使用 python 中的 psycopg2 使參數化 SQL select 語句

Question

我從psycopg2 API 中了解到如何將參數傳遞給 SQL 查詢，以便我們可以輕松地使用參數方式操作 SQL 語句。 因此，帶有%(param)s的 stringify 參數可以做到這一點。 I am wondering if we could encapsulate one common parameterized SQL statement in a python function, so we could call python function by giving any arbitrary parameter values which will be consumed by SQL statement, ultimately it would server as running as many SQL statements as possible. 但是，我不知道如何使參數化 SQL select 語句，因為我們要從本地 DB 檢索的項目每次都可以變體，所以如果 Z99938282F040718599Z41E18 語句可以參數化會很好。 我們怎樣才能做到這一點？ 有沒有辦法用 python 中的psycopg2做到這一點？ 如何做到這一點？ 有什么可能的想法嗎？

數據庫表：

這是用於重現目的的示例數據庫表：

CREATE TABLE trans_tbl(
date_received DATE,
pk_est VARCHAR,
grd_name VARCHAR,
cl_val SMALLINT,
quant_received NUMERIC,
mg_fb_price NUMERIC,
freight NUMERIC,
standard_price NUMERIC,
grd_delv_cost NUMERIC,
order_type VARCHAR,
pk_name VARCHAR,
item_type VARCHAR,
waiting_days NUMERIC,
item_name VARCHAR,
mk_price_variance NUMERIC,
);

而且，這里是示例查詢的列表，其中我需要一個參數化的 SQL 查詢語句（ select ，應參數化where子句）：

示例查詢 1

SELECT
    date_trunc('week', date_received::date) AS received_week,
    cl_val,
    item_type,
    ROUND(ROUND(SUM(quant_received * mg_fb_price)::numeric,4) / SUM(quant_received),4) AS price_1,
    ROUND(ROUND(SUM(quant_received * grd_delv_cost)::numeric,4) / SUM(quant_received),4) AS dv_price,
FROM trans_tbl
GROUP BY received_week,cl_val,item_type
ORDER BY received_week;

示例查詢 2 ：

SELECT
    date_trunc('month', date_received) AS received_month,
    ROUND(ROUND(SUM(quant_received * standard_price)::numeric,4) / SUM(quant_received),4) AS mk_price,
    ROUND(ROUND(SUM(quant_received * mg_fb_price)::numeric,4) / SUM(quant_received),4) AS price,
    ROUND(ROUND(SUM(quant_received * mk_price_variance)::numeric,4) / SUM(quant_received),4) AS fob_market_price_variance,
    ROUND(ROUND(SUM(quant_received * grd_delv_cost)::numeric,4) / SUM(quant_received),4) AS dv_cost,
    ROUND(ROUND(SUM(quant_received * freight)::numeric,4) / SUM(quant_received),4) AS weight_avg,
FROM trans_tbl

示例查詢 3 ：

SELECT
    date_trunc('week', date_received::date) AS received_week,
    grd_name,
    pk_name,
    pk_est,
    TO_CHAR(SUM(quant_received), '999G999G990D') AS received_amt
FROM trans_tbl

我想要做什么我想要一個通用的參數化 SQL 語句，這樣我就可以通過任意傳遞參數值來運行 SQL 語句，這樣它就可以與分別運行三個 Z9778840A017410CB30C98ZB7 語句一樣。 有沒有辦法用 python 中的psycopg2來完成這個？ 這樣做是否可行？ 任何想法？

更新：

也許我的嘗試不太可行，所以我願意接受至少可行的、可行的方法來減輕痛苦。 如果我想要實現的目標不太可行，那么我能做些什么有效的方法呢？ 任何想法？

Answer 1

首先，這是您之前的問題Previous的副本。 你應該在那兒繼續討論。 正如我所說，可以使用 psycopg2 中的sql模塊來執行您的操作。 作為我的一個應用程序的示例：

class NotificationReport():
    """Builds a query for finding task notifications.

    Use form_choices passed in to modify the select query for task
    notifications using psycopg2.sql module. Filter on status which is some
    combination of notify_expired and notify_cancelled.
    """

    def __init__(self, form_choices):
        self.id = "notification_report"
        self.form_choices = form_choices

    def returnQuery(self):
        flds, defaults = data.fetchFields(data.TaskNotification)
        base_sql = sql.SQL("""SELECT
            task_title, {}
        FROM
            tasks
        JOIN
            task_priority AS tp
        ON
            tasks. task_priority_fk= tp.priority_id
        JOIN
            task_type AS tt
        ON
            tasks.task_type_fk = tt.task_type_id
        LEFT JOIN
            task_notification AS tn
        ON
            tasks.task_id = tn.task_id_fk

        """).format(sql.SQL(",").join(map(sql.Identifier, flds)))
        f_choices = self.form_choices
        and_sql = None
        ops_list = []
        if f_choices:
            for choice in f_choices:
                if choice.get("status"):
                    status = choice["status"]
                    status_dict = {"open": ("notify_expired = 'f' "),
                                   "expired": ("notify_expired = 't' "),
                                   }
                    if status == "all":
                        pass
                    else:
                        ops = sql.SQL(status_dict[status])
                        ops_list.append(ops)
        if ops_list:
            and_sql = sql.Composed([base_sql, sql.SQL(" AND ")])
            additional_and = sql.SQL(" AND ").join(ops_list)
            ops_sql = sql.Composed([and_sql, additional_and])
        orderby_sql = sql.SQL("""ORDER BY
            task_title""")
        if and_sql:
            combined_sql = sql.Composed([ops_sql, orderby_sql])
        else:
            combined_sql = sql.Composed([base_sql, orderby_sql])

        return combined_sql

Output。 首先沒有提供參數來報告：

SELECT
    task_title, "task_id_fk","before_value","before_interval","every_value","every_interval","notify_note","notify_id","notify_expired"
FROM
    tasks
JOIN
    task_priority AS tp
ON
    tasks. task_priority_fk= tp.priority_id
JOIN
    task_type AS tt
ON
    tasks.task_type_fk = tt.task_type_id
LEFT JOIN
    task_notification AS tn
ON
    tasks.task_id = tn.task_id_fk

ORDER BY
    task_title

然后是狀態：

SELECT
    task_title, "task_id_fk","before_value","before_interval","every_value","every_interval","notify_note","notify_id","notify_expired"
FROM
    tasks
JOIN
    task_priority AS tp
ON
    tasks. task_priority_fk= tp.priority_id
JOIN
    task_type AS tt
ON
    tasks.task_type_fk = tt.task_type_id
LEFT JOIN
    task_notification AS tn
ON
    tasks.task_id = tn.task_id_fk

    AND notify_expired = 'f' ORDER BY
    task_title