You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@lucene.apache.org by "donnerpeter (via GitHub)" <gi...@apache.org> on 2023/11/22 11:34:18 UTC

[PR] hunspell: allow in-memory entry sorting for faster dictionary loading [lucene]

donnerpeter opened a new pull request, #12834:
URL: https://github.com/apache/lucene/pull/12834

   ### Description
   
   <!--
   If this is your first contribution to Lucene, please make sure you have reviewed the contribution guide.
   https://github.com/apache/lucene/blob/main/CONTRIBUTING.md
   -->
   


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org
For additional commands, e-mail: issues-help@lucene.apache.org


Re: [PR] hunspell: allow in-memory entry sorting for faster dictionary loading [lucene]

Posted by "donnerpeter (via GitHub)" <gi...@apache.org>.
donnerpeter merged PR #12834:
URL: https://github.com/apache/lucene/pull/12834


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org
For additional commands, e-mail: issues-help@lucene.apache.org


Re: [PR] hunspell: allow in-memory entry sorting for faster dictionary loading [lucene]

Posted by "donnerpeter (via GitHub)" <gi...@apache.org>.
donnerpeter commented on code in PR #12834:
URL: https://github.com/apache/lucene/pull/12834#discussion_r1402703007


##########
lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/SortingStrategy.java:
##########
@@ -0,0 +1,180 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+package org.apache.lucene.analysis.hunspell;
+
+import java.io.Closeable;
+import java.io.IOException;
+import java.nio.charset.StandardCharsets;
+import java.util.ArrayList;
+import java.util.Comparator;
+import java.util.List;
+import org.apache.lucene.codecs.CodecUtil;
+import org.apache.lucene.store.Directory;
+import org.apache.lucene.store.IOContext;
+import org.apache.lucene.store.IndexOutput;
+import org.apache.lucene.util.BytesRef;
+import org.apache.lucene.util.BytesRefComparator;
+import org.apache.lucene.util.IOUtils;
+import org.apache.lucene.util.OfflineSorter;
+import org.apache.lucene.util.OfflineSorter.ByteSequencesReader;
+import org.apache.lucene.util.OfflineSorter.ByteSequencesWriter;
+
+/**
+ * The strategy defining how a Hunspell dictionary should be loaded, with different tradeoffs. The
+ * entries should be sorted in a special way, and this can be done either in-memory (faster, but
+ * temporarily allocating more memory) or using disk (slower, but not needing much memory).
+ *
+ * @see #offline(Directory, String)
+ * @see #inMemory()
+ */
+public abstract class SortingStrategy {
+
+  abstract EntryAccumulator start() throws IOException;
+
+  interface EntryAccumulator {
+
+    void addEntry(String entry) throws IOException;
+
+    EntrySupplier finishAndSort() throws IOException;
+  }
+
+  interface EntrySupplier extends Closeable {
+    int wordCount();
+
+    String next() throws IOException;
+  }
+
+  /**
+   * An "offline" strategy that creates temporary files in the given directory and uses them for
+   * sorting with {@link OfflineSorter}. It's slower than {@link #inMemory()}, but doesn't need to
+   * load the entire dictionary to the memory.
+   */
+  public static SortingStrategy offline(Directory tempDir, String tempFileNamePrefix) {
+    return new SortingStrategy() {
+      @Override
+      EntryAccumulator start() throws IOException {
+        IndexOutput output = tempDir.createTempOutput(tempFileNamePrefix, "dat", IOContext.DEFAULT);
+        ByteSequencesWriter writer = new ByteSequencesWriter(output);
+        return new EntryAccumulator() {
+          int wordCount = 0;
+
+          @Override
+          public void addEntry(String entry) throws IOException {
+            wordCount++;
+            writer.write(entry.getBytes(StandardCharsets.UTF_8));
+          }
+
+          @Override
+          public EntrySupplier finishAndSort() throws IOException {
+            CodecUtil.writeFooter(output);
+            writer.close();
+            String sortedFile = sortWordsOffline();
+            ByteSequencesReader reader =
+                new ByteSequencesReader(tempDir.openChecksumInput(sortedFile), sortedFile);
+            return new EntrySupplier() {
+              boolean success = false;
+
+              @Override
+              public int wordCount() {
+                return wordCount;
+              }
+
+              @Override
+              public String next() throws IOException {
+                BytesRef scratch = reader.next();
+                if (scratch == null) {
+                  success = true;
+                  return null;
+                }
+                return scratch.utf8ToString();
+              }
+
+              @Override
+              public void close() throws IOException {
+                reader.close();
+                if (success) {
+                  tempDir.deleteFile(sortedFile);
+                } else {
+                  IOUtils.deleteFilesIgnoringExceptions(tempDir, sortedFile);

Review Comment:
   If I understand this code correctly, in both branches it deletes only the `sortedFile` inside the `tempDir`, never the dir itself. In the first case it retrhrows the possible deletion exceptions and in the second case there's already an exception in flight, so it suppresses them.



-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org
For additional commands, e-mail: issues-help@lucene.apache.org


Re: [PR] hunspell: allow in-memory entry sorting for faster dictionary loading [lucene]

Posted by "dweiss (via GitHub)" <gi...@apache.org>.
dweiss commented on code in PR #12834:
URL: https://github.com/apache/lucene/pull/12834#discussion_r1402449022


##########
lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/SortingStrategy.java:
##########
@@ -0,0 +1,180 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+package org.apache.lucene.analysis.hunspell;
+
+import java.io.Closeable;
+import java.io.IOException;
+import java.nio.charset.StandardCharsets;
+import java.util.ArrayList;
+import java.util.Comparator;
+import java.util.List;
+import org.apache.lucene.codecs.CodecUtil;
+import org.apache.lucene.store.Directory;
+import org.apache.lucene.store.IOContext;
+import org.apache.lucene.store.IndexOutput;
+import org.apache.lucene.util.BytesRef;
+import org.apache.lucene.util.BytesRefComparator;
+import org.apache.lucene.util.IOUtils;
+import org.apache.lucene.util.OfflineSorter;
+import org.apache.lucene.util.OfflineSorter.ByteSequencesReader;
+import org.apache.lucene.util.OfflineSorter.ByteSequencesWriter;
+
+/**
+ * The strategy defining how a Hunspell dictionary should be loaded, with different tradeoffs. The
+ * entries should be sorted in a special way, and this can be done either in-memory (faster, but
+ * temporarily allocating more memory) or using disk (slower, but not needing much memory).
+ *
+ * @see #offline(Directory, String)
+ * @see #inMemory()
+ */
+public abstract class SortingStrategy {
+
+  abstract EntryAccumulator start() throws IOException;
+
+  interface EntryAccumulator {
+
+    void addEntry(String entry) throws IOException;
+
+    EntrySupplier finishAndSort() throws IOException;
+  }
+
+  interface EntrySupplier extends Closeable {
+    int wordCount();
+
+    String next() throws IOException;

Review Comment:
   Maybe add a javadoc explaining next() behaves like BufferedReader (returning null on eof)?



##########
lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/SortingStrategy.java:
##########
@@ -0,0 +1,180 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+package org.apache.lucene.analysis.hunspell;
+
+import java.io.Closeable;
+import java.io.IOException;
+import java.nio.charset.StandardCharsets;
+import java.util.ArrayList;
+import java.util.Comparator;
+import java.util.List;
+import org.apache.lucene.codecs.CodecUtil;
+import org.apache.lucene.store.Directory;
+import org.apache.lucene.store.IOContext;
+import org.apache.lucene.store.IndexOutput;
+import org.apache.lucene.util.BytesRef;
+import org.apache.lucene.util.BytesRefComparator;
+import org.apache.lucene.util.IOUtils;
+import org.apache.lucene.util.OfflineSorter;
+import org.apache.lucene.util.OfflineSorter.ByteSequencesReader;
+import org.apache.lucene.util.OfflineSorter.ByteSequencesWriter;
+
+/**
+ * The strategy defining how a Hunspell dictionary should be loaded, with different tradeoffs. The
+ * entries should be sorted in a special way, and this can be done either in-memory (faster, but
+ * temporarily allocating more memory) or using disk (slower, but not needing much memory).
+ *
+ * @see #offline(Directory, String)
+ * @see #inMemory()
+ */
+public abstract class SortingStrategy {
+
+  abstract EntryAccumulator start() throws IOException;
+
+  interface EntryAccumulator {
+
+    void addEntry(String entry) throws IOException;
+
+    EntrySupplier finishAndSort() throws IOException;
+  }
+
+  interface EntrySupplier extends Closeable {
+    int wordCount();
+
+    String next() throws IOException;
+  }
+
+  /**
+   * An "offline" strategy that creates temporary files in the given directory and uses them for
+   * sorting with {@link OfflineSorter}. It's slower than {@link #inMemory()}, but doesn't need to
+   * load the entire dictionary to the memory.
+   */
+  public static SortingStrategy offline(Directory tempDir, String tempFileNamePrefix) {
+    return new SortingStrategy() {
+      @Override
+      EntryAccumulator start() throws IOException {
+        IndexOutput output = tempDir.createTempOutput(tempFileNamePrefix, "dat", IOContext.DEFAULT);
+        ByteSequencesWriter writer = new ByteSequencesWriter(output);
+        return new EntryAccumulator() {
+          int wordCount = 0;
+
+          @Override
+          public void addEntry(String entry) throws IOException {
+            wordCount++;
+            writer.write(entry.getBytes(StandardCharsets.UTF_8));
+          }
+
+          @Override
+          public EntrySupplier finishAndSort() throws IOException {
+            CodecUtil.writeFooter(output);
+            writer.close();
+            String sortedFile = sortWordsOffline();
+            ByteSequencesReader reader =
+                new ByteSequencesReader(tempDir.openChecksumInput(sortedFile), sortedFile);
+            return new EntrySupplier() {
+              boolean success = false;
+
+              @Override
+              public int wordCount() {
+                return wordCount;
+              }
+
+              @Override
+              public String next() throws IOException {
+                BytesRef scratch = reader.next();
+                if (scratch == null) {
+                  success = true;
+                  return null;
+                }
+                return scratch.utf8ToString();
+              }
+
+              @Override
+              public void close() throws IOException {
+                reader.close();
+                if (success) {
+                  tempDir.deleteFile(sortedFile);
+                } else {
+                  IOUtils.deleteFilesIgnoringExceptions(tempDir, sortedFile);
+                }
+              }
+            };
+          }
+
+          private String sortWordsOffline() throws IOException {
+            var sorter = new OfflineSorter(tempDir, tempFileNamePrefix, BytesRefComparator.NATURAL);
+
+            String sorted;
+            boolean success = false;
+            try {
+              sorted = sorter.sort(output.getName());
+              success = true;
+            } finally {
+              if (success) {
+                tempDir.deleteFile(output.getName());
+              } else {
+                IOUtils.deleteFilesIgnoringExceptions(tempDir, output.getName());
+              }
+            }
+            return sorted;
+          }
+        };
+      }
+    };
+  }
+
+  /**
+   * The strategy that loads all entries as {@link String} objects and sorts them in the memory. The

Review Comment:
   ```suggestion
      * The strategy that loads all entries as {@link String} objects and sorts them in memory. The
   ```



##########
lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/SortingStrategy.java:
##########
@@ -0,0 +1,180 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+package org.apache.lucene.analysis.hunspell;
+
+import java.io.Closeable;
+import java.io.IOException;
+import java.nio.charset.StandardCharsets;
+import java.util.ArrayList;
+import java.util.Comparator;
+import java.util.List;
+import org.apache.lucene.codecs.CodecUtil;
+import org.apache.lucene.store.Directory;
+import org.apache.lucene.store.IOContext;
+import org.apache.lucene.store.IndexOutput;
+import org.apache.lucene.util.BytesRef;
+import org.apache.lucene.util.BytesRefComparator;
+import org.apache.lucene.util.IOUtils;
+import org.apache.lucene.util.OfflineSorter;
+import org.apache.lucene.util.OfflineSorter.ByteSequencesReader;
+import org.apache.lucene.util.OfflineSorter.ByteSequencesWriter;
+
+/**
+ * The strategy defining how a Hunspell dictionary should be loaded, with different tradeoffs. The
+ * entries should be sorted in a special way, and this can be done either in-memory (faster, but
+ * temporarily allocating more memory) or using disk (slower, but not needing much memory).
+ *
+ * @see #offline(Directory, String)
+ * @see #inMemory()
+ */
+public abstract class SortingStrategy {

Review Comment:
   I think it's exactly what it is - can't think of a better name?



##########
lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/SortingStrategy.java:
##########
@@ -0,0 +1,180 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+package org.apache.lucene.analysis.hunspell;
+
+import java.io.Closeable;
+import java.io.IOException;
+import java.nio.charset.StandardCharsets;
+import java.util.ArrayList;
+import java.util.Comparator;
+import java.util.List;
+import org.apache.lucene.codecs.CodecUtil;
+import org.apache.lucene.store.Directory;
+import org.apache.lucene.store.IOContext;
+import org.apache.lucene.store.IndexOutput;
+import org.apache.lucene.util.BytesRef;
+import org.apache.lucene.util.BytesRefComparator;
+import org.apache.lucene.util.IOUtils;
+import org.apache.lucene.util.OfflineSorter;
+import org.apache.lucene.util.OfflineSorter.ByteSequencesReader;
+import org.apache.lucene.util.OfflineSorter.ByteSequencesWriter;
+
+/**
+ * The strategy defining how a Hunspell dictionary should be loaded, with different tradeoffs. The
+ * entries should be sorted in a special way, and this can be done either in-memory (faster, but
+ * temporarily allocating more memory) or using disk (slower, but not needing much memory).
+ *
+ * @see #offline(Directory, String)
+ * @see #inMemory()
+ */
+public abstract class SortingStrategy {
+
+  abstract EntryAccumulator start() throws IOException;
+
+  interface EntryAccumulator {
+
+    void addEntry(String entry) throws IOException;
+
+    EntrySupplier finishAndSort() throws IOException;
+  }
+
+  interface EntrySupplier extends Closeable {
+    int wordCount();
+
+    String next() throws IOException;
+  }
+
+  /**
+   * An "offline" strategy that creates temporary files in the given directory and uses them for
+   * sorting with {@link OfflineSorter}. It's slower than {@link #inMemory()}, but doesn't need to
+   * load the entire dictionary to the memory.
+   */
+  public static SortingStrategy offline(Directory tempDir, String tempFileNamePrefix) {
+    return new SortingStrategy() {
+      @Override
+      EntryAccumulator start() throws IOException {
+        IndexOutput output = tempDir.createTempOutput(tempFileNamePrefix, "dat", IOContext.DEFAULT);
+        ByteSequencesWriter writer = new ByteSequencesWriter(output);
+        return new EntryAccumulator() {
+          int wordCount = 0;
+
+          @Override
+          public void addEntry(String entry) throws IOException {
+            wordCount++;
+            writer.write(entry.getBytes(StandardCharsets.UTF_8));
+          }
+
+          @Override
+          public EntrySupplier finishAndSort() throws IOException {
+            CodecUtil.writeFooter(output);
+            writer.close();
+            String sortedFile = sortWordsOffline();
+            ByteSequencesReader reader =
+                new ByteSequencesReader(tempDir.openChecksumInput(sortedFile), sortedFile);
+            return new EntrySupplier() {
+              boolean success = false;
+
+              @Override
+              public int wordCount() {
+                return wordCount;
+              }
+
+              @Override
+              public String next() throws IOException {
+                BytesRef scratch = reader.next();
+                if (scratch == null) {
+                  success = true;
+                  return null;
+                }
+                return scratch.utf8ToString();
+              }
+
+              @Override
+              public void close() throws IOException {
+                reader.close();
+                if (success) {
+                  tempDir.deleteFile(sortedFile);
+                } else {
+                  IOUtils.deleteFilesIgnoringExceptions(tempDir, sortedFile);

Review Comment:
   I see it's just moved existing code... but still - trying to figure out why you're deleting tempDir here but only on failure?... Should we try to wipe the tempDir at all? I'd be tempted to do only:
   ```
   IOUtils.closeAll(reader, () -> tempDir.deleteFile(sortedFile)). 
   ```
   If one or the other fails, so be it - it'll rethrow an exception.



##########
lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/SortingStrategy.java:
##########
@@ -0,0 +1,180 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+package org.apache.lucene.analysis.hunspell;
+
+import java.io.Closeable;
+import java.io.IOException;
+import java.nio.charset.StandardCharsets;
+import java.util.ArrayList;
+import java.util.Comparator;
+import java.util.List;
+import org.apache.lucene.codecs.CodecUtil;
+import org.apache.lucene.store.Directory;
+import org.apache.lucene.store.IOContext;
+import org.apache.lucene.store.IndexOutput;
+import org.apache.lucene.util.BytesRef;
+import org.apache.lucene.util.BytesRefComparator;
+import org.apache.lucene.util.IOUtils;
+import org.apache.lucene.util.OfflineSorter;
+import org.apache.lucene.util.OfflineSorter.ByteSequencesReader;
+import org.apache.lucene.util.OfflineSorter.ByteSequencesWriter;
+
+/**
+ * The strategy defining how a Hunspell dictionary should be loaded, with different tradeoffs. The
+ * entries should be sorted in a special way, and this can be done either in-memory (faster, but
+ * temporarily allocating more memory) or using disk (slower, but not needing much memory).
+ *
+ * @see #offline(Directory, String)
+ * @see #inMemory()
+ */
+public abstract class SortingStrategy {
+
+  abstract EntryAccumulator start() throws IOException;
+
+  interface EntryAccumulator {
+
+    void addEntry(String entry) throws IOException;
+
+    EntrySupplier finishAndSort() throws IOException;
+  }
+
+  interface EntrySupplier extends Closeable {
+    int wordCount();
+
+    String next() throws IOException;
+  }
+
+  /**
+   * An "offline" strategy that creates temporary files in the given directory and uses them for
+   * sorting with {@link OfflineSorter}. It's slower than {@link #inMemory()}, but doesn't need to
+   * load the entire dictionary to the memory.

Review Comment:
   ```suggestion
      * load the entire dictionary into memory.
   ```



-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org
For additional commands, e-mail: issues-help@lucene.apache.org


Re: [PR] hunspell: allow in-memory entry sorting for faster dictionary loading [lucene]

Posted by "dweiss (via GitHub)" <gi...@apache.org>.
dweiss commented on code in PR #12834:
URL: https://github.com/apache/lucene/pull/12834#discussion_r1402711522


##########
lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/SortingStrategy.java:
##########
@@ -0,0 +1,180 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+package org.apache.lucene.analysis.hunspell;
+
+import java.io.Closeable;
+import java.io.IOException;
+import java.nio.charset.StandardCharsets;
+import java.util.ArrayList;
+import java.util.Comparator;
+import java.util.List;
+import org.apache.lucene.codecs.CodecUtil;
+import org.apache.lucene.store.Directory;
+import org.apache.lucene.store.IOContext;
+import org.apache.lucene.store.IndexOutput;
+import org.apache.lucene.util.BytesRef;
+import org.apache.lucene.util.BytesRefComparator;
+import org.apache.lucene.util.IOUtils;
+import org.apache.lucene.util.OfflineSorter;
+import org.apache.lucene.util.OfflineSorter.ByteSequencesReader;
+import org.apache.lucene.util.OfflineSorter.ByteSequencesWriter;
+
+/**
+ * The strategy defining how a Hunspell dictionary should be loaded, with different tradeoffs. The
+ * entries should be sorted in a special way, and this can be done either in-memory (faster, but
+ * temporarily allocating more memory) or using disk (slower, but not needing much memory).
+ *
+ * @see #offline(Directory, String)
+ * @see #inMemory()
+ */
+public abstract class SortingStrategy {
+
+  abstract EntryAccumulator start() throws IOException;
+
+  interface EntryAccumulator {
+
+    void addEntry(String entry) throws IOException;
+
+    EntrySupplier finishAndSort() throws IOException;
+  }
+
+  interface EntrySupplier extends Closeable {
+    int wordCount();
+
+    String next() throws IOException;
+  }
+
+  /**
+   * An "offline" strategy that creates temporary files in the given directory and uses them for
+   * sorting with {@link OfflineSorter}. It's slower than {@link #inMemory()}, but doesn't need to
+   * load the entire dictionary to the memory.
+   */
+  public static SortingStrategy offline(Directory tempDir, String tempFileNamePrefix) {
+    return new SortingStrategy() {
+      @Override
+      EntryAccumulator start() throws IOException {
+        IndexOutput output = tempDir.createTempOutput(tempFileNamePrefix, "dat", IOContext.DEFAULT);
+        ByteSequencesWriter writer = new ByteSequencesWriter(output);
+        return new EntryAccumulator() {
+          int wordCount = 0;
+
+          @Override
+          public void addEntry(String entry) throws IOException {
+            wordCount++;
+            writer.write(entry.getBytes(StandardCharsets.UTF_8));
+          }
+
+          @Override
+          public EntrySupplier finishAndSort() throws IOException {
+            CodecUtil.writeFooter(output);
+            writer.close();
+            String sortedFile = sortWordsOffline();
+            ByteSequencesReader reader =
+                new ByteSequencesReader(tempDir.openChecksumInput(sortedFile), sortedFile);
+            return new EntrySupplier() {
+              boolean success = false;
+
+              @Override
+              public int wordCount() {
+                return wordCount;
+              }
+
+              @Override
+              public String next() throws IOException {
+                BytesRef scratch = reader.next();
+                if (scratch == null) {
+                  success = true;
+                  return null;
+                }
+                return scratch.utf8ToString();
+              }
+
+              @Override
+              public void close() throws IOException {
+                reader.close();
+                if (success) {
+                  tempDir.deleteFile(sortedFile);
+                } else {
+                  IOUtils.deleteFilesIgnoringExceptions(tempDir, sortedFile);

Review Comment:
   ```
   IOUtils.deleteFilesIgnoringExceptions(tempDir, sortedFile);
   ```
   This will delete tempDir, won't it?



-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org
For additional commands, e-mail: issues-help@lucene.apache.org


Re: [PR] hunspell: allow in-memory entry sorting for faster dictionary loading [lucene]

Posted by "donnerpeter (via GitHub)" <gi...@apache.org>.
donnerpeter commented on code in PR #12834:
URL: https://github.com/apache/lucene/pull/12834#discussion_r1401913469


##########
lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/SortingStrategy.java:
##########
@@ -0,0 +1,180 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+package org.apache.lucene.analysis.hunspell;
+
+import java.io.Closeable;
+import java.io.IOException;
+import java.nio.charset.StandardCharsets;
+import java.util.ArrayList;
+import java.util.Comparator;
+import java.util.List;
+import org.apache.lucene.codecs.CodecUtil;
+import org.apache.lucene.store.Directory;
+import org.apache.lucene.store.IOContext;
+import org.apache.lucene.store.IndexOutput;
+import org.apache.lucene.util.BytesRef;
+import org.apache.lucene.util.BytesRefComparator;
+import org.apache.lucene.util.IOUtils;
+import org.apache.lucene.util.OfflineSorter;
+import org.apache.lucene.util.OfflineSorter.ByteSequencesReader;
+import org.apache.lucene.util.OfflineSorter.ByteSequencesWriter;
+
+/**
+ * The strategy defining how a Hunspell dictionary should be loaded, with different tradeoffs. The
+ * entries should be sorted in a special way, and this can be done either in-memory (faster, but
+ * temporarily allocating more memory) or using disk (slower, but not needing much memory).
+ *
+ * @see #offline(Directory, String)
+ * @see #inMemory()
+ */
+public abstract class SortingStrategy {

Review Comment:
   Better name ideas are welcome



-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org
For additional commands, e-mail: issues-help@lucene.apache.org


Re: [PR] hunspell: allow in-memory entry sorting for faster dictionary loading [lucene]

Posted by "dweiss (via GitHub)" <gi...@apache.org>.
dweiss commented on code in PR #12834:
URL: https://github.com/apache/lucene/pull/12834#discussion_r1402712071


##########
lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/SortingStrategy.java:
##########
@@ -0,0 +1,180 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+package org.apache.lucene.analysis.hunspell;
+
+import java.io.Closeable;
+import java.io.IOException;
+import java.nio.charset.StandardCharsets;
+import java.util.ArrayList;
+import java.util.Comparator;
+import java.util.List;
+import org.apache.lucene.codecs.CodecUtil;
+import org.apache.lucene.store.Directory;
+import org.apache.lucene.store.IOContext;
+import org.apache.lucene.store.IndexOutput;
+import org.apache.lucene.util.BytesRef;
+import org.apache.lucene.util.BytesRefComparator;
+import org.apache.lucene.util.IOUtils;
+import org.apache.lucene.util.OfflineSorter;
+import org.apache.lucene.util.OfflineSorter.ByteSequencesReader;
+import org.apache.lucene.util.OfflineSorter.ByteSequencesWriter;
+
+/**
+ * The strategy defining how a Hunspell dictionary should be loaded, with different tradeoffs. The
+ * entries should be sorted in a special way, and this can be done either in-memory (faster, but
+ * temporarily allocating more memory) or using disk (slower, but not needing much memory).
+ *
+ * @see #offline(Directory, String)
+ * @see #inMemory()
+ */
+public abstract class SortingStrategy {
+
+  abstract EntryAccumulator start() throws IOException;
+
+  interface EntryAccumulator {
+
+    void addEntry(String entry) throws IOException;
+
+    EntrySupplier finishAndSort() throws IOException;
+  }
+
+  interface EntrySupplier extends Closeable {
+    int wordCount();
+
+    String next() throws IOException;
+  }
+
+  /**
+   * An "offline" strategy that creates temporary files in the given directory and uses them for
+   * sorting with {@link OfflineSorter}. It's slower than {@link #inMemory()}, but doesn't need to
+   * load the entire dictionary to the memory.
+   */
+  public static SortingStrategy offline(Directory tempDir, String tempFileNamePrefix) {
+    return new SortingStrategy() {
+      @Override
+      EntryAccumulator start() throws IOException {
+        IndexOutput output = tempDir.createTempOutput(tempFileNamePrefix, "dat", IOContext.DEFAULT);
+        ByteSequencesWriter writer = new ByteSequencesWriter(output);
+        return new EntryAccumulator() {
+          int wordCount = 0;
+
+          @Override
+          public void addEntry(String entry) throws IOException {
+            wordCount++;
+            writer.write(entry.getBytes(StandardCharsets.UTF_8));
+          }
+
+          @Override
+          public EntrySupplier finishAndSort() throws IOException {
+            CodecUtil.writeFooter(output);
+            writer.close();
+            String sortedFile = sortWordsOffline();
+            ByteSequencesReader reader =
+                new ByteSequencesReader(tempDir.openChecksumInput(sortedFile), sortedFile);
+            return new EntrySupplier() {
+              boolean success = false;
+
+              @Override
+              public int wordCount() {
+                return wordCount;
+              }
+
+              @Override
+              public String next() throws IOException {
+                BytesRef scratch = reader.next();
+                if (scratch == null) {
+                  success = true;
+                  return null;
+                }
+                return scratch.utf8ToString();
+              }
+
+              @Override
+              public void close() throws IOException {
+                reader.close();
+                if (success) {
+                  tempDir.deleteFile(sortedFile);
+                } else {
+                  IOUtils.deleteFilesIgnoringExceptions(tempDir, sortedFile);

Review Comment:
   Oh, I looked at the code of deleteFilesIgnoringExceptions, nevermind - you're right.



-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org
For additional commands, e-mail: issues-help@lucene.apache.org