比如說，現(xiàn)在要下載大批量的數(shù)據(jù)，從es，放到excel中，我們說，月度，或者年度，銷售記錄，很多，比如幾千條，幾萬條，幾十萬條
其實(shí)就要用到我們之前講解的es scroll api，對(duì)大量數(shù)據(jù)批量的獲取和處理

PUT /car_shop/cars/4
{
    "brand": "寶馬",
    "name": "寶馬320",
    "price": 320000,
    "produce_date": "2017-01-01",
    "sale_price": 280000,
    "sale_date": "2017-01-25"
}

就是要看寶馬的銷售記錄
做一個(gè)演示,每個(gè)批次下載一條寶馬的銷售記錄,分幾個(gè)批次給它下載完

Settings settings = Settings.builder()
        .put("cluster.name", "my-application")
        .put("client.transport.sniff", true)
        .build();

TransportClient client = new PreBuiltTransportClient(settings).
        addTransportAddress(new TransportAddress(InetAddress.getByName("192.168.80.188"), 9300));

SearchResponse searchResponse = client.prepareSearch("car_shop")
        .setTypes("cars")
        .setQuery(QueryBuilders.termQuery("brand", "寶馬"))
        .setScroll(new TimeValue(60000))
        .setSize(1)
        .get();

int batchCount = 0;

do {
    for (SearchHit searchHit : searchResponse.getHits().getHits()) {
        System.out.println("batch: " + ++batchCount);
        System.out.println(searchHit.getSourceAsString());
    }
    searchResponse = client.prepareSearchScroll(searchResponse.getScrollId())
            .setScroll(new TimeValue(60000))
            .execute().actionGet();
} while (searchResponse.getHits().getHits().length != 0);

client.close();