BigQuery 在加入 FULL JOIN + USING 时删除重复的列名

Question

我有两个 BigQuery 表，它们的列名完全相同，需要连接在一起，同时保留正确的统计信息名称。在我的实际示例中，我的表名为offensive 和defensive，它们包含球队的进攻和防守篮球统计数据。两个表中的列名是：

team, game, fga, fgm, ast, blk, stk

我想加入team 和game 列上的列。我找到了the following example，它看起来非常适合我的用例，但是我不喜欢的一件事是该示例没有删除重复的连接列（示例中的公司、供应商、交易）。加入后，我不想在我的表中出现team 和game 中的两个。我试过了：

SELECT offensive, defensive EXCEPT (team, game)
FROM offensive FULL JOIN defensive
USING (team, game)

...并且收到以下Syntax error: Unexpected keyword EXCEPT at [40:39] 错误。有关如何删除重复的连接列的任何想法都会有所帮助。

Answer 1

以下是 BigQuery 标准 SQL

#standardSQL
SELECT team, game, 
  (SELECT AS STRUCT offensive.*  EXCEPT (team, game)) AS offensive,
  (SELECT AS STRUCT defensive.*  EXCEPT (team, game)) AS defensive
FROM offensive FULL JOIN defensive
USING (team, game)

这将导致以下架构

team game offensive.fga offensive.fgm offensive.ast offensive.blk offensive.stk defensive.fga defensive.fgm defensive.ast defensive.blk defensive.stk    

类似

[
  {
    "team": "1",
    "game": "1",
    "offensive": {
      "fga": "1",
      "fgm": "2",
      "ast": "3",
      "blk": "4",
      "stk": "5"
    },
    "defensive": {
      "fga": "21",
      "fgm": "22",
      "ast": "23",
      "blk": "24",
      "stk": "25"
    }
  }
]