X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FWwwAtomStream.pm;h=c3fbb1a7cef589479f55860474390a7a3b3e5687;hb=c91490ccae1a1775da1e816866ef0b08b234ff54;hp=5720384cf648f4f50c93f661f2ab467bf2392d77;hpb=95d4bf7aded41cb3b0040c321d315532f68633e1;p=public-inbox.git

diff --git a/lib/PublicInbox/WwwAtomStream.pm b/lib/PublicInbox/WwwAtomStream.pm
index 5720384c..c3fbb1a7 100644
--- a/lib/PublicInbox/WwwAtomStream.pm
+++ b/lib/PublicInbox/WwwAtomStream.pm
@@ -1,27 +1,27 @@
-# Copyright (C) 2016 all contributors <meta@public-inbox.org>
+# Copyright (C) 2016-2020 all contributors <meta@public-inbox.org>
 # License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
 #
 # Atom body stream for which yields getline+close methods
+# public-inbox-httpd favors "getline" response bodies to take a
+# "pull"-based approach to feeding slow clients (as opposed to a
+# more common "push" model)
 package PublicInbox::WwwAtomStream;
 use strict;
 use warnings;
 
-# FIXME: locale-independence:
 use POSIX qw(strftime);
-use Date::Parse qw(strptime);
-
+use Digest::SHA qw(sha1_hex);
 use PublicInbox::Address;
-use PublicInbox::Hval qw(ascii_html);
-use PublicInbox::MID qw/mid_clean mid2path mid_escape/;
+use PublicInbox::Hval qw(ascii_html mid_href);
+use PublicInbox::MsgTime qw(msg_timestamp);
 
 # called by PSGI server after getline:
 sub close {}
 
 sub new {
 	my ($class, $ctx, $cb) = @_;
-	$ctx->{emit_header} = 1;
 	$ctx->{feed_base_url} = $ctx->{-inbox}->base_url($ctx->{env});
-	bless { cb => $cb || *close, ctx => $ctx }, $class;
+	bless { cb => $cb || \&close, ctx => $ctx, emit_header => 1 }, $class;
 }
 
 sub response {
@@ -34,8 +34,8 @@ sub response {
 sub getline {
 	my ($self) = @_;
 	if (my $middle = $self->{cb}) {
-		my $mime = $middle->();
-		return feed_entry($self, $mime) if $mime;
+		my $smsg = $middle->($self->{ctx});
+		return feed_entry($self, $smsg) if $smsg;
 	}
 	delete $self->{cb} ? '</feed>' : undef;
 }
@@ -51,6 +51,15 @@ sub title_tag {
 	"<title$type>$title</title>";
 }
 
+sub to_uuid ($) {
+	my ($any) = @_;
+	utf8::encode($any); # really screwed up In-Reply-To fields exist
+	$any = sha1_hex($any);
+	my $h = '[a-f0-9]';
+	my (@uuid5) = ($any =~ m!\A($h{8})($h{4})($h{4})($h{4})($h{12})!o);
+	'urn:uuid:' . join('-', @uuid5);
+}
+
 sub atom_header {
 	my ($ctx, $title) = @_;
 	my $ibx = $ctx->{-inbox};
@@ -58,48 +67,55 @@ sub atom_header {
 	my $search_q = $ctx->{search_query};
 	my $self_url = $base_url;
 	my $mid = $ctx->{mid};
+	my $page_id;
 	if (defined $mid) { # per-thread
-		$self_url .= mid_escape($mid).'/t.atom';
+		$self_url .= mid_href($mid).'/t.atom';
+		$page_id = to_uuid("t\n".$mid)
 	} elsif (defined $search_q) {
 		my $query = $search_q->{'q'};
 		$title = title_tag("$query - search results");
 		$base_url .= '?' . $search_q->qs_html(x => undef);
 		$self_url .= '?' . $search_q->qs_html;
+		$page_id = to_uuid("q\n".$query);
 	} else {
 		$title = title_tag($ibx->description);
 		$self_url .= 'new.atom';
+		$page_id = "mailto:$ibx->{-primary_address}";
 	}
-	my $mtime = (stat($ibx->{mainrepo}))[9] || time;
+	my $mtime = (stat($ibx->{inboxdir}))[9] || time;
 
 	qq(<?xml version="1.0" encoding="us-ascii"?>\n) .
-	qq{<feed\nxmlns="http://www.w3.org/2005/Atom">} .
+	qq(<feed\nxmlns="http://www.w3.org/2005/Atom"\n) .
+	qq(xmlns:thr="http://purl.org/syndication/thread/1.0">) .
 	qq{$title} .
 	qq(<link\nrel="alternate"\ntype="text/html") .
 		qq(\nhref="$base_url"/>) .
 	qq(<link\nrel="self"\nhref="$self_url"/>) .
-	qq(<id>mailto:$ibx->{-primary_address}</id>) .
+	qq(<id>$page_id</id>) .
 	feed_updated(gmtime($mtime));
 }
 
 # returns undef or string
 sub feed_entry {
-	my ($self, $mime) = @_;
+	my ($self, $smsg) = @_;
 	my $ctx = $self->{ctx};
+	my $mid = $smsg->mid; # may extract Message-ID from {mime}
+	my $mime = delete $smsg->{mime};
 	my $hdr = $mime->header_obj;
-	my $mid = mid_clean($hdr->header_raw('Message-ID'));
-
-	my $uuid = mid2path($mid);
-	$uuid =~ tr!/!!d;
-	my $h = '[a-f0-9]';
-	my (@uuid5) = ($uuid =~ m!\A($h{8})($h{4})($h{4})($h{4})($h{12})!o);
-	$uuid = 'urn:uuid:' . join('-', @uuid5);
-
-	$mid = PublicInbox::Hval->new_msgid($mid);
-	my $href = $ctx->{feed_base_url} . $mid->{href}. '/';
-
-	my $date = $hdr->header('Date');
-	my @t = eval { strptime($date) } if defined $date;
-	@t = gmtime(time) unless scalar @t;
+	my $irt = PublicInbox::View::in_reply_to($hdr);
+	my $uuid = to_uuid($mid);
+	my $base = $ctx->{feed_base_url};
+	if (defined $irt) {
+		my $irt_uuid = to_uuid($irt);
+		$irt = mid_href($irt);
+		$irt = qq(<thr:in-reply-to\nref="$irt_uuid"\n).
+			qq(href="$base$irt/"/>);
+	} else {
+		$irt = '';
+	}
+	my $href = $base . mid_href($mid) . '/';
+	my $t = msg_timestamp($hdr);
+	my @t = gmtime(defined $t ? $t : time);
 	my $updated = feed_updated(@t);
 
 	my $title = $hdr->header('Subject');
@@ -113,18 +129,21 @@ sub feed_entry {
 	$email = ascii_html($email);
 
 	my $s = '';
-	if (delete $ctx->{emit_header}) {
+	if (delete $self->{emit_header}) {
 		$s .= atom_header($ctx, $title);
 	}
 	$s .= "<entry><author><name>$name</name><email>$email</email>" .
 		"</author>$title$updated" .
+		qq(<link\nhref="$href"/>).
+		"<id>$uuid</id>$irt" .
 		qq{<content\ntype="xhtml">} .
 		qq{<div\nxmlns="http://www.w3.org/1999/xhtml">} .
-		qq(<pre\nstyle="white-space:pre-wrap">) .
-		PublicInbox::View::multipart_text_as_html($mime, $href) .
-		'</pre>' .
-		qq!</div></content><link\nhref="$href"/>!.
-		"<id>$uuid</id></entry>";
+		qq(<pre\nstyle="white-space:pre-wrap">);
+	$ctx->{obuf} = \$s;
+	$ctx->{mhref} = $href;
+	PublicInbox::View::multipart_text_as_html($mime, $ctx);
+	delete $ctx->{obuf};
+	$s .= '</pre></div></content></entry>';
 }
 
 sub feed_updated {