Elasticsearch中endDate为空时计算起止日期月数的报错排查

问题场景 #

计算startDate与endDate的月份差：用户非在职时（isEmployed为false）endDate有值，计算正常；用户在职时endDate为null，尝试通过runtime_mapping将endDate替换为当前时间时，触发类型转换错误。

索引映射 #

PUT {{elasticUrl}}/user-experiences
{
    "mappings": {
        "properties": {
            "userId": {
                "type": "integer"
            },
            "experienceId": {
                "type": "integer"
            },
            "startDate": {
                "type": "date"
            },
            "endDate": {
                "type": "date"
            },
            "isEmployed": {
                "type": "boolean" // 原映射遗漏该字段，需补充否则无法访问doc['isEmployed']
            }
        }
    }
}

原错误查询 #

POST {{elasticUrl}}/user-experiences/_search
{
    "runtime_mappings": {
        "endDate": {
            "type": "date",
            "script": {
                "source": "if (doc['isEmployed'].value == true) { emit(new Date().getTime()) } else { emit (params._source['endDate']) }"
            }
        }
    },
    "size": 0,
    "aggregations": {
        "group_by": {
            "terms": {
                "field": "userId"
            },
            "aggregations": {
                "start": {
                    "sum": {
                        "field": "startDate"
                    }
                },
                "end": {
                    "sum": {
                        "field": "endDate"
                    }
                },
                "duration": {
                    "bucket_script": {
                        "buckets_path": {
                            "start": "start.value",
                            "end": "end.value"
                        },
                        "script": {
                            "params": {
                                "month_in_milliseconds": 2628000000
                            },
                            "source": "Math.round((params.end - params.start) / params.month_in_milliseconds)"
                        }
                    }
                },
                "duration_bucket_filter": {
                    "bucket_selector": {
                        "buckets_path": {
                            "durationBucket": "duration"
                        },
                        "script": {
                            "params": {
                                "number_of_months": 1
                            },
                            "source": "params.durationBucket >= params.number_of_months"
                        }
                    }
                }
            }
        }
    }
}

错误信息 #

{
    "error": {
        "root_cause": [
            {
                "type": "script_exception",
                "reason": "runtime error",
                "script_stack": [
                    "emit (params._source['endDate']) }",
                    "                    ^---- HERE"
                ],
                "script": "if (doc['isEmployed'].value == true) { emit(new Date().getTime()) } else { emit (params._source['endDate']) }",
                "lang": "painless",
                "position": {
                    "offset": 95,
                    "start": 75,
                    "end": 109
                }
            }
        ],
        "type": "search_phase_execution_exception",
        "reason": "all shards failed",
        "phase": "query",
        "grouped": true,
        "failed_shards": [
            {
                "shard": 0,
                "reason": {
                    "type": "script_exception",
                    "reason": "runtime error",
                    "script_stack": [
                        "emit (params._source['endDate']) }",
                        "                    ^---- HERE"
                    ],
                    "script": "if (doc['isEmployed'].value == true) { emit(new Date().getTime()) } else { emit (params._source['endDate']) }",
                    "lang": "painless",
                    "position": {
                        "offset": 95,
                        "start": 75,
                        "end": 109
                    },
                    "caused_by": {
                        "type": "class_cast_exception",
                        "reason": "class java.lang.String cannot be cast to class java.lang.Number (java.lang.String and java.lang.Number are in module java.base of loader 'bootstrap')"
                    }
                }
            }
        ]
    },
    "status": 400
}

问题原因 #

runtime脚本中，当isEmployed为false时，emit(params._source['endDate'])传入的是字符串格式的日期（如"2024-05-01"），而emit要求传入数字类型的时间戳（与new Date().getTime()输出类型一致），类型不匹配导致强制转换失败。

修复方案 #

修改runtime脚本，使用doc['endDate'].value.getTime()获取已解析的日期时间戳，避免字符串类型问题：

POST {{elasticUrl}}/user-experiences/_search
{
    "runtime_mappings": {
        "endDate": {
            "type": "date",
            "script": {
                "source": "if (doc['isEmployed'].value) { emit(new Date().getTime()); } else { emit(doc['endDate'].value.getTime()); }"
            }
        }
    },
    "size": 0,
    "aggregations": {
        "group_by": {
            "terms": {
                "field": "userId"
            },
            "aggregations": {
                "start": {
                    "sum": {
                        "field": "startDate"
                    }
                },
                "end": {
                    "sum": {
                        "field": "endDate"
                    }
                },
                "duration": {
                    "bucket_script": {
                        "buckets_path": {
                            "start": "start.value",
                            "end": "end.value"
                        },
                        "script": {
                            "params": {
                                "month_in_milliseconds": 2628000000
                            },
                            "source": "Math.round((params.end - params.start) / params.month_in_milliseconds)"
                        }
                    }
                },
                "duration_bucket_filter": {
                    "bucket_selector": {
                        "buckets_path": {
                            "durationBucket": "duration"
                        },
                        "script": {
                            "params": {
                                "number_of_months": 1
                            },
                            "source": "params.durationBucket >= params.number_of_months"
                        }
                    }
                }
            }
        }
    }
}

额外说明 #

• doc['endDate'].value是Elasticsearch已解析的ZonedDateTime对象，调用getTime()可直接获取毫秒级时间戳，与new Date().getTime()输出类型一致。
• 若必须使用params._source，需将字符串日期解析为时间戳：

  emit(ZonedDateTime.parse(params._source['endDate']).toInstant().toEpochMilli())
  

  此方式需确保日期格式符合ISO标准，可靠性低于直接使用doc字段。

内容的提问来源于stack exchange，提问作者Dženan Džafić