tripal/tripal_feature/tripal_feature.module

<?php

//
// Copyright 2009 Clemson University
//

/*************************************************************************
*
*/
function tripal_feature_init(){

   // add the jGCharts JS and CSS
   drupal_add_js (drupal_get_path('theme', 'tripal').'/js/tripal_feature.js'); 
   drupal_add_js (drupal_get_path('theme', 'tripal').'/js/jgcharts/jgcharts.js'); 
}
/************************************************************************
 *
 */
function tripal_feature_admin () {

   // before proceeding check to see if we have any
   // currently processing jobs. If so, we don't want
   // to give the opportunity to sync libraries
   $active_jobs = FALSE;
   if(tripal_get_module_active_jobs('tripal_feature')){
      $active_jobs = TRUE;
   }

   if(!$active_jobs){

      $form['chado_feature_accession_prefix'] = array (
         '#title'       => t('Accession Prefix'),
         '#type'        => t('textfield'),
         '#description' => t("Accession numbers for features consist of the ".
            "chado feature_id and a site specific prefix.  Set the prefix that ".
            "will be incorporated in front of each feature_id to form a unique ".
            "accession number for this site."),
         '#required'    => TRUE,
         '#default_value' => variable_get('chado_feature_accession_prefix','ID'),
      );

      $form['chado_feature_types'] = array(
         '#title'       => t('Feature Types'),
         '#type'        => 'textarea',
         '#description' => t('Enter the names of the sequence types that the ".
            "site will support with independent pages.  Pages for these data ".
            "types will be built automatically for features that exist in the ".
            "chado database.  The names listed here should be spearated by ".
            "spaces or entered separately on new lines. The names must match ".
            "exactly (spelling and case) with terms in the sequence ontology'),
         '#required'    => TRUE,
         '#default_value' => variable_get('chado_feature_types','EST contig'),
      );

      $form['browser'] = array(
         '#type' => 'fieldset',
         '#title' => t('Feature Browser')
      );
      $allowedoptions1 ['show_feature_browser'] = "Show the feature browser on the organism page. The browser loads when page loads. This may be slow for large sites.";
      $allowedoptions1 ['hide_feature_browser'] = "Hide the feature browser on the organism page. Disables the feature browser completely.";
//      $allowedoptions ['allow_feature_browser'] = "Allow loading of the feature browsing through AJAX. For large sites the initial page load will be quick with the feature browser loading afterwards.";

      $form['browser']['browse_features'] = array(
         '#title' => 'Feature Browser on Organism Page',
         '#description' => 'A feature browser can be added to an organism page to allow users to quickly '. 
            'access a feature.  This will most likely not be the ideal mechanism for accessing feature '.
            'information, especially for large sites, but it will alow users exploring the site (such '.
            'as students) to better understand the data types available on the site.',
         '#type' => 'radios',
         '#options' => $allowedoptions1,
         '#default_value'=>variable_get('tripal_feature_browse_setting',
            array('show_feature_browser')),
      );
      $form['browser']['set_browse_button'] = array(
         '#type' => 'submit',
         '#value' => t('Set Browser'),
         '#weight' => 2,
      );

      $form['summary'] = array(
         '#type' => 'fieldset',
         '#title' => t('Feature Summary')
      );
      $allowedoptions2 ['show_feature_summary'] = "Show the feature summary on the organism page. The summary loads when page loads.";
      $allowedoptions2 ['hide_feature_summary'] = "Hide the feature summary on the organism page. Disables the feature summary.";

      $form['summary']['feature_summary'] = array(
         '#title' => 'Feature Summary on Organism Page',
         '#description' => 'A feature summary can be added to an organism page to allow users to see the '.
            'type and quantity of features available for the organism.',
         '#type' => 'radios',
         '#options' => $allowedoptions2,
         '#default_value'=>variable_get('tripal_feature_summary_setting',
            array('show_feature_summary')),
      );
      $form['summary']['set_summary_button'] = array(
         '#type' => 'submit',
         '#value' => t('Set Summary'),
         '#weight' => 2,
      );

      get_tripal_feature_admin_form_sync_set($form);
      get_tripal_feature_admin_form_taxonomy_set($form);
      get_tripal_feature_admin_form_reindex_set($form);
      get_tripal_feature_admin_form_cleanup_set($form);
   } else {
      $form['notice'] = array(
         '#type' => 'fieldset',
         '#title' => t('Feature Management Temporarily Unavailable')
      );
      $form['notice']['message'] = array(
         '#value' => t('Currently, feature management jobs are waiting or ".
            "are running. Managemment features have been hidden until these ".
            "jobs complete.  Please check back later once these jobs have ".
            "finished.  You can view the status of pending jobs in the Tripal ".
            "jobs page.'),
      );
   }
   return system_settings_form($form);
}

/************************************************************************
 *
 */
function tripal_feature_admin_validate($form, &$form_state) {
   global $user;  // we need access to the user info
   $job_args = array();

   // if the user wants to sync up the chado features then
   // add the job to the management queue
   if ($form_state['values']['op'] == t('Sync all Features')) {
      tripal_add_job('Sync all features','tripal_feature',
         'tripal_feature_sync_features',$job_args,$user->uid);
   }

   if ($form_state['values']['op'] == t('Set/Reset Taxonomy for all feature nodes')) {
      tripal_add_job('Set all feature taxonomy','tripal_feature',
         'tripal_features_set_taxonomy',$job_args,$user->uid);
   }

   if ($form_state['values']['op'] == t('Reindex all feature nodes')) {
      tripal_add_job('Reindex all features','tripal_feature',
         'tripal_features_reindex',$job_args,$user->uid);
   }

   if ($form_state['values']['op'] == t('Clean up orphaned features')) {
      tripal_add_job('Cleanup orphaned features','tripal_feature',
         'tripal_features_cleanup',$job_args,$user->uid);
   }

   if ($form_state['values']['op'] == t('Set Browser')) {
      variable_set('tripal_feature_browse_setting',$form_state['values']['browse_features']);
   }

   if ($form_state['values']['op'] == t('Set Summary')) {
      variable_set('tripal_feature_summary_setting',$form_state['values']['feature_summary']);
   }
}
/************************************************************************
 *
 */
function get_tripal_feature_admin_form_cleanup_set(&$form) {
   $form['cleanup'] = array(
      '#type' => 'fieldset',
      '#title' => t('Clean Up')
   );
   $form['cleanup']['description'] = array(
       '#type' => 'item',
       '#value' => t("With Drupal and chado residing in different databases ".
          "it is possible that nodes in Drupal and features in Chado become ".
          "\"orphaned\".  This can occur if a feature node in Drupal is ".
          "deleted but the corresponding chado feature is not and/or vice ".
          "versa.  The Cleanup function will also remove nodes for features ".
          "that are not in the list of allowed feature types as specified ".
          "above.  This is helpful when a feature type needs to be ".
          "removed but was previously present as Drupal nodes. ".
          "Click the button below to resolve these discrepancies."),
       '#weight' => 1,
   );
   $form['cleanup']['button'] = array(
      '#type' => 'submit',
      '#value' => t('Clean up orphaned features'),
      '#weight' => 2,
   );
}
/************************************************************************
 *
 */
function get_tripal_feature_admin_form_reindex_set(&$form) {
   $form['reindex'] = array(
      '#type' => 'fieldset',
      '#title' => t('Reindex')
   );
   $form['reindex']['description'] = array(
       '#type' => 'item',
       '#value' => t("Reindexing of nodes is important when content for nodes ".
          "is updated external to drupal, such as external uploads to chado. ".
          "Features need to be reindexed to ensure that updates to features ".
          "are searchable. Depending on the number of features this may take ".
          "quite a while. Click the button below to begin reindexing of ".
          "features."),
       '#weight' => 1,
   );
   $form['reindex']['button'] = array(
      '#type' => 'submit',
      '#value' => t('Reindex all feature nodes'),
      '#weight' => 2,
   );
}
/************************************************************************
 *
 */
function get_tripal_feature_admin_form_taxonomy_set (&$form) {


   $form['taxonomy'] = array(
      '#type' => 'fieldset',
      '#title' => t('Set Taxonomy')
   );

   $form['taxonomy']['description'] = array(
       '#type' => 'item',
       '#value' => t("Drupal allows for assignment of \"taxonomy\" or ".
          "catagorical terms to nodes. These terms allow for advanced ".
          "filtering during searching."),
       '#weight' => 1,
   );
   $tax_options = array (
      'organism' => t('Organism name'),
      'feature_type'  => t('Feature Type (e.g. EST, mRNA, etc.)'),
      'analysis' => t('Analysis Name'),
      'library'  => t('Library Name'),
   );
   $form['taxonomy']['tax_classes'] = array (
     '#title'       => t('Available Taxonomic Classes'),
     '#type'        => t('checkboxes'),
     '#description' => t("Please select the class of terms to assign to ".
        "chado features"),
     '#required'    => FALSE,
     '#prefix'      => '<div id="taxclass_boxes">',
     '#suffix'      => '</div>',
     '#options'     => $tax_options,
     '#weight'      => 2,
     '#default_value' => variable_get('tax_classes',''),
   );
   $form['taxonomy']['button'] = array(
      '#type' => 'submit',
      '#value' => t('Set/Reset Taxonomy for all feature nodes'),
      '#weight' => 3,
   );

}
/************************************************************************
 *
 */
function get_tripal_feature_admin_form_sync_set (&$form) {

  
   // get the list of organisms which will be synced.
   $feature_sql = "SELECT * FROM {Feature} WHERE uniquename = '%s' and organism_id = %d";
   $previous_db = db_set_active('chado');
   $feature = db_fetch_object(db_query($feature_sql,$node->title,$node->organism_id));
   db_set_active($previous_db);

   // define the fieldsets
   $form['sync'] = array(
      '#type' => 'fieldset',
      '#title' => t('Sync Features')
   );

   $form['sync']['description'] = array(
      '#type' => 'item',
      '#value' => t("Click the 'Sync all Features' button to create Drupal ".
         "content for features in chado. Only features of the types listed ".
         "above in the Feature Types box will be synced. Depending on the ".
         "number of features in the chado database this may take a long ".
         "time to complete. "),
      '#weight' => 1,
   );

   $orgs = tripal_organism_get_synced();   
   $org_list = '';
   foreach($orgs as $org){
      $org_list .= "$org->genus $org->species, ";
   }
   $form['sync']['description2'] = array(
      '#type' => 'item',
      '#value' => "Only features for the following organisms will be synced: ".
         " $org_list",
      '#weight' => 1,
   );

   $form['sync']['button'] = array(
      '#type' => 'submit',
      '#value' => t('Sync all Features'),
      '#weight' => 3,
   );

}
/************************************************************************
 * Display help and module information
 * @param path which path of the site we're displaying help
 * @param arg array that holds the current path as would be returned from arg() function
 * @return help text for the path
 */
function tripal_feature_help($path, $arg) {
   $output = '';
   switch ($path) {
      case "admin/help#tripal_feature":
         $output='<p>'.t("Displays links to nodes created on this date").'</p>';
         break;
   }
   return $output;
}

/************************************************************************
 *  Provide information to drupal about the node types that we're creating
 *  in this module
 */
function tripal_feature_node_info() {
   $nodes = array();

   $nodes['chado_feature'] = array(
      'name' => t('Feature'),
      'module' => 'chado_feature',
      'description' => t('A feature from the chado database'),
      'has_title' => FALSE,
      'title_label' => t('Feature'),
      'has_body' => FALSE,
      'body_label' => t('Feature Description'),
      'locked' => TRUE
   );
   return $nodes;
}

/************************************************************************
 *  Set the permission types that the chado module uses.  Essentially we
 *  want permissionis that protect creation, editing and deleting of chado
 *  data objects
 */
function tripal_feature_perm(){
   return array(
      'access chado_feature content',
      'create chado_feature content',
      'delete chado_feature content',
      'edit chado_feature content',
   );
}

/************************************************************************
 *  Set the permission types that the module uses.
 */
function chado_feature_access($op, $node, $account) {
   if ($op == 'create') {
      return user_access('create chado_feature content', $account);
   }

   if ($op == 'update') {
      if (user_access('edit chado_feature content', $account)) {
         return TRUE;
      }
   }
   if ($op == 'delete') {
      if (user_access('delete chado_feature content', $account)) {
         return TRUE;
      }
   }
   if ($op == 'view') {
      if (user_access('access chado_feature content', $account)) {
         return TRUE;
      }
   }
   return FALSE;
}
/************************************************************************
 *  Menu items are automatically added for the new node types created
 *  by this module to the 'Create Content' Navigation menu item.  This function
 *  adds more menu items needed for this module.
 */
function tripal_feature_menu() {
   $items = array();

   // the administative settings menu
   $items['admin/tripal/tripal_feature'] = array(
     'title' => 'Features',
     'description' => 'Settings for Chado Features',
     'page callback' => 'drupal_get_form',
     'page arguments' => array('tripal_feature_admin'),
     'access arguments' => array('administer site configuration'),
     'type' => MENU_NORMAL_ITEM,
   );

   $items['admin/settings/tripal/tripal_feature/load'] = array(
     'title' => 'Bulk Load',
     'description' => 'Upload Data into Chado & Drupal',
     'page callback' => 'tripal_feature_bulkload',
     'access arguments' => array('administer site configuration'),
     'type' => MENU_NORMAL_ITEM,
   );

   return $items;
}

/************************************************************************
 *  When a new chado_feature node is created we also need to add information
 *  to our chado_feature table.  This function is called on insert of a new node
 *  of type 'chado_feature' and inserts the necessary information.
 */
function chado_feature_insert($node){
   // remove spaces, newlines from residues
   $residues = preg_replace("/[\n\r\s]/","",$node->residues);

   // If this feature already exists then don't recreate it in chado
   // TODO: the unique index in chado for this also includes the type_id. If the site
   // ever needs to have the same feature name for different types then this will break.
   $feature_sql = "SELECT * FROM {Feature} WHERE uniquename = '%s' and organism_id = %d";
   $previous_db = db_set_active('chado');
   $feature = db_fetch_object(db_query($feature_sql,$node->title,$node->organism_id));
   db_set_active($previous_db);

   // if the feature doesn't exist then let's create it in chado.
   if(!$feature){
      $sql = "INSERT INTO {feature} (organism_id, name, uniquename, residues, seqlen,".
             "    is_obsolete, type_id)".
             " VALUES(%d,'%s','%s','%s',%d, %s, ".
             "   (SELECT cvterm_id ".
             "    FROM {CVTerm} CVT ".
             "    INNER JOIN CV ON CVT.cv_id = CV.cv_id ".
             "    WHERE CV.name = 'sequence' and CVT.name = '%s'))";
      $obsolete = 'FALSE';
      if($node->is_obsolete){
         $obsolete = 'TRUE';
      }

      // use chado database
      $previous_db = db_set_active('chado');
      db_query($sql,$node->organism_id,$node->title,$node->title,
      $residues,strlen($residues),$obsolete,$node->feature_type);

      // now that we've added the feature, get the feature id for this feature
      $feature = db_fetch_object(db_query($feature_sql,$node->title,$node->organism_id));

      // now use drupal database
      db_set_active($previous_db);
   }

   // add the genbank accession and synonyms
   chado_feature_add_synonyms($node->synonyms,$feature->feature_id);

   // make sure the entry for this feature doesn't already exist in the chado_feature table
   // if it doesn't exist then we want to add it.
   $node_check_sql = "SELECT * FROM {chado_feature} ".
                     "WHERE feature_id = '%s'";
   $node_check = db_fetch_object(db_query($node_check_sql,$feature->feature_id));
   if(!$node_check){
      // next add the item to the drupal table
      $sql = "INSERT INTO {chado_feature} (nid, vid, feature_id, sync_date) ".
             "VALUES (%d, %d, %d, " . time() . ")";
      db_query($sql,$node->nid,$node->vid,$feature->feature_id);
   }
}
/************************************************************************
 */
function chado_feature_delete($node){
   // get feature_id so we can remove it from chado database
   $sql_drupal = "SELECT feature_id ".
                 "FROM {chado_feature} ".
                 "WHERE nid = %d AND vid = %d";
   $feature_id = db_result(db_query($sql_drupal, $node->nid, $node->vid));

   // remove the drupal content  
   $sql_del = "DELETE FROM {chado_feature} ".
              "WHERE nid = %d ".
              "AND vid = %d";
   db_query($sql_del, $node->nid, $node->vid);
   $sql_del = "DELETE FROM {node} ".
              "WHERE nid = %d ".
              "AND vid = %d";
   db_query($sql_del, $node->nid, $node->vid);
   $sql_del = "DELETE FROM {node_revisions} ".
              "WHERE nid = %d ".
              "AND vid = %d";
   db_query($sql_del, $node->nid, $node->vid);


   // Remove data from feature tables of chado database.  This will
   // cause a cascade delete and remove all data in referencing tables
   // for this feature
   $previous_db = db_set_active('chado');
   db_query("DELETE FROM {feature} WHERE feature_id = %d", $feature_id);
   db_set_active($previous_db);
   
   drupal_set_message("The feature and all associated data were removed from ".
      "chado");

}
/************************************************************************
 */
function chado_feature_update($node){
   if($node->revision){
      // TODO -- decide what to do about revisions
   } else {
      // get the feature for this node:
      $sql = 'SELECT feature_id FROM {chado_feature} WHERE vid = %d';
      $feature = db_fetch_object(db_query($sql, $node->vid));

      // remove spaces, newlines from residues
      $residues = preg_replace("/[\n\r\s]/","",$node->residues);

      $sql = "UPDATE {feature} ".
             " SET residues = '%s', ".
             "   name = '%s', ".
             "   uniquename = '%s', ".
             "   seqlen = %d, ".
             "   organism_id = %d, ".
             "   is_obsolete = %s, ".
             "   type_id = (SELECT cvterm_id ".
             "              FROM {CVTerm} CVT ".
             "              INNER JOIN CV ON CVT.cv_id = CV.cv_id ".
             "              WHERE CV.name = 'sequence' and CVT.name = '%s') ".
             "WHERE feature_id = %d ";
      $obsolete = 'FALSE';
      if($node->is_obsolete){
         $obsolete = 'TRUE';
      }
      $previous_db = db_set_active('chado');  // use chado database
      db_query($sql,$residues,$node->title,$node->title,
      strlen($residues),$node->organism_id,$obsolete,$node->feature_type,
      $feature->feature_id);
      db_set_active($previous_db);  // now use drupal database

      // add the genbank accession & synonyms
      // chado_feature_add_gbaccession($node->gbaccession,$feature->feature_id);
      chado_feature_add_synonyms($node->synonyms,$feature->feature_id);
   }
}
/************************************************************************
 *
 */
function chado_feature_add_synonyms($synonyms,$feature_id){

   // make sure we only have a single space between each synonym
   $synonyms = preg_replace("/[\s\n\r]+/"," ",$synonyms);
   // split the synonyms into an array based on a space as the delimieter
   $syn_array = array();
   $syn_array = explode(" ",$synonyms);

   // use the chado database
   $previous_db = db_set_active('chado');

   // remove any old synonyms
   $feature_syn_dsql = "DELETE FROM {feature_synonym} WHERE feature_id = %d";
   if(!db_query($feature_syn_dsql,$feature_id)){
      $error .= "Could not remove synonyms from feature. ";
   }

   // return if we don't have any synonmys to add
   if(!$synonyms){
      db_set_active($previous_db);
      return;
   }
   // iterate through each synonym and add it to the database
   foreach($syn_array as $syn){
      // skip this item if it's empty
      if(!$syn){ break; }

      // check to see if we have this accession number already in the database
      // if so then don't add it again. it messes up drupal if the insert fails.
      // It is possible for the accession number to be present and not the feature
      $synonym_sql = "SELECT synonym_id FROM {synonym} ".
                     "WHERE name = '%s'";
      $synonym = db_fetch_object(db_query($synonym_sql,$syn));
      if(!$synonym){
         $synonym_isql = "INSERT INTO {synonym} (name,synonym_sgml,type_id) ".
                         "VALUES ('%s','%s', ".
                         "   (SELECT cvterm_id ".
                         "    FROM {CVTerm} CVT ".
                         "    INNER JOIN CV ON CVT.cv_id = CV.cv_id ".
                         "    WHERE CV.name = 'feature_property' and CVT.name = 'synonym'))";
         if(!db_query($synonym_isql,$syn,$syn)){
            $error .= "Could not add synonym. ";
         }
         // now get the synonym we just added
         $synonym_sql = "SELECT synonym_id FROM {synonym} ".
                        "WHERE name = '%s'";
         $synonym = db_fetch_object(db_query($synonym_sql,$syn));
      }

      // now add in our new sysnonym
      $feature_syn_isql = "INSERT INTO {feature_synonym} (synonym_id,feature_id,pub_id) ".
                          "VALUES (%d,%d,1)";
      if(!db_query($feature_syn_isql,$synonym->synonym_id,$feature_id)){
         $error .= "Could not add synonyms to feature. ";
      }
   }

   // return to the drupal database
   db_set_active($previous_db);
   return $error;

}
/************************************************************************
 *
 */
function chado_feature_add_gbaccession($accession,$feature_id){

   // use chado database
   $previous_db = db_set_active('chado');

   // remove any old accession from genbank dbEST
   $fdbxref_dsql = "DELETE FROM {feature_dbxref} ".
                   "WHERE feature_id = %d and dbxref_id IN ".
                   "   (SELECT DBX.dbxref_id FROM {dbxref} DBX ".
                   "    INNER JOIN DB  ON DB.db_id = DBX.db_id ".
                   "    INNER JOIN feature_dbxref FDBX ON DBX.dbxref_id = FDBX.dbxref_id ".
                   "    WHERE DB.name = 'DB:Genbank' and FDBX.feature_id = %d)";
   if(!db_query($fdbxref_dsql,$feature_id,$feature_id)){
      $error .= "Could not remove accession from feature. ";
   }

   // if we don't have an accession number to add then just return
   if(!$accession){
      db_set_active($previous_db);
      return;
   }
   // get the db_id
   $db_sql = "SELECT db_id FROM {DB} ".
             "WHERE name = 'DB:Genbank_est'";
   $db = db_fetch_object(db_query($db_sql));

   // check to see if we have this accession number already in the database
   // if so then don't add it again. it messes up drupal if the insert fails.
   // It is possible for the accession number to be present and not the feature
   $dbxref_sql = "SELECT dbxref_id FROM {dbxref} ".
                 "WHERE db_id = %d and accession = '%s'";
   $dbxref = db_fetch_object(db_query($dbxref_sql,$db->db_id,$accession));
   if(!$dbxref){
      // add the accession number
      $dbxref_isql = "INSERT INTO {dbxref} (db_id,accession) ".
                     "  VALUES (%d, '%s') ";
      if(!db_query($dbxref_isql,$db->db_id,$accession)){
         $error .= 'Could not add accession as a database reference ';
      }
      // get the dbxref_id for the just added accession number
      $dbxref_sql = "SELECT dbxref_id FROM {dbxref} ".
                    "WHERE db_id = %d and accession = '%s'";
      $dbxref = db_fetch_object(db_query($dbxref_sql,$db->db_id,$accession));
   }


   // associate the accession number with the feature
   $feature_dbxref_isql = "INSERT INTO {feature_dbxref} (feature_id,dbxref_id) ".
                          "  VALUES (%d, %d) ";
   if(!db_query($feature_dbxref_isql,$feature_id,$dbxref->dbxref_id)){
      $error .= 'Could not add feature database reference. ';
   }

   db_set_active($previous_db);
   return $error;
}

/************************************************************************
 *
 */
function chado_feature_form ($node,$param){
   $type = node_get_types('type', $node);
   $form = array();
   $feature = $node->feature;
   $synonyms = $node->synonyms;
   $analyses = $node->analyses;
   $references = $node->references;

   // We need to pass above variables for preview to show
   $form['feature'] = array(
      '#type' => 'value',
      '#value' => $feature
   );
   // This field is read when previewing a node
   $form['synonyms'] = array(
      '#type' => 'value',
      '#value' => $synonyms
   );
   // This field is read when previewing a node
   $form['analyses'] = array(
      '#type' => 'value',
      '#value' => $analyses
   );
   // This field is read when previewing a node
   $form['references'] = array(
      '#type' => 'value',
      '#value' => $references
   );

   // keep track of the feature id if we have one.  If we do have one then
   // this would indicate an update as opposed to an insert.
   $form['feature_id'] = array(
      '#type' => 'value',
      '#value' => $feature->feature_id,
   );

   $form['title']= array(
      '#type' => 'textfield',
      '#title' => t('Unique Feature Name'),
      '#required' => TRUE,
      '#default_value' => $feature->featurename,
      '#description' => t('Enter a unique name for this feature'),
      '#weight' => 1,
      '#maxlength' => 255
   );

   // get the list of supported feature types
   $ftypes = array();
   $ftypes[''] = '';
   $supported_ftypes = split("[ \n]",variable_get('chado_feature_types','EST contig'));
   foreach($supported_ftypes as $ftype){
      $ftypes["$ftype"] = $ftype;
   }

   $form['feature_type'] = array (
     '#title'       => t('Feature Type'),
     '#type'        => t('select'),
     '#description' => t("Choose the feature type."),
     '#required'    => TRUE,
     '#default_value' => $feature->cvname,
     '#options'     => $ftypes,
     '#weight'      => 2
   );
   // get the list of organisms
   $sql = "SELECT * FROM {Organism} ORDER BY genus, species";
   $previous_db = db_set_active('chado');  // use chado database
   $org_rset = db_query($sql);
   db_set_active($previous_db);  // now use drupal database

   //
   $organisms = array();
   $organisms[''] = '';
   while($organism = db_fetch_object($org_rset)){
      $organisms[$organism->organism_id] = "$organism->genus $organism->species ($organism->common_name)";
   }

   $form['organism_id'] = array (
     '#title'       => t('Organism'),
     '#type'        => t('select'),
     '#description' => t("Choose the organism with which this feature is associated "),
     '#required'    => TRUE,
     '#default_value' => $feature->organism_id,
     '#options'     => $organisms,
     '#weight'      => 3,
   );

   // Get synonyms
   if ($synonyms) {
      if (is_array($synonyms)) {
         foreach ($synonyms as $synonym){
            $syn_text .= "$synonym->name\n";
         }
      } else {
         $syn_text = $synonyms;
      }
   }
   $form['synonyms']= array(
      '#type' => 'textarea',
      '#title' => t('Synonyms'),
      '#required' => FALSE,
      '#default_value' => $syn_text,
      '#description' => t('Enter alternate names (synonmys) for this feature to help in searching and identification. You may enter as many alternate names as needed separated by spaces or on different lines.'),
      '#weight' => 5,
   );

   $form['residues']= array(
      '#type' => 'textarea',
      '#title' => t('Residues'),
      '#required' => FALSE,
      '#default_value' => $feature->residues,
      '#description' => t('Enter the nucelotide sequences for this feature'),
      '#weight' => 6
   );

   $checked = '';
   if($feature->is_obsolete == 't'){
      $checked = '1';
   }
   $form['is_obsolete']= array(
      '#type' => 'checkbox',
      '#title' => t('Is Obsolete'),
      '#required' => FALSE,
      '#default_value' => $checked,
      '#description' => t('Check this box if this sequence should be retired and no longer included in further analysis.'),
      '#weight' => 8
   );
   return $form;
}
/************************************************************************
 *
 */
function chado_feature_validate($node){
   $result = 0;

   // if this is an update, we want to make sure that a different feature for
   // the organism doesn't already have this uniquename. We don't want to give
   // two sequences the same uniquename
   if($node->feature_id){
      $sql = "SELECT * FROM {Feature} WHERE uniquename = '%s' ".
             "  AND organism_id = %d AND NOT feature_id = %d";
      $previous_db = db_set_active('chado');
      $result = db_fetch_object(db_query($sql, $node->title,$node->organism_id,$node->feature_id));
      db_set_active($previous_db);
      if($result){
         form_set_error('title',t("Feature update cannot proceed. The feature name '$node->title' is not unique for this organism. Please provide a unique name for this feature. "));
      }
   }

   // if this is an insert then we just need to make sure this name doesn't
   // already exist for this organism if it does then we need to throw an error
   else {
      $sql = "SELECT * FROM {Feature} WHERE uniquename = '%s' AND organism_id = %d";
      $previous_db = db_set_active('chado');
      $result = db_fetch_object(db_query($sql, $node->title,$node->organism_id));
      db_set_active($previous_db);
      if($result){
         form_set_error('title',t("Feature insert cannot proceed. The feature name '$node->title' already exists for this organism. Please provide a unique name for this feature. "));
      }
   }

   // we want to remove all characters except IUPAC nucleotide characters from the
   // the residues. however, residues are not required so if blank then we'll skip
   // this step
   if($node->residues){
      $residues = preg_replace("/[^\w]/",'',$node->residues);
      if(!preg_match("/^[ACTGURYMKSWBDHVN]+$/i",$residues)){
         form_set_error('residues',t("The residues in feature $node->title contains more than the nucleotide IUPAC characters. Only the following characters are allowed: A,C,T,G,U,R,Y,M,K,S,W,B,D,H,V,N: '" . $residues ."'"));
      }
   }

   // we don't allow a genbank accession number for a contig
   if($node->feature_type == 'contig' and $node->gbaccession){
      form_set_error('gbaccession',t("Contigs cannot have a genbank accession number.  Please change the feature type or remove the accession number"));
   }



}
/************************************************************************
 *  When a node is requested by the user this function is called to allow us
 *  to add auxiliary data to the node object.
 */
function chado_feature_load($node){
   // get the feature_id for this node:
   $sql = 'SELECT feature_id FROM {chado_feature} WHERE vid = %d';
   $map = db_fetch_object(db_query($sql, $node->vid));

   $previous_db = db_set_active('chado');  // use chado database

   // get information about this organism and add it to the items in this node
   $sql = "SELECT F.feature_id, F.name as featurename, F.uniquename, ".
          "F.residues, F.seqlen, O.genus, O.species, O.common_name, ".
          "  CVT.name as cvname, O.organism_id, F.type_id, F.is_obsolete  ".
          "FROM {Feature} F ".
          "  INNER JOIN Organism O ON F.organism_id = O.organism_id ".
          "  INNER JOIN CVterm CVT ON F.type_id = CVT.cvterm_id ".
          "WHERE F.feature_id = %d";
   $feature = db_fetch_object(db_query($sql,$map->feature_id));
   $additions->feature = $feature;
   $additions->seqlen = $feature->seqlen;
   
   // get the feature synonyms
   $sql = "SELECT S.name ".
          "FROM {Feature_Synonym} FS ".
          "  INNER JOIN Synonym S ".
          "    ON FS.synonym_id = S.Synonym_id ".
          "WHERE FS.feature_id = %d";
   $results = db_query($sql,$map->feature_id);
   $synonyms = array();
   $i=0;
   while($synonym = db_fetch_object($results)){
      $synonyms[$i++] = $synonym;
   }
   $additions->synonyms = $synonyms;

   // get feature references in external databases
   $sql = "SELECT F.uniquename,F.Feature_id,DBX.accession,DB.description as dbdesc, ".
          "   DB.db_id, DB.name as db_name, DB.urlprefix ".
          "FROM {Feature} F ".
          "  INNER JOIN Feature_dbxref FDBX on F.feature_id = FDBX.feature_id ".
          "  INNER JOIN Dbxref DBX on DBX.dbxref_id = FDBX.dbxref_id ".
          "  INNER JOIN DB on DB.db_id = DBX.db_id ".
          "WHERE F.feature_id = %d";
   $results = db_query($sql,$map->feature_id);
   $references = array();
   $i=0;
   while($accession = db_fetch_object($results)){
      $references[$i++] = $accession;
      // we want to specifically pull out the genbank id
      if(preg_match("/Genbank_est/",$accession->db_name)){
         $additions->gbaccession = $accession;
      }
   }
   $additions->references = $references;
   db_set_active($previous_db);  // now use drupal database

   // get organism node nid
   $sql = "SELECT nid FROM {chado_organism} WHERE organism_id = %d";
   $org_nid = db_result(db_query($sql, $additions->feature->organism_id));
   $additions->org_nid = $org_nid;
   return $additions;
}

/************************************************************************
 *  This function customizes the view of the chado_feature node.  It allows
 *  us to generate the markup.
 */
function chado_feature_view ($node, $teaser = FALSE, $page = FALSE) {

   if (!$teaser) {
      // use drupal's default node view:
      $node = node_prepare($node, $teaser);
            
      // if we're building the node for searching then
      // we want to handle this within the module and
      // not allow theme customization.  We don't want to
      // index all items (such as DNA sequence).
      if($node->build_mode == NODE_BUILD_SEARCH_INDEX){
         $node->content['index_version'] = array(
            '#value' => theme('tripal_feature_search_index',$node),
            '#weight' => 1,
         );
      }
      else if($node->build_mode == NODE_BUILD_SEARCH_RESULT){
         $node->content['index_version'] = array(
            '#value' => theme('tripal_feature_search_results',$node),
            '#weight' => 1,
         );
      }
      else {
         // do nothing here, let the theme derived template handle display
      }
   }

   return $node;
}

/*******************************************************************************
 * Display feature information for associated organisms. This function also
 * provides contents for indexing
 */
function tripal_feature_nodeapi(&$node, $op, $teaser, $page) {

   switch ($op) {
      // Note that this function only adds feature view to an organism node.
      // The view of a feature node is controled by the theme *.tpl file
      case 'view':
         // Set the node types for showing feature information
         $types_to_show = array('chado_organism', 'chado_library');

         // Abort if this node is not one of the types we should show.
         if (!in_array($node->type, $types_to_show, TRUE)) {
            break;
         }

         // Add feature to the content item if it's not a teaser
         if (!$teaser) {

            // Show feature browser
            $node->content['tripal_feature_browser'] = array(
               '#value' => theme('tripal_feature_browser', $node),
               '#weight' => 5
            );
            $node->content['tripal_feature_org_counts'] = array(
               '#value' => theme('tripal_feature_counts', $node),
               '#weight' => 4
            );
         }
   }
}
/************************************************************************
 *  We need to let drupal know about our theme functions and their arguments.
 *  We create theme functions to allow users of the module to customize the
 *  look and feel of the output generated in this module
 */
function tripal_feature_theme () {
   return array(
      'tripal_feature_search_index' => array (
         'arguments' => array('node'),
       ),
      'tripal_feature_search_results' => array (
         'arguments' => array('node'),
      ),
      'tripal_feature_browser' => array (
         'arguments' => array('node'),
      ),
      'tripal_feature_counts' => array (
         'arguments' => array('node'),
      )
   );
}
/*******************************************************************************
 *  create a list of features for the organism and pie chart
 */
function theme_tripal_feature_counts($node){

   // don't show the summary if the settings in the admin page is turned off
   $show_browser = variable_get('tripal_feature_summary_setting',array('show_feature_summary'));

   if(strcmp($show_browser,'show_feature_summary')!=0){
      return;
   }

   // get the feature counts.  This is dependent on a materialized view
   // installed with the organism module
   $content = '';
   if ($node->organism_id && $node->type == 'chado_organism') {
      $sql = "SELECT * FROM {organism_feature_count} ".
      		 "WHERE organism_id = %d AND NOT feature_type = 'EST_match' ".
             "ORDER BY num_features desc";
      $features = array();
      $previous_db = db_set_active('chado');  // use chado database
      $results = db_query($sql,$node->organism_id);
      db_set_active($previous_db);  // now use drupal database
      $feature = db_fetch_object($results); // retrieve the first result
      if ($feature) {
         $content .= "<div class=\"tripal_feature_summary-info-box\"><br>
                             <div class=\"tripal_expandableBox\">".
                     "<h3>Feature Summary</h3>".
                     "</div>";
         $content .= "<div class=\"tripal_expandableBoxContent\">";
         $content .= "<table class=\"tripal_table_horz\">";
         $content .= "  <tr>";
         $content .= "    <th class=\"dbfieldname\">Type</th>";
         $content .= "    <th class=\"dbfieldname\">Number</th>";
         $content .= "  </tr>";
         do {
            $content .= "<tr>";
            $content .= "  <td>$feature->feature_type</td>";
            $content .= "  <td>". number_format($feature->num_features) . "</td>";
            $content .= "</tr>";
         } while($feature = db_fetch_object($results));
         $content .= "</table>";
         $content .= "
            <img class=\"tripal_cv_chart\" id=\"tripal_feature_cv_chart_$node->organism_id\" src=\"\" border=\"0\">
         ";
         $content .= "</div></div>";
      }
   }
   return $content;
}
/************************************************************************
 *
 */
function tripal_feature_cv_chart($chart_id){

  // The CV module will create the JSON array necessary for buillding a
  // pie chart using jgChart and Google Charts.  We have to pass to it
  // a table that contains count information, tell it which column 
  // contains the cvterm_id and provide a filter for getting the
  // results we want from the table.
  $organism_id = preg_replace("/^tripal_feature_cv_chart_(\d+)$/","$1",$chart_id);
  $options = array(
     count_mview      => 'organism_feature_count',
     cvterm_id_column => 'cvterm_id',
     count_column     => 'num_features',
     size             => '650x200',
     filter           => "CNT.organism_id = $organism_id AND NOT feature_type = 'EST_match' ",
  );
  return $options;
}
/************************************************************************
 *
 */
function tripal_feature_cv_tree($tree_id){

  // The CV module will create the JSON array necessary for buillding a
  // pie chart using jgChart and Google Charts.  We have to pass to it
  // a table that contains count information, tell it which column 
  // contains the cvterm_id and provide a filter for getting the
  // results we want from the table.
  $organism_id = preg_replace("/^tripal_feature_cv_tree_(\d+)$/","$1",$tree_id);
  $options = array(
     cv_id            => tripal_cv_get_cv_id('sequence'),
     count_mview      => 'organism_feature_count',
     cvterm_id_column => 'cvterm_id',
     count_column     => 'num_features',
     filter           => "CNT.organism_id = $organism_id",
     label            => 'Features',
  );
  return $options;
}
/*******************************************************************************
 *  create a simple paged feature browser
 */
function theme_tripal_feature_browser($node){

   // don't show the browser if the settings in the admin page is turned off
   $show_browser = variable_get('tripal_feature_browse_setting',array('show_feature_browser'));

   if(strcmp($show_browser,'show_feature_browser')!=0){
      return;
   }
   if ($node->organism_id && $node->type == 'chado_organism') {
      # 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.
      $allowed_types = variable_get('chado_feature_types','EST contig');
      $allowed_types = preg_replace("/[\s\n\r]+/"," ",$allowed_types);
      $so_terms = split(' ',$allowed_types);
      $where_cvt = "";
      foreach ($so_terms as $term){
         $where_cvt .= "CVT.name = '$term' OR ";
      }
      $where_cvt = substr($where_cvt,0,strlen($where_cvt)-3);  # strip trailing 'OR'

      // get the features for this organism
      $sql  = "SELECT F.name,F.feature_id,F.uniquename,CVT.name as cvname ".
              "FROM {feature} F ".
                 "  INNER JOIN {cvterm} CVT on F.type_id = CVT.cvterm_id ".
               "WHERE organism_id = $node->organism_id and ($where_cvt) ".
               "ORDER BY feature_id ASC";

      // the counting SQL
      $csql  = "SELECT count(*) ".
               "FROM {feature} F".
               "  INNER JOIN {cvterm} CVT on F.type_id = CVT.cvterm_id ".
               "WHERE organism_id = $node->organism_id and ($where_cvt) ".
               "GROUP BY organism_id ";

      $previous_db = db_set_active('chado');  // use chado database
      $features = pager_query($sql,10,0,$csql);
      db_set_active($previous_db);  // now use drupal database

      $content = "<br><div id=\"tripal_feature_box\" class=\"feature-info-box\">";
      $content .= "<div class=\"tripal_expandableBox\">".
                  "<h3>Browse Features</h3>".
                  "</div>";
      $content .= "<div class=\"tripal_expandableBoxContent\">";
      $content .= "Below are the features associated with this organism.\n";
      $content .= "<table class=\"tripal_table_horz\">";
      $content .= "  <tr>";
      $content .= "    <th>Feature Name</th>";
      $content .= "    <th>Type</th>";
      $content .= "  </tr>";

      // prepare the query that will lookup node ids
      $sql = "SELECT nid FROM {chado_feature} ".
              "WHERE feature_id = %d";
      while($feature = db_fetch_object($features)){
         $node = db_fetch_object(db_query($sql,$feature->feature_id));
         if($node){
            $name= "<a href=\"" . url("node/$node->nid") . "\">$feature->name</a>";
         } else {
            $name= "$feature->name";
         }
         $content .= "  <tr>";
         $content .= "    <td>$name</td>";
         $content .= "    <td>$feature->cvname</td>";
         $content .= "  </tr>";
      }
      $content .= "</table>";
      $content .= theme('pager');
      $content .= "</div></div>";
      return $content;
   }
}
/************************************************************************
 *  This function is an extension of the chado_feature_view by providing
 *  the markup for the feature object THAT WILL BE INDEXED.
 */
function theme_tripal_feature_search_index ($node) {
   $feature = $node->feature;
   $content = '';

   // get the accession prefix
   $aprefix = variable_get('chado_feature_accession_prefix','ID');

   $content .= "<h1>$feature->uniquename</h1>. ";
   $content .= "<strong>$aprefix$feature->feature_id.</strong> ";
   $content .= "$feature->cvname ";
   $content .= "$feature->common_name ";

   // add the synonyms of this feature to the text for searching
   $synonyms = $node->synonyms;
   if(count($synonyms) > 0){
      foreach ($synonyms as $result){
         $content .= "$result->name ";
      }
   }

   return $content;
}
/************************************************************************
 *  This function is an extension of the chado_feature_view by providing
 *  the markup for the feature object THAT WILL BE INDEXED.
 */
function theme_tripal_feature_search_results ($node) {
   $feature = $node->feature;
   $content = '';

   // get the accession prefix
   $aprefix = variable_get('chado_feature_accession_prefix','ID');

   $content .= "Feature Name: <h1>$feature->uniquename</h1>. ";
   $content .= "<strong>Accession: $aprefix$feature->feature_id.</strong>";
   $content .= "Type: $feature->cvname. ";
   $content .= "Organism: $feature->common_name. ";

   // add the synonyms of this feature to the text for searching
   $synonyms = $node->synonyms;
   if(count($synonyms) > 0){
      $content .= "Synonyms: ";
      foreach ($synonyms as $result){
         $content .= "$result->name, ";
      }
   }

   return $content;
}
/************************************************************************
 *
 */
function tripal_feature_sync_features ($max_sync = 0, $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.
   $allowed_types = variable_get('chado_feature_types','EST contig');
   $allowed_types = preg_replace("/[\s\n\r]+/"," ",$allowed_types);
   $so_terms = split(' ',$allowed_types);
   $where_cvt = "";
   foreach ($so_terms as $term){
      $where_cvt .= "CVT.name = '$term' OR ";
   }
   $where_cvt = substr($where_cvt,0,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){
      $where_org .= "F.organism_id = $org->organism_id OR ";
   }
   $where_org = substr($where_org,0,strlen($where_org)-3);  # strip trailing 'OR'

   // use this SQL statement to get the features that we're going to upload
   $sql = "SELECT feature_id ".
          "FROM {FEATURE} F ".
          "   INNER JOIN Cvterm CVT ON F.type_id = CVT.cvterm_id ".
          "WHERE ($where_cvt) AND ($where_org) ".
          "ORDER BY feature_id";
   // get the list of features
   $previous_db = db_set_active('chado');  // use chado database
   $results = db_query($sql);

   db_set_active($previous_db);  // now use drupal database

   // 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);
   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,$feature_id))){
         tripal_feature_sync_feature ($feature_id);
      }
      $i++;
   }

   return '';
}

/************************************************************************
 *
 */
function tripal_feature_sync_feature ($feature_id){

   global $user;
   $create_node = 1;   // set to 0 if the node exists and we just sync and not create

   // get the accession prefix
   $aprefix = variable_get('chado_feature_accession_prefix','ID');

   // if we don't have a feature_id then return
   if(!$feature_id){
      drupal_set_message(t("Please provide a feature_id to sync"));
      return '';
   }

   // get information about this feature
   $fsql = "SELECT F.feature_id, F.name, F.uniquename,O.genus, ".
           "    O.species,CVT.name as cvname,F.residues,F.organism_id ".
           "FROM {FEATURE} F ".
           "  INNER JOIN Cvterm CVT ON F.type_id = CVT.cvterm_id ".
           "  INNER JOIN Organism O ON F.organism_id = O.organism_ID ".
           "WHERE F.feature_id = %d";
   $previous_db = db_set_active('chado');  // use chado database
   $feature = db_fetch_object(db_query($fsql,$feature_id));
   db_set_active($previous_db);  // now use drupal database

   // check to make sure that we don't have any nodes with this feature name as a title
   // but without a corresponding entry in the chado_feature table if so then we want to
   // clean up that node.  (If a node is found we don't know if it belongs to our feature or
   // not since features can have the same name/title.)
   $tsql =  "SELECT * FROM {node} N ".
            "WHERE title = '%s'";
   $cnsql = "SELECT * FROM {chado_feature} ".
            "WHERE nid = %d";
   $nodes = db_query($tsql,$feature->name);
   // cycle through all nodes that may have this title
   while($node = db_fetch_object($nodes)){
      $feature_nid = db_fetch_object(db_query($cnsql,$node->nid));
      if(!$feature_nid){
         drupal_set_message(t("$feature_id: A node is present but the chado_feature entry is missing... correcting"));
         node_delete($node->nid);
      }
   }

   // check if this feature already exists in the chado_feature table.
   // if we have a chado feature, we want to check to see if we have a node
   $cfsql = "SELECT * FROM {chado_feature} ".
            "WHERE feature_id = %d";
   $nsql =  "SELECT * FROM {node} ".
            "WHERE nid = %d";
   $chado_feature = db_fetch_object(db_query($cfsql,$feature->feature_id));
   if($chado_feature){
      drupal_set_message(t("$feature_id: A chado_feature entry exists"));
      $node = db_fetch_object(db_query($nsql,$chado_feature->nid));
      if(!$node){
         // if we have a chado_feature but not a node then we have a problem and
         // need to cleanup
         drupal_set_message(t("$feature_id: The node is missing, but has a chado_feature entry... correcting"));
         $df_sql = "DELETE FROM {chado_feature} WHERE feature_id = %d";
         db_query($df_sql,$feature_id);
      } else {
         drupal_set_message(t("$feature_id: A corresponding node exists"));
         $create_node = 0;
      }
   }

   // if we've encountered an error then just return.
   if($error_msg = db_error()){
      //print "$error_msg\n";
      return '';
   }

   // if a drupal node does not exist for this feature then we want to
   // create one.  Note that the node_save call in this block
   // will call the hook_submit function which
   if($create_node){
      drupal_set_message(t("$feature_id: Creating node $feature->name"));
      $new_node = new stdClass();
      $new_node->type = 'chado_feature';
      $new_node->uid = $user->uid;
      $new_node->title = "$feature->name";
      $new_node->feature_id = $feature->feature_id;
      $new_node->residues = $feature->residues;
      $new_node->organism_id = $feature->organism_id;
      $new_node->feature_type = $feature->cvname;

      // validate the node and if okay then submit
      node_validate($new_node);
      if ($errors = form_get_errors()) {
         foreach($errors as $key => $msg){
            drupal_set_message($msg);
         }
         return $errors;
      } else {
         $node = node_submit($new_node);
         node_save($node);
      }

   }
   else {
      $node = $chado_feature;
   }


   // set the taxonomy for this node
   drupal_set_message(t("$feature_id ($node->nid): setting taxonomy"));
   tripal_feature_set_taxonomy($node,$feature_id);

   // reindex the node
   drupal_set_message(t("$feature_id( $node->nid): indexing"));
   tripal_feature_index_feature ($feature_id,$node->nid);

   // remove any URL alias that may already exist and recreate
   drupal_set_message(t("$feature_id ($node->nid): setting URL alias"));
   db_query("DELETE FROM {url_alias} WHERE dst = '%s'", "$aprefix$feature_id");
   path_set_alias("node/$node->nid","$aprefix$feature_id");

   return '';
}
/************************************************************************
 *
 */
function tripal_feature_set_vocabulary (){

   //include the file containing the required functions for adding taxonomy vocabs
   module_load_include('inc', 'taxonomy', 'taxonomy.admin');

   // get the vocabularies so that we make sure we don't recreate
   // the vocabs that already exist
   $vocabularies = taxonomy_get_vocabularies();
   $ft_vid = NULL;
   $op_vid = NULL;
   $lb_vid = NULL;
   $an_vid = NULL;

   // These taxonomic terms are hard coded because we
   // konw we have these relationships in the chado tables
   // through foreign key relationships.  The tripal
   // modules that correspond to these chado "modules" don't
   // need to be installed for the taxonomy to work.
   foreach($vocabularies as $vocab){
      if($vocab->name == 'Feature Type'){
         $ft_vid = $vocab->vid;
      }
      if($vocab->name == 'Organism'){
         $op_vid = $vocab->vid;
      }
      if($vocab->name == 'Library'){
         $lb_vid = $vocab->vid;
      }
      if($vocab->name == 'Analysis'){
         $an_vid = $vocab->vid;
      }
   }

   if(!$ft_vid){
      $form_state = array();
      $values = array(
         'name' => t('Feature Type'),
         'nodes' => array('chado_feature' => 'chado_feature'),
         'description' => t('The feature type (or SO cvterm for this feature).'),
         'help' => t('Select the term that matches the feature '),
         'tags' => 0,
         'hierarchy' => 1,
         'relations' => 1,
         'multiple' => 0,
         'required' => 0,
         'weight' => 1,
      );
      drupal_execute('taxonomy_form_vocabulary', $form_state,$values);
      drupal_execute('taxonomy_form_vocabulary', $form_state);
   }

   if(!$op_vid){
      $form_state = array();
      $values = array(
         'name' => t('Organism'),
         'nodes' => array('chado_feature' => 'chado_feature'),
         'description' => t('The organism to which this feature belongs.'),
         'help' => t('Select the term that matches the feature '),
         'tags' => 0,
         'hierarchy' => 1,
         'relations' => 1,
         'multiple' => 0,
         'required' => 0,
         'weight' => 2,
      );
      drupal_execute('taxonomy_form_vocabulary', $form_state,$values);
      drupal_execute('taxonomy_form_vocabulary', $form_state);
   }

   if(!$lb_vid){
      $form_state = array();
      $values = array(
         'name' => t('Library'),
         'nodes' => array('chado_feature' => 'chado_feature'),
         'description' => t('Chado features associated with a library are assigned the term associated with the library'),
         'help' => t('Select the term that matches the feature '),
         'tags' => 0,
         'hierarchy' => 1,
         'relations' => 1,
         'multiple' => 0,
         'required' => 0,
         'weight' => 3,
      );
      drupal_execute('taxonomy_form_vocabulary', $form_state, $values);
      drupal_execute('taxonomy_form_vocabulary', $form_state);
   }

   if(!$an_vid){
      $form_state = array();
      $values = array(
         'name' => t('Analysis'),
         'nodes' => array('chado_feature' => 'chado_feature'),
         'description' => t('Any analysis to which this feature belongs.'),
         'help' => t('Select the term that matches the feature '),
         'tags' => 0,
         'hierarchy' => 1,
         'relations' => 1,
         'multiple' => 1,
         'required' => 0,
         'weight' => 4,
      );
      drupal_execute('taxonomy_form_vocabulary', $form_state,$values);
      drupal_execute('taxonomy_form_vocabulary', $form_state);
   }
}
/************************************************************************
 *
 */
function tripal_feature_del_vocabulary(){
   //include the file containing the required functions for adding taxonomy vocabs
   module_load_include('inc', 'taxonomy', 'taxonomy.admin');

   // get the vocabularies
   $vocabularies = taxonomy_get_vocabularies();

   // These taxonomic terms are hard coded because we
   // know we have these relationships in the chado tables
   // through foreign key relationships.  The tripal
   // modules that correspond to these chado "modules" don't
   // need to be installed for the taxonomy to work.
   foreach($vocabularies as $vocab){
      if($vocab->name == 'Feature Type'){
         taxonomy_del_vocabulary($vocab->vid);
      }
      if($vocab->name == 'Organism'){
         taxonomy_del_vocabulary($vocab->vid);
      }
      if($vocab->name == 'Library'){
         taxonomy_del_vocabulary($vocab->vid);
      }
      if($vocab->name == 'Analysis'){
         taxonomy_del_vocabulary($vocab->vid);
      }
   }

}
/************************************************************************
 *
 */
function tripal_features_set_taxonomy($max_sync = 0,$job_id = NULL){

   // make sure our vocabularies are cleaned and reset before proceeding
   tripal_feature_del_vocabulary();
   tripal_feature_set_vocabulary();

   // iterate through all drupal feature nodes and set the taxonomy
   $results = db_query("SELECT * FROM {chado_feature}");
   $nsql =  "SELECT * FROM {node} ".
            "WHERE nid = %d";
   $i = 0;

   // load into ids array
   $count = 0;
   $chado_features = array();
   while($chado_feature = db_fetch_object($results)){
      $chado_features[$count] = $chado_feature;
      $count++;
   }

   // Iterate through features that need to be synced
   $interval = intval($count * 0.01);
   foreach($chado_features as $chado_feature){

      // update the job status every 1% features
      if($job_id and $i % $interval == 0){
         tripal_job_set_progress($job_id,intval(($i/$count)*100));
      }
      $node = db_fetch_object(db_query($nsql,$chado_feature->nid));
      tripal_feature_set_taxonomy($node,$chado_feature->feature_id);

      $i++;
   }
}
/************************************************************************
 *
 */
function tripal_feature_set_taxonomy ($node,$feature_id){

   // iterate through the taxonomy classes that have been
   // selected by the admin user and make sure we only set those
   $tax_classes = variable_get('tax_classes', '');
   $do_ft = 0;
   $do_op = 0;
   $do_lb = 0;
   $do_an = 0;
   foreach($tax_classes as $class){
      if(strcmp($class ,'organism')==0){
         $do_op = 1;
      }
      if(strcmp($class,'feature_type')==0){
         $do_ft = 1;
      }
      if(strcmp($class,'library')==0){
         $do_lb = 1;
      }
      if(strcmp($class,'analysis')==0){
         $do_an = 1;
      }
   }


   // get the list of vocabularies and find our two vocabularies of interest
   $vocabularies = taxonomy_get_vocabularies();
   $ft_vid = NULL;
   $op_vid = NULL;
   $lb_vid = NULL;
   $an_vid = NULL;
   foreach($vocabularies as $vocab){
      if($vocab->name == 'Feature Type'){
         $ft_vid = $vocab->vid;
      }
      if($vocab->name == 'Organism'){
         $op_vid = $vocab->vid;
      }
      if($vocab->name == 'Library'){
         $lb_vid = $vocab->vid;
      }
      if($vocab->name == 'Analysis'){
         $an_vid = $vocab->vid;
      }
   }

   // get the cvterm and the organism for this feature
   $sql = "SELECT CVT.name AS cvname, O.genus, O.species ".
          "FROM {CVTerm} CVT ".
          "  INNER JOIN Feature F on F.type_id = CVT.cvterm_id ".
          "  INNER JOIN Organism O ON F.organism_id = O.organism_id ".
          "WHERE F.feature_id = $feature_id";
   $previous_db = db_set_active('chado');  // use chado database
   $feature = db_fetch_object(db_query($sql));
   db_set_active($previous_db);  // now use drupal database

   // Set the feature type for this feature
   if($do_ft && $ft_vid){
      $tags["$ft_vid"] = "$feature->cvname";
   }
   // Set the organism for this feature type
   if($do_op && $op_vid){
      $tags["$op_vid"] = "$feature->genus $feature->species";
   }

   // get the library that this feature may belong to and add it as taxonomy
   if($do_lb && $lb_vid){
      $sql = "SELECT L.name ".
             "FROM {Library} L ".
             "  INNER JOIN Library_feature LF ON LF.library_id = L.library_id ".
             "WHERE LF.feature_id = %d ";
      $previous_db = db_set_active('chado');  // use chado database
      $library = db_fetch_object(db_query($sql,$feature_id));
      db_set_active($previous_db);  // now use drupal database
      $tags["$lb_vid"] = "$library->name";
   }

   // get the analysis that this feature may belong to and add it as taxonomy
   if($do_an && $an_vid){
      $sql = "SELECT A.name ".
             "FROM {Analysis} A ".
             "  INNER JOIN Analysisfeature AF ON AF.analysis_id = A.analysis_id ".
             "WHERE AF.feature_id = $feature_id ";
      $results = db_query($sql);
      $previous_db = db_set_active('chado');  // use chado database
      $analysis_terms = array();
      while($analysis=db_fetch_object($results)){
         // TODO -- how to set more than one taxonmy term ????
         $analysis_terms[] = "$analysis->name";
         //           $tags["$an_vid"] = "$analysis->name";
      }
      $tags["$an_vid"] = $analysis_terms;
      db_set_active($previous_db);  // now use drupal database
   }

   // now add the taxonomy to the node
   $terms['tags'] = $tags;
   taxonomy_node_save($node,$terms);
   //   print "Setting $node->title: " . implode(", ",$tags) . "\n";

}
/************************************************************************
 *
 */
function tripal_features_reindex ($max_sync,$job_id = NULL){
   $i = 0;

   // We register a shutdown function to ensure that the nodes
   // that are indexed will have proper entries in the search_totals
   // table.  Without these entries, the searching doesn't work
   // properly. This function may run for quite a while since
   // it must calculate the sum of the scores of all entries in
   // the search_index table.  In the case of common words like
   // 'contig', this will take quite a while
   register_shutdown_function('search_update_totals');

   // use this SQL statement to get the features that we're going to index. This
   // SQL statement is derived from the hook_search function in the Drupal API.
   // Essentially, this is the SQL statement that finds all nodes that need
   // reindexing, but adjusted to include the chado_feature
   $sql = "SELECT N.nid, N.title, CF.feature_id ".
          "FROM {node} N ".
          "  INNER JOIN chado_feature CF ON CF.nid = N.nid ";
   $results = db_query($sql);

   // load into ids array
   $count = 0;
   $chado_features = array();
   while($chado_feature = db_fetch_object($results)){
      $chado_features[$count] = $chado_feature;
      $count++;
   }

   // Iterate through features that need to be synced
   $interval = intval($count * 0.01);
   foreach($chado_features as $chado_feature){

      // update the job status every 1% features
      if($job_id and $i % $interval == 0){
         tripal_job_set_progress($job_id,intval(($i/$count)*100));
      }

      // sync only the max requested
      if($max_sync and $i == $max_sync){
         return '';
      }
      tripal_feature_index_feature ($chado_feature->feature_id,$chado_feature->nid);
      $i++;
   }

   return '';
}

/************************************************************************
 *
 */
function tripal_feature_index_feature ($feature_id,$nid){
   // return if we haven't been provided with a feature_id
   if(!$feature_id){
      return 0;
   }

   // if we only have a feature_id then let's find a corresponding
   // node.  If we can't find a node then return.
   if(!$nid){
      $nsql = "SELECT N.nid,N.title FROM {chado_feature} CF ".
              "  INNER JOIN {node} N ON N.nid = CF.nid ".
              "WHERE CF.feature_id = %d";
      $node = db_fetch_object(db_query($nsql,$feature_id));
      if(!$node){
         return 0;
      }
      $node = node_load($node->nid);
   } else {
      $node = node_load($nid);
   }

   // node load the noad, the comments and the taxonomy and
   // index
   $node->build_mode = NODE_BUILD_SEARCH_INDEX;
   $node = node_build_content($node, FALSE, FALSE);
   $node->body = drupal_render($node->content);
   node_invoke_nodeapi($node, 'view', FALSE, FALSE);
   $node->body .= module_invoke('comment', 'nodeapi', $node, 'update index');
   $node->body .= module_invoke('taxonomy','nodeapi', $node, 'update index');
   //   print "$node->title: $node->body\n";
   search_index($node->nid,'node',$node->body);
   return 1;
}
/************************************************************************
 *
 */
function tripal_features_cleanup($dummy = NULL, $job_id = NULL) {

   // build the SQL statments needed to check if nodes point to valid features
   $dsql = "SELECT * FROM {node} WHERE type = 'chado_feature' order by nid";
   $nsql = "SELECT * FROM {node} WHERE nid = %d";
   $csql = "SELECT * FROM {chado_feature} where nid = %d ";
   $cfsql= "SELECT * FROM {chado_feature}";
   $tsql = "SELECT * FROM {feature} F ".
          "  INNER JOIN CVTerm CVT ON F.type_id = CVT.cvterm_id ".
          "WHERE feature_id = %d  AND (";
   $supported_ftypes = split("[ \n]",variable_get('chado_feature_types','EST contig'));
   foreach($supported_ftypes as $ftype){
      $tsql .= " CVT.name = '$ftype' OR ";
   }
   $tsql .= " 0=1) ";  // add a 0=1 just as a filler so we don't have to remove a trailing 'OR'

   // load into nodes array
   $results = db_query($dsql);
   $count = 0;
   $nodes = array();
   while($node = db_fetch_object($results)){
      $nodes[$count] = $node;
      $count++;
   }

   // load the chado_features into an array
   $results = db_query($cfsql);
   $cnodes = array();
   while($node = db_fetch_object($results)){
      $cnodes[$count] = $node;
      $count++;
   }
   $interval = intval($count * 0.01);

   // iterate through all of the chado_feature nodes and delete those  that aren't valid
   foreach($nodes as $nid){

      // update the job status every 1% features
      if($job_id and $i % $interval == 0){
         tripal_job_set_progress($job_id,intval(($i/$count)*100));
      }

      // first check to see if the node has a corresponding entry
      // in the chado_feature table. If not then delete the node.
      $feature = db_fetch_object(db_query($csql,$nid->nid));
      if(!$feature){
         node_delete($nid->nid);
         $message = "Missing in chado_feature table.... DELETING: $nid->nid\n";
         watchdog('tripal_feature',$message,array(),WATCHDOG_WARNING);
         continue;
      }

      // second check to see if the node is for a feature of an allowed type.
      // if not, then delete the node.  This check will also take care of the
      // case when a node exists and an entry in the chado_feature table exists
      // but no feature with a matching feature_id exists
      $previous_db = db_set_active('chado');  // use chado database
      $ftype = db_fetch_object(db_query($tsql,$feature->feature_id));
      db_set_active($previous_db);  // now use drupal database

      if(!$ftype){
         node_delete($nid->nid);
         db_query("DELETE FROM {chado_feature} WHERE feature_id = $feature->feature_id");
         $message = "Node of the wrong feature type.... DELETING: $nid->nid\n";
         watchdog('tripal_feature',$message,array(),WATCHDOG_WARNING);
      }
      $i++;
   }

   // iterate through all of the chado_feature nodes and delete those  that aren't valid
   foreach($cnodes as $nid){
      // update the job status every 1% features
      if($job_id and $i % $interval == 0){
         tripal_job_set_progress($job_id,intval(($i/$count)*100));
      }
      $node = db_fetch_object(db_query($nsql,$nid->nid));
      if(!$node){
         db_query("DELETE FROM {chado_feature} WHERE nid = $nid->nid");
         $message = "chado_feature missing node.... DELETING: $nid->nid\n";
         watchdog('tripal_feature',$message,array(),WATCHDOG_WARNING);
      }

      $i++;
   }
   return '';
}

/************************************************************************
 *
 */
function tripal_feature_bulkload(){
   return drupal_get_form('tripal_feature_load_fasta_form');
}
/************************************************************************
 *
 */
function tripal_feature_load_fasta_form (&$form_state = NULL){

   // get the list of organisms
   $sql = "SELECT * FROM {Organism} ORDER BY genus,species";
   $previous_db = db_set_active('chado');  // use chado database
   $org_rset = db_query($sql);
   db_set_active($previous_db);  // now use drupal database
   $organisms = array();
   $organisms[''] = '';
   while($organism = db_fetch_object($org_rset)){
      $organisms[$organism->organism_id] = "$organism->genus $organism->species ($organism->common_name)";
   }

   // get the list of supported feature types
   $ftypes = array();
   $ftypes[''] = '';
   $supported_ftypes = split("[ \n]",variable_get('chado_feature_feature_types','EST contig'));
   foreach($supported_ftypes as $ftype){
      $ftypes["$ftype"] = $ftype;
   }

   // get the list of libraries
   // TODO !!!! Use Ajax to filter this automatically based on the organism
   // selected by the user. This will prevent mistakes from user input.
   $sql = "SELECT * FROM {Library} L ".
          "  INNER JOIN Organism O ON L.organism_id = O.organism_id ".
          "ORDER BY L.name";
   $previous_db = db_set_active('chado');  // use chado database
   $lib_rset = db_query($sql);
   db_set_active($previous_db);  // now use drupal database
   $libraries = array();
   $libraries[''] = '';
   while($library = db_fetch_object($lib_rset)){
      $libraries[$library->library_id] = "$library->name ($library->genus $library->species)";
   }

   $form['#attributes']['enctype'] = 'multipart/form-data';

   $form['organism'] = array (
      '#title'       => t('Organism'),
      '#type'        => t('select'),
      '#description' => t("Choose the organism with which these sequences are associated "),
      '#required'    => TRUE,
      '#default_vaule' => '',
      '#options'     => $organisms,
      '#weight'      => 1,
   );
   $form['library'] = array (
      '#title'       => t('Library'),
      '#type'        => t('select'),
      '#description' => t("Choose the library with from which these sequences are derived. Leave blank if not applicable."),
      '#required'    => FALSE,
      '#default_vaule' => '',
      '#options'     => $libraries,
      '#weight'      => 2,
   );
   $form['ftype'] = array (
      '#title'       => t('Feature Type'),
      '#type'        => t('select'),
      '#description' => t("Choose the category of sequences you are uploading.  All sequences in the FASTA file will be imported as this type"),
      '#required'    => TRUE,
      '#default_vaule' => '',
      '#options'     => $ftypes,
      '#weight'      => 3,
   );
   $form['fasta_file'] = array(
      '#type'       => t('file'),
      '#title'      => t('Fasta File'),
      '#description'   => t('Upload a FASTA file of sequences. The definition line should contain only the feature name. All other annotations should be removed. The file must not be larger than ' . file_upload_max_size() . ' bytes'),
      '#weight'        => 4,
   );
   $form['upload'] = array(
      '#type' => 'submit',
      '#value' => t('Upload File'),
      '#weight' => 2,
      '#executes_submit_callback' => TRUE,
      '#weight'      => 5,
   );

   return $form;
}
/************************************************************************
 *
 */
function tripal_feature_load_fasta_form_validate($form, &$form_state){

   // TODO !!! check that the fasta file is valid

   global $user;

   // we need a path within the drupal installation to temporarily use as the destination
   // after we upload, we'll move the file to analysis directory
   $upload_url = file_directory_path() . "/chado_feature_bulk_upload/$user->uid";

   // create the download directory. We do it this way rather than the
   // file_check_directory because we don't want a drupal message presented
   // the user when the directory is created.
   if (!is_dir($upload_url)) {
      mkdir($upload_url,0775,TRUE);
   }

   // upload the file and copy it to the proper location
   $validators = array();  // we don't have any validators
   if($file = file_save_upload('fasta_file',$validators,$upload_url)){
      drupal_set_message("File $file->name uploaded succesfully");
   } else {
      form_set_error('fasta_file',t('Upload Failed'));
   }
}
/************************************************************************
 *
 */
function tripal_feature_load_fasta_form_submit($form, &$form_state){
   global $user;

   // add a job to be executed
   tripal_add_job ($job_name,$type,$callback,$uid);
}
/************************************************************************
 *
 */
function tripal_feature_return_fasta($feature,$desc){   
   $fasta  = ">" . variable_get('chado_feature_accession_prefix','ID') . "$feature->feature_id|$feature->name";
   $fasta .= " $desc\n";
   $fasta .= wordwrap($feature->residues, 50, "\n", true);
   $fasta .= "\n\n";
   return $fasta;
}