MongoDB实现批量存在执行修改，不存在执行添加（BulkWrite）

最近使用mongodb实现一个需求，需要批量处理数据，并且存在则修改，不存在则添加，updateMany方法，无法达到要求，经查，发现mongodb中存在一个BulkWrite方法可以实现该需求，该操作可实现类似于mysql的insert into ... on duplicate key udpdate ...

使用如下命令创建用于测试的db和collection以及添加index并查看

use mockdb
db.mockcoll.createIndex({"flag":1})
db.mockcoll.getIndexes()

mockcoll中只存在一个字段flag，并在该字段建立唯一索引

下面为代码示例

public interface MongoCons {

    String mockdbName = "mockdb";
    String mockcollName = "mockcoll";
    
    static MongoCollection getCollection(MongoTemplate mongoTemplate){
        MongoDatabase mongoDatabase = mongoTemplate.getMongoDbFactory().getMongoDatabase(mockdbName);
        MongoCollection collection = mongoDatabase.getCollection(mockcollName);
        return collection;
    }
    
}

@Component
public class MongoBulkDao {

    private Logger logger = LoggerFactory.getLogger(MongoBulkDao.class);
    @Autowired
    private MongoTemplate mongoTemplate;

    public void addOrUpdate(List dataList) {
        try {
            MongoCollection collection = MongoCons.getCollection(mongoTemplate);

            List> list = new ArrayList<>();

            UpdateOptions upsert = new UpdateOptions().upsert(true);
            Iterator iterator = dataList.iterator();
            String next = null;
            while (iterator.hasNext()) {
                next = iterator.next();
                BasicDBObject filter = new BasicDBObject().append("flag", next);
                BasicDBObject update = new BasicDBObject().append("$set", new BasicDBObject().append("flag", next));

                UpdateOneModel um = new UpdateOneModel(filter, update, upsert);
                list.add(um);
                // 删除掉不需要的数据
                iterator.remove();
            }

            collection.bulkWrite(list, new BulkWriteOptions().ordered(false));
        } catch (Exception e) {
            logger.error("", e);
        }
    }

}