LUCENE-4055: move 3.x codec specific stuff out of IW into Lucene3xCodec

git-svn-id: https://svn.apache.org/repos/asf/lucene/dev/branches/lucene4055@1342371 13f79535-47bb-0310-9956-ffa450edef68

lucene/core/src/java/org/apache/lucene/codecs/lucene3x/Lucene3xCodec.java

@@ -18,6 +18,8 @@ package org.apache.lucene.codecs.lucene3x;
  */
 
 import java.io.IOException;
+import java.util.HashSet;
+import java.util.Set;
 
 import org.apache.lucene.codecs.Codec;
 import org.apache.lucene.codecs.DocValuesFormat;
@@ -31,7 +33,9 @@ import org.apache.lucene.codecs.SegmentInfoFormat;
 import org.apache.lucene.codecs.StoredFieldsFormat;
 import org.apache.lucene.codecs.TermVectorsFormat;
 import org.apache.lucene.codecs.lucene40.Lucene40LiveDocsFormat;
+import org.apache.lucene.index.IndexFileNames;
 import org.apache.lucene.index.PerDocWriteState;
+import org.apache.lucene.index.SegmentInfo;
 import org.apache.lucene.index.SegmentInfoPerCommit;
 import org.apache.lucene.index.SegmentReadState;
 import org.apache.lucene.store.Directory;
@@ -123,4 +127,25 @@ public class Lucene3xCodec extends Codec {
   public LiveDocsFormat liveDocsFormat() {
     return liveDocsFormat;
   }
+
+  /** Returns file names for shared doc stores, if any, else
+   * null. */
+  public static Set<String> getDocStoreFiles(SegmentInfo info) {
+    if (Lucene3xSegmentInfoFormat.getDocStoreOffset(info) != -1) {
+      final String dsName = Lucene3xSegmentInfoFormat.getDocStoreSegment(info);
+      Set<String> files = new HashSet<String>();
+      if (Lucene3xSegmentInfoFormat.getDocStoreIsCompoundFile(info)) {
+        files.add(IndexFileNames.segmentFileName(dsName, "", COMPOUND_FILE_STORE_EXTENSION));
+      } else {
+        files.add(IndexFileNames.segmentFileName(dsName, "", Lucene3xStoredFieldsReader.FIELDS_INDEX_EXTENSION));
+        files.add(IndexFileNames.segmentFileName(dsName, "", Lucene3xStoredFieldsReader.FIELDS_EXTENSION));
+        files.add(IndexFileNames.segmentFileName(dsName, "", Lucene3xTermVectorsReader.VECTORS_INDEX_EXTENSION));
+        files.add(IndexFileNames.segmentFileName(dsName, "", Lucene3xTermVectorsReader.VECTORS_FIELDS_EXTENSION));
+        files.add(IndexFileNames.segmentFileName(dsName, "", Lucene3xTermVectorsReader.VECTORS_DOCUMENTS_EXTENSION));
+      }
+      return files;
+    } else {
+      return null;
+    }
+  }
 }

lucene/core/src/java/org/apache/lucene/index/IndexWriter.java

@@ -33,6 +33,7 @@ import java.util.concurrent.atomic.AtomicInteger;
 
 import org.apache.lucene.analysis.Analyzer;
 import org.apache.lucene.codecs.Codec;
+import org.apache.lucene.codecs.lucene3x.Lucene3xCodec;
 import org.apache.lucene.codecs.lucene3x.Lucene3xSegmentInfoFormat;
 import org.apache.lucene.index.DocumentsWriterPerThread.FlushedSegment;
 import org.apache.lucene.index.FieldInfos.FieldNumbers;
@@ -2369,33 +2370,23 @@ public class IndexWriter implements Closeable, TwoPhaseCommit {
       newDsName = segName;
     }
 
-    Set<String> codecDocStoreFiles = new HashSet<String>();
+    Set<String> docStoreFiles3xOnly = Lucene3xCodec.getDocStoreFiles(info.info);
-    final boolean hasSharedDocStore = Lucene3xSegmentInfoFormat.getDocStoreOffset(info.info) != -1;
 
-    // copy the attributes map, we modify it for the preflex case
     final Map<String,String> attributes;
-    if (info.info.attributes() == null) {
+    if (docStoreFiles3xOnly != null) {
-      attributes = new HashMap<String,String>();
-    } else {
-      attributes = new HashMap<String,String>(info.info.attributes());
-    }
-    if (hasSharedDocStore) {
       // only violate the codec this way if it's preflex &
       // shares doc stores
-      // nocommit what to do....
+      // change docStoreSegment to newDsName
-      // cant we determine a file is a 3.x shared doc store file if hasSharedDocStore=true
+      // copy the attributes map, we modify it below:
-      // and the segment prefix != info.info.name instead of this stuff?
+      if (info.info.attributes() == null) {
-      if (Lucene3xSegmentInfoFormat.getDocStoreIsCompoundFile(info.info)) {
+        attributes = new HashMap<String,String>();
-        codecDocStoreFiles.add(IndexFileNames.segmentFileName(dsName, "", "cfx"));
       } else {
-        codecDocStoreFiles.add(IndexFileNames.segmentFileName(dsName, "", "fdt"));
+        attributes = new HashMap<String,String>(info.info.attributes());
-        codecDocStoreFiles.add(IndexFileNames.segmentFileName(dsName, "", "fdx"));
-        codecDocStoreFiles.add(IndexFileNames.segmentFileName(dsName, "", "tvx"));
-        codecDocStoreFiles.add(IndexFileNames.segmentFileName(dsName, "", "tvf"));
-        codecDocStoreFiles.add(IndexFileNames.segmentFileName(dsName, "", "tvd"));
       }
       // change docStoreSegment to newDsName
       attributes.put(Lucene3xSegmentInfoFormat.DS_NAME_KEY, newDsName);
+    } else {
+      attributes = info.info.attributes();
     }
 
     //System.out.println("copy seg=" + info.info.name + " version=" + info.info.getVersion());
@@ -2412,7 +2403,7 @@ public class IndexWriter implements Closeable, TwoPhaseCommit {
     // before writing SegmentInfo:
     for (String file: info.files()) {
       final String newFileName;
-      if (codecDocStoreFiles.contains(file)) {
+      if (docStoreFiles3xOnly != null && docStoreFiles3xOnly.contains(file)) {
         newFileName = newDsName + IndexFileNames.stripSegmentName(file);
       } else {
         newFileName = segName + IndexFileNames.stripSegmentName(file);
@@ -2438,7 +2429,7 @@ public class IndexWriter implements Closeable, TwoPhaseCommit {
     for (String file: info.files()) {
 
       final String newFileName;
-      if (codecDocStoreFiles.contains(file)) {
+      if (docStoreFiles3xOnly != null && docStoreFiles3xOnly.contains(file)) {
         newFileName = newDsName + IndexFileNames.stripSegmentName(file);
         if (dsFilesCopied.contains(newFileName)) {
           continue;