Cassandra中的预准备语句与绑定语句？

Question

我想知道使用边界语句比准备语句有什么好处？

PreparedStatement statement = session.prepare(
                      "INSERT INTO simplex.songs " +
                      "(id, title, album, artist) " +
                      "VALUES (?, ?, ?, ?);");

BoundStatement boundStatement = new BoundStatement(statement);
            session.execute(boundStatement.bind(
                  UUID.fromString("756716f7-2e54-4715-9f00-91debea6cf50"),
                  "La Petite Tonkinoise",
                  "Bye Bye Blackbird",
                  "Joséphine Baker");

最简单的方法是：

PreparedStatement ps = session.prepare(
                      "INSERT INTO simplex.songs " +
                      "(id, title, album, artist, tags) " +
                      "VALUES (?, ?, ?, ?, ?);");
ps.bind(UUID.fromString("756716f7-2e54-4715-9f00-91debea6cf50"),
                      "La Petite Tonkinoise",
                      "Bye Bye Blackbird",
                      "Joséphine Baker");

如您所见，我可以将数据绑定到< code>preparedStatement而无需< code>boundStatements。< code>boundStatement在哪里有用？

Answer 1

对我来说，当您在准备语句上调用bind（…）时，会自动创建边界语句。您也可以直接创建实例Bound语句。

准备语句和绑定语句具有不同的行为，因为准备语句.bind（） 返回新的实例绑定语句，而绑定语句.bind（） 返回自己。

这是多线程环境中的重要细节。调用方法。共享边界上的bind()语句结果危险

// shared instance BoundStatement on few threads 
BoundStatement bs1 = 
// bs2 is the same as bs1
// param1, param2, ... are different for every thread
BoundStatement bs2 = bs1.bind(param1, param2, ...);
// add some time to wait so other thread can modify your params
// Thread.sleep(RandomUtils.nextInt(100));        
// result2 sometimes may be with incorrect result 
results2 = session.execute(bs2); 

当您在PreparedStatement上调用bind时，您将获得对象的不同实例，并且它是线程安全的。(在与上述相同的场景中)

PreparedStatement ps1 = 
BoundStatement bs2 = ps1.bind(param1, param2, ...);
results2 = session.execute(bs2); 

Answer 2

没有优势:BoundStatement只不过是一个变量有界的PreparedStatement。实际上，PreparedStatements的< code>bind()方法返回一个BoundStatement。