/**
* Update a record table for a set of harvesters and repositories
*
* @param string $table
* @param array $allowedRepositoryIdStrings
* @param array $authGroupIdStrings The ids of the groups to check view authorization for,
* May be one of the following or another group Id string:
* edu.middlebury.agents.everyone
* edu.middlebury.agents.all_agents
* If empty, all assets in the specified repositories will be added regardless of
* their visibility.
*
* @return void
* @access public
* @since 3/9/07
*/
function updateTable($table, $allowedRepositoryIdStrings, $authGroupIdStrings)
{
ArgumentValidator::validate($table, StringValidatorRule::getRule());
ArgumentValidator::validate($allowedRepositoryIdStrings, ArrayValidatorRuleWithRule::getRule(StringValidatorRule::getRule()));
ArgumentValidator::validate($authGroupIdStrings, ArrayValidatorRuleWithRule::getRule(StringValidatorRule::getRule()));
$harmoni = Harmoni::instance();
$config = $harmoni->getAttachedData('OAI_CONFIG');
$repositoryManager = Services::getService('Repository');
$authorizationManager = Services::getService('AuthZ');
$idManager = Services::getService("IdManager");
$dbc = Services::getService("DatabaseManager");
$authGroupIds = array();
foreach ($authGroupIdStrings as $id) {
$authGroupIds[] = $idManager->getId($id);
}
$baseCheckQuery = new SelectQuery();
$baseCheckQuery->addTable('oai_' . $table);
$baseCheckQuery->addColumn('datestamp');
$baseCheckQuery->addColumn('deleted');
$baseUpdateQuery = new UpdateQuery();
$baseUpdateQuery->setTable('oai_' . $table);
$baseUpdateColumns = array('datestamp', 'deleted', 'oai_set', 'dc_title', 'dc_description');
$dcUpdateColumns = array('datestamp', 'deleted', 'oai_set', 'dc_title', 'dc_description', 'dc_creator', 'dc_subject', 'dc_contributor', 'dc_publisher', 'dc_date', 'dc_type', 'dc_format', 'dc_identifier', 'dc_source', 'dc_language', 'dc_relation', 'dc_coverage', 'dc_rights');
$baseInsertQuery = new InsertQuery();
$baseInsertQuery->setTable('oai_' . $table);
$baseInsertColumns = array('datestamp', 'oai_identifier', 'deleted', 'oai_set', 'dc_title', 'dc_description');
$dcInsertColumns = array('datestamp', 'oai_identifier', 'deleted', 'oai_set', 'dc_title', 'dc_description', 'dc_creator', 'dc_subject', 'dc_contributor', 'dc_publisher', 'dc_date', 'dc_type', 'dc_format', 'dc_identifier', 'dc_source', 'dc_language', 'dc_relation', 'dc_coverage', 'dc_rights');
$baseDeleteQuery = new UpdateQuery();
$baseDeleteQuery->setTable('oai_' . $table);
$baseDeleteQuery->addValue('deleted', 'true');
$baseDeleteQuery->addRawValue('datestamp', 'NOW()');
$baseUndeleteQuery = new UpdateQuery();
$baseUndeleteQuery->setTable('oai_' . $table);
$baseUndeleteQuery->addValue('deleted', 'false');
$baseUndeleteQuery->addRawValue('datestamp', 'NOW()');
$forceUpdate = false;
$repositories = $repositoryManager->getRepositories();
$r = 0;
if (count($allowedRepositoryIdStrings)) {
$numR = count($allowedRepositoryIdStrings);
} else {
$numR = $repositories->count();
}
$numUpdates = 0;
$numDeleted = 0;
$message = _('Updating OAI records for repository (%1 of %2) : ');
$message = str_replace('%2', $numR, $message);
$instituteId = $idManager->getId('edu.middlebury.agents.users');
$viewId = $idManager->getId('edu.middlebury.authorization.view');
require_once HARMONI . "/utilities/Timer.class.php";
$timer = new Timer();
$timer->start();
$existingRepositoryIds = array();
while ($repositories->hasNext()) {
$updatesInRepository = 0;
$repository = $repositories->next();
$repositoryId = $repository->getId();
// Only work with allowed repositories
if (count($allowedRepositoryIdStrings) && !in_array($repositoryId->getIdString(), $allowedRepositoryIdStrings)) {
continue;
}
$r++;
$existingRepositoryIds[] = $repositoryId->getIdString();
$assets = $repository->getAssets();
$status = new CLIStatusStars(str_replace('%1', $r, $message) . $repository->getDisplayName());
$status->initializeStatistics($assets->count());
$existingAssetIds = array();
while ($assets->hasNext()) {
$asset = $assets->next();
$assetId = $asset->getId();
$existingAssetIds[] = $assetId->getIdString();
try {
$modificationDate = $asset->getModificationDate();
} catch (UnimplementedException $e) {
$modificationDate = DateAndTime::now();
}
$query = $baseCheckQuery->copy();
$query->addWhereEqual("oai_set", $repositoryId->getIdString());
$query->addWhereEqual("oai_identifier", $assetId->getIdString());
$result = $dbc->query($query, $config->getProperty('OAI_DBID'));
if (!$result->getNumberOfRows()) {
// printpre("Doesn't exist:\t".$asset->getDisplayName()."");
$query = $baseInsertQuery->copy();
$query->addValue('oai_set', $repositoryId->getIdString());
$query->addValue('oai_identifier', $assetId->getIdString());
} else {
// printpre("Exists:\t".$asset->getDisplayName()."");
if ($modificationDate->isGreaterThan(DateAndTime::fromString($result->field('datestamp'))) || $forceUpdate) {
// printpre("\tUpdating:\t".$asset->getDisplayName());
$query = $baseUpdateQuery->copy();
$query->addWhereEqual("oai_set", $repositoryId->getIdString());
$query->addWhereEqual("oai_identifier", $assetId->getIdString());
} else {
$query = null;
}
}
if ($query) {
$query->addRawValue('datestamp', 'NOW()');
}
$isCurrentlyDeleted = $result->getNumberOfRows() && $result->field('deleted') == 'true' ? true : false;
$result->free();
if (!count($authGroupIds)) {
$isVisible = true;
} else {
$isVisible = false;
try {
foreach ($authGroupIds as $id) {
if ($authorizationManager->isAuthorized($id, $viewId, $assetId)) {
$isVisible = true;
break;
}
}
} catch (UnknownIdException $e) {
$isVisible = true;
}
}
if ($query) {
//Add the data fields
// Deleted
if ($isVisible) {
$query->addValue('deleted', 'false');
} else {
$query->addValue('deleted', 'true');
}
$query->addValue('dc_title', $asset->getDisplayName());
$query->addValue('dc_description', $asset->getDescription());
$this->addDublinCoreValues($asset, $query);
$dbc->query($query, $config->getProperty('OAI_DBID'));
$updatesInRepository++;
$numUpdates++;
} else {
if ($isCurrentlyDeleted && $isVisible) {
$query = $baseUndeleteQuery->copy();
} else {
if (!$isCurrentlyDeleted && !$isVisible) {
$query = $baseDeleteQuery->copy();
} else {
$query = null;
}
}
if ($query) {
$query->addWhereEqual("oai_set", $repositoryId->getIdString());
$query->addWhereEqual("oai_identifier", $assetId->getIdString());
$dbc->query($query, $config->getProperty('OAI_DBID'));
$updatesInRepository++;
$numUpdates++;
}
}
$status->updateStatistics();
}
// Update any missing assets as deleted
$query = $baseDeleteQuery->copy();
$query->addWhereEqual("oai_set", $repositoryId->getIdString());
if (count($existingAssetIds)) {
$query->addWhereEqual("deleted", "false");
$query->addWhereNotIn("oai_identifier", $existingAssetIds);
}
$result = $dbc->query($query, $config->getProperty('OAI_DBID'));
if ($result->getNumberOfRows()) {
$updatesInRepository = $updatesInRepository + $result->getNumberOfRows();
$numUpdates = $numUpdates + $result->getNumberOfRows();
}
print OAI_UPDATE_OUTPUT_HTML ? "<pre>" : "\n";
print "Elapsed Time:\t";
$timer->end();
printf("%1.2f", $timer->printTime());
print " seconds";
print OAI_UPDATE_OUTPUT_HTML ? "</pre>" : "";
print OAI_UPDATE_OUTPUT_HTML ? "<pre>" : "\n";
print "Updates: " . $updatesInRepository;
print OAI_UPDATE_OUTPUT_HTML ? "</pre>" : "\n";
}
// Update any missing repositories as deleted
$query = $baseDeleteQuery->copy();
$query->addWhereEqual("deleted", "false");
if (count($existingRepositoryIds)) {
$query->addWhereNotIn("oai_set", $existingRepositoryIds);
}
$result = $dbc->query($query, $config->getProperty('OAI_DBID'));
if ($result->getNumberOfRows()) {
$updatesInRepository = $updatesInRepository + $result->getNumberOfRows();
$numUpdates = $numUpdates + $result->getNumberOfRows();
}
print OAI_UPDATE_OUTPUT_HTML ? "<pre>" : "\n";
print "Total Updates:\t" . $numUpdates;
print OAI_UPDATE_OUTPUT_HTML ? "</pre>" : "\n";
}
