hook_update() to create photo nodes out of untracked files

example.install

/**
 * Move files from files dir & add image node.
 */
function hook_update_N(&$sandbox) {
  if (!isset($sandbox['progress'])) {
    $sandbox['progress'] = 0;
    $sandbox['current_nid'] = 0;
    $sandbox['max'] = db_query("SELECT COUNT(entity_id) FROM {field_data_field_body} WHERE field_body_value LIKE '%src=\"https://www.example.com/sites/default/files/%'")->fetchField();
    $sandbox['messages'] = array();
    $sandbox['csv'] = array();
  }

  // Select all NIDs where we have an old reference.
  $nids = db_select('field_data_field_body', 'n')
    ->fields('n', array('entity_id'))
    ->condition('field_body_value', '%src="https://www.example.com/sites/default/files/%', 'LIKE')
    ->condition('entity_id', $sandbox['current_nid'], '>')
    ->range(0, 5)
    ->orderBy('entity_id', 'ASC')
    ->execute();

  foreach ($nids as $nid) {
    $images_unique = array();
    $node_id = $nid->entity_id;
    $node = node_load($node_id);
    $body = $node->field_body[LANGUAGE_NONE][0]['value'];

    // Find all hardcoded image tags.
    preg_match_all('~(<img[^>]+example\.com.*?>)+~', $body, $images);
    foreach ($images[0] as $image) {
      if (!in_array($image, $images_unique)) {
        $image_path = array();
        preg_match_all('~(\/sites\/default\/files\/.+?(?="|\'))~', $image, $image_path);
        if (!empty($image_path[0])) {
          $images_unique[] = array(
            'tag' => $image,
            'path' => $image_path[0][0],
          );
        }
      }
    }

    // Loop through relevant unique image tags, update them in the body field.
    foreach ($images_unique as $image) {
      $image_nid = '';
      $image_url = 'https://www.example.com' . $image['path'];
      $image_base = str_replace('/sites/default/files/', '', $image['path']);

      // If photo is in the file_managed table, add shortcode to correct node.
      $image_file = entity_load('file', FALSE, array('filename' => basename($image_url)));

      if (!empty($image_file)) {
        $image_file = reset($image_file);
        if ($image_file->filesize > 0) {
          // Check if file is attached to a photo node.
          $images = db_select('field_data_field_image', 'n')
            ->fields('n', array('entity_id'))
            ->condition('field_image_fid', $image_file->fid, '=')
            ->range(0, 1)
            ->execute();
          foreach ($images as $nid) {
            // If file is attached to a photo node, get the photo nid.
            $image_nid = $nid->entity_id;
            $message = "Fid & photo node found for hardcoded image reference. Shortcode for photo " . $image_nid . " added to article " . $node_id;
          }

          // If not attached to a photo node, add new photo node.
          if (empty($image_nid)) {
            $values = array(
              'type' => 'photo',
              'status' => 1,
              'comment' => 0,
              'promote' => 0,
            );

            $entity = entity_create('node', $values);
            $entity_w = entity_metadata_wrapper('node', $entity);
            $entity_w->title->set(basename($image_url));
            $entity_w->field_image = (array) $image_file;
            $entity_w->save();

            $image_nid = $entity_w->nid->value();
            $message = "Fid found. New photo node created. Shortcode for photo " . $image_nid . " added to article " . $node_id;
          }
        }
        // Filesize is 0
        else {
          // Log, handle as needed later.
          $message = "Fid found. Image size = 0 for image on article " . $node_id;
        }
      }
      // If photo is not in the file_managed table.
      else {
        // Get the image file.
        if ($image_file = system_retrieve_file($image_url, file_default_scheme() . '://' . $image_base, TRUE)) {
          if ($image_file->filesize > 0) {
            $values = array(
              'type' => 'photo',
              'status' => 1,
              'comment' => 0,
              'promote' => 0,
            );
            $entity = entity_create('node', $values);
            $entity_w = entity_metadata_wrapper('node', $entity);
            $entity_w->title->set(basename($image_url));
            $entity_w->field_image = (array) $image_file;
            $entity_w->save();

            $image_nid = $entity_w->nid->value();
            $message = "New photo node created. Shortcode for photo " . $image_nid . " added to article " . $node_id;
          }
          // Filesize is 0
          else {
            // Log, handle as needed later.
            $message = "Image size = 0 for image on article " . $node_id;
          }
        }
        // If image does not exist.
        else {
          // Log, handle as needed later.
          $message = "Hardcoded image reference not found in file structure for article " . $node_id;
        }
      }

      if (!empty($image_nid)) {
        // Replace image tag with new image shortcode.
        $body = str_replace($image['tag'], '[photo align=center | size=original]' . $image_nid . '[/photo]', $body);
        $sandbox['csv'][] = array(
          'article_nid' => $node_id,
          'image_nid' => $image_nid,
          'image_path' => $image_base,
          'message' => $message,
        );
      }
      else {
        $message = "Something went wrong with image for article " . $node_id . ". " . $message;
        $sandbox['csv'][] = array(
          'article_nid' => $node_id,
          'image_nid' => 0,
          'image_path' => $image_base,
          'message' => $message,
        );
      }
      $sandbox['messages'][] = $message;
    }

    // If body changed, save node.
    if ($node->field_body[LANGUAGE_NONE][0]['value'] !== $body) {
      $node->field_body[LANGUAGE_NONE][0]['value'] = $body;
      node_save($node);

      $message = "Article " . $node_id . " updated with new shortcode image(s).";
      $sandbox['csv'][] = array(
        'article_nid' => $node_id,
        'image_nid' => 0,
        'image_path' => 'n/a',
        'message' => $message,
      );
      $sandbox['messages'][] = $message;
    };

    $sandbox['progress']++;
    $sandbox['current_nid'] = $node_id;
  }

  $sandbox['#finished'] = ($sandbox['progress'] >= $sandbox['max']) ? TRUE : ($sandbox['progress'] / $sandbox['max']);

  if ($sandbox['progress'] >= $sandbox['max']) {
    // Output urls to csv file.
    $csv_headers = array(
      'article_nid' => 'article_nid',
      'image_nid' => 'image_nid',
      'image_path' => 'image_path',
      'message' => 'message',
    );
    $csv_title = 'hard-coded-image-references';
    _create_csv($sandbox['csv'], $csv_headers, $csv_title);
    // Display final message;
    $final_message = implode("\n", $sandbox['messages']);
    return t('@max # of entities with example.com image references: @message', array(
        '@max' => $sandbox['max'],
        '@message' => $final_message,
      )
    );
  }
}