Merge pull request #6 from aleksandr-isaev/master

Various bug fixes and new script to count total number of links for each algorithm

src/main/java/uk/ac/standrews/cs/population_linkage/aleks/LinksCounter.java

@@ -0,0 +1,77 @@
+/*
+ * Copyright 2022 Systems Research Group, University of St Andrews:
+ * <https://github.com/stacs-srg>
+ *
+ * This file is part of the module population-linkage.
+ *
+ * population-linkage is free software: you can redistribute it and/or modify it under the terms of the GNU General Public
+ * License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later
+ * version.
+ *
+ * population-linkage is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied
+ * warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License along with population-linkage. If not, see
+ * <http://www.gnu.org/licenses/>.
+ */
+package uk.ac.standrews.cs.population_linkage.aleks;
+import org.neo4j.driver.Result;
+import uk.ac.standrews.cs.neoStorr.impl.Store;
+import uk.ac.standrews.cs.neoStorr.util.NeoDbCypherBridge;
+import uk.ac.standrews.cs.population_linkage.graph.Query;
+
+import java.util.List;
+
+public class LinksCounter {
+
+    public static void main(String[] args) {
+        NeoDbCypherBridge bridge = Store.getInstance().getBridge();
+
+        System.out.println("Number of links for each algorithm:");
+        long[] numLinks = countLinks(bridge);
+        calculateBasicStats(numLinks);
+    }
+
+    private static long[] countLinks(NeoDbCypherBridge bridge) {
+        List <String> queries = Query.getAllLinkQueries();
+        List <String> algorithms = Query.getAllLinkerNames();
+        long[] numLinks = new long[queries.size()];
+
+        for (int i = 0; i < queries.size(); i++) {
+            Result result = bridge.getNewSession().run(queries.get(i));
+            List<Long> counts = result.list(r -> r.get("link_count").asLong());
+            if (!counts.isEmpty()) {
+                long count = counts.get(0);
+                numLinks[i] = count;
+                System.out.println((i+1) + ". " + algorithms.get(i) + " -> " + count);
+            }
+        }
+
+        return numLinks;
+    }
+
+    private static void calculateBasicStats(long[] numLinks) {
+        List <String> algorithms = Query.getAllLinkerNames();
+        long lowestNum = 999999999;
+        int lowestIndex = -1;
+        long highestNum = 0;
+        int highestIndex = -1;
+        long totalNum = 0;
+
+        for (int i = 0; i < numLinks.length; i++) {
+            if(numLinks[i] < lowestNum) {
+                lowestNum = numLinks[i];
+                lowestIndex = i;
+            }else if(numLinks[i] > highestNum) {
+                highestNum = numLinks[i];
+                highestIndex = i;
+            }
+
+            totalNum += numLinks[i];
+        }
+
+        System.out.println("Lowest number of links: " + algorithms.get(lowestIndex) + " -> " + lowestNum);
+        System.out.println("Highest number of links: " + algorithms.get(highestIndex) + " -> " + highestNum);
+        System.out.println("Average number of links: " + totalNum / numLinks.length);
+    }
+}

src/main/java/uk/ac/standrews/cs/population_linkage/endToEnd/builders/BirthBrideOwnMarriageBuilder.java

@@ -42,11 +42,11 @@ public static void main(String[] args) throws Exception {
         String number_of_records = args[1]; // e.g. EVERYTHING or 10000 etc.
 
         try(
-            BirthBrideSiblingLinkageRecipe linkageRecipe = new BirthBrideSiblingLinkageRecipe(sourceRepo, number_of_records, BirthBrideSiblingBundleBuilder.class.getName(), null) ) {
+            BirthBrideIdentityLinkageRecipe linkageRecipe = new BirthBrideIdentityLinkageRecipe(sourceRepo, number_of_records, BirthBrideSiblingBundleBuilder.class.getName(), null) ) {
 
             BitBlasterLinkageRunner runner = new BitBlasterLinkageRunner();
 
-            int linkage_fields = BirthBrideIdentityLinkageRecipe.ALL_LINKAGE_FIELDS;
+            int linkage_fields = linkageRecipe.ALL_LINKAGE_FIELDS;
             int half_fields = linkage_fields - (linkage_fields / 2);
 
             while (linkage_fields >= half_fields) {

src/main/java/uk/ac/standrews/cs/population_linkage/endToEnd/builders/BirthDeathSiblingBundleBuilder.java

@@ -40,7 +40,7 @@ public static void main(String[] args) throws Exception {
         String number_of_records = args[1]; // e.g. EVERYTHING or 10000 etc.
 
         try(
-            BirthDeathSiblingLinkageRecipe linkageRecipe = new BirthDeathSiblingLinkageRecipe(sourceRepo, number_of_records, BirthBrideSiblingBundleBuilder.class.getName(), null) ) {
+            BirthDeathSiblingLinkageRecipe linkageRecipe = new BirthDeathSiblingLinkageRecipe(sourceRepo, number_of_records, BirthDeathSiblingBundleBuilder.class.getName(), null) ) {
 
             BitBlasterLinkageRunner runner = new BitBlasterLinkageRunner();
 

src/main/java/uk/ac/standrews/cs/population_linkage/endToEnd/builders/DeathBrideOwnMarriageBuilder.java

@@ -27,6 +27,7 @@
 import uk.ac.standrews.cs.population_linkage.supportClasses.LinkageQuality;
 import uk.ac.standrews.cs.population_linkage.supportClasses.LinkageResult;
 import uk.ac.standrews.cs.population_records.record_types.Birth;
+import uk.ac.standrews.cs.population_records.record_types.Death;
 import uk.ac.standrews.cs.population_records.record_types.Marriage;
 
 /**
@@ -61,7 +62,7 @@ public static void main(String[] args) throws Exception {
     @Override
     public void makePersistent(LinkageRecipe recipe, Link link) {
         try {
-            final String std_id1 = link.getRecord1().getReferend(Birth.class).getString(Birth.STANDARDISED_ID);
+            final String std_id1 = link.getRecord1().getReferend(Death.class).getString(Death.STANDARDISED_ID);
             final String std_id2 = link.getRecord2().getReferend(Marriage.class).getString(Marriage.STANDARDISED_ID);
 
             if( ! Query.DMDeathBrideOwnMarriageReferenceExists(recipe.getBridge(), std_id1, std_id2, recipe.getLinksPersistentName())) {