Merge pull request #7 from kortemik/refactor-1

Refactor to byte[] version

.gitignore

@@ -3,3 +3,5 @@
 .flattened-pom.xml
 .idea/**
 rlp_01.iml
+
+src/test/**/internal/

pom.xml

@@ -159,6 +159,9 @@
             <exclude>README.md</exclude>
             <!-- astyle -->
             <exclude>astyle-options.ini</exclude>
+            <!-- perf-test-asset -->
+            <exclude>src/test/resources/base64.txt</exclude>
+            <exclude>src/test/resources/base64-8m.txt</exclude>
           </excludes>
         </configuration>
       </plugin>

src/main/java/com/teragrep/blf_01/ConcatenatedToken.java

@@ -0,0 +1,85 @@
+/*
+ * Teragrep Bloom Filter Library BLF-01
+ * Copyright (C) 2019, 2020, 2021, 2022, 2023 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://github.com/teragrep/teragrep/blob/main/LICENSE>.
+ *
+ *
+ * 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.blf_01;
+
+import java.nio.ByteBuffer;
+import java.util.ArrayList;
+
+public class ConcatenatedToken {
+
+    private ByteBuffer concatenatedBuffer;
+
+    public ConcatenatedToken() {
+        this.concatenatedBuffer = ByteBuffer.allocateDirect(256*1024);
+    }
+
+    byte[] concatenate(ArrayList<Token> tokens) {
+        concatenatedBuffer.clear();
+        for (Token token : tokens) {
+            if (concatenatedBuffer.position() + token.bytes.length >= concatenatedBuffer.capacity()) {
+                int size = 0;
+                for (Token tokenForSize : tokens) {
+                    size = size + tokenForSize.bytes.length;
+                }
+                concatenatedBuffer = extendBuffer(concatenatedBuffer, size);
+            }
+            concatenatedBuffer.put(token.bytes);
+        }
+
+        concatenatedBuffer.flip();
+        byte[] rv = new byte[concatenatedBuffer.remaining()];
+        concatenatedBuffer.get(rv);
+        return rv;
+    }
+
+    private ByteBuffer extendBuffer(ByteBuffer byteBuffer, int size) {
+        ByteBuffer newBuffer = ByteBuffer.allocateDirect(byteBuffer.capacity() + size);
+        byteBuffer.flip();
+        newBuffer.put(byteBuffer);
+        return newBuffer;
+    }
+}

src/main/java/com/teragrep/blf_01/Delimiter.java

@@ -0,0 +1,90 @@
+/*
+ * Teragrep Bloom Filter Library BLF-01
+ * Copyright (C) 2019, 2020, 2021, 2022, 2023 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://github.com/teragrep/teragrep/blob/main/LICENSE>.
+ *
+ *
+ * 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.blf_01;
+
+import java.nio.ByteBuffer;
+import java.nio.charset.StandardCharsets;
+import java.util.HashMap;
+import java.util.Objects;
+
+public class Delimiter {
+    public final boolean isStub;
+
+    public final ByteBuffer delimiterBuffer;
+
+    Delimiter() {
+        this.isStub = true;
+        this.delimiterBuffer = ByteBuffer.allocate(0);
+    }
+
+    Delimiter(String delimiter) {
+        this.isStub = false;
+        byte[] bytes = delimiter.getBytes(StandardCharsets.UTF_8);
+        this.delimiterBuffer = ByteBuffer.allocateDirect(bytes.length);
+        this.delimiterBuffer.put(bytes);
+        this.delimiterBuffer.flip();
+    }
+
+    @Override
+    public boolean equals(Object o) {
+        if (this == o) return true;
+        if (o == null || getClass() != o.getClass()) return false;
+        Delimiter delimiter = (Delimiter) o;
+        return Objects.equals(delimiterBuffer, delimiter.delimiterBuffer);
+    }
+
+    @Override
+    public int hashCode() {
+        return Objects.hash(delimiterBuffer);
+    }
+
+    public HashMap<ByteBuffer, Delimiter> asMap() {
+        HashMap<ByteBuffer, Delimiter> map = new HashMap<>();
+        map.put(delimiterBuffer, this);
+        return map;
+    }
+}

src/test/java/com/teragrep/blf_01/FilterTest.java → src/main/java/com/teragrep/blf_01/Delimiters.java

@@ -1,8 +1,6 @@
-package com.teragrep.blf_01;
-
 /*
  * Teragrep Bloom Filter Library BLF-01
- * Copyright (C) 2019, 2020, 2021, 2022  Suomen Kanuuna Oy
+ * Copyright (C) 2019, 2020, 2021, 2022, 2023 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
@@ -46,15 +44,11 @@
  * a licensee so wish it.
  */
 
-import org.junit.jupiter.api.Test;
-
-import java.io.IOException;
+package com.teragrep.blf_01;
 
-public class FilterTest {
+import java.nio.ByteBuffer;
+import java.util.HashMap;
 
-    @Test
-    public void testFilter() throws IOException {
-        FilterManager filterManager = new FilterManager();
-        filterManager.select();
-    }
+public interface Delimiters {
+    HashMap<ByteBuffer, Delimiter> getDelimiters();
 }

src/main/java/com/teragrep/blf_01/Entanglement.java

@@ -0,0 +1,119 @@
+/*
+ * Teragrep Bloom Filter Library BLF-01
+ * Copyright (C) 2019, 2020, 2021, 2022, 2023 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://github.com/teragrep/teragrep/blob/main/LICENSE>.
+ *
+ *
+ * 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.blf_01;
+
+import java.util.ArrayList;
+import java.util.ListIterator;
+
+public class Entanglement {
+
+    private final ArrayList<Token> endWindowScanTokens;
+    private final ArrayList<Token> allTokens;
+    private final ArrayList<Token> forwardScanTokens;
+    private final ArrayList<Token> backwardsScanTokens;
+    private final ConcatenatedToken concatenatedToken;
+    public Entanglement() {
+        this.endWindowScanTokens = new ArrayList<>(512);
+        this.allTokens = new ArrayList<>(512);
+        this.forwardScanTokens = new ArrayList<>(512);
+        this.backwardsScanTokens = new ArrayList<>(512);
+        this.concatenatedToken = new ConcatenatedToken();
+    }
+
+    public ArrayList<Token> entangle(ArrayList<Token> tokens) {
+        //System.out.println("entangling> " + tokens + " results into " + rv);
+        return startWindowScan(tokens);
+
+    }
+
+
+    /**
+     * Iterates Token list in forward order,
+     * starting from the largest subList and going to smaller ones
+     * and processes reverse order ones within
+     */
+    private ArrayList<Token> startWindowScan(ArrayList<Token> tokenList)  {
+        allTokens.clear();
+        for (int i = 0; i < tokenList.size(); i++) {
+            ListIterator<Token> forwardIterator = tokenList.listIterator(i);
+            // +++++ task
+            forwardScanTokens.clear();
+            while (forwardIterator.hasNext()) {
+                forwardScanTokens.add(forwardIterator.next());
+            }
+            // +++++ subtask endWindowScan
+            allTokens.addAll(endWindowScan(forwardScanTokens));
+            // -----
+            Token concatenated = new Token(concatenatedToken.concatenate(forwardScanTokens));
+            allTokens.add(concatenated);
+            // -----
+        }
+
+        return allTokens;
+    }
+
+    /**
+     * Iterates Token list in reverse order,
+     * starting from the largest subList and going to smaller ones
+     */
+    private ArrayList<Token> endWindowScan(ArrayList<Token> tokenList) {
+        endWindowScanTokens.clear();
+
+        for (int i = tokenList.size() - 1; i > 0; i--) {
+            ListIterator<Token> backwardIterator = tokenList.listIterator(i);
+            // +++++ task
+            backwardsScanTokens.clear();
+            while (backwardIterator.hasPrevious()) {
+                backwardsScanTokens.add(0, backwardIterator.previous());
+            }
+            endWindowScanTokens.add(new Token(concatenatedToken.concatenate(backwardsScanTokens)));
+            // ----- task
+        }
+
+        return endWindowScanTokens;
+    }
+}