Bugfix in size estimates for filtered tarfile outputs.

[cumulus.git] / scandir.cc

diff --git a/scandir.cc b/scandir.cc
index cfa29ca..f8f6127 100644 (file)
--- a/scandir.cc
+++ b/scandir.cc
@@ -131,22 +131,28 @@ int64_t dumpfile(int fd, dictionary &file_info)
 
         hash.process(block_buf, bytes);
 
-        // tarstore processing
-        LbsObject *o = new LbsObject;
-        o->set_group("data");
-        o->set_data(block_buf, bytes);
-        o->write(tss);
-        object_list.push_back(o->get_name());
-        segment_list.insert(o->get_ref().get_segment());
-
-        // Index this block so it can be used by future snapshots
+        // Either find a copy of this block in an already-existing segment, or
+        // index it so it can be re-used in the future
         SHA1Checksum block_hash;
         block_hash.process(block_buf, bytes);
-        db->StoreObject(o->get_ref(), block_hash.checksum_str(), bytes);
+        string block_csum = block_hash.checksum_str();
+        ObjectReference ref = db->FindObject(block_csum, bytes);
+
+        // Store a copy of the object if one does not yet exist
+        if (ref.get_segment().size() == 0) {
+            LbsObject *o = new LbsObject;
+            o->set_group("data");
+            o->set_data(block_buf, bytes);
+            o->write(tss);
+            ref = o->get_ref();
+            db->StoreObject(ref, block_csum, bytes);
+            delete o;
+        }
 
+        object_list.push_back(ref.to_string());
+        segment_list.insert(ref.get_segment());
+        db->UseObject(ref);
         size += bytes;
-
-        delete o;
     }
 
     file_info["checksum"] = hash.checksum_str();
@@ -351,10 +357,6 @@ int main(int argc, char *argv[])
 
     tss = new TarSegmentStore(backup_dest);
 
-    string database_path = backup_dest + "/localdb.sqlite";
-    db = new LocalDb;
-    db->Open(database_path.c_str());
-
     /* Write a backup descriptor file, which says which segments are needed and
      * where to start to restore this snapshot.  The filename is based on the
      * current time. */
@@ -367,6 +369,13 @@ int main(int argc, char *argv[])
     string desc_filename = backup_dest + "/" + desc_buf + ".lbs";
     std::ofstream descriptor(desc_filename.c_str());
 
+    /* Open the local database which tracks all objects that are stored
+     * remotely, for efficient incrementals.  Provide it with the name of this
+     * snapshot. */
+    string database_path = backup_dest + "/localdb.sqlite";
+    db = new LocalDb;
+    db->Open(database_path.c_str(), desc_buf);
+
     try {
         scanfile(".");
     } catch (IOException e) {