search: force large mbox result downloads to POST

This should prevent crawlers (including most robots.txt ignoring
ones) from burning our CPU time without severely compromising
usability for humans.

1 files changed, 18 insertions, 4 deletions

diff --git a/lib/PublicInbox/WWW.pm b/lib/PublicInbox/WWW.pm
index f3c702e9..3fd77d42 100644
--- a/lib/PublicInbox/WWW.pm
+++ b/lib/PublicInbox/WWW.pm
@@ -53,10 +53,14 @@ sub call {
         my $path_info = $env->{PATH_INFO};
         my $method = $env->{REQUEST_METHOD};
 
-        if ($method eq 'POST' &&
-                 $path_info =~ m!$INBOX_RE/(git-upload-pack)\z!) {
-                my $path = $2;
-                return invalid_inbox($ctx, $1) || serve_git($ctx, $path);
+        if ($method eq 'POST') {
+                if ($path_info =~ m!$INBOX_RE/(git-upload-pack)\z!) {
+                        my $path = $2;
+                        return invalid_inbox($ctx, $1) ||
+                                serve_git($ctx, $path);
+                } elsif ($path_info =~ m!$INBOX_RE/!o) {
+                        return invalid_inbox($ctx, $1) || mbox_results($ctx);
+                }
         }
         elsif ($method !~ /\AGET|HEAD\z/) {
                 return r(405, 'Method Not Allowed');
@@ -400,6 +404,16 @@ sub serve_git {
         PublicInbox::GitHTTPBackend::serve($ctx->{env}, $ctx->{git}, $path);
 }
 
+sub mbox_results {
+        my ($ctx) = @_;
+        if ($ctx->{env}->{QUERY_STRING} =~ /(?:\A|[&;])q=/) {
+                searcher($ctx) or return need_search($ctx);
+                require PublicInbox::SearchView;
+                return PublicInbox::SearchView::mbox_results($ctx);
+        }
+        r404();
+}
+
 sub serve_mbox_range {
         my ($ctx, $inbox, $range) = @_;
         invalid_inbox($ctx, $inbox) || eval {