Post-pend md5 to prebuilt indexes (#2412)

castorini · Mar 21, 2024 · 1dd9139 · 1dd9139
1 parent d48a40f
commit 1dd9139
Show file tree

Hide file tree

Showing 2 changed files with 6 additions and 1 deletion.
diff --git a/src/main/java/io/anserini/index/IndexInfo.java b/src/main/java/io/anserini/index/IndexInfo.java
@@ -19,7 +19,7 @@
 public enum IndexInfo {
   CACM("cacm",
       "Lucene index of the CACM corpus.",
-      "lucene-index.cacm.tar.gz",
+      "lucene-index.cacm.20221005.252b5e.tar.gz",
       new String[] {
           "https://github.com/castorini/anserini-data/raw/master/CACM/lucene-index.cacm.20221005.252b5e.tar.gz" },
       "cfe14d543c6a27f4d742fb2d0099b8e0"),

diff --git a/src/main/java/io/anserini/util/PrebuiltIndexHandler.java b/src/main/java/io/anserini/util/PrebuiltIndexHandler.java
@@ -196,7 +196,12 @@ public String decompressIndex() throws Exception {
     }
 
     System.out.println("Index decompressed successfully!");
+
+    // postpend md5 to decompressed index
+    Path oldIndexPath = Paths.get(indexFolder);
+    indexFolder += "." + info.md5;
     this.indexFolderPath = Paths.get(indexFolder);
+    Files.move(oldIndexPath, this.indexFolderPath);
     return indexFolder;
   }