Python to Postgres interface with real prepared statements?

I've been trying to find a postgres interface for python 2.x that supports real prepared statements, but can't seem to find anything. I don't want one that just escapes quotes in the params you pass in and then interpolates them into the query before 开发者_JS百科executing it. Anyone have any suggestions?

Either py-postgresql for Python3 or pg_proboscis for Python2 will do this.

Python-pgsql will also do this but is not threadsafe. Notably, SQLAlchemy does not make use of prepared statements.

have a look at web.py's db module

examples can be found at

1. http://webpy.org/cookbook/select
2. http://webpy.org/cookbook/update
3. http://webpy.org/cookbook/delete
4. http://webpy.org/Insert

These links hint at the answer when using psycopg2. You don't need special API extensions.

• Re: psycopg2 and prepared statements
• Prepared Statements in Postgresql
• Transparently execute SQL queries as prepared statements with
  Postgresql (Python recipe)

Here's an example that I played with. A word of caution though, it didn't give me the expected performance increase I had hoped for. In fact, it was even slower (just slightly) in a contrived case where I tried to read the whole table of one million rows, one row at a time.

cur.execute('''
    PREPARE prepared_select(text, int) AS
        SELECT * FROM test
        WHERE (name = $1 and rowid > $2) or name > $1
        ORDER BY name, rowid
        LIMIT 1
''')
name = ''
rowid = 0
cur.execute('EXECUTE prepared_select(%s, %s)', (name, rowid))