From 12ab006599ee71c1921e7a3c762224bb0e4470ba Mon Sep 17 00:00:00 2001
From: Florent Torregrosa <florent.torregrosa@gmail.com>
Date: Tue, 14 Apr 2015 16:36:49 +0200
Subject: [PATCH] Issue #2470553: Allows search_api_attachments to parse txt
 files.

---
 .../processor/FilesFieldsProcessorPlugin.php       |   67 +++++++++++++++++++-
 1 file changed, 66 insertions(+), 1 deletion(-)

diff --git a/src/Plugin/search_api/processor/FilesFieldsProcessorPlugin.php b/src/Plugin/search_api/processor/FilesFieldsProcessorPlugin.php
index 8fd68d1..b0d5a7f 100644
--- a/src/Plugin/search_api/processor/FilesFieldsProcessorPlugin.php
+++ b/src/Plugin/search_api/processor/FilesFieldsProcessorPlugin.php
@@ -6,6 +6,8 @@ use Drupal\Core\TypedData\DataDefinition;
 use Drupal\field\Entity\FieldConfig;
 use Drupal\search_api\Datasource\DatasourceInterface;
 use Drupal\search_api\Processor\ProcessorPluginBase;
+use Drupal\Core\StreamWrapper\StreamWrapperInterface;
+use Drupal\Component\Utility\Xss;
 
 /**
  * @SearchApiProcessor(
@@ -46,11 +48,37 @@ class FilesFieldsProcessorPlugin extends ProcessorPluginBase {
         if (!($field = $item->getField('search_api_attachments_' . $field_name))) {
           continue;
         }
-        $field->addValue('test test');
+
+        // Need to retrieve the files.
+        $entity = $item->getOriginalObject()->getValue();
+        $filefield_values = $entity->get($field_name)->getValue();
+
+        $fids = array();
+        foreach ($filefield_values as $filefield_value) {
+          $fids[] = $filefield_value['target_id'];
+        }
+
+        // Retrieve the files.
+        $files = entity_load_multiple('file', $fids);
+
+        // Parse the files.
+        $value = '';
+        foreach ($files as $file) {
+          $value .= $this->extract_simple($file);
+        }
+
+        // Add the value in the indexed data.
+        $field->addValue($value);
       }
     }
   }
 
+  /**
+   * Helper function.
+   *
+   * @return array
+   *  An array of file fields.
+   */
   protected function getFileFields() {
     $file_fields = array();
     // Retrieve file fields of indexed bundles.
@@ -66,4 +94,41 @@ class FilesFieldsProcessorPlugin extends ProcessorPluginBase {
     return $file_fields;
   }
 
+  /**
+   * Helper function.
+   *
+   * @param $file
+   *   A file object that implements FileInterface.
+   * @return string
+   *   The text extracted from txt file.
+   */
+  protected function extract_simple(FileInterface $file) {
+    $text = file_get_contents($this->get_realpath($file->getFileUri()));
+    $text = iconv("UTF-8", "UTF-8//IGNORE", $text);
+    $text = Xss::filter(str_replace(array('<', '>'), array(' <', '> '), $text), array());
+    $text = htmlspecialchars(html_entity_decode($text, ENT_NOQUOTES, 'UTF-8'), ENT_NOQUOTES, 'UTF-8');
+    $text = trim($text);
+    return $text;
+  }
+
+  /**
+   * Helper function.
+   *
+   * @param $uri
+   *   The URI of the file, e.g. public://directory/file.jpg.
+   * @return mixed
+   *   The real path to the file if it is a local file. An URL otherwise.
+   */
+  protected function get_realpath($uri) {
+    $wrapper = \Drupal::service('stream_wrapper_manager')->getViaUri($uri);
+    $scheme = file_uri_scheme($uri);
+    $local_wrappers = \Drupal::service('stream_wrapper_manager')->getWrappers(StreamWrapperInterface::LOCAL);
+    if (in_array($scheme, array_keys($local_wrappers))) {
+      return $wrapper->realpath();
+    }
+    else {
+      return $wrapper->getExternalUrl();
+    }
+  }
+
 }
-- 
1.7.10.4