Related topics

File

tripal_feature/includes/tripal_feature.sync_features.inc, line 300

@todo Add file header description

Code

function tripal_feature_sync_features($max_sync = 0, $organism_id = NULL, 
$feature_types = NULL, $job_id = NULL) {

  $i = 0;

  // get the list of available sequence ontology terms for which
  // we will build drupal pages from features in chado.  If a feature
  // is not one of the specified typse we won't build a node for it.
  if (!$feature_types) {
    $allowed_types = variable_get('chado_sync_feature_types', 'gene contig');
  }
  else {
    $allowed_types = $feature_types;
  }
  $allowed_types = preg_replace("/[\s\n\r]+/", " ", $allowed_types);

  print "Looking for features of type: $allowed_types\n";

  $so_terms = split(' ', $allowed_types);
  $where_cvt = "";
  foreach ($so_terms as $term) {
    $where_cvt .= "CVT.name = '$term' OR ";
  }
  $where_cvt = drupal_substr($where_cvt, 0, drupal_strlen($where_cvt) -3); # strip trailing 'OR'

  // get the list of organisms that are synced and only include features from
  // those organisms
  $orgs = tripal_organism_get_synced();
  $where_org = "";
  foreach ($orgs as $org) {
    if ($organism_id) {
      if ($org->organism_id and $org->organism_id == $organism_id) {
        $where_org .= "F.organism_id = $org->organism_id OR ";
      }
    }
    else {
      if ($org->organism_id) {
        $where_org .= "F.organism_id = $org->organism_id OR ";
      }
    }
  }
  $where_org = drupal_substr($where_org, 0, drupal_strlen($where_org) -3); # strip trailing 'OR'

  // use this SQL statement to get the features that we're going to upload
  $sql = "SELECT feature_id " .
    "FROM {FEATURE} F " .
    "  INNER JOIN {Cvterm} CVT ON F.type_id = CVT.cvterm_id " .
    "  INNER JOIN {CV} on CV.cv_id = CVT.cv_id " .
    "WHERE ($where_cvt) AND ($where_org) AND CV.name = 'sequence' " .
    "ORDER BY feature_id";

  // get the list of features
  $results = chado_query($sql);

  // load into ids array
  $count = 0;
  $ids = array();
  while ($id = db_fetch_object($results)) {
    $ids[$count] = $id->feature_id;
    $count++;
  }

  // make sure our vocabularies are set before proceeding
  tripal_feature_set_vocabulary();

  // pre-create the SQL statement that will be used to check
  // if a feature has already been synced.  We skip features
  // that have been synced
  $sql = "SELECT * FROM {chado_feature} WHERE feature_id = %d";

  // Iterate through features that need to be synced
  $interval = intval($count * 0.01);
  if ($interval < 1) {
    $interval = 1;
  }
  $num_ids = sizeof($ids);
  $i = 0;
  foreach ($ids as $feature_id) {
    // update the job status every 1% features
    if ($job_id and $i % $interval == 0) {
      tripal_job_set_progress($job_id, intval(($i / $count) * 100));
    }
    // if we have a maximum number to sync then stop when we get there
    // if not then just continue on
    if ($max_sync and $i == $max_sync) {
      return '';
    }
    if (!db_fetch_object(db_query($sql, $feature_id))) {

      # parsing all the features can cause memory overruns
      # we are not sure why PHP does not clean up the memory as it goes
      # to avoid this problem we will call this script through an
      # independent system call
      print "$i of $num_ids Syncing feature id: $feature_id\n";
      $cmd = "php " . drupal_get_path('module', 'tripal_feature') . "/includes/tripal_feature.sync_features.inc -f $feature_id -t chado_feature";
      system($cmd);

    }
    $i++;
  }

  return '';
}