Subject: Re: psycopg3, prepared statements
To: Daniele Varrazzo <daniele.varrazzo@gmail.com>,
 Vladimir Ryabtsev <greatvovan@gmail.com>
Cc: psycopg@postgresql.org
References: 
 <CA+mi_8aAuORkdXZ9bG_GpU34iPCGktMO83ktn+ODjL+Y=Y=e3Q@mail.gmail.com>
 <25a137a0-9ad3-bb9f-b008-263dcc81a645@aklaver.com>
 <CA+mi_8YoZ6_0NM4TbfqE8SQY8-b=RgaBH-jUVxzbWQdp+4oZGQ@mail.gmail.com>
 <331f273f-fc13-8d80-b209-c94c0960268b@aklaver.com>
 <CAMqTPq=G1E=b32JxNo6dGk1-9qGHSCP7Puo5FCqf68nakiBo_Q@mail.gmail.com>
 <CA+mi_8YhQVo8xouWnc6hCR6RvE1THRB0Wz0OzHhMHsuvdLNYnw@mail.gmail.com>
 <CA+mi_8akdCJC=q9_EH5n5Prj32ttLKCY8-=yncxRr7K8V2mZhw@mail.gmail.com>
 <c07974b0-471b-340d-ee6e-1a145fbd8999@aklaver.com>
 <CAMqTPq=-CQ6Pm3cjn9j77W7zEqiwz2atxxj+JtbxKYY_xzV-7w@mail.gmail.com>
 <CA+mi_8aoK_wbSQimZ4WHHXZT2YTBPfxLj9Zsbfbx8o=gfBWyaQ@mail.gmail.com>
From: Adrian Klaver <adrian.klaver@aklaver.com>
Message-ID: <3a5a5655-51f6-4251-bbb9-64964e423e5b@aklaver.com>
Date: Thu, 24 Dec 2020 08:41:25 -0800
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101
 Thunderbird/78.6.0
MIME-Version: 1.0
In-Reply-To: 
 <CA+mi_8aoK_wbSQimZ4WHHXZT2YTBPfxLj9Zsbfbx8o=gfBWyaQ@mail.gmail.com>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Language: en-US
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 12/23/20 4:14 PM, Daniele Varrazzo wrote:
> On Wed, 23 Dec 2020 at 23:23, Vladimir Ryabtsev <greatvovan@gmail.com> wrote:
>>
>> Cause (query, types) can give more combinations than (query,)?
> 
> Yes, that's the reason
> 
>      In [1]: import psycopg3
>      In [2]: cnn = psycopg3.connect()
>      In [3]: cnn.prepare_threshold = 2
> 
>      In [4]: cnn.execute("select 1 + %s", [1]).fetchone()
>      Out[4]: (2,)
>      In [5]: cnn.execute("select 1 + %s", [None]).fetchone()
>      Out[5]: (None,)
>      In [7]: cnn.execute("select 1 + %s", [2]).fetchone()
>      Out[7]: (3,)

Alright, I was misunderstanding. I thought you where referring to 
something like:

cur.execute("select * from some table where id > %s and user_name = %s 
and current = %s", [2, 'aklaver', True])

> 
> After 3 times the expression should have been prepared, but the tally
> has been spread in two values (0 is unknown oid, 20 is int oid).
> 
>      In [8]: cnn._prepared_statements
>      Out[8]: OrderedDict([((b'select 1 + $1', (0,)), 1), ((b'select 1 +
> $1', (20,)), 2)])
> 
>      In [9]: cnn.execute("select 1 + %s", [3]).fetchone()
>      Out[9]: (4,)
> 
> However, when either key passes the threshold, eventually preparation happens.
> 
>      In [10]: cnn._prepared_statements
>      Out[10]:
>      OrderedDict([((b'select 1 + $1', (0,)), 1),
>                  ((b'select 1 + $1', (20,)), b'_pg3_0')])
> 
> _pg3_0 is the name under which that combination of query and types is
> now prepared (it is local per session).
> 
> -- Daniele
> 


-- 
Adrian Klaver
adrian.klaver@aklaver.com