BloomFilter equals tests and refactoring (#448)

* Add equals tests to BloomFilter classes, rename ModeFromBloomContext, refactor ToBloomFilter

* Fix equals tests in BloomFilter classes

* Rename ToBloomFilter to BloomFilterBlob

pom.xml

@@ -120,6 +120,12 @@
       <version>2.0.0</version>
       <scope>test</scope>
     </dependency>
+    <dependency>
+      <groupId>nl.jqno.equalsverifier</groupId>
+      <artifactId>equalsverifier</artifactId>
+      <version>3.17.4</version>
+      <scope>test</scope>
+    </dependency>
     <dependency>
       <groupId>com.teragrep</groupId>
       <artifactId>pth_06</artifactId>

src/main/java/com/teragrep/pth10/ast/commands/transformstatement/TeragrepTransformation.java

@@ -53,7 +53,7 @@
 import com.teragrep.pth10.ast.bo.Token;
 import com.teragrep.pth10.ast.commands.logicalstatement.LogicalStatementCatalyst;
 import com.teragrep.pth10.ast.commands.logicalstatement.LogicalStatementXML;
-import com.teragrep.pth10.ast.commands.transformstatement.teragrep.ModeFromBloomContext;
+import com.teragrep.pth10.ast.commands.transformstatement.teragrep.ContextBloomMode;
 import com.teragrep.pth10.ast.commands.transformstatement.teragrep.EstimateColumnFromBloomContext;
 import com.teragrep.pth10.ast.commands.transformstatement.teragrep.InputColumnFromBloomContext;
 import com.teragrep.pth10.ast.commands.transformstatement.teragrep.OutputColumnFromBloomContext;
@@ -480,7 +480,7 @@ public Node visitT_bloomModeParameter(final DPLParser.T_bloomModeParameterContex
     @Override
     public Node visitT_bloomOptionParameter(final DPLParser.T_bloomOptionParameterContext ctx) {
         // values from context
-        final ContextValue<TeragrepBloomStep.BloomMode> mode = new ModeFromBloomContext(ctx);
+        final ContextValue<TeragrepBloomStep.BloomMode> mode = new ContextBloomMode(ctx);
         final ContextValue<String> inputCol = new InputColumnFromBloomContext(ctx);
         final ContextValue<String> outputCol = new OutputColumnFromBloomContext(ctx, inputCol.value());
         final ContextValue<String> estimateCol = new EstimateColumnFromBloomContext(ctx, inputCol.value());

src/main/java/com/teragrep/pth10/ast/commands/transformstatement/teragrep/ModeFromBloomContext.java → src/main/java/com/teragrep/pth10/ast/commands/transformstatement/teragrep/ContextBloomMode.java

@@ -49,11 +49,11 @@
 import com.teragrep.pth10.steps.teragrep.TeragrepBloomStep;
 import com.teragrep.pth_03.antlr.DPLParser;
 
-public final class ModeFromBloomContext implements ContextValue<TeragrepBloomStep.BloomMode> {
+public final class ContextBloomMode implements ContextValue<TeragrepBloomStep.BloomMode> {
 
     private final DPLParser.T_bloomOptionParameterContext ctx;
 
-    public ModeFromBloomContext(final DPLParser.T_bloomOptionParameterContext ctx) {
+    public ContextBloomMode(final DPLParser.T_bloomOptionParameterContext ctx) {
         this.ctx = ctx;
     }
 

src/main/java/com/teragrep/pth10/steps/teragrep/bloomfilter/BloomFilterBlob.java

@@ -0,0 +1,94 @@
+/*
+ * Teragrep Data Processing Language (DPL) translator for Apache Spark (pth_10)
+ * Copyright (C) 2019-2024 Suomen Kanuuna Oy
+ *
+ * This program is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU Affero General Public License as published by
+ * the Free Software Foundation, either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * This program 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 Affero General Public License for more details.
+ *
+ * You should have received a copy of the GNU Affero General Public License
+ * along with this program.  If not, see <https://www.gnu.org/licenses/>.
+ *
+ *
+ * Additional permission under GNU Affero General Public License version 3
+ * section 7
+ *
+ * If you modify this Program, or any covered work, by linking or combining it
+ * with other code, such other code is not for that reason alone subject to any
+ * of the requirements of the GNU Affero GPL version 3 as long as this Program
+ * is the same Program as licensed from Suomen Kanuuna Oy without any additional
+ * modifications.
+ *
+ * Supplemented terms under GNU Affero General Public License version 3
+ * section 7
+ *
+ * Origin of the software must be attributed to Suomen Kanuuna Oy. Any modified
+ * versions must be marked as "Modified version of" The Program.
+ *
+ * Names of the licensors and authors may not be used for publicity purposes.
+ *
+ * No rights are granted for use of trade names, trademarks, or service marks
+ * which are in The Program if any.
+ *
+ * Licensee must indemnify licensors and authors for any liability that these
+ * contractual assumptions impose on licensors and authors.
+ *
+ * To the extent this program is licensed as part of the Commercial versions of
+ * Teragrep, the applicable Commercial License may apply to this file if you as
+ * a licensee so wish it.
+ */
+package com.teragrep.pth10.steps.teragrep.bloomfilter;
+
+import org.apache.spark.util.sketch.BloomFilter;
+
+import java.io.ByteArrayInputStream;
+import java.io.IOException;
+import java.util.Arrays;
+
+/**
+ * BloomFilter from byte[] used in a constructor of TeragrepBloomFilter
+ * 
+ * @see TeragrepBloomFilter
+ */
+public final class BloomFilterBlob {
+
+    private final byte[] bytes;
+
+    public BloomFilterBlob(final byte[] bytes) {
+        this.bytes = bytes;
+    }
+
+    public BloomFilter toBloomFilter() {
+        final BloomFilter filter;
+        try (ByteArrayInputStream bais = new ByteArrayInputStream(bytes)) {
+            filter = BloomFilter.readFrom(bais);
+        }
+        catch (IOException e) {
+            throw new RuntimeException("Error reading bytes to filter: " + e.getMessage());
+        }
+        return filter;
+    }
+
+    @Override
+    public boolean equals(final Object object) {
+        if (this == object)
+            return true;
+        if (object == null)
+            return false;
+        if (object.getClass() != this.getClass())
+            return false;
+        final BloomFilterBlob cast = (BloomFilterBlob) object;
+        return Arrays.equals(this.bytes, cast.bytes);
+    }
+
+    @Override
+    public int hashCode() {
+        return Arrays.hashCode(bytes);
+    }
+}

src/main/java/com/teragrep/pth10/steps/teragrep/bloomfilter/BloomFilterForeachPartitionFunction.java

@@ -51,6 +51,7 @@
 
 import java.sql.Connection;
 import java.util.Iterator;
+import java.util.Objects;
 
 public final class BloomFilterForeachPartitionFunction implements ForeachPartitionFunction<Row> {
 
@@ -101,4 +102,18 @@ public void call(final Iterator<Row> iter) throws Exception {
             conn.commit();
         }
     }
+
+    @Override
+    public boolean equals(final Object o) {
+        if (o == null || getClass() != o.getClass())
+            return false;
+        final BloomFilterForeachPartitionFunction cast = (BloomFilterForeachPartitionFunction) o;
+        return filterTypes.equals(cast.filterTypes) && lazyConnection.equals(cast.lazyConnection)
+                && overwrite == cast.overwrite && tableName.equals(cast.tableName) && regex.equals(cast.regex);
+    }
+
+    @Override
+    public int hashCode() {
+        return Objects.hash(filterTypes, lazyConnection, overwrite, tableName, regex);
+    }
 }

src/main/java/com/teragrep/pth10/steps/teragrep/bloomfilter/TeragrepBloomFilter.java

@@ -79,7 +79,7 @@ public TeragrepBloomFilter(
             String tableName,
             String regex
     ) {
-        this(partition, new ToBloomFilter(bytes), connection, filterTypes, tableName, regex);
+        this(partition, new BloomFilterBlob(bytes).toBloomFilter(), connection, filterTypes, tableName, regex);
     }
 
     public TeragrepBloomFilter(