feat: introduce TreeEntry hierarchy for hot tree representation

README.md

@@ -122,7 +122,7 @@ public class MyTranslator implements BlobTranslator {
 - `--stream-size-limit=<num>{,K,M,G}`: increase the stream size limit.
 - `--no-notes`: Stop noting the source commit ID to the commits in the target repository.
 - `--no-pack`: Stop packing objects after transformation finished.
-- `--alternates`: Share source objects via Git alternates to skip writing unchanged objects. Significantly speeds up transformations where many objects are unchanged. The target repository will depend on the source's object store until repacked (`git repack -a -d`).
+- `--alternates`: Share source objects via Git alternates to skip writing unchanged objects, which speeds up transformations where many objects are unchanged. The target repository will depend on the source's object store until repacked.
 - `--no-composite`: Stop composing multiple blob translators.
 - `--extra-attributes`: Allow opportunity to rewrite the encoding and the signature fields in commits.
 - `--cache=<level>,...`: Specify the object types for caching (`commit`, `blob`, `tree`. See [Incremental transformation](#incremental-transformation) for the details). Default: none. `commit` is recommended.
@@ -266,6 +266,7 @@ Anonymizes filenames, blob content, commit messages, branch/tag names, and autho
 Options:
 - `--all`: Enable all anonymization options.
 - `--tree`: Anonymize directory names.
+- `--link`: Anonymize link names.
 - `--blob`: Anonymize file names.
 - `--content`: Anonymize file contents.
 - `--message`: Anonymize commit/tag messages.

src/main/java/jp/ac/titech/c/se/stein/app/Anonymize.java

@@ -5,7 +5,9 @@
 
 import jp.ac.titech.c.se.stein.entry.Entry;
 import jp.ac.titech.c.se.stein.entry.AnyHotEntry;
-import jp.ac.titech.c.se.stein.entry.HotEntry;
+import jp.ac.titech.c.se.stein.entry.AnyColdEntry;
+import jp.ac.titech.c.se.stein.entry.BlobEntry;
+import jp.ac.titech.c.se.stein.entry.TreeEntry;
 import jp.ac.titech.c.se.stein.util.HashUtils;
 import lombok.ToString;
 import lombok.extern.slf4j.Slf4j;
@@ -28,6 +30,9 @@ public class Anonymize extends RepositoryRewriter {
     @Option(names = "--tree", description = "anonymize tree name")
     protected boolean isTreeNameEnabled;
 
+    @Option(names = "--link", description = "anonymize link name")
+    protected boolean isLinkNameEnabled;
+
     @Option(names = "--blob", description = "anonymize blob name")
     protected boolean isBlobNameEnabled;
 
@@ -53,6 +58,7 @@ public class Anonymize extends RepositoryRewriter {
     @Option(names = "--all", description = "anonymize all")
     protected void setAllEnabled(boolean isEnabled) {
         isTreeNameEnabled = isEnabled;
+        isLinkNameEnabled = isEnabled;
         isBlobNameEnabled = isEnabled;
         isBlobContentEnabled = isEnabled;
         isMessageEnabled = isEnabled;
@@ -91,6 +97,8 @@ public String convert(final String name) {
 
     private final NameMap treeNameMap = new NameMap("directory", "t");
 
+    private final NameMap linkNameMap = new NameMap("link", "l");
+
     private final NameMap blobNameMap = new NameMap("file", "f");
 
     private final NameMap branchNameMap = new NameMap("branch", "b");
@@ -105,7 +113,7 @@ public String rewriteMessage(final String message, final Context c) {
     }
 
     @Override
-    public AnyHotEntry rewriteBlobEntry(HotEntry entry, final Context c) {
+    public AnyHotEntry rewriteBlobEntry(BlobEntry entry, final Context c) {
         if (isBlobContentEnabled) {
             entry = entry.update(entry.getId().name());
         }
@@ -116,12 +124,21 @@ public AnyHotEntry rewriteBlobEntry(HotEntry entry, final Context c) {
     }
 
     @Override
-    public String rewriteName(final String name, final Context c) {
-        final Entry entry = c.getEntry();
-        if (entry.isTree()) {
-            return isTreeNameEnabled ? treeNameMap.convert(name) : name;
+    protected AnyColdEntry rewriteTreeEntry(TreeEntry entry, Context c) {
+        final AnyColdEntry result = super.rewriteTreeEntry(entry, c);
+        if (isTreeNameEnabled && result instanceof Entry) {
+            final Entry e = (Entry) result;
+            return Entry.of(e.mode, treeNameMap.convert(e.name), e.id, e.directory);
         }
-        return name;
+        return result;
+    }
+
+    @Override
+    protected AnyColdEntry rewriteLinkEntry(Entry entry, Context c) {
+        if (isLinkNameEnabled) {
+            return Entry.of(entry.mode, linkNameMap.convert(entry.name), entry.id, entry.directory);
+        }
+        return entry;
     }
 
     @Override

src/main/java/jp/ac/titech/c/se/stein/app/blob/ConvertBlob.java

@@ -3,6 +3,7 @@
 import jp.ac.titech.c.se.stein.core.Context;
 import jp.ac.titech.c.se.stein.core.Try;
 import jp.ac.titech.c.se.stein.entry.AnyHotEntry;
+import jp.ac.titech.c.se.stein.entry.BlobEntry;
 import jp.ac.titech.c.se.stein.entry.HotEntry;
 import jp.ac.titech.c.se.stein.rewriter.BlobTranslator;
 import jp.ac.titech.c.se.stein.rewriter.NameFilter;
@@ -77,7 +78,7 @@ protected String[] makeCommand(final String inputFile) {
     }
 
     @Override
-    public AnyHotEntry rewriteBlobEntry(final HotEntry entry, final Context c) {
+    public AnyHotEntry rewriteBlobEntry(final BlobEntry entry, final Context c) {
         if (!filter.accept(entry)) {
             return entry;
         }
@@ -92,7 +93,7 @@ public AnyHotEntry rewriteBlobEntry(final HotEntry entry, final Context c) {
         }
     }
 
-    protected HotEntry processCommandlineFilter(final HotEntry entry, final Context c) {
+    protected BlobEntry processCommandlineFilter(final BlobEntry entry, final Context c) {
         try {
             final Process proc = new ProcessBuilder()
                     .command(makeCommand(entry.getName()))
@@ -131,7 +132,7 @@ protected HotEntry processCommandlineFilter(final HotEntry entry, final Context
         }
     }
 
-    protected AnyHotEntry processCommandline(final HotEntry entry, final Context c) {
+    protected AnyHotEntry processCommandline(final BlobEntry entry, final Context c) {
         try (final TemporaryFile tmp = TemporaryFile.directoryOf("_stein")) {
             // write input
             final Path inputPath = tmp.getPath().resolve(entry.getName());
@@ -163,7 +164,7 @@ protected AnyHotEntry processCommandline(final HotEntry entry, final Context c)
         }
     }
 
-    protected HotEntry processEndpoint(final HotEntry entry, final Context c) {
+    protected BlobEntry processEndpoint(final BlobEntry entry, final Context c) {
         try {
             final HttpURLConnection conn = (HttpURLConnection) options.endpoint.openConnection();
             conn.setRequestMethod("POST");

src/main/java/jp/ac/titech/c/se/stein/app/blob/Cregit.java

@@ -2,6 +2,7 @@
 
 import jp.ac.titech.c.se.stein.core.Context;
 import jp.ac.titech.c.se.stein.entry.AnyHotEntry;
+import jp.ac.titech.c.se.stein.entry.BlobEntry;
 import jp.ac.titech.c.se.stein.entry.HotEntry;
 import jp.ac.titech.c.se.stein.rewriter.BlobTranslator;
 import jp.ac.titech.c.se.stein.rewriter.NameFilter;
@@ -81,7 +82,7 @@ protected void setLanguage(final String language) {
     protected String language;
 
     @Override
-    public AnyHotEntry rewriteBlobEntry(HotEntry entry, Context c) {
+    public AnyHotEntry rewriteBlobEntry(BlobEntry entry, Context c) {
         if (!filter.accept(entry)) {
             return entry;
         }

src/main/java/jp/ac/titech/c/se/stein/app/blob/FilterBlob.java

@@ -1,7 +1,7 @@
 package jp.ac.titech.c.se.stein.app.blob;
 
 import jp.ac.titech.c.se.stein.entry.AnyHotEntry;
-import jp.ac.titech.c.se.stein.entry.HotEntry;
+import jp.ac.titech.c.se.stein.entry.BlobEntry;
 import jp.ac.titech.c.se.stein.rewriter.BlobTranslator;
 import jp.ac.titech.c.se.stein.rewriter.NameFilter;
 import lombok.ToString;
@@ -30,7 +30,7 @@ public class FilterBlob implements BlobTranslator {
     protected long maxSize = -1L;
 
     @Override
-    public AnyHotEntry rewriteBlobEntry(final HotEntry entry, final Context c) {
+    public AnyHotEntry rewriteBlobEntry(final BlobEntry entry, final Context c) {
         // name
         if (nameFilter.getPatterns() != null) {
             if (!nameFilter.accept(entry)) {

src/main/java/jp/ac/titech/c/se/stein/app/blob/Historage.java

@@ -4,6 +4,7 @@
 import com.google.gson.reflect.TypeToken;
 import jp.ac.titech.c.se.stein.core.*;
 import jp.ac.titech.c.se.stein.entry.AnyHotEntry;
+import jp.ac.titech.c.se.stein.entry.BlobEntry;
 import jp.ac.titech.c.se.stein.entry.HotEntry;
 import jp.ac.titech.c.se.stein.rewriter.BlobTranslator;
 import jp.ac.titech.c.se.stein.rewriter.NameFilter;
@@ -59,7 +60,7 @@ public class Historage implements BlobTranslator {
     protected Set<String> moduleKinds;
 
     @Override
-    public AnyHotEntry rewriteBlobEntry(final HotEntry entry, final Context c) {
+    public AnyHotEntry rewriteBlobEntry(final BlobEntry entry, final Context c) {
         if (!filter.accept(entry)) {
             return entry;
         }

src/main/java/jp/ac/titech/c/se/stein/app/blob/HistorageViaJDT.java

@@ -9,6 +9,7 @@
 import jp.ac.titech.c.se.stein.entry.AnyHotEntry;
 import jp.ac.titech.c.se.stein.core.SourceText;
 import jp.ac.titech.c.se.stein.core.SourceText.Fragment;
+import jp.ac.titech.c.se.stein.entry.BlobEntry;
 import jp.ac.titech.c.se.stein.entry.HotEntry;
 import jp.ac.titech.c.se.stein.rewriter.BlobTranslator;
 import jp.ac.titech.c.se.stein.rewriter.NameFilter;
@@ -86,7 +87,7 @@ public class HistorageViaJDT implements BlobTranslator {
     protected boolean parsable = false;
 
     @Override
-    public AnyHotEntry rewriteBlobEntry(final HotEntry entry, final Context c) {
+    public AnyHotEntry rewriteBlobEntry(final BlobEntry entry, final Context c) {
         if (!JAVA.accept(entry)) {
             return entry;
         }

src/main/java/jp/ac/titech/c/se/stein/app/blob/Tokenize.java

@@ -3,7 +3,7 @@
 import jp.ac.titech.c.se.stein.core.Context;
 import jp.ac.titech.c.se.stein.entry.AnyHotEntry;
 import jp.ac.titech.c.se.stein.core.SourceText;
-import jp.ac.titech.c.se.stein.entry.HotEntry;
+import jp.ac.titech.c.se.stein.entry.BlobEntry;
 import jp.ac.titech.c.se.stein.rewriter.BlobTranslator;
 import lombok.ToString;
 import picocli.CommandLine.Command;
@@ -30,7 +30,7 @@ public class Tokenize implements BlobTranslator {
     ));
 
     @Override
-    public AnyHotEntry rewriteBlobEntry(final HotEntry entry, final Context c) {
+    public AnyHotEntry rewriteBlobEntry(final BlobEntry entry, final Context c) {
         final String text = SourceText.of(entry.getBlob()).getContent();
         return entry.update(encode(text));
     }

src/main/java/jp/ac/titech/c/se/stein/app/blob/TokenizeViaJDT.java

@@ -4,7 +4,7 @@
 
 import jp.ac.titech.c.se.stein.entry.AnyHotEntry;
 import jp.ac.titech.c.se.stein.core.SourceText;
-import jp.ac.titech.c.se.stein.entry.HotEntry;
+import jp.ac.titech.c.se.stein.entry.BlobEntry;
 import jp.ac.titech.c.se.stein.rewriter.BlobTranslator;
 import lombok.ToString;
 import lombok.extern.slf4j.Slf4j;
@@ -27,7 +27,7 @@
 @Command(name = "@tokenize-jdt", description = "Encode Java source files to linetoken format via JDT")
 public class TokenizeViaJDT implements BlobTranslator {
     @Override
-    public AnyHotEntry rewriteBlobEntry(final HotEntry entry, final Context c) {
+    public AnyHotEntry rewriteBlobEntry(final BlobEntry entry, final Context c) {
         if (!HistorageViaJDT.JAVA.accept(entry)) {
             return entry;
         }

src/main/java/jp/ac/titech/c/se/stein/app/blob/Untokenize.java

@@ -3,7 +3,7 @@
 import jp.ac.titech.c.se.stein.core.Context;
 import jp.ac.titech.c.se.stein.entry.AnyHotEntry;
 import jp.ac.titech.c.se.stein.core.SourceText;
-import jp.ac.titech.c.se.stein.entry.HotEntry;
+import jp.ac.titech.c.se.stein.entry.BlobEntry;
 import jp.ac.titech.c.se.stein.rewriter.BlobTranslator;
 import jp.ac.titech.c.se.stein.rewriter.NameFilter;
 import lombok.ToString;
@@ -23,7 +23,7 @@ public class Untokenize implements BlobTranslator {
     @Mixin
     protected final NameFilter filter = new NameFilter();
     @Override
-    public AnyHotEntry rewriteBlobEntry(final HotEntry entry, final Context c) {
+    public AnyHotEntry rewriteBlobEntry(final BlobEntry entry, final Context c) {
         if (!filter.accept(entry)) {
             return entry;
         }

src/main/java/jp/ac/titech/c/se/stein/entry/AnyColdEntry.java

@@ -33,7 +33,7 @@ public interface AnyColdEntry extends Serializable {
      * Normalizes this entry. A {@link Set} of size 0 becomes {@link Empty},
      * size 1 is unwrapped to its sole {@link Entry}, and others remain as-is.
      */
-    default AnyColdEntry pack() {
+    default AnyColdEntry normalize() {
         return this;
     }
 
@@ -67,7 +67,7 @@ static Empty empty() {
 
     /**
      * A collection of multiple {@link Entry} instances.
-     * Use {@link #pack()} to normalize after construction.
+     * Use {@link #normalize()} to normalize after construction.
      */
     @NoArgsConstructor
     @EqualsAndHashCode
@@ -101,7 +101,7 @@ public String toString() {
         }
 
         @Override
-        public AnyColdEntry pack() {
+        public AnyColdEntry normalize() {
             switch (size()) {
                 case 0:
                     return empty();

src/main/java/jp/ac/titech/c/se/stein/entry/AnyHotEntry.java

@@ -3,6 +3,7 @@
 import jp.ac.titech.c.se.stein.core.Context;
 import jp.ac.titech.c.se.stein.core.RepositoryAccess;
 import lombok.Getter;
+import org.eclipse.jgit.lib.ObjectId;
 
 import java.util.ArrayList;
 import java.util.Arrays;
@@ -29,6 +30,20 @@ public interface AnyHotEntry {
      */
     int size();
 
+    /**
+     * Returns the first entry as a {@link BlobEntry}.
+     */
+    default BlobEntry asBlob() {
+        return (BlobEntry) stream().findFirst().orElseThrow();
+    }
+
+    /**
+     * Returns the first entry as a {@link TreeEntry}.
+     */
+    default TreeEntry asTree() {
+        return (TreeEntry) stream().findFirst().orElseThrow();
+    }
+
     /**
      * Converts this Hot entry to a Cold entry by writing blob data to the target repository.
      */
@@ -98,8 +113,9 @@ public int size() {
         public AnyColdEntry fold(RepositoryAccess target, Context c) {
             return AnyColdEntry.set(stream()
                     .map(e -> e.fold(target, c))
+                    .filter(e -> !e.getId().equals(ObjectId.zeroId()))
                     .collect(Collectors.toList()))
-                    .pack();
+                    .normalize();
         }
 
         @Override