asyncsqlite3 - asynchronous driver for sqlite3.
Asynchronous interface with context managers for sqlite database:
async with asyncsqlite3.connect(...) as conn:
async with conn.execute("INSERT INTO some_table ..."):
await conn.commit()
async with conn.execute("SELECT * FROM some_table") as cursor:
rows = await cursor.fetchall()
And without context managers:
conn = await asyncsqlite3.connect(...)
cursor = await conn.execute("SELECT * FROM some_table")
rows = await cursor.fetchall()
await cursor.close()
await conn.close()
Connection pool and transaction:
async with asyncsqlite3.create_pool(..., min_size=1, max_size=2) as pool:
async with pool.acquire(timeout=1.5) as conn: # You can use a timeout when getting a connection from queue.
async with conn.transaction():
async with conn.execute("INSERT INTO some_table ..."):
...
async with pool.acquire() as conn:
async with conn.execute("SELECT id, name FROM some_table") as cursor:
async for row in cursor:
print(row) # <Record id=1 name='alex'>
print(row[0], row["id"], row.get("id")) # 1 1 1
print(dict(row)) # {'id': 1, 'name': 'alex'}
print(tuple(row), list(row), set(row)) # (1, 'alex') [1, 'alex'] {1, 'alex'}
Installation
asyncsqlite3 is compatible with Python 3.7 and newer. Use pip to install:
$ pip install asyncsqlite3
You can speed up asyncsqlite3 as follows:
$ pip install asyncsqlite3[uvloop]
Details
How you see this module very look like aiosqlite and asyncpg.
Implicit transactions are turned off, but you can use Connection method transaction
.
License
MIT