bulk使用updateRequest().upsert()方法保存子类doc 与父类doc对应不上？
Elasticsearch | 作者 codeKing007 | 发布于2019年03月25日 | 阅读数：5434
环境：
es5.3.3
client：transport
1.索引情况：
我的代码：
保存子类type customfield 关联有parentId的doc到es中（前提是parentType：medicalrecord的doc已经存在于es中）
对于data参数如下，其中每个参数的id一直是固定的，value会根据不同的records而不同；
问题描述：
执行完后，medicalrecordid子类中的customfield doc不全，部分customfield doc会关联到其他parenId 的medicalrecordid。
例如：
customfield 中 乔治4应该对应 medicalrecordid 中的 记录张三，实际对应的是medicalrecordid的 记录李四；
在张三这条记录中只有传入五个子类的其中几个，比如：严选4就可能存在；
希望大家帮忙解决一下
es5.3.3
client：transport
1.索引情况：
{
"casefolder": {
"customfield": {
"_parent": {
"type": "medicalrecord"
},
"_routing": {
"required": true
},
"properties": {
"fieldValue": {
"type": "keyword",
"ignore_above": 10922
},
"id": {
"type": "text",
"fields": {
"keyword": {
"type": "keyword",
"ignore_above": 256
}
}
}
}
},
"medicalrecord": {
"properties": {
"address": {
"type": "keyword",
"copy_to": [
"fullFieldName"
]
},
"age": {
"type": "keyword",
"copy_to": [
"fullFieldName"
]
},
"ageUnit": {
"type": "keyword"
},
"basicInformation": {
"type": "keyword",
"copy_to": [
"fullFieldName"
]
},
"caseCode": {
"type": "keyword",
"copy_to": [
"fullFieldName"
]
},
"caseCodeType": {
"type": "keyword"
},
"cell": {
"type": "keyword",
"copy_to": [
"fullFieldName"
]
},
"contactPersonName": {
"type": "keyword",
"copy_to": [
"fullFieldName"
]
},
"createTimeStr": {
"type": "date",
"ignore_malformed": true,
"format": "yyyy-MM-dd"
},
"email": {
"type": "text",
"copy_to": [
"fullFieldName"
]
},
"encounterTime": {
"type": "date",
"ignore_malformed": true
},
"fullFieldName": {
"type": "keyword"
},
"gender": {
"type": "keyword",
"copy_to": [
"fullFieldName"
]
},
"iDCardNumber": {
"type": "keyword",
"copy_to": [
"fullFieldName"
]
},
"id": {
"type": "text",
"fields": {
"keyword": {
"type": "keyword",
"ignore_above": 256
}
}
},
"introducer": {
"type": "keyword",
"copy_to": [
"fullFieldName"
]
},
"otherCaseCode": {
"type": "keyword",
"copy_to": [
"fullFieldName"
]
},
"otherCaseCodeType": {
"type": "keyword"
},
"otherMemo": {
"type": "keyword",
"copy_to": [
"fullFieldName"
]
},
"patientName": {
"type": "keyword",
"copy_to": [
"fullFieldName"
]
},
"patientOccupation": {
"type": "keyword",
"copy_to": [
"fullFieldName"
]
},
"tel": {
"type": "text",
"copy_to": [
"fullFieldName"
]
},
"uid": {
"type": "text",
"fields": {
"keyword": {
"type": "keyword",
"ignore_above": 256
}
}
},
"updateTimeStr": {
"type": "date",
"ignore_malformed": true
},
"userId": {
"type": "integer"
},
"version": {
"type": "long"
}
}
}
}
}
}我的代码：
保存子类type customfield 关联有parentId的doc到es中（前提是parentType：medicalrecord的doc已经存在于es中）
public <T extends IndexVo> boolean batchIndex(String index, String type, List<String> parentIdList, List<T> data) {
if (parentIdList.size() != data.size()) {
throw new SearchException("批量数据与parentId数量不匹配！");
}
BulkRequestBuilder bulkRequest = esBase.getClient().prepareBulk();
for (int i = 0; i < data.size(); i++) {
T t = data.get(i);
String json = JSONObject.toJSONString(t);
IndexRequest indexRequest = new IndexRequest(index, type, t.get_id()).source(json).parent(parentIdList.get(i));
UpdateRequest updateRequest = new UpdateRequest(index, type, t.get_id()).parent(parentIdList.get(i)).doc(json).upsert(indexRequest);
bulkRequest.add(updateRequest);
}
BulkResponse bulkResponse = bulkRequest.execute().actionGet();
.......
}对于data参数如下，其中每个参数的id一直是固定的，value会根据不同的records而不同；
CustomFieldIndexVo customFieldIndexVo = new CustomFieldIndexVo();
customFieldIndexVo.setFieldValue("144");
customFieldIndexVo.set_id("68e5c41e-2fbd-4dfa-bef5-901ceadece28");
indexVos.add(customFieldIndexVo);
CustomFieldIndexVo customFieldIndexVo1 = new CustomFieldIndexVo();
customFieldIndexVo1.setFieldValue("224");
customFieldIndexVo1.set_id("0edcf173-8af2-4818-9056-4c2d1ceafb0b");
indexVos.add(customFieldIndexVo1);
CustomFieldIndexVo customFieldIndexVo2 = new CustomFieldIndexVo();
customFieldIndexVo2.setFieldValue("乔治4");
customFieldIndexVo2.set_id("dc038e90-df9c-4ccb-b50a-ed050f803b3a");
indexVos.add(customFieldIndexVo2);
CustomFieldIndexVo customFieldIndexVo3 = new CustomFieldIndexVo();
customFieldIndexVo3.setFieldValue("严选4");
customFieldIndexVo3.set_id("15ed58c9-94b7-4763-9333-b0abf7198415");
indexVos.add(customFieldIndexVo3);
CustomFieldIndexVo customFieldIndexVo4 = new CustomFieldIndexVo();
customFieldIndexVo4.setFieldValue("跨栏4");
customFieldIndexVo4.set_id("662b002c-c6f1-482f-9e49-9c101d967d22");
indexVos.add(customFieldIndexVo4);["D5286126-B1FC-4A3B-AE17-776799239C6F",
"D5286126-B1FC-4A3B-AE17-776799239C6F",
"D5286126-B1FC-4A3B-AE17-776799239C6F",
"D5286126-B1FC-4A3B-AE17-776799239C6F",
"D5286126-B1FC-4A3B-AE17-776799239C6F"]"问题描述：
执行完后，medicalrecordid子类中的customfield doc不全，部分customfield doc会关联到其他parenId 的medicalrecordid。
例如：
customfield 中 乔治4应该对应 medicalrecordid 中的 记录张三，实际对应的是medicalrecordid的 记录李四；
在张三这条记录中只有传入五个子类的其中几个，比如：严选4就可能存在；
希望大家帮忙解决一下
1 个回复
laoyang360 - 《一本书讲透Elasticsearch》作者，Elastic认证工程师 [死磕Elasitcsearch]知识星球地址：http://t.cn/RmwM3N9；微信公众号：铭毅天下; 博客：https://elastic.blog.csdn.net
赞同来自: