如何在BigQuery中使用offset和ordinal从包含不同数据类型的数组中选择一个元素？

Question

基于这个页面accessing array elements，我创建了一个名为some_numbers的数组，它有两种数据类型（字符串和整数）。这是从数组中选择一个元素的查询，但它不起作用。

WITH sequences AS
  (SELECT array<struct<subject as string, marks as int64>>[("mat",87),("phy",45),("che",45),("zoo",78),("eng",87)] AS some_numbers
   UNION ALL SELECT array<struct<subject as string, marks as int64>>[("mat",87),("phy",45),("che",45),("zoo",78),("eng",87)] AS some_numbers
   UNION ALL SELECT array<struct<subject as string, marks as int64>>[("mat",87),("phy",45),("che",45),("zoo",78),("eng",87)] AS some_numbers)
SELECT some_numbers,
       some_numbers[OFFSET(1)] AS offset_1,
       some_numbers[ORDINAL(1)] AS ordinal_1
FROM sequences;

我有以下错误：

错误：语法错误：预期“，”或“>”但在[3:32]获得关键字AS -

Answer 1

几乎相同的查询，但工作：

WITH sequences AS (
   SELECT           [STRUCT("mat" AS subject,87 AS marks),("phy",45),("che",45),("zoo",78),("eng",87)] AS some_numbers
   UNION ALL SELECT [STRUCT("mat" AS subject,87 AS marks),("phy",45),("che",45),("zoo",78),("eng",87)] AS some_numbers
   UNION ALL SELECT [STRUCT("mat" AS subject,87 AS marks),("phy",45),("che",45),("zoo",78),("eng",87)] AS some_numbers
)
SELECT some_numbers,
       some_numbers[OFFSET(1)] AS offset_1,
       some_numbers[ORDINAL(1)] AS ordinal_1
FROM sequences;

（对于将来的问题，请包括获得的错误和期望的结果）