pydynamodb 0.2.0

Basic usage

from pydynamodb import connect

cursor = connect(aws_access_key_id="aws_access_key_id",
                aws_secret_access_key="aws_secret_access_key"
                 region_name="region_name").cursor()
cursor.execute('SELECT * FROM "ddb_table_name"')
print(cursor.fetchall())

Cursor iteration

from pydynamodb import connect

cursor = connect(aws_access_key_id="aws_access_key_id",
                aws_secret_access_key="aws_secret_access_key"
                 region_name="region_name").cursor()
cursor.execute('SELECT * FROM "ddb_table_name"')
rows = cursor.fetchall()
for row in rows:
    print(row)

Query with parameters

PyDynamoDB is able to serialize the parameters which passed to DDB and deserialize the response to Python built-in types.

from pydynamodb import connect
cursor = connect(aws_access_key_id="aws_access_key_id",
                aws_secret_access_key="aws_secret_access_key"
                 region_name="region_name").cursor()
cursor.execute("""INSERT INTO "ddb_table_name" VALUE {
                    'partition_key' = ?,
                    'sort_key' = ?,
                    'col_str' = ?,
                    'col_num' = ?,
                    'col_byte' = ?,
                    'col_ss' = ?,
                    'col_ns' = ?,
                    'col_bs' = ?,
                    'col_list' = ?,
                    'col_map' = ?,
                    'col_nested' = ?
                }""", ["pkey_value", "skey_value", "str", 100, b"ABC", # String, Number, Bytes
                        {"str", "str"}, {100, 100}, {b"A", b"B"}, # String/Numnber/Bytes Set
                        ["str", 100, b"ABC"],  # List
                        {"key1": "val", "key2": "val"}, # Map
                        ["str", 100, {"key1": "val"}] # Nested Structure
                    ])

cursor.execute('SELECT * FROM "ddb_table_name" WHERE partition_key = ?', ["key_value"])
print(cursor.fetchall())

Description of Result Set

DDB is a NoSQL database. That means except key schema, the data in each row may have flexible columns or types. PyDynamoDB cannot get a completed result set description before fetching all result data. So you have to use fetch* method to iterate the whole result set, then call cursor.description to get the full columns description.

from pydynamodb import connect

cursor = connect(aws_access_key_id="aws_access_key_id",
                aws_secret_access_key="aws_secret_access_key"
                 region_name="region_name").cursor()
cursor.execute('SELECT * FROM "ddb_table_name"')
print(cursor.fetchall())
print(cursor.description)

Dict Cursor and Result Set

Using DictCursor, you can get a dict result set with column name and value pair. This type of cursor has better performance and manipulate result data easily. But cursor.description will return empty with this way.

from pydynamodb import connect
from pydynamodb.cursor import DictCursor

cursor = connect(aws_access_key_id="aws_access_key_id",
                aws_secret_access_key="aws_secret_access_key"
                 region_name="region_name").cursor(cursor=DictCursor)
cursor.execute('SELECT * FROM "ddb_table_name"')
print(cursor.fetchall())

Transaction

Transaction is partially supported also. connection.rollback() is not implemented. Regarding information and restrictions of DDB transaction, please see the page: Performing transactions with PartiQL for DynamoDB

from pydynamodb import connect

conn = connect(aws_access_key_id="aws_access_key_id",
                aws_secret_access_key="aws_secret_access_key"
                 region_name="region_name")
cursor = conn.cursor()

conn.begin()
cursor.execute("""INSERT INTO "ddb_table_name" VALUE {'key_partition': ?, 'key_sort': ?, 'col1': ?}""",
                ["pk1", "sk1", "test"])
cursor.execute("""INSERT INTO "ddb_table_name" VALUE {'key_partition': ?, 'key_sort': ?, 'col1': ?}""",
                ["pk2", "sk2", "test"])
conn.commit()

Test with local DynamoDB

Install Local DDB, please see: Deploying DynamoDB locally on your computer. If you want to run tests with local DDB, please make sure environment variables are set properly.

USE_LOCAL_DDB=true
LOCAL_DDB_ENDPOINT_URL=http://localhost:8000