Work on #14.

config/services.yaml

@@ -8,9 +8,15 @@ parameters:
     app.service.detailmanager.delimiter: ';'
 
     # To not register any fetchresourcelist plugins, use app.plugins.fetchresourcelist []
-    app.plugins.fetchresourcelist: ['app:riprap:plugin:fetchresourcelist:from:file']
+    # app.plugins.fetchresourcelist: ['app:riprap:plugin:fetchresourcelist:from:file']
     app.plugins.fetchresourcelist.from.file.paths: ['resources/riprap_resource_ids.txt']
-    app.plugins.fetchresourcelist.from.drupal.baseurl: ['http://localhost:8000']
+
+    # Settings for the app:riprap:plugin:fetchresourcelist:from:drupal plugin.
+    app.plugins.fetchresourcelist: ['app:riprap:plugin:fetchresourcelist:from:drupal']
+    app.plugins.fetchresourcelist.from.drupal.baseurl: 'http://localhost:8000'
+    app.plugins.fetchresourcelist.from.drupal.authorization_headers: ['Authorization: Basic YWRtaW46aXNsYW5kb3Jh'] # admin:islandora
+    app.plugins.fetchresourcelist.from.drupal.content_types: ['islandora_object']
+    app.plugins.fetchresourcelist.from.drupal.media_tags: ['/taxonomy/term/2']
 
     # To not register any fetchdigest plugins, use app.plugins.fetchdigest [].
     # Currently we only allow one fetchdigest plugin per fixity event (i.e., this

src/Command/PluginFetchResourceListFromDrupal.php

@@ -9,7 +9,8 @@
 use Symfony\Component\DependencyInjection\ParameterBag\ParameterBagInterface;
 
 use Psr\Log\LoggerInterface;
-
+use GuzzleHttp\Client;
+use GuzzleHttp\Exception\RequestException;
 use App\Entity\Event;
 use App\Service\FixityEventDetailManager;
 
@@ -23,7 +24,12 @@ public function __construct(
         FixityEventDetailManager $event_detail = null
     ) {
         $this->params = $params;
-        $this->input_files = $this->params->get('app.plugins.fetchresourcelist.from.drupal.baseurl');
+        $this->drupal_base_url = $this->params->get('app.plugins.fetchresourcelist.from.drupal.baseurl');
+        // An array, we need to loop through and add to guzzle request.
+        $this->jsonapi_authorization_headers = $this->params->get('app.plugins.fetchresourcelist.from.drupal.authorization_headers');
+        // For now we only use the first one, not sure how to handle multiple content types.
+        $this->drupal_content_types = $this->params->get('app.plugins.fetchresourcelist.from.drupal.content_types');
+        $this->media_tags = $this->params->get('app.plugins.fetchresourcelist.from.drupal.media_tags');
 
         $this->logger = $logger;
         $this->event_detail = $event_detail;
@@ -42,20 +48,59 @@ protected function execute(InputInterface $input, OutputInterface $output)
     {
         // @todo: See https://github.com/mjordan/riprap/issues/14.
 
-/*
-        foreach ($this->input_files as $input_file) {
-            $resource_ids = file($input_file, FILE_IGNORE_NEW_LINES);
-            foreach ($resource_ids as $resource_id) {
-                // This is a string containing one resource ID (URL) per line;
-                $output->writeln($resource_id);
+        $client = new \GuzzleHttp\Client();
+        $url = $this->drupal_base_url . '/jsonapi/node/' . $this->drupal_content_types[0];
+        // First, get JSON:API's first page of nodes, then loop through each one and print it to output.
+        // curl -v -H 'Authorization: Basic YWRtaW46aXNsYW5kb3Jh'  "http://localhost:8000/jsonapi/node/islandora_object?page[offset]=2&page[limit]=1"
+        $response = $client->request('GET', $url, [
+            'http_errors' => false,
+            'headers' => [$this->jsonapi_authorization_headers[0]], // @todo: Loop through this array and add each header. 
+            'query' => ['page[offset]' => '1', 'page[limit]' => '3']
+        ]);
+        $status_code = $response->getStatusCode();
+        // var_dump($status_code);
+        $node_list = (string) $response->getBody();
+        $node_list_array = json_decode($node_list, true);
+        // var_dump($node_list_array);
+
+        foreach ($node_list_array['data'] as $node) {
+            $nid = $node['attributes']['nid'];
+            // var_dump($nid);            
+            // Get the media associated with this node using the Islandora-supplied Manage Media View.
+            $media_url = $this->drupal_base_url . '/node/' . $nid . '/media';
+            $media_response = $client->request('GET', $url, [
+                'http_errors' => false,
+                // @todo: Split this header out into its own config parameter.
+                'headers' => [$this->jsonapi_authorization_headers[0]], 
+                'query' => ['_format' => 'json']
+            ]);
+            $status_code = $media_response->getStatusCode();
+            // var_dump($status_code);
+            $media_list = (string) $response->getBody();
+            $media_list = json_decode($media_list, true);
+
+            // Loop through all the media and pick the ones that
+            // are tagged with terms in $taxonomy_terms_to_check.
+            foreach ($media_list as $media) {
+              var_dump($media);
+              if (count($media->field_tags)) {
+                foreach ($media->field_tags as $term) {
+                  if (in_array($term->url, $taxonomy_terms_to_check)) {
+                    // @todo: Convert to the equivalent Fedora URL and add to the plugin's output.
+                    // @todo: Add option to not convert to Fedora URL if the site doesn't use Fedora.
+                    // In that case, we need to figure out how to get Drupal's checksum for the file over HTTP.
+                    // var_dump($media->field_media_image[0]->url);
+                    $output->writeln($resource_id);
+                  }
+                }
+              }
             }
-        }
-*/
 
+        }
 
         // $this->logger is null while testing.
         if ($this->logger) {
-            $this->logger->info("PluginPersistToDatabase executed");
+            $this->logger->info("PluginFetchResourceListFromDrupal executed");
         }
     }
 }