OCA\FaceRecognition\BackgroundJob\Tasks\CreateClustersTask::createClusterIfNeeded()

Complexity

Conditions	11
Paths	34

Size

Total Lines	121
Code Lines	64

Duplication

Lines	0
Ratio	0 %

Code Coverage

Tests	47
CRAP Score	12.7127

Importance

Changes	4
Bugs	0	Features	0

<?php
	private function createClusterIfNeeded(string $userId) {
		$modelId = $this->settingsService->getCurrentFaceModel();

		// Depending on whether we already have clusters, decide if we should create/recreate them.
		//
		$hasPersons = $this->personMapper->countPersons($userId, $modelId) > 0;
		if ($hasPersons) {
			$forceRecreate = $this->needRecreateBySettings($userId);
			$haveEnoughFaces = $this->hasNewFacesToRecreate($userId, $modelId);
			$haveStaled = $this->hasStalePersonsToRecreate($userId, $modelId);

			if ($forceRecreate) {
				$this->logInfo('Clusters already exist, but there was some change that requires recreating the clusters');
			}
			else if ($haveEnoughFaces || $haveStaled) {
				$this->logInfo('Face clustering will be recreated with new information or changes');
			}
			else {
				// If there is no invalid persons, and there is no recent new faces, no need to recreate cluster
				$this->logInfo('Clusters already exist, estimated there is no need to recreate them');
				return;
			}
		}
		else {
			// User should not be able to use this directly, used in tests
			$forceTestCreation = $this->settingsService->_getForceCreateClusters($userId);
			$needCreate = $this->needCreateFirstTime($userId, $modelId);

			if ($forceTestCreation) {
				$this->logInfo('Force the creation of clusters for testing');
			}
			else if ($needCreate) {
				$this->logInfo('Face clustering will be created for the first time.');
			}
			else {
				$this->logInfo(
					'Skipping cluster creation, not enough data (yet) collected. ' .
					'For cluster creation, you need either one of the following:');
				$this->logInfo('* have 1000 faces already processed');
				$this->logInfo('* or you need to have 95% of you images processed');
				$this->logInfo('Use stats command to track progress');
				return;
			}
		}

		// Ok. If we are here, the clusters must be recreated.
		//

		$min_face_size = $this->settingsService->getMinimumFaceSize();
		$min_confidence = $this->settingsService->getMinimumConfidence();

		$faces = $this->faceMapper->getGroupableFaces($userId, $modelId, $min_face_size, $min_confidence);

		$facesCount = count($faces);
		$this->logInfo('There are ' . $facesCount . ' faces for clustering.');

		// The default slice is just one for the total.
		$noSlices = 1;
		$sliceSize = $facesCount;

		// Now calculate it if there is a batch size configured.
		$batchSize = $this->settingsService->getClusterigBatchSize();
		if ($facesCount > 0 && $batchSize > 0) {
			// The minimum batch size is 2000 faces.
			$batchSize = max($batchSize, 2000);
			// The maximun batch size is the faces count.
			$batchSize = min($batchSize, $facesCount);

			// Calculate the number of slices and their sizes.
			$noSlices = intval($facesCount / $batchSize) + 1;
			$sliceSize = ceil($facesCount / $noSlices);
		}

		$this->logDebug('We will cluster these with ' . $noSlices . ' batch(es) of ' . $sliceSize . ' faces.');

		$newClusters = [];
		// Obtain the clusters in batches and append them.
		for ($i = 0; $i < $noSlices ; $i++) {
			// Get the batches.
			$facesSliced = array_slice($faces, $i * $sliceSize, $sliceSize);

$i * $sliceSize of type double is incompatible with the type integer expected by parameter $offset of array_slice().

It seems like $sliceSize can also be of type double; however, parameter $length of array_slice() does only seem to accept integer|null, maybe add an additional type check?

			// Get the indices, obtain the partial clusters and incorporate them.
			$faceIds = array_map(function ($face) { return $face['id']; }, $facesSliced);
			$facesDescripted = $this->faceMapper->findDescriptorsBathed($faceIds);
			$newClusters = array_merge($newClusters, $this->getNewClusters($facesDescripted));
			// Discard variables aggressively to improve memory consumption.
			unset($facesDescripted);
			unset($facesSliced);
		}

		// Append non groupable faces on a single step.
		$nonGroupables = $this->faceMapper->getNonGroupableFaces($userId, $modelId, $min_face_size, $min_confidence);
		$this->logInfo('We will add '. count($nonGroupables) . ' faces that cannot be grouped.');
		$newClusters = array_merge($newClusters, $this->getFakeClusters($nonGroupables));

		// Cluster is associative array where key is person ID.
		// Value is array of face IDs. For old clusters, person IDs are some existing person IDs,
		// and for new clusters is whatever chinese whispers decides to identify them.
		//
		$currentClusters = $this->getCurrentClusters(array_merge($faces, $nonGroupables));
		$this->logInfo(count($newClusters) . ' clusters found after clustering');

		// Discard variables aggressively to improve memory consumption.
		unset($faces);
		unset($nonGroupables);

		// New merge
		$mergedClusters = $this->mergeClusters($currentClusters, $newClusters);

		$this->personMapper->mergeClusterToDatabase($userId, $currentClusters, $mergedClusters);

		// Remove all orphaned persons (those without any faces)
		// NOTE: we will do this for all models, not just for current one, but this is not problem.
		$orphansDeleted = $this->personMapper->deleteOrphaned($userId);
		if ($orphansDeleted > 0) {
			$this->logInfo('Deleted ' . $orphansDeleted . ' persons without faces');
		}

		// Prevents not create/recreate the clusters unnecessarily.

		$this->settingsService->setNeedRecreateClusters(false, $userId);
		$this->settingsService->_setForceCreateClusters(false, $userId);
	}