ElasticSearch 7.13.2 版本中过滤嵌套数据满足条件的文档

我有一个索引，存储的是每个货物在每个仓库中的出入库流水信息，mapping如下： { "mappings": { "properties": { // 出入库流水信息 "details": { "type": "nested", "properties": { // 出入库流水审核时间 "auditDate": { "type": "date", "format": "yyyy-MM-dd HH:mm:ss" }, // 审核后库存 "closingStock": { "type": "integer" }, // 审核前库存 "openingStock": { "type": "integer" }, // 出入库数量 "quantity": { "type": "integer" } } }, "id": { "type": "keyword" }, // 存货编码 "inventoryCode": { "type": "keyword" }, "lastAuditDate": { "type": "date", "format": "yyyy-MM-dd HH:mm:ss" }, "lastClosingStock": { "type": "integer" }, // 仓库ID "warehouseId": { "type": "long" } } } }

我想过滤出，满足某个日期前的最后一条出入库流水的期末库存数量大于0的文档，我想使用script进行出入库流水轮询时，结果doc并不包含嵌套类型的数据：

GET bec_erp_warehouse_stock/_search { "query": { "bool": { "filter": [ { "script": { "script": { "source": """ def details = doc['details']; if (details.size() > 0) { def lastDetail = details[details.size() - 1]; return lastDetail['closingStock'].value > 0; } return false; """ } } } ] } } }

"failed_shards" : [ { "shard" : 0, "index" : "bec_erp_warehouse_stock_1646793913013456898", "node" : "tzYpyrZHQJuBSJLJ0BZHXw", "reason" : { "type" : "script_exception", "reason" : "runtime error", "script_stack" : [ "org.elasticsearch.search.lookup.LeafDocLookup.get(LeafDocLookup.java:65)", "org.elasticsearch.search.lookup.LeafDocLookup.get(LeafDocLookup.java:27)", "details = doc['details'];\n ", " ^---- HERE" ], "script" : " ...", "lang" : "painless", "position" : { "offset" : 35, "start" : 21, "end" : 63 }, "caused_by" : { "type" : "illegal_argument_exception", "reason" : "No field found for [details] in mapping" } } } ]

请问还有什么其他方式能够实现我的期望吗？