さくらVPSでCassandra 0.7のbatch_mutate（削除処理）を試す

---

---

既に「さくらVPSでApache Cassandra 0.7でいろいろデータを取得してみる」で
batch_mutateは利用してみたが削除処理がまだだったので試してみた
　

batch_mutateメソッドを利用する

「Cassandra--The-Definitive-Guide」のBatchDeleteExample.javaを参考に
試してみました。
　

テストデータに関して

Keyspace1のカラムファミリー「Standard2」に下記の様なデータが存在しているとします。
　

list Standard2;
Using default limit of 100
-------------------
RowKey: michibatajessica3
-------------------
RowKey: xxx
=> (column=birth_date, value=00000000000007cf, timestamp=1295514962094)
=> (column=blood, value=4f, timestamp=1295514962094)
=> (column=full_name, value=5461726f2059616d616461, timestamp=1295514962094)
-------------------
RowKey: yyy
=> (column=birth_date, value=00000000000007c6, timestamp=1295514962094)
=> (column=blood, value=4142, timestamp=1295514962094)
=> (column=full_name, value=49636869726f2053757a756b69, timestamp=1295514962094)
-------------------
RowKey: michibatajessica
-------------------
RowKey: sasakinozomi3
=> (column=age, value=3232, timestamp=1295423487396)
=> (column=first, value=6e6f7a6f6d69, timestamp=1295423487396)
=> (column=last, value=736173616b69, timestamp=1295423487396)

5 Rows Returned.

batch_mutateメソッドを実行してみる

ByteBuffer key = ByteBufferUtil.bytes("xxx"); // this is the row key

long ts = System.currentTimeMillis();

SlicePredicate delPred = new SlicePredicate();
SlicePredicate delPred2 = new SlicePredicate();
SlicePredicate delPred3 = new SlicePredicate();
List<ByteBuffer> delCols = new ArrayList<ByteBuffer>();
List<ByteBuffer> delCols2 = new ArrayList<ByteBuffer>();
List<ByteBuffer> delCols3 = new ArrayList<ByteBuffer>();

// let's delete the column named 'b', though we could add more
delCols.add(ByteBufferUtil.bytes("birth_date"));
delPred.column_names = delCols;
delCols2.add(ByteBufferUtil.bytes("blood"));
delPred2.column_names = delCols2;
delCols3.add(ByteBufferUtil.bytes("full_name"));
delPred3.column_names = delCols3;

Deletion deletion = new Deletion();
deletion.predicate = delPred;
deletion.timestamp = ts;
Mutation mutation = new Mutation();
mutation.deletion = deletion;

Deletion deletion2 = new Deletion();
deletion2.predicate = delPred2;
deletion2.timestamp = ts;
Mutation mutation2 = new Mutation();
mutation2.deletion = deletion2;

Deletion deletion3 = new Deletion();
deletion3.predicate = delPred3;
deletion3.timestamp = ts;
Mutation mutation3 = new Mutation();
mutation3.deletion = deletion3;

Map<ByteBuffer, Map<String, List<Mutation>>> mutationMap = new HashMap<ByteBuffer, Map<String, List<Mutation>>>();

List<Mutation> mutationList = new ArrayList<Mutation>();
mutationList.add(mutation);
mutationList.add(mutation2);
mutationList.add(mutation3);

Map<String, List<Mutation>> m = new HashMap<String, List<Mutation>>();
m.put(columnFamily, mutationList);

// just for this row key, though we could add more
mutationMap.put(key, m);
client.batch_mutate(mutationMap, ConsistencyLevel.ALL);
System.out.println("Delete mutation done.");

List mutationListにひたすら削除したいカラム名を加えていく。

実行結果は、下記のような形となる。

list Standard2;
Using default limit of 100
-------------------
RowKey: michibatajessica3
-------------------
RowKey: xxx
-------------------
RowKey: yyy
=> (column=birth_date, value=00000000000007c6, timestamp=1295516237209)
=> (column=blood, value=4142, timestamp=1295516237209)
=> (column=full_name, value=49636869726f2053757a756b69, timestamp=1295516237209)
-------------------
RowKey: michibatajessica
-------------------
RowKey: sasakinozomi3
=> (column=age, value=3232, timestamp=1295423487396)
=> (column=first, value=6e6f7a6f6d69, timestamp=1295423487396)
=> (column=last, value=736173616b69, timestamp=1295423487396)

5 Rows Returned.

実行しているときは良いが、対象クラスが多いので素の状態で利用する時は
結構、戸惑ってしまいました。
　

---