FitService.java

package edu.columbia.biology.service;

import org.springframework.security.core.Authentication;

import edu.columbia.biology.model.FitDetails;

public interface FitService {

	/**
	 * Given fitId, retrieve the fitDetails
	 * 
	 * @return
	 */
	public FitDetails retrieveFitDetails(int fitId, boolean isSiteAdmin);

	public String retrieveFitScoringJson(int fitId);

	/**
	 * Determine if fit is user accessible
	 * 
	 * @param principal
	 * @param fitId
	 * @return
	 */
	public boolean isFitUserAccessible(Authentication authentication, int fitId);
}

FitServiceImpl.java

package edu.columbia.biology.service;

import java.util.Arrays;

import java.util.Collection;
import java.util.LinkedList;

import java.util.List;
import java.util.stream.Collectors;

import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.beans.factory.annotation.Qualifier;
import org.springframework.security.core.Authentication;
import org.springframework.stereotype.Service;

import edu.columbia.biology.mapper.FitsRepositoryMapper;
import edu.columbia.biology.model.FitDetails;
import edu.columbia.biology.model.FitsMetadataInformation;
import edu.columbia.biology.model.PostprocContainerTO;
import edu.columbia.biology.model.PostprocEntryTO;
import edu.columbia.biology.model.PostprocItem;
import edu.columbia.biology.processor.CellxPrincipalProcessor;
import edu.columbia.biology.processor.PostprocProcessor;
import edu.columbia.biology.repository.CellxGlobalFitCurationCommentRepository;
import edu.columbia.biology.repository.FitsRepository;
import edu.columbia.biology.view_model.PostprocComponent;

@Service
public class FitServiceImpl implements FitService {

	@Autowired
	private PostprocProcessor postprocProcessor;

	@Autowired
	private CellxPrincipalProcessor cellxPrincipalProcessor;

	@Autowired
	private FitsRepository fitsRepository;

	@Autowired
	private FitsRepositoryMapper fitsRepositoryMapper;

	@Autowired
	private StudyService studyService;

	@Autowired
	private PostprocService postprocService;

	@Qualifier("publicAccessiblePostprocs")
	@Autowired
	private List<String> publicAccessiblePostprocs;

	/**
	 * TODO: The code here could be optimized for faster loading. Also perhaps
	 * postprocs could be loaded asyncronously by sending the postproc entries to
	 * the page, then asyncronosuly loading the postproc on expand? Therefore, it
	 * could also be cacheable?
	 */
	@Override
	public FitDetails retrieveFitDetails(int fitId, boolean isSiteAdmin) {
		FitsMetadataInformation fitsMetadataInformation = retrieveFitsMetadataInformation(fitId);
		List<PostprocItem> postprocItems = postprocService.retrievePostprocItems(fitId);

		// Non site admins only have access to basicModelViewer and resourceMotif
		if (!isSiteAdmin) {
			postprocItems = postprocItems.stream()
					.filter(postprocItem -> publicAccessiblePostprocs.contains(postprocItem.getPostMethodName()))
					.collect(Collectors.toList());
		}
		return new FitDetails(fitsMetadataInformation, postprocItems);
	}

	@Override
	public boolean isFitUserAccessible(Authentication authentication, int fitId) {
		if (cellxPrincipalProcessor.isSiteAdmin(authentication)) {
			return true;
		} else {
			List<String> fitStudyNames = studyService.retrieveFitAssociatedStudies(fitId);
			List<String> userAccessibleStudies = studyService.retrieveUserAccessibleStudies(authentication);
			return userAccessibleStudies.containsAll(fitStudyNames);
		}
	}

	@Override
	public String retrieveFitScoringJson(int fitId) {
		return fitsRepository.retrieveFitScoringJson(fitId);
	}

	public boolean isFitGlobalAccessible(Authentication authentication, int fitId) {
		if (cellxPrincipalProcessor.isSiteAdmin(authentication)) {
			return true;
		} else {
			List<String> fitStudyNames = studyService.retrieveFitAssociatedStudies(fitId);
			List<String> userAccessibleStudies = studyService.retrieveUserAccessibleStudies(authentication);
			return userAccessibleStudies.containsAll(fitStudyNames);
		}
	}

	/**
	 * Given the fitId, retrieve the fit's metadata information
	 * 
	 * @param fitId
	 * @return
	 */
	private FitsMetadataInformation retrieveFitsMetadataInformation(int fitId) {
		return fitsRepositoryMapper.retrieveFitMetadataMapper(fitsRepository.retrieveFitMetadata(fitId));
	}

}

FitsRepository.java

package edu.columbia.biology.repository;

import org.springframework.transaction.annotation.Transactional;

import java.util.Collection;
import java.util.List;

import org.springframework.data.jpa.repository.JpaRepository;
import org.springframework.data.jpa.repository.JpaSpecificationExecutor;
import org.springframework.data.jpa.repository.Query;
import org.springframework.stereotype.Repository;

import edu.columbia.biology.entity.Fits;

/**
 * Generated by Spring Data Generator on 14/01/2019
 */
@Repository
@Transactional
public interface FitsRepository extends JpaRepository<Fits, Integer>, JpaSpecificationExecutor<Fits> {

	// TODO: [CONVENIENT] All added queries and repository methods somehow needs to
	// be externalized to a properties/yaml file.
	@Query("SELECT f from edu.columbia.biology.entity.Fits f WHERE f.regressorId = ?1")
	Collection<Fits> findAllByRegressorId(Integer regressorId);

	/**
	 * Retrieve every fit - used for administrator view
	 * 
	 * TODO: Is this method still used? 3/12/21 YES I believe for admins?
	 * 
	 * @deprecated
	 */
	@Deprecated
	@Query(value = "SELECT fits.fit_id, cellx_global_fit_curation.curated,"
			+ "CAST(ARRAY_AGG(DISTINCT experiment.study_name) AS text) AS study_name_list, "
			+ "CAST(ARRAY_AGG(DISTINCT study.study_display_name) AS text) AS study_display_name_list,\r\n"
			+ "CAST(ARRAY_AGG(DISTINCT experiment.experiment_name) AS text) AS experiment_name_list, "
			+ "CAST(ARRAY_AGG(DISTINCT factor.gene_symbol) AS text) AS gene_symbol_list, "
			+ "CAST(ARRAY_AGG(DISTINCT factor.factor_id) AS text)  AS factor_id_list, "
			+ "CAST(ARRAY_AGG(DISTINCT factor.tax_id) AS text) AS tax_id_list FROM fits, "
			+ "INNER JOIN count_fit ON fits.fit_id = count_fit.fit_id "
			+ "INNER JOIN count_table ON count_fit.count_table_id = count_table.count_table_id "
			+ "INNER JOIN experiment ON count_table.study_name = experiment.study_name AND count_table.experiment_name = experiment.experiment_name "
			+ "INNER JOIN factor_complex ON experiment.complex_id = factor_complex.complex_id "
			+ "INNER JOIN factor ON factor_complex.factor_id = factor.factor_id "
			+ "LEFT JOIN cellx_global_fit_curation ON fits.fit_id = cellx_global_fit_curation.fit_id "
			+ "WHERE fits.iscomplete=TRUE "
			+ "GROUP BY fits.fit_id, cellx_global_fit_curation.curated ", nativeQuery = true)
	Collection<Object[]> retrieveAllFits();

	/**
	 * Retrieve every fit in a public study - used for administrator view
	 */
	@Query(value = "SELECT fits.fit_id, fits.fit_path, cellx_global_fit_curation.curated,\r\n "
			+ "CAST(ARRAY_AGG(DISTINCT cellx_global_fit_curation_tag.tag) AS text) AS tag_list,\r\n"
			+ "CAST(ARRAY_AGG(DISTINCT experiment.study_name) AS text) AS study_name_list,\r\n"
			+ "CAST(ARRAY_AGG(DISTINCT study.study_display_name) AS text) AS study_display_name_list,\r\n"
			+ "CAST(ARRAY_AGG(DISTINCT experiment.experiment_name) AS text) AS experiment_name_list, "
			+ "CAST(ARRAY_AGG(DISTINCT factor.gene_symbol) AS text) AS gene_symbol_list, "
			+ "CAST(ARRAY_AGG(DISTINCT factor.factor_id) AS text)  AS factor_id_list, "
			+ "CAST(ARRAY_AGG(DISTINCT factor.tax_id) AS text) AS tax_id_list, " + "fits.ranking_score, "
			+ "fit_group_curated.group_id AS pipeline_group_id, " + "fit_group_curated.metric AS pipeline_metric, "
			+ "fit_group_curated.metric_value AS pipeline_metric_value, "
			+ "fit_group_curated.iscurated AS pipeline_curated FROM fits "
			+ "INNER JOIN fit_group_curated ON fits.fit_id = fit_group_curated.fit_id "
			+ "INNER JOIN count_fit ON fits.fit_id = count_fit.fit_id "
			+ "INNER JOIN count_table ON count_fit.count_table_id = count_table.count_table_id "
			+ "INNER JOIN experiment ON count_table.study_name = experiment.study_name AND count_table.experiment_name = experiment.experiment_name "
			+ "INNER JOIN study ON study.study_name = experiment.study_name "
			+ "INNER JOIN factor_complex ON experiment.complex_id = factor_complex.complex_id "
			+ "INNER JOIN factor ON factor_complex.factor_id = factor.factor_id "
			+ "LEFT JOIN cellx_global_fit_curation ON fits.fit_id = cellx_global_fit_curation.fit_id "
			+ "LEFT JOIN cellx_global_fit_curation_tag ON fits.fit_id = cellx_global_fit_curation_tag.fit_id "
			+ "WHERE fits.iscomplete=TRUE " + "GROUP BY fits.fit_id, fits.fit_path, cellx_global_fit_curation.curated, "
			+ "fit_group_curated.group_id, fit_group_curated.metric, fit_group_curated.metric_value, fit_group_curated.iscurated "
			+ "HAVING NOT true=any(ARRAY_AGG(DISTINCT study.isembargoed))", nativeQuery = true)
	Collection<Object[]> retrievePublicFits();

	/**
	 * Retrieve every fit in a public study that is also curated - for users
	 */
	@Query(value = "SELECT fits.fit_id, fits.fit_path, cellx_global_fit_curation.curated, \r\n"
			+ "CAST(ARRAY_AGG(DISTINCT cellx_global_fit_curation_tag.tag) AS text) AS tag_list,\r\n"
			+ "CAST(ARRAY_AGG(DISTINCT experiment.study_name) AS text) AS study_name_list,\r\n"
			+ "CAST(ARRAY_AGG(DISTINCT study.study_display_name) AS text) AS study_display_name_list,\r\n"
			+ "CAST(ARRAY_AGG(DISTINCT experiment.experiment_name) AS text) AS experiment_name_list, "
			+ "CAST(ARRAY_AGG(DISTINCT factor.gene_symbol) AS text) AS gene_symbol_list, "
			+ "CAST(ARRAY_AGG(DISTINCT factor.factor_id) AS text)  AS factor_id_list, "
			+ "CAST(ARRAY_AGG(DISTINCT factor.tax_id) AS text) AS tax_id_list, " + "fits.ranking_score, "
			+ "fit_group_curated.group_id AS pipeline_group_id, " + "fit_group_curated.metric AS pipeline_metric, "
			+ "fit_group_curated.metric_value AS pipeline_metric_value, "
			+ "fit_group_curated.iscurated AS pipeline_curated FROM fits "
			+ "INNER JOIN fit_group_curated ON fits.fit_id = fit_group_curated.fit_id "
			+ "INNER JOIN count_fit ON fits.fit_id = count_fit.fit_id "
			+ "INNER JOIN count_table ON count_fit.count_table_id = count_table.count_table_id "
			+ "INNER JOIN experiment ON count_table.study_name = experiment.study_name AND count_table.experiment_name = experiment.experiment_name "
			+ "INNER JOIN study ON study.study_name = experiment.study_name "
			+ "INNER JOIN factor_complex ON experiment.complex_id = factor_complex.complex_id "
			+ "INNER JOIN factor ON factor_complex.factor_id = factor.factor_id "
			+ "LEFT JOIN cellx_global_fit_curation ON fits.fit_id = cellx_global_fit_curation.fit_id "
			+ "LEFT JOIN cellx_global_fit_curation_tag ON fits.fit_id = cellx_global_fit_curation_tag.fit_id "
			+ "WHERE ((fits.iscomplete=TRUE) AND (cellx_global_fit_curation.curated IS NOT FALSE) AND ((cellx_global_fit_curation.curated IS TRUE) OR ((cellx_global_fit_curation.curated IS NULL) AND (fit_group_curated.iscurated IS TRUE)))) "
			+ "GROUP BY fits.fit_id, fits.fit_path, cellx_global_fit_curation.curated, "
			+ "fit_group_curated.group_id, fit_group_curated.metric, fit_group_curated.metric_value, fit_group_curated.iscurated "
			+ "HAVING NOT true=any(ARRAY_AGG(DISTINCT study.isembargoed)) AND ARRAY_AGG(DISTINCT cellx_global_fit_curation_tag.tag)  @> '{\"GOOD\"}'", nativeQuery = true)
	Collection<Object[]> retrievePublicCuratedFits();

	// TODO: Replace the cellx tables with the new version in celldb-devtest version
	/**
	 * Retrieve every fit that a user has access to
	 * 
	 * @Deprecated - method not yet used
	 */
	@Deprecated
	@Query(value = "SELECT fits.fit_id, fits.fit_path, postproc.postproc_id, postproc.output_string, "
			+ "CAST(ARRAY_AGG(DISTINCT experiment.study_name) AS text) AS study_name_list, "
			+ "CAST(ARRAY_AGG(DISTINCT study.study_display_name) AS text) AS study_display_name_list,\r\n"
			+ "CAST(ARRAY_AGG(DISTINCT experiment.experiment_name) AS text) AS experiment_name_list, "
			+ "CAST(ARRAY_AGG(DISTINCT factor.gene_symbol) AS text) AS gene_symbol_list, "
			+ "CAST(ARRAY_AGG(DISTINCT factor.factor_id) AS text)  AS factor_id_list, "
			+ "CAST(ARRAY_AGG(DISTINCT factor.tax_id) AS text) AS tax_id_list FROM fits "
			+ "INNER JOIN count_fit ON fits.fit_id = count_fit.fit_id "
			+ "INNER JOIN count_table ON count_fit.count_table_id = count_table.count_table_id "
			+ "INNER JOIN experiment ON count_table.study_name = experiment.study_name AND count_table.experiment_name = experiment.experiment_name "
			+ "INNER JOIN study ON study.study_name = experiment.study_name "
			+ "LEFT JOIN cellx_group_study ON study.study_name = cellx_group_study.study_name "
			+ "LEFT JOIN cellx_group ON cellx_group_study.group_id = cellx_group.group_id "
			+ "LEFT JOIN cellx_group_user ON cellx_group.group_id = cellx_group_user.group_id "
			+ "INNER JOIN factor_complex ON experiment.complex_id = factor_complex.complex_id "
			+ "INNER JOIN factor ON factor_complex.factor_id = factor.factor_id "
			+ "INNER JOIN fit_post ON fits.fit_id = fit_post.fit_id "
			+ "INNER JOIN postproc ON fit_post.postproc_id = postproc.postproc_id "
			+ "WHERE postproc.post_method_name='summaryQC' AND fits.iscomplete=TRUE AND (cellx_group_user.user_id = ?1 OR cellx_group_user.user_id IS NULL)"
			+ "GROUP BY fits.fit_id, postproc.postproc_id, postproc.output_string "
			+ "HAVING NOT false=any(ARRAY_AGG(DISTINCT CASE WHEN study.isembargoed = false OR cellx_group_user.user_id = ?1 THEN true ELSE false END))", nativeQuery = true)
	Collection<Object[]> retrieveUserAccessibleFits(Integer userId);

	/**
	 * Retrieve studies associated with a fit
	 */
	@Query("SELECT DISTINCT countTable.studyName FROM edu.columbia.biology.entity.Fits fits "
			+ "INNER JOIN edu.columbia.biology.entity.CountFit countFit ON fits.fitId = countFit.fitId "
			+ "INNER JOIN edu.columbia.biology.entity.CountTable countTable ON countFit.countTableId = countTable.countTableId "
			+ "WHERE fits.fitId = ?1")
	Collection<String> retrieveFitStudies(Integer fitId);

	@Query("SELECT fits.scoringJson FROM  edu.columbia.biology.entity.Fits fits WHERE fits.fitId = ?1")
	String retrieveFitScoringJson(Integer fitId);

	/**
	 * Retrieve fit metadata
	 */
	@Query(value = "SELECT fits.fit_id, fits.fit_path,\r\n"
			+ "CAST(ARRAY_AGG(DISTINCT experiment.study_name) AS text) AS study_name_list,\r\n"
			+ "CAST(ARRAY_AGG(DISTINCT study.study_display_name) AS text) AS study_display_name_list,\r\n"
			+ "CAST(ARRAY_AGG(DISTINCT experiment.experiment_name) AS text) AS experiment_name_list,\r\n"
			+ "CAST(ARRAY_AGG(DISTINCT factor.gene_symbol) AS text) AS gene_symbol_list, \r\n"
			+ "CAST(ARRAY_AGG(DISTINCT factor.factor_id) AS text)  AS factor_id_list, \r\n"
			+ "CAST(ARRAY_AGG(DISTINCT factor.tax_id) AS text) AS tax_id_list \r\n" + "FROM fits\r\n"
			+ "INNER JOIN count_fit ON fits.fit_id = count_fit.fit_id\r\n"
			+ "INNER JOIN count_table ON count_fit.count_table_id = count_table.count_table_id \r\n"
			+ "INNER JOIN experiment ON count_table.study_name = experiment.study_name AND count_table.experiment_name = experiment.experiment_name \r\n"
			+ "INNER JOIN study ON experiment.study_name = study.study_name\r\n"
			+ "INNER JOIN factor_complex ON experiment.complex_id = factor_complex.complex_id \r\n"
			+ "INNER JOIN factor ON factor_complex.factor_id = factor.factor_id \r\n" + "WHERE fits.fit_id = ?1\r\n"
			+ "GROUP BY fits.fit_id, fits.fit_path\r\n" + "LIMIT 1", nativeQuery = true)
	List<Object[]> retrieveFitMetadata(Integer fitId);

}