treewide: "require" + "use" cleanup and docs

[public-inbox.git] / lib / PublicInbox / WwwAtomStream.pm

diff --git a/lib/PublicInbox/WwwAtomStream.pm b/lib/PublicInbox/WwwAtomStream.pm
index 5a10034ba8ecfb33eb90f5210b43af648998fd97..9430dd97679eed5049d65de63bc96129b8159b27 100644 (file)
--- a/lib/PublicInbox/WwwAtomStream.pm
+++ b/lib/PublicInbox/WwwAtomStream.pm
@@ -1,17 +1,20 @@
-# Copyright (C) 2016 all contributors <meta@public-inbox.org>
+# Copyright (C) 2016-2019 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;
 
 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 mid_escape/;
+use PublicInbox::MID qw(mid_escape);
+use PublicInbox::MsgTime qw(msg_timestamp);
 
 # called by PSGI server after getline:
 sub close {}
@@ -33,8 +36,8 @@ 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 $self->{cb} ? '</feed>' : undef;
 }
@@ -50,6 +53,15 @@ sub title_tag {
        "<title$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};
@@ -57,18 +69,22 @@ 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';
+               $page_id = to_uuid("t\n".$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"\n) .
@@ -77,30 +93,22 @@ sub atom_header {
        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));
 }
 
-sub mid2uuid ($) {
-       my ($mid) = @_;
-       utf8::encode($mid); # really screwed up In-Reply-To fields exist
-       $mid = sha1_hex($mid);
-       my $h = '[a-f0-9]';
-       my (@uuid5) = ($mid =~ m!\A($h{8})($h{4})($h{4})($h{4})($h{12})!o);
-       'urn:uuid:' . join('-', @uuid5);
-}
-
 # returns undef or string
 sub feed_entry {
-       my ($self, $mime) = @_;
+       my ($self, $smsg) = @_;
        my $ctx = $self->{ctx};
+       my $mime = $smsg->{mime};
        my $hdr = $mime->header_obj;
-       my $mid = mid_clean($hdr->header_raw('Message-ID'));
+       my $mid = $smsg->mid;
        my $irt = PublicInbox::View::in_reply_to($hdr);
-       my $uuid = mid2uuid($mid);
+       my $uuid = to_uuid($mid);
        my $base = $ctx->{feed_base_url};
        if (defined $irt) {
-               my $irt_uuid = mid2uuid($irt);
+               my $irt_uuid = to_uuid($irt);
                $irt = mid_escape($irt);
                $irt = qq(<thr:in-reply-to\nref="$irt_uuid"\n).
                        qq(href="$base$irt/"/>);
@@ -108,9 +116,8 @@ sub feed_entry {
                $irt = '';
        }
        my $href = $base . mid_escape($mid) . '/';
-       my $date = $hdr->header('Date');
-       my @t = eval { strptime($date) } if defined $date;
-       @t = gmtime(time) unless scalar @t;
+       my $t = msg_timestamp($hdr);
+       my @t = gmtime(defined $t ? $t : time);
        my $updated = feed_updated(@t);
 
        my $title = $hdr->header('Subject');
@@ -129,13 +136,13 @@ sub feed_entry {
        }
        $s .= "<entry><author><name>$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>$uuid</id>$irt</entry>";
+               PublicInbox::View::multipart_text_as_html($mime, $href, $ctx) .
+               '</pre></div></content></entry>';
 }
 
 sub feed_updated {