function get_data_objects_count($batch_size = 100000) { $image_id = DataType::image()->id; $text_id = DataType::text()->id; $video_id = DataType::video()->id; $sound_id = DataType::sound()->id; $flash_id = DataType::flash()->id; $youtube_id = DataType::youtube()->id; $iucn_id = DataType::iucn()->id; $data_type_label[$text_id] = 'text'; $data_type_label[$video_id] = 'video'; $data_type_label[$sound_id] = 'sound'; $data_type_label[$flash_id] = 'flash'; $data_type_label[$youtube_id] = 'youtube'; $data_type_label[$iucn_id] = 'iucn'; $data_type_order_in_file = array("text", "video", "sound", "flash", "youtube", "iucn"); $trusted_id = Vetted::trusted()->id; $untrusted_id = Vetted::untrusted()->id; $unreviewed_id = Vetted::unknown()->id; $raw_stats = array(); $concept_info_items = array(); $concept_references = array(); for ($i = $this->min_taxon_concept_id; $i <= $this->max_taxon_concept_id; $i += $batch_size) { $this->print_status($i, $batch_size); $sql = "SELECT do.guid,\n dotc.taxon_concept_id,\n do.data_type_id,\n doii.info_item_id,\n dor.ref_id,\n REPLACE(REPLACE(do.description, '\\\\n', ' '), '\\\\r', ' '),\n dohe.vetted_id,\n do.id\n FROM data_objects_taxon_concepts dotc\n STRAIGHT_JOIN data_objects do ON (dotc.data_object_id = do.id)\n JOIN data_objects_hierarchy_entries dohe ON (do.id=dohe.data_object_id)\n LEFT JOIN data_objects_info_items doii ON (do.id = doii.data_object_id)\n LEFT JOIN data_objects_refs dor ON (do.id = dor.data_object_id)\n WHERE do.published = 1 AND dohe.visibility_id = " . Visibility::visible()->id . " AND do.data_type_id != {$image_id}"; if ($this->test_taxon_concept_ids) { $sql .= " AND dotc.taxon_concept_id IN (" . $this->test_taxon_concept_ids . ")"; } else { $sql .= " AND dotc.taxon_concept_id BETWEEN {$i} AND " . ($i + $batch_size); } $counted_data_objects = array(); foreach ($this->mysqli_slave->iterate_file($sql) as $row_number => $row) { $taxon_concept_id = trim($row[0]); $data_type_id = trim($row[1]); $info_item_id = trim($row[2]); $ref_id = trim($row[3]); $description = trim($row[4]); $vetted_id = trim($row[5]); $data_object_id = trim($row[6]); if (isset($counted_data_objects[$taxon_concept_id][$data_object_id])) { continue; } $counted_data_objects[$taxon_concept_id][$data_object_id] = 1; $label = @$data_type_label[$data_type_id]; $words_count = str_word_count(strip_tags($description), 0); @$raw_stats[$taxon_concept_id][$label]['total']++; @($raw_stats[$taxon_concept_id][$label]['total_w'] += $words_count); if ($vetted_id == $trusted_id) { @$raw_stats[$taxon_concept_id][$label]['t']++; @($raw_stats[$taxon_concept_id][$label]['t_w'] += $words_count); } elseif ($vetted_id == $untrusted_id) { @$raw_stats[$taxon_concept_id][$label]['ut']++; @($raw_stats[$taxon_concept_id][$label]['ut_w'] += $words_count); } elseif ($vetted_id == $unreviewed_id) { @$raw_stats[$taxon_concept_id][$label]['ur']++; @($raw_stats[$taxon_concept_id][$label]['ur_w'] += $words_count); } $concept_info_items[$taxon_concept_id][$info_item_id] = ''; $concept_references[$taxon_concept_id][$ref_id] = ''; } foreach ($raw_stats as $taxon_concept_id => $stats) { $new_value = ""; # the stats need to go into the file in a certain order to be imported into the MySQL table foreach ($data_type_order_in_file as $data_type) { $new_value = @$stats[$data_type]['total']; $new_value .= "\t" . @$stats[$data_type]['t']; $new_value .= "\t" . @$stats[$data_type]['ut']; $new_value .= "\t" . @$stats[$data_type]['ur']; $new_value .= "\t" . @$stats[$data_type]['total_w']; $new_value .= "\t" . @$stats[$data_type]['t_w']; $new_value .= "\t" . @$stats[$data_type]['ut_w']; $new_value .= "\t" . @$stats[$data_type]['ur_w']; } $raw_stats[$taxon_concept_id] = $new_value; } $this->save_category_stats($raw_stats, "get_data_objects_count"); $raw_stats = array(); if ($this->test_taxon_concept_ids) { break; } } // $this->save_to_json_file($concept_info_items, "concept_info_items"); // unset($concept_info_items); // // $this->save_to_json_file($concept_references, "concept_references"); // unset($concept_references); }
public static function read_taxon_xml($t, $resource) { $t_dc = $t->children("http://purl.org/dc/elements/1.1/"); $t_dcterms = $t->children("http://purl.org/dc/terms/"); $t_dwc = $t->children("http://rs.tdwg.org/dwc/dwcore/"); $taxon_parameters = array(); $taxon_parameters["identifier"] = Functions::import_decode($t_dc->identifier); $taxon_parameters["source_url"] = Functions::import_decode($t_dc->source); $taxon_parameters["kingdom"] = Functions::import_decode($t_dwc->Kingdom); $taxon_parameters["phylum"] = Functions::import_decode($t_dwc->Phylum); $taxon_parameters["class"] = Functions::import_decode($t_dwc->Class); $taxon_parameters["order"] = Functions::import_decode($t_dwc->Order); $taxon_parameters["family"] = Functions::import_decode($t_dwc->Family); $taxon_parameters["genus"] = Functions::import_decode($t_dwc->Genus); $taxon_parameters["scientific_name"] = Functions::import_decode($t_dwc->ScientificName); $taxon_parameters["rank"] = Rank::find_or_create_by_translated_label(Functions::import_decode($t->rank)); $taxon_parameters["taxon_created_at"] = trim($t_dcterms->created); $taxon_parameters["taxon_modified_at"] = trim($t_dcterms->modified); if ($taxon_parameters["scientific_name"]) { $taxon_parameters["name"] = Name::find_or_create_by_string($taxon_parameters["scientific_name"]); } else { if ($name = $taxon_parameters["genus"]) { $taxon_parameters["scientific_name"] = $name; $taxon_parameters["name"] = Name::find_or_create_by_string($name); $taxon_parameters["genus"] = ""; } elseif ($name = $taxon_parameters["family"]) { $taxon_parameters["scientific_name"] = $name; $taxon_parameters["name"] = Name::find_or_create_by_string($name); $taxon_parameters["family"] = ""; } elseif ($name = $taxon_parameters["order"]) { $taxon_parameters["scientific_name"] = $name; $taxon_parameters["name"] = Name::find_or_create_by_string($name); $taxon_parameters["order"] = ""; } elseif ($name = $taxon_parameters["class"]) { $taxon_parameters["scientific_name"] = $name; $taxon_parameters["name"] = Name::find_or_create_by_string($name); $taxon_parameters["class"] = ""; } elseif ($name = $taxon_parameters["phylum"]) { $taxon_parameters["scientific_name"] = $name; $taxon_parameters["name"] = Name::find_or_create_by_string($name); $taxon_parameters["phylum"] = ""; } elseif ($name = $taxon_parameters["kingdom"]) { $taxon_parameters["scientific_name"] = $name; $taxon_parameters["name"] = Name::find_or_create_by_string($name); $taxon_parameters["kingdom"] = ""; } else { return; } } $taxon_parameters["common_names"] = array(); foreach ($t->commonName as $c) { $common_name = Functions::import_decode((string) $c); if (!$common_name) { continue; } $xml_attr = $c->attributes("http://www.w3.org/XML/1998/namespace"); $params = array("name" => $common_name, "language" => Language::find_or_create_for_parser(@Functions::import_decode($xml_attr["lang"]))); $taxon_parameters["common_names"][] = $params; } $taxon_parameters["synonyms"] = array(); foreach ($t->synonym as $s) { $synonym = Functions::import_decode((string) $s); if (!$synonym) { continue; } $attr = $s->attributes(); if (!@$attr["relationship"]) { $attr["relationship"] = 'synonym'; } $params = array("name" => Name::find_or_create_by_string($synonym), "synonym_relation" => SynonymRelation::find_or_create_by_translated_label(trim($attr["relationship"]))); $taxon_parameters["synonyms"][] = $params; } $taxon_parameters["agents"] = array(); foreach ($t->agent as $a) { $agent_name = Functions::import_decode((string) $a); if (!$agent_name) { continue; } $attr = $a->attributes(); $params = array("full_name" => Functions::import_decode((string) $a, 0, 0), "homepage" => @Functions::import_decode($attr["homepage"]), "logo_url" => @Functions::import_decode($attr["logoURL"]), "agent_role" => AgentRole::find_or_create_by_translated_label(@trim($attr["role"]))); $taxon_parameters["agents"][] = $params; unset($params); } $taxon_parameters["refs"] = array(); foreach ($t->reference as $r) { $reference = Functions::import_decode((string) $r, 0, 0); if (!$reference) { continue; } $ref = Reference::find_or_create_by_full_reference($reference); $taxon_parameters["refs"][] = $ref; $id_labels = array("bici", "coden", "doi", "eissn", "handle", "issn", "isbn", "lsid", "oclc", "sici", "url", "urn"); $attr = $r->attributes(); foreach ($id_labels as $label) { if ($id = @Functions::import_decode($attr[$label], 0, 0)) { $type = RefIdentifierType::find_or_create_by_label($label); $ref->add_ref_identifier(@$type->id ?: 0, $id); } } } $taxon_parameters["data_objects"] = array(); foreach ($t->dataObject as $d) { $d_dc = $d->children("http://purl.org/dc/elements/1.1/"); $d_dcterms = $d->children("http://purl.org/dc/terms/"); $d_geo = $d->children("http://www.w3.org/2003/01/geo/wgs84_pos#"); $data_object = new DataObject(); $data_object->identifier = Functions::import_decode($d_dc->identifier); $data_object->data_type = DataType::find_or_create_by_schema_value(Functions::import_decode($d->dataType)); $data_object->mime_type = MimeType::find_or_create_by_translated_label(Functions::import_decode($d->mimeType)); $data_object->object_created_at = Functions::import_decode($d_dcterms->created); $data_object->object_modified_at = Functions::import_decode($d_dcterms->modified); $data_object->object_title = Functions::import_decode($d_dc->title, 0, 0); $data_object->language = Language::find_or_create_for_parser(Functions::import_decode($d_dc->language)); $data_object->license = License::find_or_create_for_parser(Functions::import_decode($d->license)); $data_object->rights_statement = Functions::import_decode($d_dc->rights, 0, 0); $data_object->rights_holder = Functions::import_decode($d_dcterms->rightsHolder, 0, 0); $data_object->bibliographic_citation = Functions::import_decode($d_dcterms->bibliographicCitation, 0, 0); $data_object->source_url = Functions::import_decode($d_dc->source); $data_object->description = Functions::import_decode($d_dc->description, 0, 0); $data_object->object_url = Functions::import_decode($d->mediaURL); $data_object->thumbnail_url = Functions::import_decode($d->thumbnailURL); $data_object->location = Functions::import_decode($d->location, 0, 0); if (@$d->additionalInformation) { $data_object->additional_information = (array) $d->additionalInformation; } if ($r = (string) @$d->additionalInformation->rating) { if (is_numeric($r) && $r > 0 && $r <= 5) { $data_object->data_rating = $r; } } if ($subtype = @$d->additionalInformation->subtype) { if ($dt = DataType::find_or_create_by_schema_value(Functions::import_decode($subtype))) { $data_object->data_subtype_id = $dt->id; } } $data_object_parameters = array(); if (!$data_object->language) { $xml_attr = $d_dc->description->attributes("http://www.w3.org/XML/1998/namespace"); $data_object->language = Language::find_or_create_for_parser(@Functions::import_decode($xml_attr["lang"])); } if (!$data_object->language && $resource->language) { $data_object->language = $resource->language; } //TODO - update this if ($data_object->mime_type && $data_object->mime_type->equals(MimeType::flash()) && $data_object->is_video()) { $data_object->data_type = DataType::youtube(); $data_object->data_type_id = DataType::youtube()->id; } //take the taxon's source_url if none present if (!@$data_object->source_url && @$taxon_parameters["source_url"]) { $data_object->source_url = $taxon_parameters["source_url"]; } // Turn newlines into paragraphs $data_object->description = str_replace("\n", "</p><p>", $data_object->description); /* Checking requirements*/ //if text: must have description if ($data_object->data_type->equals(DataType::text()) && !$data_object->description) { continue; } //if image, movie or sound: must have object_url if (($data_object->data_type->equals(DataType::video()) || $data_object->data_type->equals(DataType::sound()) || $data_object->data_type->equals(DataType::image())) && !$data_object->object_url) { continue; } $data_object->latitude = 0; $data_object->longitude = 0; $data_object->altitude = 0; foreach ($d_geo->Point as $p) { $p_geo = $p->children("http://www.w3.org/2003/01/geo/wgs84_pos#"); $data_object->latitude = Functions::import_decode($p_geo->lat); $data_object->longitude = Functions::import_decode($p_geo->long); $data_object->altitude = Functions::import_decode($p_geo->alt); } $data_object_parameters["agents"] = array(); foreach ($d->agent as $a) { $agent_name = Functions::import_decode((string) $a); if (!$agent_name) { continue; } $attr = $a->attributes(); $params = array("full_name" => Functions::import_decode((string) $a, 0, 0), "homepage" => @Functions::import_decode($attr["homepage"]), "logo_url" => @Functions::import_decode($attr["logoURL"]), "agent_role" => AgentRole::find_or_create_by_translated_label(@trim($attr["role"]))); $data_object_parameters["agents"][] = $params; unset($params); } $data_object_parameters["audiences"] = array(); foreach ($d->audience as $a) { $data_object_parameters["audiences"][] = Audience::find_or_create_by_translated_label(trim((string) $a)); } $data_object_parameters["info_items"] = array(); foreach ($d->subject as $s) { $data_object_parameters["info_items"][] = InfoItem::find_or_create_by_schema_value(trim((string) $s)); } if ($subject = @$d->additionalInformation->subject) { if ($ii = InfoItem::find_or_create_by_schema_value(trim((string) $subject))) { $data_object_parameters["info_items"] = array($ii); } } // EXCEPTIONS if ($data_object->is_text()) { if ($resource->title == "BOLD Systems Resource") { // EXCEPTION - overriding the subject for BOLD $data_object_parameters["info_items"] = array(InfoItem::find_or_create_by_schema_value('http://www.eol.org/voc/table_of_contents#Barcode')); } elseif ($resource->title == "Wikipedia") { // EXCEPTION - overriding the subject for Wikipedia $data_object_parameters["info_items"] = array(InfoItem::find_or_create_by_schema_value('http://www.eol.org/voc/table_of_contents#Wikipedia')); } elseif ($resource->title == "IUCN Red List") { if ($data_object->object_title == "IUCNConservationStatus") { // EXCEPTION - overriding the data type for IUCN text $data_object->data_type_id = DataType::iucn()->id; $data_object->data_type = DataType::iucn(); } } } $data_object_parameters["refs"] = array(); foreach ($d->reference as $r) { $reference = Functions::import_decode((string) $r, 0, 0); if (!$reference) { continue; } $ref = Reference::find_or_create_by_full_reference($reference); $data_object_parameters["refs"][] = $ref; $id_labels = array("bici", "coden", "doi", "eissn", "handle", "issn", "isbn", "lsid", "oclc", "sici", "url", "urn"); $attr = $r->attributes(); foreach ($id_labels as $label) { if ($id = @Functions::import_decode($attr[$label], 0, 0)) { $type = RefIdentifierType::find_or_create_by_label($label); $ref->add_ref_identifier(@$type->id ?: 0, $id); } } } $taxon_parameters["data_objects"][] = array($data_object, $data_object_parameters); unset($data_object); } return $taxon_parameters; }
function get_data_objects_count($batch_size = 100000) { $time_start = time_elapsed(); $concept_data_object_counts = array(); $concept_data_object_maps = array(); $concept_info_items = array(); $concept_references = array(); $image_id = DataType::image()->id; $map_id = DataType::map()->id; $text_id = DataType::text()->id; $video_id = DataType::video()->id; $sound_id = DataType::sound()->id; $flash_id = DataType::flash()->id; $youtube_id = DataType::youtube()->id; $iucn_id = DataType::iucn()->id; $data_type_label[$image_id] = 'image'; $data_type_label[$sound_id] = 'sound'; $data_type_label[$text_id] = 'text'; $data_type_label[$video_id] = 'video'; $data_type_label[$iucn_id] = 'iucn'; $data_type_label[$flash_id] = 'flash'; $data_type_label[$youtube_id] = 'youtube'; $trusted_id = Vetted::trusted()->id; $untrusted_id = Vetted::untrusted()->id; $unreviewed_id = Vetted::unknown()->id; for ($i = $this->min_taxon_concept_id; $i <= $this->max_taxon_concept_id; $i += $batch_size) { print "\n dataObjects, its infoItems, its references [2 of 14] {$i} \n"; $sql = "SELECT dotc.taxon_concept_id tc_id, do.data_type_id, doii.info_item_id, dor.ref_id, do.description, dohe.vetted_id, do.data_subtype_id\r\n FROM data_objects_taxon_concepts dotc \r\n JOIN data_objects do ON dotc.data_object_id = do.id \r\n LEFT JOIN data_objects_info_items doii ON do.id = doii.data_object_id \r\n LEFT JOIN data_objects_refs dor ON do.id = dor.data_object_id \r\n JOIN data_objects_hierarchy_entries dohe on do.id = dohe.data_object_id\r\n WHERE do.published=1 AND dohe.visibility_id=" . Visibility::visible()->id . " AND dohe.vetted_id != {$untrusted_id} "; //." AND do.data_type_id <> $image_id "; this has to be removed to count maps if (isset($GLOBALS['test_taxon_concept_ids'])) { $sql .= " and dotc.taxon_concept_id IN (" . implode(",", $GLOBALS['test_taxon_concept_ids']) . ")"; } else { $sql .= " AND dotc.taxon_concept_id BETWEEN {$i} AND " . ($i + $batch_size); } $sql .= "\r\n UNION\r\n SELECT dotc.taxon_concept_id tc_id, do.data_type_id, doii.info_item_id, dor.ref_id, do.description, udo.vetted_id, do.data_subtype_id\r\n FROM data_objects_taxon_concepts dotc \r\n JOIN data_objects do ON dotc.data_object_id = do.id \r\n LEFT JOIN data_objects_info_items doii ON do.id = doii.data_object_id \r\n LEFT JOIN data_objects_refs dor ON do.id = dor.data_object_id \r\n JOIN users_data_objects udo on do.id = udo.data_object_id\r\n WHERE do.published=1 AND udo.visibility_id=" . Visibility::visible()->id . "\r\n "; if (isset($GLOBALS['test_taxon_concept_ids'])) { $sql .= " and dotc.taxon_concept_id IN (" . implode(",", $GLOBALS['test_taxon_concept_ids']) . ")"; } else { $sql .= " AND dotc.taxon_concept_id BETWEEN {$i} AND " . ($i + $batch_size); } $outfile = $this->mysqli_slave->select_into_outfile($sql); $FILE = fopen($outfile, "r"); if (!$FILE) { print "!! ERROR: Could not read {$outfile}"; debug("!! ERROR: Could not read {$outfile}"); return; } $num_rows = 0; while (!feof($FILE)) { if ($line = fgets($FILE)) { $num_rows++; $line = trim($line); $fields = explode("\t", $line); $tc_id = trim($fields[0]); $data_type_id = trim($fields[1]); $info_item_id = trim($fields[2]); $ref_id = trim($fields[3]); $description = trim($fields[4]); $vetted_id = trim($fields[5]); $data_subtype_id = trim($fields[6]); $label = @$data_type_label[$data_type_id]; if ($data_subtype_id != $map_id) { $words_count = str_word_count(strip_tags($description), 0); @$concept_data_object_counts[$tc_id][$label]['total']++; @($concept_data_object_counts[$tc_id][$label]['total_w'] += $words_count); if ($vetted_id == $trusted_id) { @$concept_data_object_counts[$tc_id][$label]['t']++; @($concept_data_object_counts[$tc_id][$label]['t_w'] += $words_count); } elseif ($vetted_id == $untrusted_id) { @$concept_data_object_counts[$tc_id][$label]['ut']++; @($concept_data_object_counts[$tc_id][$label]['ut_w'] += $words_count); } elseif ($vetted_id == $unreviewed_id) { @$concept_data_object_counts[$tc_id][$label]['ur']++; @($concept_data_object_counts[$tc_id][$label]['ur_w'] += $words_count); } $concept_info_items[$tc_id][$info_item_id] = ''; $concept_references[$tc_id][$ref_id] = ''; } else { @$concept_data_object_maps[$tc_id][$label]['total']++; if ($vetted_id == $trusted_id) { @$concept_data_object_maps[$tc_id][$label]['t']++; } elseif ($vetted_id == $untrusted_id) { @$concept_data_object_maps[$tc_id][$label]['ut']++; } elseif ($vetted_id == $unreviewed_id) { @$concept_data_object_maps[$tc_id][$label]['ur']++; } } } } fclose($FILE); unlink($outfile); print "\n num_rows: {$num_rows}"; } self::save_to_json_file($concept_info_items, "concept_info_items"); unset($concept_info_items); self::save_to_json_file($concept_references, "concept_references"); unset($concept_references); //save map data to be accessed later self::save_to_json_file($concept_data_object_maps, "map_counts"); unset($concept_data_object_maps); //convert associative array to a regular array $data_type_order_in_file = array("text", "video", "sound", "flash", "youtube", "iucn"); foreach ($concept_data_object_counts as $taxon_concept_id => $taxon_object_counts) { $new_value = ""; foreach ($data_type_order_in_file as $data_type) { $new_value .= "\t" . @$taxon_object_counts[$data_type]['total']; $new_value .= "\t" . @$taxon_object_counts[$data_type]['t']; $new_value .= "\t" . @$taxon_object_counts[$data_type]['ut']; $new_value .= "\t" . @$taxon_object_counts[$data_type]['ur']; $new_value .= "\t" . @$taxon_object_counts[$data_type]['total_w']; $new_value .= "\t" . @$taxon_object_counts[$data_type]['t_w']; $new_value .= "\t" . @$taxon_object_counts[$data_type]['ut_w']; $new_value .= "\t" . @$taxon_object_counts[$data_type]['ur_w']; } $concept_data_object_counts[$taxon_concept_id] = $new_value; } print "\n get_data_objects_count():" . (time_elapsed() - $time_start) / 60 . " minutes"; self::save_totals_to_cumulative_txt($concept_data_object_counts, "tpm_data_objects"); unset($concept_data_object_counts); }