Transactions¶
Create / interact with Google Cloud Datastore transactions.
- class google.cloud.datastore.transaction.Transaction(client, read_only=False, read_time=None, begin_later=False)[source]¶
Bases:
google.cloud.datastore.batch.Batch
An abstraction representing datastore Transactions.
Transactions can be used to build up a bulk mutation and ensure all or none succeed (transactionally).
For example, the following snippet of code will put the two
save
operations (eitherinsert
orupsert
) into the same mutation, and execute those within a transaction:>>> entity1 = datastore.Entity(client.key('EntityKind', 1234)) >>> entity2 = datastore.Entity(client.key('EntityKind', 2345)) >>> with client.transaction(): ... client.put_multi([entity1, entity2])
Because it derives from
Batch
,Transaction
also providesput()
anddelete()
methods:>>> with client.transaction() as xact: ... xact.put(entity1) ... xact.delete(entity2.key)
By default, the transaction is rolled back if the transaction block exits with an error:
>>> def do_some_work(): ... return >>> class SomeException(Exception): ... pass >>> with client.transaction(): ... do_some_work() ... raise SomeException # rolls back Traceback (most recent call last): ... SomeException
If the transaction block exits without an exception, it will commit by default.
Warning
Inside a transaction, automatically assigned IDs for entities will not be available at save time! That means, if you try:
>>> with client.transaction(): ... thing1 = datastore.Entity(key=client.key('Thing')) ... client.put(thing1)
thing1
won’t have a complete key until the transaction is committed.Once you exit the transaction (or call
commit()
), the automatically generated ID will be assigned to the entity:>>> with client.transaction(): ... thing2 = datastore.Entity(key=client.key('Thing')) ... client.put(thing2) ... print(thing2.key.is_partial) # There is no ID on this key. ... True >>> print(thing2.key.is_partial) # There *is* an ID. False
If you don’t want to use the context manager you can initialize a transaction manually:
>>> transaction = client.transaction() >>> transaction.begin() >>> >>> thing3 = datastore.Entity(key=client.key('Thing')) >>> transaction.put(thing3) >>> >>> transaction.commit()
- Parameters
client (
google.cloud.datastore.client.Client
) – the client used to connect to datastore.read_only (bool) – indicates the transaction is read only.
read_time (datetime) – (Optional) Time at which the transaction reads entities. Only allowed when
read_only=True
. This feature is in private preview.begin_later (bool) – (Optional) If True, the transaction will be started lazily (i.e. when the first RPC is made). If False, the transaction will be started as soon as the context manager is entered. self.begin() can also be called manually to begin the transaction at any time. Default is False.
- Raises
ValueError
if read_time is specified whenread_only=False
.
- begin(retry=None, timeout=None)[source]¶
Begins a transaction.
This method is called automatically when entering a with statement, however it can be called explicitly if you don’t want to use a context manager.
- Parameters
retry (
google.api_core.retry.Retry
) – A retry object used to retry requests. IfNone
is specified, requests will be retried using a default configuration.timeout (float) – Time, in seconds, to wait for the request to complete. Note that if
retry
is specified, the timeout applies to each individual attempt.
- Raises
ValueError
if the transaction has already begun.
- commit(retry=None, timeout=None)[source]¶
Commits the transaction.
This is called automatically upon exiting a with statement, however it can be called explicitly if you don’t want to use a context manager.
This method has necessary side-effects:
Sets the current transaction’s ID to None.
- Parameters
retry (
google.api_core.retry.Retry
) – A retry object used to retry requests. IfNone
is specified, requests will be retried using a default configuration.timeout (float) – Time, in seconds, to wait for the request to complete. Note that if
retry
is specified, the timeout applies to each individual attempt.
- current()[source]¶
Return the topmost transaction.
Note
If the topmost element on the stack is not a transaction, returns None.
- Return type
- Returns
The current transaction (if any are active).
- property database¶
Getter for database in which the batch will run.
- Return type
- Returns
The database in which the batch will run.
- delete(key)¶
Remember a key to be deleted during
commit()
.- Parameters
key (
google.cloud.datastore.key.Key
) – the key to be deleted.- Raises
ValueError
if the batch is not in progress, if key is not complete, or if the key’sproject
does not match ours.
- property id¶
Getter for the transaction ID.
- Return type
bytes or None
- Returns
The ID of the current transaction, or None if not started.
- property mutations¶
Getter for the changes accumulated by this batch.
Every batch is committed with a single commit request containing all the work to be done as mutations. Inside a batch, calling
put()
with an entity, ordelete()
with a key, builds up the request by adding a new mutation. This getter returns the protobuf that has been built-up so far.- Return type
iterable
- Returns
The list of
datastore_pb2.Mutation
protobufs to be sent in the commit request.
- property namespace¶
Getter for namespace in which the batch will run.
- Return type
- Returns
The namespace in which the batch will run.
- property project¶
Getter for project in which the batch will run.
- Return type
- Returns
The project in which the batch will run.
- put(entity)[source]¶
Adds an entity to be committed.
Ensures the transaction is not marked readonly. Please see documentation at
put()
- Parameters
entity (
Entity
) – the entity to be saved.- Raises
RuntimeError
if the transaction is marked ReadOnly
- rollback(retry=None, timeout=None)[source]¶
Rolls back the current transaction.
This method has necessary side-effects:
Sets the current transaction’s ID to None.
- Parameters
retry (
google.api_core.retry.Retry
) – A retry object used to retry requests. IfNone
is specified, requests will be retried using a default configuration.timeout (float) – Time, in seconds, to wait for the request to complete. Note that if
retry
is specified, the timeout applies to each individual attempt.