API¶

High Level API¶

DynamoDB Models for PynamoDB

class pynamodb.models.Model(hash_key: Optional[Any] = None, range_key: Optional[Any] = None, _user_instantiated: bool = True, **attributes)¶

Defines a PynamoDB Model

This model is backed by a table in DynamoDB. You can create the table by with the create_table method.

exception DoesNotExist(msg: Optional[str] = None, cause: Optional[Exception] = None)¶: Raised when an item queried does not exist

classmethod batch_get(items: Iterable[Union[Any, Iterable[Any]]], consistent_read: Optional[bool] = None, attributes_to_get: Optional[Sequence[str]] = None, settings: pynamodb.settings.OperationSettings = <pynamodb.settings.OperationSettings object>) → Iterator[_T]¶

BatchGetItem for this model

Parameters:	items – Should be a list of hash keys to retrieve, or a list of tuples if range keys are used.

classmethod batch_write(auto_commit: bool = True, settings: pynamodb.settings.OperationSettings = <pynamodb.settings.OperationSettings object>) → pynamodb.models.BatchWrite[~_T][_T]¶

Returns a BatchWrite context manager for a batch operation.

Parameters:	auto_commit – If true, the context manager will commit writes incrementally as items are written to as necessary to honor item count limits in the DynamoDB API (see BatchWrite). Regardless of the value passed here, changes automatically commit on context exit (whether successful or not).

classmethod count(hash_key: Optional[Any] = None, range_key_condition: Optional[pynamodb.expressions.condition.Condition] = None, filter_condition: Optional[pynamodb.expressions.condition.Condition] = None, consistent_read: bool = False, index_name: Optional[str] = None, limit: Optional[int] = None, rate_limit: Optional[float] = None, settings: pynamodb.settings.OperationSettings = <pynamodb.settings.OperationSettings object>) → int¶

Provides a filtered count

Parameters:	hash_key – The hash key to query. Can be None. range_key_condition – Condition for range key filter_condition – Condition used to restrict the query results consistent_read – If True, a consistent read is performed index_name – If set, then this index is used rate_limit – If set then consumed capacity will be limited to this amount per second

classmethod create_table(wait: bool = False, read_capacity_units: Optional[int] = None, write_capacity_units: Optional[int] = None, billing_mode: Optional[str] = None, ignore_update_ttl_errors: bool = False) → Any¶

Create the table for this model

Parameters:	wait – If set, then this call will block until the table is ready for use read_capacity_units – Sets the read capacity units for this table write_capacity_units – Sets the write capacity units for this table billing_mode – Sets the billing mode ‘PROVISIONED’ (default) or ‘PAY_PER_REQUEST’ for this table

delete(condition: Optional[pynamodb.expressions.condition.Condition] = None, settings: pynamodb.settings.OperationSettings = <pynamodb.settings.OperationSettings object>, *, add_version_condition: bool = True) → Any¶

Deletes this object from DynamoDB.

Parameters:	add_version_condition – For models which have a `VersionAttribute`, specifies whether the item should only be deleted if its current version matches the expected one. Set to False for a ‘delete anyway’ strategy.
Raises:	pynamodb.exceptions.DeleteError – If the record can not be deleted

classmethod delete_table() → Any¶: Delete the table for this model

classmethod describe_table() → Any¶: Returns the result of a DescribeTable operation on this model’s table

deserialize(attribute_values: Dict[str, Dict[str, Any]]) → None¶: Sets attributes sent back from DynamoDB on this object

classmethod exists() → bool¶: Returns True if this table exists, False otherwise

classmethod from_raw_data(data: Dict[str, Any]) → _T¶

Returns an instance of this class from the raw data

Parameters:	data – A serialized DynamoDB object

classmethod get(hash_key: Any, range_key: Optional[Any] = None, consistent_read: bool = False, attributes_to_get: Optional[Sequence[str]] = None, settings: pynamodb.settings.OperationSettings = <pynamodb.settings.OperationSettings object>) → _T¶

Returns a single object using the provided keys

Parameters:	hash_key – The hash key of the desired item range_key – The range key of the desired item, only used when appropriate. consistent_read – attributes_to_get –
Raises:	ModelInstance.DoesNotExist – if the object to be updated does not exist

classmethod query(hash_key: Any, range_key_condition: Optional[pynamodb.expressions.condition.Condition] = None, filter_condition: Optional[pynamodb.expressions.condition.Condition] = None, consistent_read: bool = False, index_name: Optional[str] = None, scan_index_forward: Optional[bool] = None, limit: Optional[int] = None, last_evaluated_key: Optional[Dict[str, Dict[str, Any]]] = None, attributes_to_get: Optional[Iterable[str]] = None, page_size: Optional[int] = None, rate_limit: Optional[float] = None, settings: pynamodb.settings.OperationSettings = <pynamodb.settings.OperationSettings object>) → pynamodb.pagination.ResultIterator[~_T][_T]¶

Provides a high level query API

Parameters:

hash_key – The hash key to query
range_key_condition – Condition for range key
filter_condition – Condition used to restrict the query results
consistent_read – If True, a consistent read is performed
index_name – If set, then this index is used
limit – Used to limit the number of results returned
scan_index_forward – If set, then used to specify the same parameter to the DynamoDB API. Controls descending or ascending results
last_evaluated_key – If set, provides the starting point for query.
attributes_to_get – If set, only returns these elements
page_size – Page size of the query to DynamoDB
rate_limit – If set then consumed capacity will be limited to this amount per second

refresh(consistent_read: bool = False, settings: pynamodb.settings.OperationSettings = <pynamodb.settings.OperationSettings object>) → None¶

Retrieves this object’s data from dynamodb and syncs this local object

Parameters:	consistent_read – If True, then a consistent read is performed. settings – per-operation settings
Raises:	ModelInstance.DoesNotExist – if the object to be updated does not exist

save(condition: Optional[pynamodb.expressions.condition.Condition] = None, settings: pynamodb.settings.OperationSettings = <pynamodb.settings.OperationSettings object>, *, add_version_condition: bool = True) → Dict[str, Any]¶: Save this object to dynamodb

classmethod scan(filter_condition: Optional[pynamodb.expressions.condition.Condition] = None, segment: Optional[int] = None, total_segments: Optional[int] = None, limit: Optional[int] = None, last_evaluated_key: Optional[Dict[str, Dict[str, Any]]] = None, page_size: Optional[int] = None, consistent_read: Optional[bool] = None, index_name: Optional[str] = None, rate_limit: Optional[float] = None, attributes_to_get: Optional[Sequence[str]] = None, settings: pynamodb.settings.OperationSettings = <pynamodb.settings.OperationSettings object>) → pynamodb.pagination.ResultIterator[~_T][_T]¶

Iterates through all items in the table

Parameters:

filter_condition – Condition used to restrict the scan results
segment – If set, then scans the segment
total_segments – If set, then specifies total segments
limit – Used to limit the number of results returned
last_evaluated_key – If set, provides the starting point for scan.
page_size – Page size of the scan to DynamoDB
consistent_read – If True, a consistent read is performed
index_name – If set, then this index is used
rate_limit – If set then consumed capacity will be limited to this amount per second
attributes_to_get – If set, specifies the properties to include in the projection expression

serialize(null_check: bool = True) → Dict[str, Dict[str, Any]]¶: Serialize attribute values for DynamoDB

update(actions: List[pynamodb.expressions.update.Action], condition: Optional[pynamodb.expressions.condition.Condition] = None, settings: pynamodb.settings.OperationSettings = <pynamodb.settings.OperationSettings object>, *, add_version_condition: bool = True) → Any¶

Updates an item using the UpdateItem operation.

Parameters:

actions – a list of Action updates to apply
condition – an optional Condition on which to update
settings – per-operation settings
add_version_condition – For models which have a VersionAttribute, specifies whether only to update if the version matches the model that is currently loaded. Set to False for a ‘last write wins’ strategy. Regardless, the version will always be incremented to prevent “rollbacks” by concurrent save() calls.

Raises:

ModelInstance.DoesNotExist – if the object to be updated does not exist
pynamodb.exceptions.UpdateError – if the condition is not met

classmethod update_ttl(ignore_update_ttl_errors: bool) → None¶: Attempt to update the TTL on the table. Certain implementations (eg: dynalite) do not support updating TTLs and will fail.

PynamoDB attributes

class pynamodb.attributes.Attribute(hash_key: bool = False, range_key: bool = False, null: Optional[bool] = None, default: Union[_T, Callable[[...], _T], None] = None, default_for_new: Union[Any, Callable[[...], _T], None] = None, attr_name: Optional[str] = None)¶

An attribute of a model

deserialize(value: Any) → Any¶: Performs any needed deserialization on the value

serialize(value: Any) → Any¶: This method should return a dynamodb compatible value

class pynamodb.attributes.BinaryAttribute(hash_key: bool = False, range_key: bool = False, null: Optional[bool] = None, default: Union[_T, Callable[[...], _T], None] = None, default_for_new: Union[Any, Callable[[...], _T], None] = None, attr_name: Optional[str] = None)¶

A binary attribute

deserialize(value)¶: Returns a decoded byte string from a base64 encoded value

serialize(value)¶: Returns a base64 encoded binary string

class pynamodb.attributes.BinarySetAttribute(hash_key: bool = False, range_key: bool = False, null: Optional[bool] = None, default: Union[_T, Callable[[...], _T], None] = None, default_for_new: Union[Any, Callable[[...], _T], None] = None, attr_name: Optional[str] = None)¶

A binary set

deserialize(value)¶: Returns a set of decoded byte strings from base64 encoded values.

serialize(value)¶: Returns a list of base64 encoded binary strings. Encodes empty sets as “None”.

class pynamodb.attributes.BooleanAttribute(hash_key: bool = False, range_key: bool = False, null: Optional[bool] = None, default: Union[_T, Callable[[...], _T], None] = None, default_for_new: Union[Any, Callable[[...], _T], None] = None, attr_name: Optional[str] = None)¶

A class for boolean attributes

deserialize(value)¶: Performs any needed deserialization on the value

serialize(value)¶: This method should return a dynamodb compatible value

class pynamodb.attributes.DiscriminatorAttribute(attr_name: Optional[str] = None)¶

deserialize(value)¶: Returns the class corresponding to the given discriminator value.

serialize(value)¶: Returns the discriminator value corresponding to the given class.

class pynamodb.attributes.DynamicMapAttribute(**attributes)¶

A map attribute that supports declaring attributes (like an AttributeContainer) but will also store any other values that are set on it (like a raw MapAttribute).

>>> class MyDynamicMapAttribute(DynamicMapAttribute):
>>>     a_date_time = UTCDateTimeAttribute()  # raw map attributes cannot serialize/deserialize datetime values
>>>
>>> dynamic_map = MyDynamicMapAttribute()
>>> dynamic_map.a_date_time = datetime.utcnow()
>>> dynamic_map.a_number = 5
>>> dynamic_map.serialize()  # {'a_date_time': {'S': 'xxx'}, 'a_number': {'N': '5'}}

deserialize(values)¶: Decode as a dict.

serialize(values, *, null_check: bool = True)¶: This method should return a dynamodb compatible value

class pynamodb.attributes.JSONAttribute(hash_key: bool = False, range_key: bool = False, null: Optional[bool] = None, default: Union[_T, Callable[[...], _T], None] = None, default_for_new: Union[Any, Callable[[...], _T], None] = None, attr_name: Optional[str] = None)¶

A JSON Attribute

Encodes JSON to unicode internally

deserialize(value)¶: Deserializes JSON

serialize(value) → Optional[str]¶: Serializes JSON to unicode

class pynamodb.attributes.ListAttribute(hash_key: bool = False, range_key: bool = False, null: Optional[bool] = None, default: Union[Any, Callable[[...], Any], None] = None, attr_name: Optional[str] = None, of: Optional[Type[_T]] = None)¶

deserialize(values)¶: Decode from list of AttributeValue types.

serialize(values, *, null_check: bool = True)¶: Encode the given list of objects into a list of AttributeValue types.

class pynamodb.attributes.MapAttribute(**attributes)¶

A Map Attribute

The MapAttribute class can be used to store a JSON document as “raw” name-value pairs, or it can be subclassed and the document fields represented as class attributes using Attribute instances.

To support the ability to subclass MapAttribute and use it as an AttributeContainer, instances of MapAttribute behave differently based both on where they are instantiated and on their type. Because of this complicated behavior, a bit of an introduction is warranted.

Models that contain a MapAttribute define its properties using a class attribute on the model. For example, below we define “MyModel” which contains a MapAttribute “my_map”:

class MyModel(Model):: my_map = MapAttribute(attr_name=”dynamo_name”, default={})

When instantiated in this manner (as a class attribute of an AttributeContainer class), the MapAttribute class acts as an instance of the Attribute class. The instance stores data about the attribute (in this example the dynamo name and default value), and acts as a data descriptor, storing any value bound to it on the attribute_values dictionary of the containing instance (in this case an instance of MyModel).

Unlike other Attribute types, the value that gets bound to the containing instance is a new instance of MapAttribute, not an instance of the primitive type. For example, a UnicodeAttribute stores strings in the attribute_values of the containing instance; a MapAttribute does not store a dict but instead stores a new instance of itself. This difference in behavior is necessary when subclassing MapAttribute in order to access the Attribute data descriptors that represent the document fields.

For example, below we redefine “MyModel” to use a subclass of MapAttribute as “my_map”:

class MyMapAttribute(MapAttribute):: my_internal_map = MapAttribute()
class MyModel(Model):: my_map = MyMapAttribute(attr_name=”dynamo_name”, default = {})

In order to set the value of my_internal_map on an instance of MyModel we need the bound value for “my_map” to be an instance of MapAttribute so that it acts as a data descriptor:

MyModel().my_map.my_internal_map = {‘foo’: ‘bar’}

That is the attribute access of “my_map” must return a MyMapAttribute instance and not a dict.

When an instance is used in this manner (bound to an instance of an AttributeContainer class), the MapAttribute class acts as an AttributeContainer class itself. The instance does not store data about the attribute, and does not act as a data descriptor. The instance stores name-value pairs in its internal attribute_values dictionary.

Thus while MapAttribute multiply inherits from Attribute and AttributeContainer, a MapAttribute instance does not behave as both an Attribute AND an AttributeContainer. Rather an instance of MapAttribute behaves EITHER as an Attribute OR as an AttributeContainer, depending on where it was instantiated.

So, how do we create this dichotomous behavior? All MapAttribute instances are initialized as AttributeContainers only. During construction of AttributeContainer classes (subclasses of MapAttribute and Model), any instances that are class attributes are transformed from AttributeContainers to Attributes (via the _make_attribute method call).

deserialize(values)¶: Decode as a dict.

serialize(values, *, null_check: bool = True)¶: This method should return a dynamodb compatible value

class pynamodb.attributes.NullAttribute(hash_key: bool = False, range_key: bool = False, null: Optional[bool] = None, default: Union[_T, Callable[[...], _T], None] = None, default_for_new: Union[Any, Callable[[...], _T], None] = None, attr_name: Optional[str] = None)¶

deserialize(value)¶: Performs any needed deserialization on the value

serialize(value)¶: This method should return a dynamodb compatible value

class pynamodb.attributes.NumberAttribute(hash_key: bool = False, range_key: bool = False, null: Optional[bool] = None, default: Union[_T, Callable[[...], _T], None] = None, default_for_new: Union[Any, Callable[[...], _T], None] = None, attr_name: Optional[str] = None)¶

A number attribute

deserialize(value)¶: Decode numbers from JSON

serialize(value)¶: Encode numbers as JSON

class pynamodb.attributes.NumberSetAttribute(hash_key: bool = False, range_key: bool = False, null: Optional[bool] = None, default: Union[_T, Callable[[...], _T], None] = None, default_for_new: Union[Any, Callable[[...], _T], None] = None, attr_name: Optional[str] = None)¶

A number set attribute

deserialize(value)¶: Returns a set from a JSON list of numbers.

serialize(value)¶: Encodes a set of numbers as a JSON list. Encodes empty sets as “None”.

class pynamodb.attributes.TTLAttribute(hash_key: bool = False, range_key: bool = False, null: Optional[bool] = None, default: Union[_T, Callable[[...], _T], None] = None, default_for_new: Union[Any, Callable[[...], _T], None] = None, attr_name: Optional[str] = None)¶

A time-to-live attribute that signifies when the item expires and can be automatically deleted. It can be assigned with a timezone-aware datetime value (for absolute expiry time) or a timedelta value (for expiry relative to the current time), but always reads as a UTC datetime value.

deserialize(value)¶: Deserializes a timestamp (Unix time) as a UTC datetime.

serialize(value)¶: Serializes a datetime as a timestamp (Unix time).

class pynamodb.attributes.UTCDateTimeAttribute(hash_key: bool = False, range_key: bool = False, null: Optional[bool] = None, default: Union[_T, Callable[[...], _T], None] = None, default_for_new: Union[Any, Callable[[...], _T], None] = None, attr_name: Optional[str] = None)¶

An attribute for storing a UTC Datetime

deserialize(value)¶: Takes a UTC datetime string and returns a datetime object

serialize(value)¶: Takes a datetime object and returns a string

class pynamodb.attributes.UnicodeAttribute(hash_key: bool = False, range_key: bool = False, null: Optional[bool] = None, default: Union[_T, Callable[[...], _T], None] = None, default_for_new: Union[Any, Callable[[...], _T], None] = None, attr_name: Optional[str] = None)¶: A unicode attribute

class pynamodb.attributes.UnicodeSetAttribute(hash_key: bool = False, range_key: bool = False, null: Optional[bool] = None, default: Union[_T, Callable[[...], _T], None] = None, default_for_new: Union[Any, Callable[[...], _T], None] = None, attr_name: Optional[str] = None)¶

A unicode set

deserialize(value)¶: Returns a set from a list of strings.

serialize(value)¶: Returns a list of strings. Encodes empty sets as “None”.

class pynamodb.attributes.VersionAttribute(hash_key: bool = False, range_key: bool = False, null: Optional[bool] = None, default: Union[_T, Callable[[...], _T], None] = None, default_for_new: Union[Any, Callable[[...], _T], None] = None, attr_name: Optional[str] = None)¶

A number attribute that implements optimistic locking.

deserialize(value)¶: Decode numbers from JSON and cast to int.

serialize(value)¶: Cast value to int then encode as JSON

PynamoDB Indexes

class pynamodb.indexes.AllProjection¶: An ALL projection

class pynamodb.indexes.GlobalSecondaryIndex¶: A global secondary index

class pynamodb.indexes.IncludeProjection(non_attr_keys: Optional[List[str]] = None)¶: An INCLUDE projection

class pynamodb.indexes.Index¶

Base class for secondary indexes

classmethod count(hash_key: Any, range_key_condition: Optional[pynamodb.expressions.condition.Condition] = None, filter_condition: Optional[pynamodb.expressions.condition.Condition] = None, consistent_read: bool = False, limit: Optional[int] = None, rate_limit: Optional[float] = None) → int¶: Count on an index

classmethod query(hash_key: Any, range_key_condition: Optional[pynamodb.expressions.condition.Condition] = None, filter_condition: Optional[pynamodb.expressions.condition.Condition] = None, consistent_read: Optional[bool] = False, scan_index_forward: Optional[bool] = None, limit: Optional[int] = None, last_evaluated_key: Optional[Dict[str, Dict[str, Any]]] = None, attributes_to_get: Optional[List[str]] = None, page_size: Optional[int] = None, rate_limit: Optional[float] = None) → pynamodb.pagination.ResultIterator[~_M][_M]¶: Queries an index

classmethod scan(filter_condition: Optional[pynamodb.expressions.condition.Condition] = None, segment: Optional[int] = None, total_segments: Optional[int] = None, limit: Optional[int] = None, last_evaluated_key: Optional[Dict[str, Dict[str, Any]]] = None, page_size: Optional[int] = None, consistent_read: Optional[bool] = None, rate_limit: Optional[float] = None, attributes_to_get: Optional[List[str]] = None) → pynamodb.pagination.ResultIterator[~_M][_M]¶: Scans an index

class pynamodb.indexes.KeysOnlyProjection¶: Keys only projection

class pynamodb.indexes.LocalSecondaryIndex¶: A local secondary index

class pynamodb.indexes.Projection¶: A class for presenting projections

class pynamodb.transactions.TransactGet(*args, **kwargs)¶

get(model_cls: Type[_M], hash_key: Any, range_key: Optional[Any] = None) → pynamodb.models._ModelFuture[~_M][_M]¶

Adds the operation arguments for an item to list of models to get returns a _ModelFuture object as a placeholder

Parameters:	model_cls – hash_key – range_key –
Returns:

class pynamodb.transactions.TransactWrite(client_request_token: Optional[str] = None, return_item_collection_metrics: Optional[str] = None, **kwargs)¶

class pynamodb.transactions.Transaction(connection: pynamodb.connection.base.Connection, return_consumed_capacity: Optional[str] = None)¶: Base class for a type of transaction operation

class pynamodb.pagination.PageIterator(operation: Callable, args: Any, kwargs: Dict[str, Any], rate_limit: Optional[float] = None, settings: pynamodb.settings.OperationSettings = <pynamodb.settings.OperationSettings object>)¶

PageIterator handles Query and Scan result pagination.

http://docs.aws.amazon.com/amazondynamodb/latest/developerguide/Query.html#Query.Pagination http://docs.aws.amazon.com/amazondynamodb/latest/developerguide/Scan.html#Scan.Pagination

class pynamodb.pagination.RateLimiter(rate_limit: float, time_module: Optional[Any] = None)¶

RateLimiter limits operations to a pre-set rate of units/seconds

Example:

Initialize a RateLimiter with the desired rate: rate_limiter = RateLimiter(rate_limit)
Now, every time before calling an operation, call acquire(): rate_limiter.acquire()
And after an operation, update the number of units consumed: rate_limiter.consume(units)

acquire() → None¶

Sleeps the appropriate amount of time to follow the rate limit restriction

Returns:	None

consume(units: int) → None¶

Records the amount of units consumed.

Parameters:	units – Number of units consumed
Returns:	None

rate_limit¶: A limit of units per seconds

class pynamodb.pagination.ResultIterator(operation: Callable, args: Any, kwargs: Dict[str, Any], map_fn: Optional[Callable] = None, limit: Optional[int] = None, rate_limit: Optional[float] = None, settings: pynamodb.settings.OperationSettings = <pynamodb.settings.OperationSettings object>)¶

ResultIterator handles Query and Scan item pagination.

http://docs.aws.amazon.com/amazondynamodb/latest/developerguide/Query.html#Query.Pagination http://docs.aws.amazon.com/amazondynamodb/latest/developerguide/Scan.html#Scan.Pagination

Low Level API¶

PynamoDB lowest level connection

class pynamodb.connection.Connection(region: Optional[str] = None, host: Optional[str] = None, read_timeout_seconds: Optional[float] = None, connect_timeout_seconds: Optional[float] = None, max_retry_attempts: Optional[int] = None, base_backoff_ms: Optional[int] = None, max_pool_connections: Optional[int] = None, extra_headers: Optional[Mapping[str, str]] = None, aws_access_key_id: Optional[str] = None, aws_secret_access_key: Optional[str] = None, aws_session_token: Optional[str] = None)¶

A higher level abstraction over botocore

add_meta_table(meta_table: pynamodb.connection.base.MetaTable) → None¶: Adds information about the table’s schema.

batch_get_item(table_name: str, keys: Sequence[str], consistent_read: Optional[bool] = None, return_consumed_capacity: Optional[str] = None, attributes_to_get: Optional[Any] = None, settings: pynamodb.settings.OperationSettings = <pynamodb.settings.OperationSettings object>) → Dict[KT, VT]¶: Performs the batch get item operation

batch_write_item(table_name: str, put_items: Optional[Any] = None, delete_items: Optional[Any] = None, return_consumed_capacity: Optional[str] = None, return_item_collection_metrics: Optional[str] = None, settings: pynamodb.settings.OperationSettings = <pynamodb.settings.OperationSettings object>) → Dict[KT, VT]¶: Performs the batch_write_item operation

client¶: Returns a botocore dynamodb client

create_table(table_name: str, attribute_definitions: Optional[Any] = None, key_schema: Optional[Any] = None, read_capacity_units: Optional[int] = None, write_capacity_units: Optional[int] = None, global_secondary_indexes: Optional[Any] = None, local_secondary_indexes: Optional[Any] = None, stream_specification: Optional[Dict[KT, VT]] = None, billing_mode: str = 'PROVISIONED', tags: Optional[Dict[str, str]] = None) → Dict[KT, VT]¶: Performs the CreateTable operation

delete_item(table_name: str, hash_key: str, range_key: Optional[str] = None, condition: Optional[pynamodb.expressions.condition.Condition] = None, return_values: Optional[str] = None, return_consumed_capacity: Optional[str] = None, return_item_collection_metrics: Optional[str] = None, settings: pynamodb.settings.OperationSettings = <pynamodb.settings.OperationSettings object>) → Dict[KT, VT]¶: Performs the DeleteItem operation and returns the result

delete_table(table_name: str) → Dict[KT, VT]¶: Performs the DeleteTable operation

describe_table(table_name: str) → Dict[KT, VT]¶: Performs the DescribeTable operation

dispatch(operation_name: str, operation_kwargs: Dict[KT, VT], settings: pynamodb.settings.OperationSettings = <pynamodb.settings.OperationSettings object>) → Dict[KT, VT]¶

Dispatches operation_name with arguments operation_kwargs

Raises TableDoesNotExist if the specified table does not exist

get_attribute_type(table_name: str, attribute_name: str, value: Optional[Any] = None) → str¶: Returns the proper attribute type for a given attribute name :param value: The attribute value an be supplied just in case the type is already included

get_consumed_capacity_map(return_consumed_capacity: str) → Dict[KT, VT]¶: Builds the consumed capacity map that is common to several operations

get_exclusive_start_key_map(table_name: str, exclusive_start_key: str) → Dict[KT, VT]¶: Builds the exclusive start key attribute map

get_identifier_map(table_name: str, hash_key: str, range_key: Optional[str] = None, key: str = 'Key') → Dict[KT, VT]¶: Builds the identifier map that is common to several operations

get_item(table_name: str, hash_key: str, range_key: Optional[str] = None, consistent_read: bool = False, attributes_to_get: Optional[Any] = None, settings: pynamodb.settings.OperationSettings = <pynamodb.settings.OperationSettings object>) → Dict[KT, VT]¶: Performs the GetItem operation and returns the result

get_item_attribute_map(table_name: str, attributes: Any, item_key: str = 'Item', pythonic_key: bool = True) → Dict[KT, VT]¶: Builds up a dynamodb compatible AttributeValue map

get_item_collection_map(return_item_collection_metrics: str) → Dict[KT, VT]¶: Builds the item collection map

get_meta_table(table_name: str) → pynamodb.connection.base.MetaTable¶: Returns information about the table’s schema.

get_return_values_map(return_values: str) → Dict[KT, VT]¶: Builds the return values map that is common to several operations

get_return_values_on_condition_failure_map(return_values_on_condition_failure: str) → Dict[KT, VT]¶: Builds the return values map that is common to several operations

list_tables(exclusive_start_table_name: Optional[str] = None, limit: Optional[int] = None) → Dict[KT, VT]¶: Performs the ListTables operation

parse_attribute(attribute: Any, return_type: bool = False) → Any¶: Returns the attribute value, where the attribute can be a raw attribute value, or a dictionary containing the type: {‘S’: ‘String value’}

put_item(table_name: str, hash_key: str, range_key: Optional[str] = None, attributes: Optional[Any] = None, condition: Optional[pynamodb.expressions.condition.Condition] = None, return_values: Optional[str] = None, return_consumed_capacity: Optional[str] = None, return_item_collection_metrics: Optional[str] = None, settings: pynamodb.settings.OperationSettings = <pynamodb.settings.OperationSettings object>) → Dict[KT, VT]¶: Performs the PutItem operation and returns the result

query(table_name: str, hash_key: str, range_key_condition: Optional[pynamodb.expressions.condition.Condition] = None, filter_condition: Optional[Any] = None, attributes_to_get: Optional[Any] = None, consistent_read: bool = False, exclusive_start_key: Optional[Any] = None, index_name: Optional[str] = None, limit: Optional[int] = None, return_consumed_capacity: Optional[str] = None, scan_index_forward: Optional[bool] = None, select: Optional[str] = None, settings: pynamodb.settings.OperationSettings = <pynamodb.settings.OperationSettings object>) → Dict[KT, VT]¶: Performs the Query operation and returns the result

scan(table_name: str, filter_condition: Optional[Any] = None, attributes_to_get: Optional[Any] = None, limit: Optional[int] = None, return_consumed_capacity: Optional[str] = None, exclusive_start_key: Optional[str] = None, segment: Optional[int] = None, total_segments: Optional[int] = None, consistent_read: Optional[bool] = None, index_name: Optional[str] = None, settings: pynamodb.settings.OperationSettings = <pynamodb.settings.OperationSettings object>) → Dict[KT, VT]¶: Performs the scan operation

session¶: Returns a valid botocore session

transact_get_items(get_items: Sequence[Dict[KT, VT]], return_consumed_capacity: Optional[str] = None, settings: pynamodb.settings.OperationSettings = <pynamodb.settings.OperationSettings object>) → Dict[KT, VT]¶: Performs the TransactGet operation and returns the result

transact_write_items(condition_check_items: Sequence[Dict[KT, VT]], delete_items: Sequence[Dict[KT, VT]], put_items: Sequence[Dict[KT, VT]], update_items: Sequence[Dict[KT, VT]], client_request_token: Optional[str] = None, return_consumed_capacity: Optional[str] = None, return_item_collection_metrics: Optional[str] = None, settings: pynamodb.settings.OperationSettings = <pynamodb.settings.OperationSettings object>) → Dict[KT, VT]¶: Performs the TransactWrite operation and returns the result

update_item(table_name: str, hash_key: str, range_key: Optional[str] = None, actions: Optional[Sequence[pynamodb.expressions.update.Action]] = None, condition: Optional[pynamodb.expressions.condition.Condition] = None, return_consumed_capacity: Optional[str] = None, return_item_collection_metrics: Optional[str] = None, return_values: Optional[str] = None, settings: pynamodb.settings.OperationSettings = <pynamodb.settings.OperationSettings object>) → Dict[KT, VT]¶: Performs the UpdateItem operation

update_table(table_name: str, read_capacity_units: Optional[int] = None, write_capacity_units: Optional[int] = None, global_secondary_index_updates: Optional[Any] = None) → Dict[KT, VT]¶: Performs the UpdateTable operation

update_time_to_live(table_name: str, ttl_attribute_name: str) → Dict[KT, VT]¶: Performs the UpdateTimeToLive operation

class pynamodb.connection.TableConnection(table_name: str, region: Optional[str] = None, host: Optional[str] = None, connect_timeout_seconds: Optional[float] = None, read_timeout_seconds: Optional[float] = None, max_retry_attempts: Optional[int] = None, base_backoff_ms: Optional[int] = None, max_pool_connections: Optional[int] = None, extra_headers: Optional[Mapping[str, str]] = None, aws_access_key_id: Optional[str] = None, aws_secret_access_key: Optional[str] = None, aws_session_token: Optional[str] = None, *, meta_table: Optional[pynamodb.connection.base.MetaTable] = None)¶

A higher level abstraction over botocore

batch_get_item(keys: Sequence[str], consistent_read: Optional[bool] = None, return_consumed_capacity: Optional[str] = None, attributes_to_get: Optional[Any] = None, settings: pynamodb.settings.OperationSettings = <pynamodb.settings.OperationSettings object>) → Dict[KT, VT]¶: Performs the batch get item operation

batch_write_item(put_items: Optional[Any] = None, delete_items: Optional[Any] = None, return_consumed_capacity: Optional[str] = None, return_item_collection_metrics: Optional[str] = None, settings: pynamodb.settings.OperationSettings = <pynamodb.settings.OperationSettings object>) → Dict[KT, VT]¶: Performs the batch_write_item operation

create_table(attribute_definitions: Optional[Any] = None, key_schema: Optional[Any] = None, read_capacity_units: Optional[int] = None, write_capacity_units: Optional[int] = None, global_secondary_indexes: Optional[Any] = None, local_secondary_indexes: Optional[Any] = None, stream_specification: Optional[Dict[KT, VT]] = None, billing_mode: str = 'PROVISIONED', tags: Optional[Dict[str, str]] = None) → Dict[KT, VT]¶: Performs the CreateTable operation and returns the result

delete_item(hash_key: str, range_key: Optional[str] = None, condition: Optional[pynamodb.expressions.condition.Condition] = None, return_values: Optional[str] = None, return_consumed_capacity: Optional[str] = None, return_item_collection_metrics: Optional[str] = None, settings: pynamodb.settings.OperationSettings = <pynamodb.settings.OperationSettings object>) → Dict[KT, VT]¶: Performs the DeleteItem operation and returns the result

delete_table() → Dict[KT, VT]¶: Performs the DeleteTable operation and returns the result

describe_table() → Dict[KT, VT]¶: Performs the DescribeTable operation and returns the result

get_item(hash_key: str, range_key: Optional[str] = None, consistent_read: bool = False, attributes_to_get: Optional[Any] = None, settings: pynamodb.settings.OperationSettings = <pynamodb.settings.OperationSettings object>) → Dict[KT, VT]¶: Performs the GetItem operation and returns the result

get_meta_table() → pynamodb.connection.base.MetaTable¶: Returns a MetaTable

put_item(hash_key: str, range_key: Optional[str] = None, attributes: Optional[Any] = None, condition: Optional[pynamodb.expressions.condition.Condition] = None, return_values: Optional[str] = None, return_consumed_capacity: Optional[str] = None, return_item_collection_metrics: Optional[str] = None, settings: pynamodb.settings.OperationSettings = <pynamodb.settings.OperationSettings object>) → Dict[KT, VT]¶: Performs the PutItem operation and returns the result

query(hash_key: str, range_key_condition: Optional[pynamodb.expressions.condition.Condition] = None, filter_condition: Optional[Any] = None, attributes_to_get: Optional[Any] = None, consistent_read: bool = False, exclusive_start_key: Optional[Any] = None, index_name: Optional[str] = None, limit: Optional[int] = None, return_consumed_capacity: Optional[str] = None, scan_index_forward: Optional[bool] = None, select: Optional[str] = None, settings: pynamodb.settings.OperationSettings = <pynamodb.settings.OperationSettings object>) → Dict[KT, VT]¶: Performs the Query operation and returns the result

scan(filter_condition: Optional[Any] = None, attributes_to_get: Optional[Any] = None, limit: Optional[int] = None, return_consumed_capacity: Optional[str] = None, segment: Optional[int] = None, total_segments: Optional[int] = None, exclusive_start_key: Optional[str] = None, consistent_read: Optional[bool] = None, index_name: Optional[str] = None, settings: pynamodb.settings.OperationSettings = <pynamodb.settings.OperationSettings object>) → Dict[KT, VT]¶: Performs the scan operation

update_item(hash_key: str, range_key: Optional[str] = None, actions: Optional[Sequence[pynamodb.expressions.update.Action]] = None, condition: Optional[pynamodb.expressions.condition.Condition] = None, return_consumed_capacity: Optional[str] = None, return_item_collection_metrics: Optional[str] = None, return_values: Optional[str] = None, settings: pynamodb.settings.OperationSettings = <pynamodb.settings.OperationSettings object>) → Dict[KT, VT]¶: Performs the UpdateItem operation

update_table(read_capacity_units: Optional[int] = None, write_capacity_units: Optional[int] = None, global_secondary_index_updates: Optional[Any] = None) → Dict[KT, VT]¶: Performs the UpdateTable operation and returns the result

update_time_to_live(ttl_attr_name: str) → Dict[KT, VT]¶: Performs the UpdateTimeToLive operation and returns the result

Exceptions¶

exception pynamodb.exceptions.PynamoDBConnectionError(msg: Optional[str] = None, cause: Optional[Exception] = None)¶: A base class for connection errors

exception pynamodb.exceptions.DeleteError(msg: Optional[str] = None, cause: Optional[Exception] = None)¶: Raised when an error occurs deleting an item

exception pynamodb.exceptions.QueryError(msg: Optional[str] = None, cause: Optional[Exception] = None)¶: Raised when queries fail

exception pynamodb.exceptions.ScanError(msg: Optional[str] = None, cause: Optional[Exception] = None)¶: Raised when a scan operation fails

exception pynamodb.exceptions.PutError(msg: Optional[str] = None, cause: Optional[Exception] = None)¶: Raised when an item fails to be created

exception pynamodb.exceptions.UpdateError(msg: Optional[str] = None, cause: Optional[Exception] = None)¶: Raised when an item fails to be updated

exception pynamodb.exceptions.GetError(msg: Optional[str] = None, cause: Optional[Exception] = None)¶: Raised when an item fails to be retrieved

exception pynamodb.exceptions.TableError(msg: Optional[str] = None, cause: Optional[Exception] = None)¶: An error involving a dynamodb table operation

exception pynamodb.exceptions.TableDoesNotExist(table_name: str)¶: Raised when an operation is attempted on a table that doesn’t exist

exception pynamodb.exceptions.DoesNotExist(msg: Optional[str] = None, cause: Optional[Exception] = None)¶: Raised when an item queried does not exist

exception pynamodb.exceptions.TransactWriteError(msg: Optional[str] = None, cause: Optional[Exception] = None)¶: Raised when a TransactWrite operation fails

exception pynamodb.exceptions.TransactGetError(msg: Optional[str] = None, cause: Optional[Exception] = None)¶: Raised when a TransactGet operation fails

exception pynamodb.exceptions.InvalidStateError(msg: Optional[str] = None, cause: Optional[Exception] = None)¶: Raises when the internal state of an operation context is invalid

exception pynamodb.exceptions.AttributeDeserializationError(attr_name: str, attr_type: str)¶: Raised when attribute type is invalid

exception pynamodb.exceptions.AttributeNullError(attr_name: str)¶

class pynamodb.exceptions.CancellationReason(*, code: str, message: Optional[str])¶: A reason for a transaction cancellation.