th + PublicInbox::SearchThread::thread($msgs, *sort_ts, $srch); } sub missing_thread { @@ -770,13 +755,37 @@ sub _msg_date { sub fmt_ts { POSIX::strftime('%Y-%m-%d %k:%M', gmtime($_[0])) } -sub _skel_header { - my ($ctx, $smsg, $level) = @_; +sub dedupe_subject { + my ($prev_subj, $subj, $val) = @_; + + my $omit = ''; # '"' denotes identical text omitted + my (@prev_pop, @curr_pop); + while (@$prev_subj && @$subj && $subj->[-1] eq $prev_subj->[-1]) { + push(@prev_pop, pop(@$prev_subj)); + push(@curr_pop, pop(@$subj)); + $omit ||= $val; + } + pop @$subj if @$subj && $subj->[-1] =~ /^re:\s*/i; + if (scalar(@curr_pop) == 1) { + $omit = ''; + push @$prev_subj, @prev_pop; + push @$subj, @curr_pop; + } + $omit; +} + +sub skel_dump { + my ($ctx, $level, $node) = @_; + my $smsg = $node->{smsg} or return _skel_ghost($ctx, $level, $node); my $dst = $ctx->{dst}; my $cur = $ctx->{cur}; my $mid = $smsg->{mid}; + my $f = ascii_html($smsg->from_name); + my $obfs_ibx = $ctx->{-obfs_ibx}; + obfuscate_addrs($obfs_ibx, $f) if $obfs_ibx; + my $d = fmt_ts($smsg->{ts}) . ' ' . indent_for($level) . th_pfx($level); my $attr = $f; $ctx->{first_level} ||= $level; @@ -802,24 +811,30 @@ sub _skel_header { # 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 = $smsg->subject; - my $h = $ctx->{srch}->subject_path($s); - if ($ctx->{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 = dedupe_subject($prev_subj, \@subj, '" '); + my $end; + if (@subj) { + my $subj = join(' ', @subj); + $subj = ascii_html($subj); + obfuscate_addrs($obfs_ibx, $subj) if $obfs_ibx; + $end = "$subj $omit$f\n" } else { - $ctx->{seen}->{$h} = 1; - $s = PublicInbox::Hval->new($s); - $s = $s->as_html; + $end = "$f\n"; } my $m; my $id = ''; my $mapping = $ctx->{mapping}; - my $end = defined($s) ? "$s $f\n" : "$f\n"; if ($mapping) { my $map = $mapping->{$mid}; $id = id_compress($mid, 1); $m = '#m'.$id; - $map->[1] = "$d$end"; + $map->[0] = "$d$end"; $id = "\nid=r".$id; } else { $m = $ctx->{-upfx}.mid_escape($mid).'/'; @@ -827,32 +842,29 @@ sub _skel_header { $$dst .= $d . "" . $end; } -sub skel_dump { +sub _skel_ghost { my ($ctx, $level, $node) = @_; - if (my $smsg = $node->{smsg}) { - _skel_header($ctx, $smsg, $level); + + 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] = $d . qq{<$html>\n}; + $d .= qq{<$html>\n}; } else { - my $mid = $node->{id}; - my $dst = $ctx->{dst}; - my $mapping = $ctx->{mapping}; - my $map = $mapping->{$mid} if $mapping; - 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 = $m->as_html; - - if ($map) { - my $id = id_compress($mid, 1); - $map->[1] = $d . qq{<$html>\n}; - $d .= qq{<$html>\n}; - } else { - $d .= qq{<$html>\n}; - } - $$dst .= $d; + $d .= qq{<$html>\n}; } + my $dst = $ctx->{dst}; + $$dst .= $d; } sub sort_ts { @@ -909,6 +921,9 @@ 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) { @@ -917,6 +932,7 @@ sub dump_topics { 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); @@ -937,12 +953,17 @@ sub dump_topics { " $ts UTC $n - $mbox / $atom\n"; for (my $i = 0; $i < scalar(@ex); $i += 2) { my $level = $ex[$i]; - my $sub = $ex[$i + 1]; - $mid = delete $seen->{$sub}; - $sub = PublicInbox::Hval->new($sub)->as_html; + my $subj = $ex[$i + 1]; + $mid = delete $seen->{$subj}; + my @subj = split(/ /, $srch->subject_normalized($subj)); + my @next_prev = @subj; # full copy + my $omit = dedupe_subject($prev_subj, \@subj, ' "'); + $prev_subj = \@next_prev; + $subj = ascii_html(join(' ', @subj)); + obfuscate_addrs($obfs_ibx, $subj) if $obfs_ibx; $href = mid_escape($mid); $s .= indent_for($level) . TCHILD; - $s .= "$sub\n"; + $s .= "$subj$omit\n"; } push @out, $s; } @@ -980,11 +1001,12 @@ sub index_topics { my $opts = { offset => $off, limit => 200 }; $ctx->{order} = []; - my $sres = $ctx->{srch}->query('', $opts); + my $srch = $ctx->{srch}; + my $sres = $srch->query('', $opts); my $nr = scalar @{$sres->{msgs}}; if ($nr) { - $sres = load_results($sres); - walk_thread(thread_results($sres), $ctx, *acc_topic); + $sres = load_results($srch, $sres); + walk_thread(thread_results($sres, $srch), $ctx, *acc_topic); } $ctx->{-next_o} = $off+ $nr; $ctx->{-cur_o} = $off;