跳转到主要内容
eLearner.app
模块 15 · 第 2 课(共 2)课程中的57/57~15 min
模块课程(2/2)

JSON 聚合函数

JSON Aggregation Functions

The integration is bidirectional. The database doesn't have to limit itself to parsing JSON: it can build it. Imagine if your API and Node.js didn't have to do the dirty, intensive work of serializing nested objects into JSON, because the DB delivered them already built straight from disk.

Building Objects and Arrays

Instead of the classic mass aggregation or string concatenation (String_agg), we turn to these functions:

  • json_build_object(key, value, ...) or jsonb_build_object(...)
  • json_agg(...) or jsonb_agg(...)
SQL
SELECT json_build_object('name', first_name, 'city', city) AS data
FROM customers;

You'll see rows printed containing { "name": "...", "city": "..." }, perfectly compliant.

锻炼#sql.m15.l2.e1
尝试:0加载中...

From the 'customers' table, return the data as JSON rows, mapped with \`json_build_object\`. Map two keys: 'name' bound to 'first_name', and 'mail' bound to 'email'. Use \`AS customer_doc\` for the column.

正在加载编辑器...
显示提示

json_build_object('name', first_name, 'mail', email)

3 次尝试后可用的解决方案

锻炼#sql.m15.l2.e2
尝试:0加载中...

For each order (\`order_id\`), build in a single unit all the \`product_id\`s it includes. From the 'order_items' table, run \`GROUP BY order_id\`. Then select \`order_id\` and accumulate into a JSON array (JSON_AGG) just the \`product_id\` data. Name the array \`products\`.

正在加载编辑器...
显示提示

JSON_AGG(product_id) plus a classic group by to gather the data.

3 次尝试后可用的解决方案