URLVerify.pm; docs; url shuffling

Signed-off-by: Bill Erickson <berick@esilibrary.com>

diff --git a/Open-ILS/src/perlmods/lib/OpenILS/Application/URLVerify.pm b/Open-ILS/src/perlmods/lib/OpenILS/Application/URLVerify.pm
index 64ec64b..5a99a8f 100644 (file)
--- a/Open-ILS/src/perlmods/lib/OpenILS/Application/URLVerify.pm
+++ b/Open-ILS/src/perlmods/lib/OpenILS/Application/URLVerify.pm
@@ -55,8 +55,6 @@ sub validate_session {
     my ($self, $client, $auth, $session_id, $url_ids, $options) = @_;
     $options ||= {};
 
-    # loop through list of URLs / session URLs
-    # add a sleep (org setting) to the multisession handler
     # TODO: Avoid testing URLs having the same domain sequentially
 
     my $e = new_editor(authtoken => $auth, xact => 1);
@@ -70,6 +68,8 @@ sub validate_session {
 
     if (!$url_ids) {
 
+        # No URLs provided, load all URLs for the requested session
+
         my $query = {
             select => {uvu => ['id']},
             from => {
@@ -86,13 +86,14 @@ sub validate_session {
         };
 
         if ($attempt_id) {
-            $logger->info("url: resuming attempt $attempt_id");
 
             # when resuming an existing attempt (that presumably failed
             # mid-processing), we only want to process URLs that either
             # have no linked url_verification or have an un-completed
             # url_verification.
 
+            $logger->info("url: resuming attempt $attempt_id");
+
             $query->{from}->{uvu}->{uvuv} = {
                 type => 'left',
                 filter => {attempt => $attempt_id}
@@ -109,7 +110,7 @@ sub validate_session {
 
         } else {
 
-            # this is a new run, so we only want to process URLs that
+            # this is a new attempt, so we only want to process URLs that
             # originated from the source records and not from redirects.
 
             $query->{where} = {
@@ -119,6 +120,17 @@ sub validate_session {
 
         my $ids = $e->json_query($query);
         $url_ids = [ map {$_->{id}} @$ids ];
+
+        # this is kinda hinky and probably an abuse of order_by, but 
+        # shuffling is good for spreading out domains and this avoids 
+        # the necessity of loading all the URLs (could be lots) and 
+        # shuffling them here.
+
+        $url_ids = $e->search_url_verify_url(
+            [   {id => $url_ids},
+                {order_by => {uvu => 'RANDOM()'}}
+            ],  {idlist => 1}
+        );
     }
 
     my $url_count = scalar(@$url_ids);