123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390391392393394395396397398399400401402403404405406407408409410411412413414415416417418419420421422423424425426427428429430431432433434435436437438439440441442443444445446447448449450451452453454455456457458459460461462463464465466467468469470471472473474475476477478479480481482483484485486487488489490491492493494495496497498499500501502503504 |
- <?php
- abstract class TripalFieldDownloader {
- /**
- * Sets the label shown to the user describing this formatter. It
- * should be a short identifier. Use the $full_label for a more
- * descriptive label.
- */
- static public $label = 'Generic';
- /**
- * A more verbose label that better describes the formatter.
- */
- static public $full_label = 'Generic File format';
- /**
- * Indicates the default extension for the outputfile.
- */
- static public $default_extension = 'txt';
- /**
- * The data collection assigned to this downloader.
- */
- protected $collection = NULL;
- /**
- * The collection ID
- */
- protected $collection_id = NULL;
- /**
- * An array of collection_bundle records for the content types that
- * belong to this data collection.
- */
- protected $collection_bundles = NULL;
- /**
- * The output file URI.
- */
- protected $outfile = '';
- /**
- * An array of printable fields. Because fields can come from multiple
- * bundles and those bundles can be from multiple sites, it is possible that
- * 1) two bundles use the same field and we want to conslidate to a
- * single printable field; and 2) that a remote site may use the same term
- * for a field as a bundle on the local site. The only way to sort out this
- * mess is to use the term accession numbers. Therefore, the array contains
- * a unique list of printable fields using their accession numbers as keys
- * and a field label as the value.
- *
- */
- protected $printable_fields = array();
- /**
- * The remote site json data returned for the entity
- */
- protected $remote_entity = '';
- /**
- * An array that associates a field ID with a term.
- *
- * The first-level key is the site ID. For the local site this will be
- * the word 'local' for all others it will be the numeric id. The second
- * level key is the bundle bundle name. For local bundles this will
- * always be bio_data_xxxx. Third, are two subkeys: by_field and
- * by_accession. To lookup a field's term you use the 'by_field' subkey
- * with the field_id as the next level. To lookup the field ID for a term
- * use the 'by_accession' subkey with the accession as the next level. Below
- * is an example of the structure of this array.
- *
- * @code
- Array (
- [local] => Array(
- [bio_data_7] => Array(
- [by_field] => Array(
- [56] => data:2091,
- [57] => OBI:0100026,
- [17] => schema:name,
- [58] => data:2044,
- [34] => data:0842,
- [67] => schema:alternateName,
- ),
- [by_accession] => Array (
- [data:2091] => 56,
- [OBI:0100026] => 57,
- [schema:name] => 17,
- [data:2044] => 58,
- [data:0842] => 34,
- [schema:alternateName] => 67,
- ),
- ),
- ),
- )
- * @endcode
- */
- protected $fields2terms = array();
- /**
- * A list of field and instance items, indexed first by site_id with 'local'
- * being the key for local fields and the numeric site_id for remote
- * fields. The second-levle key is the bundle_name and the the field_id.
- * Below the field_id are the keys 'field' or 'instance' where the
- * value of each is the field or instance details respectively.
- */
- protected $fields = array();
- /**
- * The file handle for an opeend file using during writing.
- */
- protected $fh;
- /**
- * Constructs a new instance of the TripalFieldDownloader class.
- *
- * @param $collection_id
- * The ID for the collection.
- * @param $outfile_name
- * The name of the output file to create. The name should not include
- * a path.
- */
- public function __construct($collection_id, $outfile_name) {
- if (!$outfile_name) {
- throw new Exception("Please provide an outputfilename");
- }
- // Get the collection record and store it.
- $collection = db_select('tripal_collection', 'tc')
- ->fields('tc')
- ->condition('collection_id', $collection_id, '=')
- ->execute()
- ->fetchObject();
- if (!$collection) {
- throw new Exception(t("Cannot find a collection that matches the provided id: !id", array('!id' => $collection_id)));
- }
- $this->collection = $collection;
- // Make sure the user directory exists
- $user = user_load($this->collection->uid);
- $user_dir = tripal_get_user_files_dir($user);
- if (!tripal_is_user_files_dir_writeable($user)) {
- throw new Exception(t("The user's data directory is not writeable: !user_dir.", array('!user_dir' => $user_dir)));
- }
- // Set the collection ID of the collection that this downloader will use.
- $this->collection_id = $collection_id;
-
- // Make sure the data_collections directory exists.
- $collections_dir = $user_dir . '/data_collections';
- if (!file_exists($collections_dir)) {
- if (!file_prepare_directory($collections_dir, [FILE_CREATE_DIRECTORY])) {
- throw new Exception(t("The data_collection directory cannot be created: !collections_dir.", array('!collections_dir' => $collections_dir)));
- }
- }
- $this->outfile = $collections_dir . '/' . $outfile_name;
- // A data collection may have multiple bundles. We'll need to get
- // them all and store them.
- $collection_bundles = db_select('tripal_collection_bundle')
- ->fields('tripal_collection_bundle')
- ->condition('collection_id', $collection_id, '=')
- ->execute();
- while ($collection_bundle = $collection_bundles->fetchObject()) {
- $collection_bundle->ids = unserialize($collection_bundle->ids);
- $collection_bundle->fields = unserialize($collection_bundle->fields);
- $this->collection_bundles[] = $collection_bundle;
- }
-
- // Map the fields to their term accessions.
- $this->setFields();
- $this->setFields2Terms();
- $this->setPrintableFields();
- }
- /**
- * Inidcates if a given field is supported by this Downloader class.
- *
- * @param $field
- * A field info array.
- */
- public function isFieldSupported($field, $instance) {
- $field_name = $field['field_name'];
- $field_type = $field['type'];
- $this_formatter = get_class($this);
- // If a field is a TripalField then check its supported downloaders.
- if (tripal_load_include_field_class($field_type)) {
- $formatters = $field_type::$download_formatters;
- if (in_array($this_formatter, $formatters)) {
- return TRUE;
- }
- }
- // If this is a remote field then check that differently.
- if ($field['storage']['type'] == 'tripal_remote_field' ) {
- if (in_array($this_formatter, $instance['formatters'])) {
- return TRUE;
- }
- }
- }
- /**
- * Retrieves the URL for the downloadable file.
- */
- public function getURL() {
- return $this->outfile;
- }
- /**
- * Removes the downloadable file.
- */
- public function delete() {
- $fid = db_select('file_managed', 'fm')
- ->fields('fm', array('fid'))
- ->condition('uri', $this->outfile)
- ->execute()
- ->fetchField();
- if ($fid) {
- $file = file_load($fid);
- file_usage_delete($file, 'tripal', 'data-collection');
- file_delete($file, TRUE);
- }
- }
- /**
- *
- * @param TripalJob $job
- */
- public function writeInit(TripalJob $job = NULL) {
- $user = user_load($this->collection->uid);
- $this->fh = fopen(drupal_realpath($this->outfile), "w");
- if (!$this->fh) {
- throw new Exception("Cannout open collection file: " . $this->outfile);
- }
- // Add the headers to the file.
- $headers = $this->getHeader();
- if ($headers) {
- foreach ($headers as $line) {
- fwrite($this->fh, $line . "\r\n");
- }
- }
- }
- /**
- * Write a single entity to the file.
- *
- * Before calling this function call the initWrite() function to
- * establish the file and write headers.
- *
- * @param $entity
- * The Entity to write.
- * @param TripalJob $job
- */
- public function writeEntity($entity, TripalJob $job = NULL){
- $lines = $this->formatEntity($entity);
- foreach ($lines as $line) {
- fwrite($this->fh, $line . "\r\n");
- }
- }
- /**
- * Closes the output file once writing of all entities is completed.
- *
- * @param TripalJob $job
- */
- public function writeDone(TripalJob $job = NULL) {
- fclose($this->fh);
- $user = user_load($this->collection->uid);
- $file = new stdClass();
- $file->uri = $this->outfile;
- $file->filename = basename($this->outfile);
- $file->filemime = file_get_mimetype($this->outfile);
- $file->uid = $user->uid;
- $file->status = FILE_STATUS_PERMANENT;
- // Check if this file already exists. If it does then just update
- // the stats.
- $fid = db_select('file_managed', 'fm')
- ->fields('fm', array('fid'))
- ->condition('uri', $this->outfile)
- ->execute()
- ->fetchField();
- if ($fid) {
- $file->fid = $fid;
- $file = file_save($file);
- }
- else {
- $file = file_save($file);
- $fid = $file->fid;
- $file = file_load($fid);
- }
-
- // Associate this file with data collections if it isn't already
- $usage = file_usage_list($file);
- if (!array_key_exists('tripal', $usage) or
- !array_key_exists('data_collection', $usage['tripal'])) {
- file_usage_add($file, 'tripal', 'data_collection', $this->collection_id);
- }
- }
- /**
- * Creates the downloadable file.
- *
- * @param $job
- * If this function is run as a Tripal Job then this argument can be
- * set to the Tripaljob object for keeping track of progress.
- */
- public function write(TripalJob $job = NULL) {
- $this->initWrite($job);
- $num_handled = 0;
- foreach ($this->collection_bundles as $bundle_collection) {
- $collection_bundle_id = $bundle_collection->collection_bundle_id;
- $bundle_name = $bundle_collection->bundle_name;
- $entity_ids = $bundle_collection->ids;
- $fields = $bundle_collection->fields;
- $site_id = $bundle_collection->site_id;
- foreach ($entity_ids as $entity_id) {
- $num_handled++;
- if ($job) {
- $job->setItemsHandled($num_handled);
- }
- // if we have a site_id then we need to get the entity from the
- // remote service. Otherwise create the entity from the local system.
- if ($site_id) {
- $entity = $this->loadRemoteEntity($entity_id, $site_id, $bundle_name);
- if (!$entity) {
- continue;
- }
- }
- else {
- $result = tripal_load_entity('TripalEntity', array($entity_id), FALSE, $fields, FALSE);
- $entity = $result[$entity_id];
- }
- if (!$entity) {
- continue;
- }
- $lines = $this->formatEntity($entity);
- foreach ($lines as $line) {
- fwrite($this->fh, $line . "\r\n");
- }
- }
- }
- $this->finishWrite($job);
- }
- /**
- * Setups a download stream for the file.
- */
- public function download() {
- }
- /**
- * A helper function for the setFields() function.
- *
- * Adds local fields to the list of fields.
- */
- private function setLocalFields() {
- foreach ($this->collection_bundles as $collection_bundle) {
- $bundle_name = $collection_bundle->bundle_name;
- if ($collection_bundle->site_id) {
- continue;
- }
- foreach ($collection_bundle->fields as $field_id) {
- $field = field_info_field_by_id($field_id);
- $instance = field_info_instance('TripalEntity', $field['field_name'], $bundle_name);
- $this->fields['local'][$bundle_name][$field_id]['field'] = $field;
- $this->fields['local'][$bundle_name][$field_id]['instance'] = $instance;
- }
- }
- }
- /**
- * A helper function for the setFields() function.
- *
- * Adds remote fields to the list of fields.
- */
- private function setRemoteFields() {
- // We can't use the Tripal ws API extensions if the
- // tripal_ws module is not enabled.
- if (!module_exists('tripal_ws')) {
- return;
- }
- foreach ($this->collection_bundles as $collection_bundle) {
- $bundle_name = $collection_bundle->bundle_name;
- $site_id = $collection_bundle->site_id;
- // Skip local fields.
- if (!$site_id) {
- continue;
- }
- // Iterate through the fields of this collection and get the
- // info for each one from the class. We will create "fake" field and
- // instance info arrays.
- foreach ($collection_bundle->fields as $field_id) {
- $field = tripal_get_remote_field_info($site_id, $bundle_name, $field_id);
- $instance = tripal_get_remote_field_instance_info($site_id, $bundle_name, $field_id);
- $this->fields[$site_id][$bundle_name][$field_id]['field'] = $field;
- $this->fields[$site_id][$bundle_name][$field_id]['instance'] = $instance;
- }
- }
- }
- /**
- * Sets the fields array
- */
- protected function setFields() {
- $this->setLocalFields();
- $this->setRemoteFields();
- }
- /**
- * Sets the fields2term array.
- *
- * The fields2term array provides an easy lookup for mapping a term
- * to it's accession number.
- **/
- protected function setFields2Terms() {
- foreach ($this->fields as $site => $bundles) {
- foreach ($bundles as $bundle_name => $bundle_fields) {
- foreach ($bundle_fields as $field_id => $info) {
- $instance = $info['instance'];
- $accession = $instance['settings']['term_vocabulary'] . ':' . $instance['settings']['term_accession'];
- $this->fields2terms[$site][$bundle_name]['by_field'][$field_id] = $accession;
- $this->fields2terms[$site][$bundle_name]['by_accession'][$accession] = $field_id;
- }
- }
- }
- }
- /**
- * Conslidates all the fields into a single list of accession numbers.
- *
- * The array of printable fields will contain an array containing the
- * accession number and the label. The title used is from the first
- * occurance of an accession.
- */
- protected function setPrintableFields() {
- foreach ($this->fields as $site => $bundles) {
- foreach ($bundles as $bundle_name => $bundle_fields) {
- foreach ($bundle_fields as $field_id => $info) {
- $field = $info['field'];
- $instance = $info['instance'];
- $accession = $instance['settings']['term_vocabulary'] . ':' . $instance['settings']['term_accession'];
- if (!array_key_exists($accession, $this->printable_fields)) {
- // Only include fields that support this downloader type in
- // or list of printable fields.
- if ($this->isFieldSupported($field, $instance)) {
- $this->printable_fields[$accession] = $instance['label'];
- }
- }
- }
- }
- }
- }
- /**
- * Formats the entity and the specified fields for output.
- *
- * This function should be implemented by a child class. It should iterate
- * over the fields for the entity and return the appropriate format. It may
- * return multiple lines of output if necessary.
- *
- * @param $entity
- * The entity object. The fields that should be formatted are already
- * loaded.
- *
- * @return
- * An array of strings (one per line of output.
- */
- abstract protected function formatEntity($entity);
- /**
- * Retrieves header lines
- *
- * This function should be implemented by a child class. It should return
- * the header lines for an output file.
- */
- abstract protected function getHeader();
- }
|