ting_subsearch_suggestions.module

<?php

/**
 * @file
 * Ting subsearch using suggestion services: Opensuggestion and KPI-Webtrekk.
 */

use GuzzleHttp\Client as HttpClient;
use GuzzleHttp\Exception\ClientException as GuzzleClientException;
use Ting\Search\NullSearchResult;

module_load_include('inc', 'ting_subsearch_suggestions', 'ting_subsearch_suggestions.admin');

define('TING_SUBSEARCH_SUGGESTIONS_WEEKLY_CRON', 604800);

/**
 * Implements hook_menu().
 */
function ting_subsearch_suggestions_menu() {
  $items['subsearch_suggestions'] = [
    'title' => 'Subsearch Suggestions',
    'description' => 'AJAXify request for subsearch suggestions',
    'page callback' => 'ting_subsearch_suggestions_ajax_callback',
    'access callback' => TRUE,
  ];

  return $items;
}

/**
 * Custom AJAX menu callback.
 */
function ting_subsearch_suggestions_ajax_callback() {
  $message = '';
  if (!empty($_POST)) {
    $keys = $_POST['keys'];
    $conditions = $_POST['conditions'];
    $results = unserialize($_POST['results']);

    $message = ting_subsearch_suggestions_get_message($keys, $conditions, $results);
  }

  drupal_json_output($message);
  drupal_exit();
}

/**
 * Generation of suggestion message.
 *
 * @param $keys
 * @param $conditions
 * @param $results
 *
 * @return string|void
 * @throws \TingClientException
 * @throws \Ting\Search\SearchProviderException
 */
function ting_subsearch_suggestions_get_message($keys, $conditions, $results) {
  $service = variable_get('ting_subsearch_suggestions_service', 'kpi');
  if ($results instanceof NullSearchResult) {
    return;
  }

  $num_total_objects = $results->getNumTotalObjects();

  if (!empty($service)) {
    $message = '';

    // When a zero-hit search is executed and we're able to match "correct"
    // keys, a new search is done with the "correct" keys.
    // Then use the query-parameter original-keys to display what was done to
    // the visitor.
    $parameters = drupal_get_query_parameters();

    if (isset($parameters['original-search'])) {
      $msg = t('See results for "!keys", the search for "!original-keys" returned 0 hits.', [
        '!original-keys' => $parameters['original-search'],
        '!keys' => $keys,
      ]);

      $message = '<div id="subsearch-suggestions-first">' . $msg . '</div>';
    }

    if ($num_total_objects == 0) {
      // Find suggestions for "correct" search phrase.
      $suggestion = ting_subsearch_common_suggested_keys($keys, 'ting_subsearch_suggestions_suggest_keys');

      if (strpos($suggestion, 'ting_search:') !== FALSE) {
        $clear_suggestion = explode(':', $suggestion);
        $suggestion = $clear_suggestion[1];
      }

      if (!empty($suggestion)) {
        // Do secondary search for "correct" keys.
        $secondary_result = ting_subsearch_common_do_secondary_search($suggestion);
        // Only redirect to "correct" keys search if it has a positive result.
        if (!empty($secondary_result) && $secondary_result->numTotalObjects > 0) {
          $additional_query = [
            'original-search' => $keys,
          ];
          $params['query'] = ting_subsearch_common_query_params(variable_get("ting_subsearch_{$service}_additional_url_params_0_hit", ""), $additional_query);
          global $base_url;
          $url = $base_url . '/search/ting/' . strtolower($suggestion) . '?' . drupal_http_build_query($params['query']);

          header('Location: ' . $url, TRUE, 301);
        }
      }
    }

    if ($num_total_objects <= variable_get("ting_subsearch_{$service}_trigger_limit", 5)) {
      $suggestion = ting_subsearch_common_suggested_keys($keys, 'ting_subsearch_suggestions_suggest_keys');

      if ($service == 'kpi') {
        if (strpos($suggestion, 'ting_search:') !== FALSE) {
          $clear_suggestion = explode(':', $suggestion);
          $suggestion = $clear_suggestion[1];
        }
      }
      if ($suggestion) {
        // Do additional search.
        $suggested_result = ting_subsearch_common_do_secondary_search($suggestion);
        $sug_total_number = $suggested_result->numTotalObjects;
        if (!empty($suggested_result) && $sug_total_number >= variable_get("ting_subsearch_{$service}_message_limit", 10)) {
          $ratio = $sug_total_number / $num_total_objects;

          if ($ratio > variable_get('ting_subsearch_suggestions_ps_factor')) {
            $options = [];
            if (module_exists('ting_field_search')) {
              $profile = ting_field_search_get_active_profile();
              if ($profile) {
                $options['query'] = ['profile' => $profile->name];
              }
            }

            if (!empty(variable_get("ting_subsearch_{$service}_additional_url_params_inlimit", "")) && !empty($options)) {
              $options['query'] += ting_subsearch_common_query_params(variable_get("ting_subsearch_{$service}_additional_url_params_inlimit", ""));
            }
            else {
              if (empty($options)) {
                $options['query'] = ting_subsearch_common_query_params(variable_get("ting_subsearch_{$service}_additional_url_params_inlimit", ""));
              }
            }

            // Set message!
            $msg = t('Search for <strong>"!suggested-key"</strong> (!suggested-num-results)', [
              '!suggested-key' => strtolower($suggestion),
              '!suggested-num-results' => $suggested_result->numTotalObjects,
            ]);

            $wrapper = '<div id="subsearch-suggestions-second">' . $msg . '</div>';

            $url = url('/search/ting/' . strtolower($suggestion), [
              'query' => $options['query'],
              'absolute' => TRUE,
            ]);

            $message = l($wrapper, $url, [
              'html' => TRUE,
              'absolute' => TRUE,
              'attributes' => ['target' => 'blank'],
            ]);
          }
        }
      }
    }

    return $message;
  }
}

/**
 * Implements hook_ting_search_results_prefix().
 */
function ting_subsearch_suggestions_ting_search_results_prefix($keys, $conditions, $results) {
  if ($results->getNumTotalObjects() != 0) {
    drupal_add_js(
      [
        'subsearch_suggestions' => [
          'keys' => $keys,
          'conditions' => $conditions,
          'results' => serialize($results),
        ],
      ],
      'setting'
    );

    drupal_add_js(drupal_get_path('module', 'ting_subsearch_suggestions') . '/js/subsearch_suggestions.js', [
      'type' => 'file',
      'scope' => 'footer',
    ]);
  }
  else {
    ting_subsearch_suggestions_get_message($keys, $conditions, $results);
  }
}

/**
 * Find the first suggestion for given keys.
 *
 * @param string $keys
 *   The original keys that you want to find suggestion for.
 *
 * @return string
 *   The suggestion.
 *
 * @throws Exception
 */
function ting_subsearch_suggestions_suggest_keys($keys) {
  $service = variable_get('ting_subsearch_suggestions_service', 'kpi');
  switch ($service) {
    case 'os':
      if (!variable_get('opensearch_search_autocomplete_suggestion_url', FALSE)) {
        return FALSE;
      }

      $settings = opensearch_search_autocomplete_settings();
      $url = variable_get('opensearch_search_autocomplete_suggestion_url', FALSE) . 'rest/facetSpell';

      $params = $options['query'] = ['query' => check_plain($keys)] + $settings;
      $options['maxTime'] = 300;

      try {
        $client = new HttpClient();
        $response = $client->get(
          $url,
          [
            'query' => $params,
            'headers' => [
              'Accept' => 'application/json',
            ],
          ]
        );
        $result = json_decode($response->getBody());
      } catch (GuzzleClientException $e) {
        throw new Exception($e->getMessage(), $e->getCode(), $e);
      }

      $items = [];
      // First child of $result changes name from suggestions to suggestion when
      // switching between version 3.0 and 3.5.
      if (!empty(current($result))) {
        foreach (current($result) as $suggestion) {
          return $suggestion->phrase;
        }
      }
      if (empty($items)) {
        return FALSE;
      }
      break;

    case 'kpi':
      $keys_lowered = strtolower($keys);
      $ting_kpi_minimum_searches = variable_get('ting_subsearch_kpi_minimum_searches', 100);
      $ting_kpi_distance = variable_get('ting_subsearch_kpi_distance', 0);
      $weeks_selected = variable_get('ting_subsearch_kpi_week_nrs', 1);

      $weeks_in_db = db_query('SELECT DISTINCT tyw.weekno, tyw.year FROM {ting_subsearch_kpi_index_popular_keys} tyw ORDER BY tyw.year DESC, tyw.weekno DESC')->fetchAll();

      $weeks_for_processing = array_slice($weeks_in_db, 0, $weeks_selected + 1);

      $grouped_array = [];
      foreach ($weeks_for_processing as $item) {
        $grouped_array[$item->year][] = $item->weekno;
      }

      $items = [];

      foreach ($grouped_array as $year => $weeks) {
        $items[] = db_query('SELECT k.popular_keys FROM {ting_subsearch_kpi_index_popular_keys} k WHERE k.numtotalsearches > :numtotalsearches AND k.year = :year AND k.weekno IN (:weekno) ORDER BY k.weekno DESC', [
          ':numtotalsearches' => $ting_kpi_minimum_searches,
          ':year' => $year,
          ':weekno' => $weeks,
        ])->fetchCol();
      }

      $best_results = [];
      foreach ($items as $item) {
        if (!empty($item)) {
          foreach ($item as $i) {
            $best_results[] = $i;
          }
        }
      }

      foreach ($best_results as $best_result) {
        if (strpos($best_result, '_')) {
          $popular_keys = strtolower(str_replace('_', '.', $best_result));
        }
        else {
          $popular_keys = $best_result;
        }
        $distance = levenshtein($keys_lowered, $popular_keys);

        $raw_array[$popular_keys] = $distance;
      }

      $prepared_array = array_filter($raw_array, function ($distance, $key) use ($ting_kpi_distance) {
        if ($distance > 0 && $distance <= $ting_kpi_distance) {
          return $prepared_array[$key] = $distance;
        }
      }, ARRAY_FILTER_USE_BOTH);

      if (!empty($prepared_array)) {
        $result = array_keys($prepared_array, min($prepared_array));
      }

      return isset($result[0]) ? $result[0] : '';

      break;
  }
}

/**
 * Implements hook_cron().
 */
function ting_subsearch_suggestions_cron() {
  $service = variable_get('ting_subsearch_suggestions_service', 'kpi');
  $prepared = FALSE;

  if (!empty($service) && $service == 'kpi') {
    // Ensure that local data is only renewed once a day.
    if (!(variable_get('subsearch_kpi_cron_run', 0) < REQUEST_TIME)) {
      return;
    }

    $url = variable_get('ting_subsearch_kpi_webservice_endpoint', NULL);

    if (empty($url)) {
      watchdog('ting_subsearch_opensuggestions', 'KPI: Module not configured properly', [], WATCHDOG_WARNING);
      return;
    }

    $temp_files_folder = 'public://subsearch_kpi/';

    if (!is_dir($temp_files_folder)) {
      file_prepare_directory($temp_files_folder, FILE_CREATE_DIRECTORY);
    }

    $path_to_file = drupal_realpath("public://subsearch_kpi/temp_file.csv");
    copyfile_chunked($url, $path_to_file);
    print_r("File downloaded.\n");

    $file = fopen(drupal_realpath("public://subsearch_kpi/temp_file.csv"), "r");

    print_r("Insert feed data into database.\n");
    while (($data = fgetcsv($file))) {
      if ($data[0] != 'year') {
        $normalized_string = mb_convert_encoding($data[2], 'UTF-8', 'ISO-8859-15');
        /** @var \InsertQuery $query */
        $query = db_insert('ting_subsearch_kpi_index_popular_keys')->fields(
          [
            'popular_keys' => $normalized_string,
            'numtotalsearches' => (int) $data[3],
            'weekno' => $data[1],
            'year' => $data[0],
          ]
        );
        $query->execute();
      }
    }
    $prepared = TRUE;
    fclose($file);

    if (!empty($prepared)) {
      print_r("Start database cleaning.\n");
      // Get general "meta" present in db.
      $min_year = db_query("SELECT MIN(year) FROM {ting_subsearch_kpi_index_popular_keys}")->fetchField();
      $min_week = db_query("SELECT MIN(weekno) FROM {ting_subsearch_kpi_index_popular_keys} WHERE year = {$min_year}")->fetchField();
      $max_year = db_query("SELECT MAX(year) FROM {ting_subsearch_kpi_index_popular_keys}")->fetchField();
      $max_week = db_query("SELECT MAX(weekno) FROM {ting_subsearch_kpi_index_popular_keys} WHERE year = {$max_year}")->fetchField();

      foreach (range($min_week, $max_week - 1) as $week) {
        db_delete('ting_subsearch_kpi_index_popular_keys')
          ->condition('weekno', $week)
          ->condition('year', $min_year)
          ->execute();
      }

      variable_set('subsearch_kpi_cron_run', REQUEST_TIME + TING_SUBSEARCH_SUGGESTIONS_WEEKLY_CRON);
      watchdog('ting_subsearch_suggestions', 'Data fetched from KPI. Will re-run in a week.', [], WATCHDOG_INFO);
    }
    else {
      watchdog('ting_subsearch_suggestions', 'No data fetched from feed', [], WATCHDOG_WARNING);
    }
  }
}

/**
 * Copy remote file over HTTP one small chunk at a time.
 *
 * @param string $infile
 *   The full URL to the remote file
 * @param string $outfile
 *   The path where to save the file
 *
 * @return bool|false|int $cnt
 */
function copyfile_chunked($infile, $outfile) {
  $chunksize = 10 * (1024 * 1024); // 10 Megs

  /**
   * parse_url breaks a part a URL into it's parts, i.e. host, path,
   * query string, etc.
   */
  $parts = parse_url($infile);
  $i_handle = fsockopen($parts['host'], 80, $errstr, $errcode, 5);
  $o_handle = fopen($outfile, 'wb');

  if ($i_handle == FALSE || $o_handle == FALSE) {
    return FALSE;
  }

  if (!empty($parts['query'])) {
    $parts['path'] .= '?' . $parts['query'];
  }

  /**
   * Send the request to the server for the file
   */
  $request = "GET {$parts['path']} HTTP/1.1\r\n";
  $request .= "Host: {$parts['host']}\r\n";
  $request .= "User-Agent: Mozilla/5.0\r\n";
  $request .= "Keep-Alive: 115\r\n";
  $request .= "Connection: keep-alive\r\n\r\n";
  fwrite($i_handle, $request);

  /**
   * Now read the headers from the remote server. We'll need
   * to get the content length.
   */
  $headers = [];
  while (!feof($i_handle)) {
    $line = fgets($i_handle);
    if ($line == "\r\n") {
      break;
    }
    $headers[] = $line;
  }

  /**
   * Look for the Content-Length header, and get the size
   * of the remote file.
   */
  $length = 0;
  foreach ($headers as $header) {
    if (stripos($header, 'Content-Length:') === 0) {
      $length = (int) str_replace('Content-Length: ', '', $header);
      break;
    }
  }

  /**
   * Start reading in the remote file, and writing it to the
   * local file one chunk at a time.
   */
  $cnt = 0;
  while (!feof($i_handle)) {
    $buf = '';
    $buf = fread($i_handle, $chunksize);
    $bytes = fwrite($o_handle, $buf);
    if ($bytes == FALSE) {
      return FALSE;
    }
    $cnt += $bytes;

    /**
     * We're done reading when we've reached the content length
     */
    if ($cnt >= $length) {
      break;
    }
  }

  fclose($i_handle);
  fclose($o_handle);
  return $cnt;
}