fputfasta.php

<?php
/**
 * Add a fasta line to a file
 *
 * @param $FILE file handle
 * @param $uniquename the unique identifier
 * @param $data any data on the first line after the unique identifier
 * @param $sequence the sequence to print under the first line
 */
function fputfasta($FILE, $uniquename, $data, $sequence) {
  fwrite($FILE, ">$uniquename $data\n");
  fwrite($FILE, "$sequence\n\n");
}

/**
 * Generate a fasta file from chado
 *
 * @param $variables
 * @param null $job_id
 */
function trpdownload_formats_feature_fasta_generate_file($variables, $job_id = NULL) {

  // Create the file and ready it for writting to.
  $filepath = variable_get('trpdownload_fullpath', '') . $variables['filename'];
  drush_print("File: " . $filepath);
  $FILE = fopen($filepath, 'w') or die ('Uable to create file to write to');

  // Determine which where criteria to include based on what is in query parameters.
  $where = array();
  $where_args = array();
  // feature.name
  if (isset($variables['q']['name']) AND !empty($variables['q']['name'])) {
    $where[] = "feature.name ~ :name";
    $where_args[':name'] = $variables['q']['name'];
  }
  // feature.uniquename
  if (isset($variables['q']['uniquename']) AND !empty($variables['q']['uniquename'])) {
    $where[] = "feature.uniquename ~ :uniquename";
    $where_args[':uniquename'] = $variables['q']['uniquename'];
  }
  // feature.type_id
  if (isset($variables['q']['type_id']) AND !empty($variables['q']['type_id'])) {
    $where[] = "feature.type_id = :type_id";
    $where_args[':type_id'] = $variables['q']['type_id'];
  }
  // organism.common_name
  if (isset($variables['q']['organism'])
    AND !empty($variables['q']['organism'])
    AND $variables['q']['organism'] != 'All'
  ) {
    $where[] = "organism.common_name = :organism";
    $where_args[':organism'] = $variables['q']['organism'];
  }

  // Query copied from the views interface with where arguments determined above.
  $query = "SELECT
              node_chado_feature.nid AS node_chado_feature_nid,
              feature.uniquename AS feature_uniquename,
              feature.name AS feature_name, cvterm.name AS cvterm_name,
              organism.common_name AS organism_common_name,
              feature.seqlen AS feature_seqlen,
              feature.is_obsolete AS feature_is_obsolete,
              feature.residues AS feature_residues
            FROM 
              {feature} feature
            LEFT JOIN [chado_feature] chado_feature ON feature.feature_id = chado_feature.feature_id
            LEFT JOIN [node] node_chado_feature ON chado_feature.nid = node_chado_feature.nid
            LEFT JOIN {cvterm} cvterm ON feature.type_id = cvterm.cvterm_id
            LEFT JOIN {organism} organism ON feature.organism_id = organism.organism_id
            WHERE " . implode('AND', $where) . "
            ORDER BY organism_common_name ASC, cvterm_name ASC, feature_name ASC";

  // Determine the total number of lines resulting from the query
  // for tracking progress.
  $count_query = preg_replace('/SELECT.*FROM/s', 'SELECT count(*) as num_lines FROM', $query);
  $count_query = preg_replace('/ORDER BY .*$/', '', $count_query);
  $total_lines = chado_query($count_query, $where_args)->fetchField();
  drush_print('Total Lines: ' . $total_lines);

  // Execute the original query to get the results.
  $resource = chado_query($query, $where_args);

  // For each db result write a CSV line to the file.
  $cur_line = 0;
  foreach ($resource as $row) {

    // Output the progress.
    $cur_line++;
    $percent = $cur_line / $total_lines * 100;
    if ($percent % 5 == 0) {
      drush_print(round($percent, 2) . '% Complete.');
      db_query('UPDATE {tripal_jobs} SET progress=:percent WHERE job_id=:id',
        array(':percent' => round($percent), ':id' => $job_id));
    }

    // Don't forget to write the line to the file ;-).
    fputfasta($FILE, $row->feature_uniquename, ' ', $row->feature_residues);
  }

  // Finally, close the file.
  fclose($FILE);
}