All Downloads are FREE. Search and download functionalities are using the official Maven repository.
  • Log In
  • Register

    • prerna.util.usertracking.reactors.recommendations.DatabaseRecommendationsReactor Maven / Gradle / Ivy

    The newest version!
    package prerna.util.usertracking.reactors.recommendations;

import java.io.File;
import java.util.ArrayList;
import java.util.HashMap;
import java.util.List;
import java.util.Map;

import org.apache.logging.log4j.Logger;

import com.google.gson.Gson;
import com.google.gson.reflect.TypeToken;

import prerna.auth.User;
import prerna.auth.utils.SecurityEngineUtils;
import prerna.ds.r.RSyntaxHelper;
import prerna.reactor.frame.r.AbstractRFrameReactor;
import prerna.sablecc2.om.GenRowStruct;
import prerna.sablecc2.om.PixelDataType;
import prerna.sablecc2.om.PixelOperationType;
import prerna.sablecc2.om.ReactorKeysEnum;
import prerna.sablecc2.om.execptions.SemossPixelException;
import prerna.sablecc2.om.nounmeta.NounMetadata;
import prerna.util.Utility;

public class DatabaseRecommendationsReactor extends AbstractRFrameReactor {
	protected static final String CLASS_NAME = DatabaseRecommendationsReactor.class.getName();
	public static final String COMMUNITIES = "communities";

	public DatabaseRecommendationsReactor() {
		this.keysToGet = new String[] { COMMUNITIES, ReactorKeysEnum.ACCESS.getKey() };
	}

	@Override
	public NounMetadata execute() {
		init();
		organizeKeys();
		Logger logger = getLogger(CLASS_NAME);
		HashMap recommendations = new HashMap();
		String[] packages = { "igraph", "jsonlite" };
		String userName = "null";
		User user = this.insight.getUser();
		String baseFolder = Utility.getBaseFolder();

		// Check to make sure that the necessary files exist
		File itemsim = new File(baseFolder + DIR_SEPARATOR + "R" + DIR_SEPARATOR + "Recommendations" + DIR_SEPARATOR + "dataitem-itemsim.rds");
		File usersim = new File(baseFolder + DIR_SEPARATOR + "R" + DIR_SEPARATOR + "Recommendations" + DIR_SEPARATOR + "dataitem-usersim.rds");
		if (!itemsim.exists() || !usersim.exists()) {
			String message = "Necessary files missing to generate recommendations. Please run UpdateQueryData().";
			NounMetadata noun = new NounMetadata(message, PixelDataType.CONST_STRING, PixelOperationType.ERROR);
			SemossPixelException exception = new SemossPixelException(noun);
			exception.setContinueThreadOfExecution(false);
			throw exception;
		}

		if (user == null) {
			String message = "Please login to enable recommendation features.";
			NounMetadata noun = new NounMetadata(message, PixelDataType.CONST_STRING, PixelOperationType.ERROR);
			SemossPixelException exception = new SemossPixelException(noun);
			exception.setContinueThreadOfExecution(false);
			throw exception;
		}

		if (rJavaTranslator.checkPackages(packages, logger) && user != null) {
			// Step 1:
			// Run an R script to generate all communities, then get
			// additional data for each engine that exists on this machine,
			// package it up as a map to add to a list of outputs for the FE.
			boolean accessFlag = getAccessBool();
			userName = user.getAccessToken(user.getLogins().get(0)).getId();
			StringBuilder esb = new StringBuilder();
			esb.append(RSyntaxHelper.loadPackages(packages));
			esb.append("source(\"" + baseFolder + "\\R\\Recommendations\\db_recom.r\"); ");
			esb.append("fileroot<-\"" + baseFolder + "\\R\\Recommendations\\dataitem\" ; ");
			String varR = esb.toString().replace("\\", "/");
			
			// run communities script
			String communityOutput = Utility.getRandomString(8);
			List myEngines = SecurityEngineUtils.getFullUserEngineIds(this.insight.getUser());
			String items = "";
			for (int row = 0; row < myEngines.size(); row++) {
				String dbid = myEngines.get(row);
				String dbname = SecurityEngineUtils.getEngineAliasForId(dbid);
				String oneItem = dbid + "$" + dbname + "$";
				if(row < myEngines.size() - 1) {
					items += "\"" + oneItem + "\", ";
				}else {
					items += "\"" + oneItem + "\"";
				}
			}
			
			StringBuilder rsb = new StringBuilder();
			rsb.append(communityOutput + "<- locate_data_communities(fileroot,items=c(" + items + "));");
			rsb.append(communityOutput + "<- jsonlite::toJSON(" + communityOutput + "[3]);");
			
			// Step 2:
			// Run another R script to generate user specific recommendations,
			// add additional data and package as a map to be added to the list
			// for the FE.
			String userSpecificOutput = Utility.getRandomString(8);
			//rsb.append(userSpecificOutput + "<- dataitem_recom_mgr(\"" + userName + "\",fileroot);");
			rsb.append(userSpecificOutput + "<- dataitem_recom_mgr(\"" + userName + "\",fileroot,items=c(" + items + "));");
			rsb.append(userSpecificOutput + "<- jsonlite::toJSON(as.data.table(" + userSpecificOutput
					+ "[2])[,1:2], byrow = TRUE, colNames = TRUE);");
			
			String funcR = rsb.toString();
			this.rJavaTranslator.runR(varR + funcR);
			List enginesWithAccess = SecurityEngineUtils.getFullUserEngineIds(this.insight.getUser());
			ArrayList communitiesList = new ArrayList();
			String communityJson = this.rJavaTranslator.getString(communityOutput);
			// the script failed or they dont have the historical data
			if (communityJson != null) {
				Gson gson = new Gson();
				ArrayList>> myList = gson.fromJson(communityJson,
						new TypeToken>>>() {
						}.getType());

				// parse R json response for final recommendation data
				for (int i = 0; i < myList.size(); i++) {
					HashMap> map = myList.get(i);
					for (String key : map.keySet()) {
						ArrayList communityMembers = map.get(key);
						ArrayList> convertedMembers = new ArrayList>();
						for (String member : communityMembers) {
							String[] id = member.split("\\$");
							// limit to 10 dbs. Id and name have to be included
							// to be used
							if (id != null && id.length > 1 && communitiesList.size() < 10) {
								// only send 10 and make sure they exist on this
								// machine
								HashMap engineDetail = new HashMap();
								String alias = id[1];
								String engId = id[0];
								boolean access = enginesWithAccess.contains(id[0]);
								if (accessFlag && !access) {
									continue;
								}
								String dbType = "";
								if (access) {
									dbType = (Utility.getDatabase(engId)).getDatabaseType() + "";
								}
								engineDetail.put("appName", alias);
								engineDetail.put("appId", engId);
								engineDetail.put("appType", dbType);
								engineDetail.put("access", access);
								convertedMembers.add(engineDetail);
							}
						}
						if (!convertedMembers.isEmpty()) {
							communitiesList.add(convertedMembers);
						}
					}
				}
			}
			recommendations.put("Communities", communitiesList);
			// parse R json response for final recommendation data
			String userSpecificJson = this.rJavaTranslator.getString(userSpecificOutput);
			// the script failed or they dont have the historical data
			ArrayList recommendationsFinal = new ArrayList();
			if (userSpecificJson != null) {
				Gson gson = new Gson();
				ArrayList> recList = gson.fromJson(userSpecificJson,
						new TypeToken>>() {
						}.getType());
				for (int i = 0; i < recList.size(); i++) {
					Map itemMap = recList.get(i);
					if (itemMap.isEmpty() || itemMap.get("item") == null) {
						continue;
					}
					String item = itemMap.get("item");
					String[] vals = item.split("\\$");
					// limit to 10 dbs. Id and name have to be included to be
					// used
					if (vals != null && vals.length > 1 && recommendationsFinal.size() < 10) {
						// only send 10 and make sure they exist on this machine
						ArrayList> convertedMembers = new ArrayList>();
						HashMap engineDetail = new HashMap();
						String engId = vals[0];
						String freq = recList.get(i).get("score");
						String alias = vals[1];
						boolean access = enginesWithAccess.contains(vals[0]);
						if (accessFlag && !access) {
							continue;
						}
						String dbType = "";
						if (access) {
							dbType = (Utility.getDatabase(engId)).getDatabaseType() + "";
						}
						// only return databases the user can access
						engineDetail.put("appName", alias);
						engineDetail.put("appId", engId);
						engineDetail.put("appType", dbType);
						engineDetail.put("freq", freq);
						engineDetail.put("access", access);
						convertedMembers.add(engineDetail);
						recommendationsFinal.add(convertedMembers);
					}
				}
			}
			recommendations.put("Recommendations", recommendationsFinal);
			// garbage cleanup -- R script might already do this
			String gc = "rm(" + communityOutput + ", " + userSpecificOutput
					+ ",blend_mgr, data_domain_mgr, read_datamatrix, exec_tfidf, "
					+ "remove_files, fileroot, blend_tracking_semantic, get_userdata, "
					+ "dataitem_history, get_dataitem_rating, assign_unique_concepts, "
					+ "populate_ratings, build_sim, cosine_jaccard_sim, cosine_sim, "
					+ "jaccard_sim, dataitem_recom_mgr, "
					+ "get_item_recom, get_user_recom, hop_away_recom_mgr, hop_away_mgr,"
					+ "locate_user_communities, drilldown_communities, locate_data_communities, "
					+ "get_items_users, refresh_base);";
			this.rJavaTranslator.runR(gc);
		}
		return new NounMetadata(recommendations, PixelDataType.CUSTOM_DATA_STRUCTURE,
				PixelOperationType.RECOMMENDATION);
	}

	private boolean getAccessBool() {
		GenRowStruct boolGrs = this.store.getNoun(ReactorKeysEnum.ACCESS.getKey());
		if (boolGrs != null) {
			if (boolGrs.size() > 0) {
				List val = boolGrs.getValuesOfType(PixelDataType.BOOLEAN);
				return (boolean) val.get(0);
			}
		}
		return false;
	}
	
	public String getName()
	{
		return "DatabaseRecommendations";
	}

}
    
    
    
    

    




    
    
    © 2015 - 2025 Weber Informatics LLC | Privacy Policy