MagicStack / MagicStack/asyncpg

Feature request: A way to inspect the query string run by asyncpg

未关闭
#914 1 条评论 1 个 reaction 已指派 0 人 在 GitHub 查看

还没有人认领这个 Issue。

主要语言
Python
星标
8.1k
派生
468
PR 合并指标
30 天内没有已合并 PR

描述

I have hasked this before SO (https://stackoverflow.com/questions/69814471/is-there-a-way-to-inspect-the-query-string-run-by-asyncpg) but I am getting no answer, so this is a feature request.


When I use asyncpg to run a query on my database (and I have to run several queries),

I always split up things like this:

  • the string of the query, with parameters placeholders:
    QUERY_1 = """
    UPDATE TABLE my_table 
    SET field_1 = $1
    WHERE field_id = $2;
    """
    
  • a dictionary with the query parameters:
    query_params_dict = { 
        "field_1" : "value_1",
        "field_id" : "10"
    }
    
  • and the function that will get the query string, replace the placeholders with the parameters values, run the query with connection.execute, and manage the errors:
    async def run_query_1(query_params_dict):
    	connection = await Utils.get_connection_to_db(**conf.asyncpg_db_conn) # custom function to open the connection to my db
    	query = QUERY_1
    	try:
    		await connection.execute(
    			query, query_params_dict["field_1"], int(query_params_dict["field_id"])
    		)
    		return None
    	except (Exception, asyncpg.UniqueViolationError) as integrError:
    		logger.error("Violated unique constraint: {}".format(integrError), exc_info=True)
    		return "{}".format(integrError)
    	except (Exception, asyncpg.ConnectionFailureError) as error:
    		logger.error("Failed connection: {}".format(error), exc_info=True)
    		return "{}".format(error)
    	finally:
    		if (connection):
    			await Utils.close_connection(connection)
    

But the queries I have to run often require many mores parameters, so that I always make some mistake in assigning values to the variables,

so that the query fails/returns unwanted results,

and the only way I know to determine what is wrong with it, is to check carefully the value assigned to every parameter of query_params_dict.

This takes a lot of time.

Instead, I could understand quickly what is wrong by checking the query run by asyncpg.

So, is there a way to inspect the query string run by asyncpg?

I would expect a command of asyncpg taking the same arguments of connection.execute as input, and returning as output the string of the query filled up with parameters, without running any query on the database.

Let's suppose this command exists and is connection.expected_query, in my case it would work like this:

connection.expected_query(
    			query, query_params_dict["field_1"], int(query_params_dict["field_id"])
>>>

"""
UPDATE TABLE my_table 
SET field_1 = 'value_1'
WHERE field_id = 10;
"""

Could you please implement a feature like this?
Thanks in advance!

贡献指南

这个仓库没有索引到贡献指南

从这里开始

  1. 先读完整个 Issue,再读项目的贡献指南。
  2. 在 Issue 下留言说明你要接手 —— 这能避免两个人做同样的事。
  3. Fork 仓库,在一个分支上完成修改。
  4. 提交 Pull Request,并在描述里引用这个 Issue 编号。

调研方向

No repository files or tests are named. Start by tracing the Connection.execute entry point and how asyncpg handles query parameters; done would be a defined API that accepts the execute arguments, returns the parameterized query representation without executing it, and has tests covering the requested behavior.

由索引模型根据 Issue 内容生成。

评估

技术栈
postgresql, python
领域
databases
Issue 类型
功能
难度
5/5
预计耗时
一周以上
活跃度
停滞
描述清晰度
基本清楚
新手友好度
35/100

把新 issue 发到你的邮箱

精选适合新手参与的 GitHub issue 摘要。