Question

我正在尝试在Google BigQuery上合并具有相同架构的两个数据库。

我在这里关注合并示例：https://cloud.google.com/bigquery/docs/reference/standard-sql/dml-syntax#merge_statement

但是，我的表具有嵌套列，即“ service.id”或“ service.description”

到目前为止，我的代码是：

MERGE combined_table
USING table1
ON table1.id = combined_table.id
WHEN NOT MATCHED THEN
  INSERT(id, service.id, service.description)
  VALUES(id, service.id, service.description)

但是，我得到了错误消息：Syntax error: Expected ")" or "," but got "."，在.id行的INSERT(...)下有一个红色的弯曲下划线。

这是表的一部分架构的视图：

[
    {
        "name": "id",
        "type": "STRING"
    },
    {
        "name": "service",
        "type": "RECORD",
        "fields": [
            {
                "name": "id",
                "type": "STRING"
            },
            {
                "name": "description",
                "type": "STRING"
            }
        ]
    },
    {
        "name": "cost",
        "type": "FLOAT"
    }
...
]

如何正确构造此INSERT（...）语句，以便可以包括嵌套列？

Answer 1

语法错误：预期为“）”或“，”，但显示为“。”

看起来您的方向正确，请在文档中注意如何将值插入REPEATED列中，

您需要定义结构来指导BigQuery期望什么，例如：

STRUCT<created DATE, comment STRING>

这是文档中的完整示例

MERGE dataset.DetailedInventory T
USING dataset.Inventory S
ON T.product = S.product
WHEN NOT MATCHED AND quantity < 20 THEN
  INSERT(product, quantity, supply_constrained, comments)
   -- insert values like this
   VALUES(product, quantity, true, ARRAY<STRUCT<created DATE, comment STRING>>[(DATE('2016-01-01'), 'comment1')])

WHEN NOT MATCHED THEN
  INSERT(product, quantity, supply_constrained)
  VALUES(product, quantity, false)

Answer 2

我找到了答案。

当引用STRUCT的顶层时，BigQuery也会引用所有嵌套列。因此，如果我想插入const list = {"data":[{"type":"A","date":"2018-05","value":"153"},{"type":"B","date":"2018-05","value":"888"},{"type":"C","date":"2018-05","value":"999"},{"type":"D","date":"2018-05","value":"555"},{"type":"A","date":"2018-06","value":"148"},{"type":"B","date":"2018-06","value":"222"},{"type":"C","date":"2018-06","value":"555"},{"type":"D","date":"2018-06","value":"666"},{"type":"A","date":"2018-07","value":"156"},{"type":"B","date":"2018-07","value":"111"},{"type":"C","date":"2018-07","value":"333"},{"type":"D","date":"2018-07","value":"999"}],"number":"111-111"}; // const dates = [...new Set(list.data.map(item => item.date))]; const dates = Array.from(list.data.map(item => item.date)); console.log(dates); // const types = [...new Set(list.data.map(item => item.type))]; const types = Array.from(list.data.map(item => item.type)); console.log(types) const res = dates.map(date => { const obj = {}; types.map(type => { obj[type] = list.data.filter(item => item.date === date && item.type === type)[0].value; }); obj.date = date; return obj; }); console.log(res);及其所有子列（service.id和service.description），则只需在service语句中包含service。

以下代码有效：

INSERT(...)

这将合并所有子列，包括... WHEN NOT MATCHED THEN INSERT(id, service) VALUES(id, service)和service.id。

如何使用嵌套名称语法（即“ item.description”）插入列？

2 个答案: