使用 pg8000 在 sql 脚本中使用位置参数的问题

Question

我有以下 SQL 查询：

 query_string = "SELECT sum(unmatched), " \
                "TIMESTAMP WITH TIME ZONE 'epoch' + INTERVAL '1 second' * " \
                "round(extract('epoch' from time_window) / {}) * {} as time_window " \
                "FROM aggregate_counts WHERE reconciliation_name = %s " \
                "GROUP BY round(extract('epoch' from time_window) / {})".format(interval_sec, interval_sec, interval_sec)

 cur.execute(query_string, (reconciliation_name))

它工作正常，除非我想避免对 "interval_sec" 使用字符串替换，而是使用位置参数，就像我对其他参数所做的那样。问题是，如果我这样做：

 query_string = "SELECT sum(unmatched), " \
                "TIMESTAMP WITH TIME ZONE 'epoch' + INTERVAL '1 second' * " \
                "round(extract('epoch' from time_window) / %s) * %s as time_window " \
                "FROM aggregate_counts WHERE reconciliation_name = %s " \
                "GROUP BY round(extract('epoch' from time_window) / %s)"

 cur.execute(query_string, (interval_sec, interval_sec, reconciliation_name, interval_sec))

我收到以下错误：

Error handler middleware caught the following exception: {'S': 'ERROR', 'V': 'ERROR', 'C': '42803', 'M': 'column "aggregate_counts.time_window" must appear in the GROUP BY clause or be used in an aggregate function', 'P': '177', 'F': 'parse_agg.c', 'L': '1344', 'R': 'check_ungrouped_columns_walker'}

File File "pg8000/core.py", line 1829, in execute ps = cache['ps'][key] KeyError: ("SELECT sum(unmatched), TIMESTAMP WITH TIME ZONE 'epoch' + INTERVAL '1 second' * round(extract('epoch' from time_window) / %s) * %s as time_window FROM aggregate_counts WHERE reconciliation_name = %s GROUP BY round(extract('epoch' from time_window) / %s)", ((701, 1, ), (701, 1, ), (705, 0, .text_out at 0x10c58cea0>)))

位置参数只能用于比较（=、>=、< 等...）吗？

Answer 1

所以基本不可能有这个。原因是 select 子句中的 %s 将被转换为比方说 $x 位置参数，而 group by 中的 %s 将被转换为 $y（x 和 y 是各自的位置）。现在 postgres 有没办法知道解决后，这两个会是一样的。因此它假定 "aggregate_counts.time_window" 不存在于 GROUP BY 中。我知道这不是完美的解释，但这就是正在发生的事情。

使用 pg8000 在 sql 脚本中使用位置参数的问题

Issue using positional parameters in sql script, using pg8000

python

postgresql

pg8000