Python gRPC Client for EventStoreDB

This Python package provides a Python gRPC client for EventStoreDB.

This client has been developed in collaboration with the EventStoreDB team. Although not all the features of EventStoreDB are supported by this client, many of the most useful ones are presented in an easy-to-use interface.

This client has been tested to work with EventStoreDB LTS versions 21.10, without and without SSL/TLS, and with Python versions 3.7 to 3.11. There is 100% test coverage. The code has typing annotations, checked with mypy. The code is formatted with black and isort, and checked with flake8. Poetry is used for package management during development, and for building and publishing distributions to PyPI.

Synopsis

The ESDBClient class can be imported from the esdbclient package.

To run the client, you will need a connection string URI. And, to connect to a "secure" EventStoreDB server, you will also need an SSL/TLS certificate.

Probably the three most useful methods of ESDBClient are:

• append_events() This method can be used to record events in a particular "stream". This is useful when executing a command in an application. Either all or none of the events will be recorded.

• read_stream_events() This method can be used to retrieve all the recorded events in a "stream". This is useful for reconstructing an aggregate before executing a command in an application.

• subscribe_all_events() This method can be used to receive all recorded events across all "streams". This is useful in event-processing components, and supports processing events with "exactly-once" semantics.

The example below uses an "insecure" EventStoreDB server running locally on port 2114.

import esdbclient, uuid


# Construct ESDBClient with an EventStoreDB URI.

client = esdbclient.ESDBClient(uri="esdb://localhost:2114?Tls=false")


# Append events to a new stream.

stream_name = str(uuid.uuid4())

event1 = esdbclient.NewEvent(type='OrderCreated', data=b'data1')

client.append_events(
    stream_name=stream_name,
    expected_position=None,
    events=[event1],
)


# Append more events to an existing stream.

event2 = esdbclient.NewEvent(type='OrderUpdated', data=b'data2')
event3 = esdbclient.NewEvent(type='OrderDeleted', data=b'data3')

client.append_events(
    stream_name=stream_name,
    expected_position=0,
    events=[event2, event3],
)


# Read all events recorded in a stream.

recorded = client.read_stream_events(
    stream_name=stream_name
)

assert len(recorded) == 3
assert recorded[0].data == event1.data
assert recorded[1].data == event2.data
assert recorded[2].data == event3.data
assert recorded[0].type == event1.type
assert recorded[1].type == event2.type
assert recorded[2].type == event3.type


# In an event-processing component, use a "catch-up" subscription
# to receive all events across all streams, including events that
# have not yet been recorded, starting from the component's last
# saved "commit position".

last_saved_commit_position = 0

subscription = client.subscribe_all_events(
    commit_position=last_saved_commit_position
)

# To implement "exactly-once" semantics, iterate over the
# "catch-up" subscription. Process each received event,
# in turn, through an event-processing policy. Save the
# value of the commit_position attribute of the processed
# event with new state generated by the policy in the same
# atomic transaction. Use the last saved "commit position"
# when restarting the "catch-up" subscription.

received = []
for event in subscription:
    received.append(event)
    if event.id == event3.id:
        break

assert received[-3].data == event1.data
assert received[-2].data == event2.data
assert received[-1].data == event3.data
assert received[-3].type == event1.type
assert received[-2].type == event2.type
assert received[-1].type == event3.type

assert received[-3].commit_position > 0
assert received[-2].commit_position > received[-3].commit_position
assert received[-1].commit_position > received[-2].commit_position


# Close the client after use.

client.close()

See below for more details.

For an example of usage, see the eventsourcing-eventstoredb package.

Table of contents

• Install package
  • From PyPI
  • With Poetry
• EventStoreDB server
  • Run container
  • Stop container
• EventStoreDB client
  • Import class
  • Construct client
• Streams
  • Append events
  • Append event
  • Idempotent append operations
  • Read stream events
  • Read all events
  • Get current stream position
  • Get current commit position
  • Get stream metadata
  • Set stream metadata
  • Delete stream
  • Tombstone stream
• Catch-up subscriptions
  • How to implement exactly-once event processing
  • Subscribe all events
  • Subscribe stream events
• Persistent subscriptions
  • Create subscription
  • Read subscription
  • Get subscription info
  • List subscriptions
  • Delete subscription
  • Create stream subscription
  • Read stream subscription
  • Get stream subscription info
  • List stream subscriptions
  • Delete stream subscription
• Connection
  • Reconnect
  • Close
• Notes
  • Connection strings
  • Regular expression filters
  • New event objects
  • Recorded event objects
  • Asyncio client
• Contributors
  • Install Poetry
  • Setup for PyCharm users
  • Setup from command line
  • Project Makefile commands

Install package

It is recommended to install Python packages into a Python virtual environment.

From PyPI

You can use pip to install this package directly from the Python Package Index.

$ pip install esdbclient

With Poetry

You can use Poetry to add this package to your pyproject.toml and install it.

$ poetry add esdbclient

EventStoreDB server

The EventStoreDB server can be run locally using the official Docker container image.

Run container

For development, you can run a "secure" EventStoreDB server using the following command.

$ docker run -d --name eventstoredb-secure -it -p 2113:2113 --env "HOME=/tmp" eventstore/eventstore:21.10.9-buster-slim --dev

As we will see, the client needs an EventStoreDB connection string URI as the value of its uri constructor argument. See the Notes section below for detailed information about EventStoreDB connection string URIs.

The connection string for this "secure" EventStoreDB server would be:

esdb://admin:changeit@localhost:2113

To connect to a "secure" server, you will usually need to include a "username" and a "password" in the connection string, so that the server can authenticate the client. The default username is "admin" and the default password is "changeit".

When connecting to a "secure" server, the client also needs an SSL/TLS certificate as the value of its root_certificates constructor argument. To connect to a "secure" server you will need an SSL/TLS certificate so that the client can authenticate the server. For development, you can either use the SSL/TLS certificate of the certificate authority used to create the server's certificate, or when using a single-node cluster, you can use the server certificate itself. You can get the server certificate with the following Python code.

import ssl

server_certificate = ssl.get_server_certificate(addr=('localhost', 2113))

You can also start an "insecure" server using the following command.

$ docker run -d --name eventstoredb-insecure -it -p 2114:2113 eventstore/eventstore:21.10.9-buster-slim --insecure

The connection string URI for this "insecure" server would be:

esdb://localhost:2114?Tls=false

As we will see, when connecting to an "insecure" server, there is no need to include a "username" and a "password" in the connection string. If you do, these values will be ignored by the client, so that they will not be sent to the server over an insecure channel.

Please note, the "insecure" connection string uses a query string with the field-value Tls=false. The value of this field is by default true. See the Notes section below for more information about EventStoreDB connection strings and the fields that can be used in the query string to specify connection options.

Stop container

To stop and remove the "secure" container, use the following Docker commands.

$ docker stop eventstoredb-secure
$ docker rm eventstoredb-secure

To stop and remove the "insecure" container, use the following Docker commands.

$ docker stop eventstoredb-insecure
$ docker rm eventstoredb-insecure

EventStoreDB client

This EventStoreDB client is implemented in the esdbclient package with the ESDBClient class.

Import class

The ESDBClient class can be imported from the esdbclient package.

from esdbclient import ESDBClient

Construct client

The ESDBClient class can be constructed with a uri argument, which is required. And, to connect to a "secure" EventStoreDB server, the optional root_certificates argument is also required.

The uri argument is expected to be an EventStoreDB connection string URI that conforms with the standard EventStoreDB "esdb" or "esdb+discover" URI schemes. The syntax and semantics of EventStoreDB connection strings are explained in the Notes section below.

For example, the following connection string specifies that the client should attempt to create a "secure" connection to port 2113 on "localhost", and use the client credentials "username" and "password" when making calls to the server.

esdb://username:password@localhost:2113?Tls=true

The client must be configured to create a "secure" connection to a "secure" server, or alternatively an "insecure" connection to an "insecure" server. By default, the client will attempt to create a "secure" connection. And so, when connecting to an "insecure" server, the connection string must specify that the client should attempt to make an "insecure" connection.

The following connection string specifies that the client should attempt to create an "insecure" connection to port 2114 on "localhost". When connecting to an "insecure" server, the client will ignore any username and password information included in the connection string, so that usernames and passwords are not sent over an "insecure" connection.

esdb://localhost:2114?Tls=false

Please note, the "insecure" connection string uses a query string with the field-value Tls=false. The value of this field is by default true. Unless the connection string URI includes the field-value Tls=false in the query string, the root_certificates constructor argument is also required.

When connecting to a "secure" server, the root_certificates argument is expected to be a Python str containing PEM encoded SSL/TLS root certificates. This value is passed directly to grpc.ssl_channel_credentials(). It is used for authenticating the server to the client. It is commonly the certificate of the certificate authority that was responsible for generating the SSL/TLS certificate used by the EventStoreDB server. But, alternatively for development, you can use the server's certificate itself.

In the example below, the constructor argument values are taken from the operating system environment. This is a typical arrangement in a production environment. It is done this way here so that the code in this documentation can be tested with both a "secure" and an "insecure" server.

import os

client = ESDBClient(
    uri=os.getenv("ESDB_URI"),
    root_certificates=os.getenv("ESDB_ROOT_CERTIFICATES"),
)

See the Notes section below for detailed information about EventStoreDB connection strings and the fields that can be used in the query string to specify connection options.

Streams

In EventStoreDB, a "stream" is a sequence of recorded events that all have the same "stream name". There will normally be many streams in a database. Each recorded event has a "stream position" in its stream, and a "commit position" in the database. The stream positions of the recorded events in a stream is a gapless sequence starting from zero. The commit positions of the recorded events in the database form a sequence that is not gapless.

The methods append_events(), read_stream_events() and read_all_events() can be used to record and read events in the database.

Append events

requires leader

The append_events() method can be used to write a sequence of new events atomically to a "stream". Writing new events either creates a stream, or appends events to the end of a stream. This method is idempotent (see below).

This method can be used to record atomically all the new events that are generated when executing a command in an application.

Three arguments are required, stream_name, expected_position and events.

The stream_name argument is required, and is expected to be a Python str that uniquely identifies the stream in the database.

The expected_position argument is required, is expected to be: None if events are being written to a new stream, and otherwise an Python int equal to the position in the stream of the last recorded event in the stream.

The events argument is required, and is expected to be a sequence of new event objects to be appended to the named stream. The NewEvent class should be used to construct new event objects (see below).

This method takes an optional timeout argument, which is a Python float that sets a deadline for the completion of the gRPC operation.

Streams are created by writing events. The correct value of the expected_position argument when writing the first event of a new stream is None. Please note, it is not possible to somehow create an "empty" stream in EventStoreDB.

The stream positions of recorded events in a stream start from zero, and form a gapless sequence of integers. The stream position of the first recorded event in a stream is 0. And so when appending the second new event to a stream that has one recorded event, the correct value of the expected_position argument is 0. Similarly, the stream position of the second recorded event in a stream is 1, and so when appending the third new event to a stream that has two recorded events, the correct value of the expected_position argument is 1. And so on... (There is a theoretical maximum number of recorded events that any stream can have, but I'm not sure what it is; maybe 9,223,372,036,854,775,807 because it is implemented as a long in C#?)

If there is a mismatch between the given value of the expected_position argument and the position of the last recorded event in a stream, then an ExpectedPositionError exception will be raised. This effectively accomplishes optimistic concurrency control.

If you wish to disable optimistic concurrency control when appending new events, you can set the expected_position to a negative integer.

If you need to discover the current position of the last recorded event in a stream, you can use the get_stream_position() method (see below).

Please note, the append events operation is atomic, so that either all or none of the given new events will be recorded. By design, it is only possible with EventStoreDB to atomically record new events in one stream.

In the example below, a new event is appended to a new stream.

from uuid import uuid4

from esdbclient import NewEvent

# Construct new event object.
event1 = NewEvent(type='OrderCreated', data=b'data1')

# Define stream name.
stream_name1 = str(uuid4())

# Append list of events to new stream.
commit_position1 = client.append_events(
    stream_name=stream_name1,
    expected_position=None,
    events=[event1],
)

In the example below, two subsequent events are appended to an existing stream.

event2 = NewEvent(type='OrderUpdated', data=b'data2')
event3 = NewEvent(type='OrderDeleted', data=b'data3')

commit_position2 = client.append_events(
    stream_name=stream_name1,
    expected_position=0,
    events=[event2, event3],
)

If the operation is successful, this method returns an integer representing the overall "commit position" as it was when the operation was completed. Otherwise, an exception will be raised.

A "commit position" is a monotonically increasing integer representing the position of the recorded event in a "total order" of all recorded events in the database across all streams. It is the actual position of the event record on disk, and there are usually large differences between successive commits. In consequence, the sequence of commit positions is not gapless. Indeed, there are usually large differences between the commit positions of successive recorded events.

The "commit position" returned by append_events() is that of the last recorded event in the given batch of new events.

The "commit position" returned in this way can therefore be used to wait for a downstream component to have processed all the events that were recorded.

For example, consider a user interface command that results in the recording of new events, and a query into an eventually consistent materialized view in a downstream component that is updated from these events. If the new events have not yet been processed, the view would be stale. The "commit position" can be used by the user interface to poll the downstream component until it has processed those new events, after which time the view will not be stale.

Append event

requires leader

The append_event() method can be used to write a single new event to a stream.

Three arguments are required, stream_name, expected_position and event.

This method works in the same way as append_events(), however event is expected to be a single NewEvent.

This method takes an optional timeout argument, which is a Python float that sets a deadline for the completion of the gRPC operation.

Since the handling of a command in your application may result in one or many new events, and the results of handling a command should be recorded atomically, and the writing of new events generated by a command handler is usually a concern that is factored out and used everywhere in a project, it is quite usual in a project to only use append_events() to record new events. For this reason, an example is not provided here.

Idempotent append operations

Sometimes it may happen that a new event is successfully recorded and then somehow the connection to the database gets interrupted before the successful call can return successfully to the client. In case of an error when appending an event, it may be desirable to retry appending the same event at the same position. If the event was in fact successfully recorded, it is convenient for the retry to return successfully without raising an error due to optimistic concurrency control (as described above).

The example below shows the append_events() method being called again with event3 and expected_position=2. We can see that repeating the call to append_events() returns successfully.

# Retry appending event3.
commit_position_retry = client.append_events(
    stream_name=stream_name1,
    expected_position=0,
    events=[event2, event3],
)

We can see that the same commit position is returned as above.

assert commit_position_retry == commit_position2

By calling read_stream_events(), we can also see the stream has been unchanged despite the append_events() method having been called twice with the same arguments. That is, there are still only three events in the stream.

events = client.read_stream_events(
    stream_name=stream_name1
)

assert len(events) == 3

This idempotent behaviour is activated because the NewEvent class has an id attribute that, by default, is assigned a new and unique version-4 UUID when an instance of NewEvent is constructed. If events with the same id are appended at the same expected_position, the stream will be unchanged, the operation will complete successfully, and the same commit position will be returned to the caller.

from uuid import UUID


assert isinstance(event1.id, UUID)
assert isinstance(event2.id, UUID)
assert isinstance(event3.id, UUID)

assert event1.id != event2.id
assert event2.id != event3.id

assert events[0].id == event1.id
assert events[1].id == event2.id
assert events[2].id == event3.id

It is possible to set the id constructor argument of NewEvent when instantiating the NewEvent class, but in the examples above we have been using the default behaviour, which is that the id value is generated when the NewEvent class is instantiated.

Read stream events

The read_stream_events() method can be used to obtain the recorded events in a stream. It returns a sequence of recorded events objects. The received recorded event object are instances of the RecordedEvent class (see below).

This method has one required argument, stream_name, which is the name of the stream from which to read events. By default, the recorded events in the stream are returned in the order they were recorded.

The method read_stream_events() also supports four optional arguments, stream_position, backwards, limit, and timeout.

The optional stream_position argument is an optional integer that can be used to indicate the position in the stream from which to start reading. This argument is None by default, which means the stream will be read either from the start of the stream (the default behaviour), or from the end of the stream if backwards is True (see below). When reading a stream from a specific position in the stream, the recorded event at that position will be included, both when reading forwards from that position, and when reading backwards from that position.

The optional backwards argument is a boolean, by default False, which means the stream will be read forwards by default, so that events are returned in the order they were appended, If backwards is True, the stream will be read backwards, so that events are returned in reverse order.

The optional limit argument is an integer which limits the number of events that will be returned. The default value is sys.maxint.

The optional timeout argument is a Python float which sets a deadline for the completion of the gRPC operation.

The example below shows how to read the recorded events of a stream forwards from the start of the stream to the end of the stream. The name of a stream is given when calling the method.

events = client.read_stream_events(
    stream_name=stream_name1
)

Now that we have a sequence of event objects, we can check we got the three events that were appended to the stream, and that they are ordered exactly as they were appended.

assert len(events) == 3

assert events[0].stream_name == stream_name1
assert events[0].stream_position == 0
assert events[0].type == event1.type
assert events[0].data == event1.data

assert events[1].stream_name == stream_name1
assert events[1].stream_position == 1
assert events[1].type == event2.type
assert events[1].data == event2.data

assert events[2].stream_name == stream_name1
assert events[2].stream_position == 2
assert events[2].type == event3.type
assert events[2].data == event3.data

The example below shows how to read recorded events in a stream forwards from a specific stream position to the end of the stream.

events = client.read_stream_events(
    stream_name=stream_name1,
    stream_position=1,
)

assert len(events) == 2

assert events[0].stream_name == stream_name1
assert events[0].stream_position == 1
assert events[0].type == event2.type
assert events[0].data == event2.data

assert events[1].stream_name == stream_name1
assert events[1].stream_position == 2
assert events[1].type == event3.type
assert events[1].data == event3.data

The example below shows how to read the recorded events in a stream backwards from the end of the stream to the start of the stream.

events = client.read_stream_events(
    stream_name=stream_name1,
    backwards=True,
)

assert len(events) == 3

assert events[0].stream_name == stream_name1
assert events[0].stream_position == 2
assert events[0].type == event3.type
assert events[0].data == event3.data

assert events[1].stream_name == stream_name1
assert events[1].stream_position == 1
assert events[1].type == event2.type
assert events[1].data == event2.data

The example below shows how to read a limited number (two) of the recorded events in a stream forwards from the start of the stream.

events = client.read_stream_events(
    stream_name=stream_name1,
    limit=2,
)

assert len(events) == 2

assert events[0].stream_name == stream_name1
assert events[0].stream_position == 0
assert events[0].type == event1.type
assert events[0].data == event1.data

assert events[1].stream_name == stream_name1
assert events[1].stream_position == 1
assert events[1].type == event2.type
assert events[1].data == event2.data

The example below shows how to read a limited number (one) of the recorded events in a stream backwards from a given stream position.

events = client.read_stream_events(
    stream_name=stream_name1,
    stream_position=2,
    backwards=True,
    limit=1,
)

assert len(events) == 1

assert events[0].stream_name == stream_name1
assert events[0].stream_position == 2
assert events[0].type == event3.type
assert events[0].data == event3.data

Read all events

The method read_all_events() can be used to read all recorded events in the database in the order they were recorded. It returns an iterable of events that have been recorded in the database. Iterating over this iterable object will stop when it has yielded the last recorded event. The received recorded event object are instances of the RecordedEvent class (see below).

This method supports seven optional arguments, commit_position, backwards, filter_exclude, filter_include, filter_by_stream_name, limit, and timeout.

The optional commit_position argument is an optional integer that can be used to specify the commit position from which to start reading. This argument is None by default, meaning that all the events will be read either from the start, or from the end if backwards is True (see below). Please note, if specified, the specified position must be an actually existing commit position, because any other number will result in a server error (at least in EventStoreDB v21.10).

The optional backwards argument is a boolean which is by default False meaning the events will be read forwards by default, so that events are returned in the order they were committed, If backwards is True, the events will be read backwards, so that events are returned in reverse order.

The optional filter_exclude argument is a sequence of regular expressions that match recorded events that should not be included. This argument is ignored if filter_include is set to a non-empty sequence. By default, this argument is set to match the event types of "system events", so that EventStoreDB system events will not normally be included. See the Notes section below for more information about filter expressions.

The optional filter_include argument is a sequence of regular expressions that match recorded events that should be included. By default, this argument is an empty tuple. If this argument is set to a non-empty sequence, the filter_exclude argument is ignored.

The optional filter_by_stream_name argument is a boolean value that indicates whether the filter will apply to event types or stream names. By default, this value is False and so the filtering will apply to the event type strings of recorded events.

The optional limit argument is an integer which limits the number of events that will be returned. The default value is sys.maxint.

The optional timeout argument is a Python float which sets a deadline for the completion of the gRPC operation.

The filtering of events is done on the EventStoreDB server. The limit argument is applied on the server after filtering. See below for more information about filter regular expressions.

When reading forwards from a specific commit position, the event at the specified position WILL be included. However, when reading backwards, the event at the specified position will NOT be included. (This non-inclusive behaviour, of excluding the specified commit position when reading all streams backwards, differs from the behaviour when reading a stream backwards from a specific stream position.)

The example below shows how to read all events in the database in the order they were recorded.

events = list(client.read_all_events())

assert len(events) >= 3

The example below shows how to read all recorded events from a specific commit position.

events = list(
    client.read_all_events(
        commit_position=commit_position1
    )
)

assert len(events) == 3

assert events[0].stream_name == stream_name1
assert events[0].stream_position == 0
assert events[0].type == event1.type
assert events[0].data == event1.data

assert events[1].stream_name == stream_name1
assert events[1].stream_position == 1
assert events[1].type == event2.type
assert events[1].data == event2.data

assert events[2].stream_name == stream_name1
assert events[2].stream_position == 2
assert events[2].type == event3.type
assert events[2].data == event3.data

The example below shows how to read all recorded events in reverse order.

events = list(
    client.read_all_events(
        backwards=True
    )
)

assert len(events) >= 3

assert events[0].stream_name == stream_name1
assert events[0].stream_position == 2
assert events[0].type == event3.type
assert events[0].data == event3.data

assert events[1].stream_name == stream_name1
assert events[1].stream_position == 1
assert events[1].type == event2.type
assert events[1].data == event2.data

assert events[2].stream_name == stream_name1
assert events[2].stream_position == 0
assert events[2].type == event1.type
assert events[2].data == event1.data

The example below shows how to read a limited number (one) of the recorded events in the database forwards from a specific commit position. Please note, when reading all events forwards from a specific commit position, the event at the specified position WILL be included.

events = list(
    client.read_all_events(
        commit_position=commit_position1,
        limit=1,
    )
)

assert len(events) == 1

assert events[0].stream_name == stream_name1
assert events[0].stream_position == 0
assert events[0].type == event1.type
assert events[0].data == event1.data

assert events[0].commit_position == commit_position1

The example below shows how to read a limited number (one) of the recorded events in the database backwards from the end. This gives the last recorded event.

events = list(
    client.read_all_events(
        backwards=True,
        limit=1,
    )
)

assert len(events) == 1

assert events[0].stream_name == stream_name1
assert events[0].stream_position == 2
assert events[0].type == event3.type
assert events[0].data == event3.data

The example below shows how to read a limited number (one) of the recorded events in the database backwards from a specific commit position. Please note, when reading all events backwards from a specific commit position, the event at the specified position WILL NOT be included.

events = list(
    client.read_all_events(
        commit_position=commit_position2,
        backwards=True,
        limit=1,
    )
)

assert len(events) == 1

assert events[0].commit_position < commit_position2

Get current stream position

The get_stream_position() method can be used to get the "stream position" of the last recorded event in a stream.

This method has a stream_name argument, which is required.

This method also takes an optional timeout argument, that is expected to be a Python float, which sets a deadline for the completion of the gRPC operation.

The sequence of positions in a stream is gapless. It is zero-based, so that a stream with one recorded event has a current stream position of 0. The current stream position is 1 when a stream has two events, and it is 2 when there are events, and so on.

In the example below, the current stream position is obtained of the stream to which events were appended in the examples above. Because the sequence of stream positions is zero-based, and because three events were appended, so the current stream position is 2.

stream_position = client.get_stream_position(
    stream_name=stream_name1
)

assert stream_position == 2

If a stream does not exist, the returned stream position value is None, which matches the required expected position when appending the first event of a new stream (see above).

stream_position = client.get_stream_position(
    stream_name=str(uuid4())
)

assert stream_position == None

This method takes an optional argument timeout which is a Python float that sets a deadline for the completion of the gRPC operation.

Get current commit position

The method get_commit_position() can be used to get the current commit position of the database.

commit_position = client.get_commit_position()

This method takes an optional argument timeout which is a Python float that sets a deadline for the completion of the gRPC operation.

This method can be useful to measure progress of a downstream component that is processing all recorded events, by comparing the current commit position with the recorded commit position of the last successfully processed event in a downstream component.

The value of the commit_position argument when reading events either by using the read_all_events() method or by using a catch-up subscription would usually be determined by the recorded commit position of the last successfully processed event in a downstream component.

Get stream metadata

The method get_stream_metadata() gets the metadata for a stream, along with the version of the stream metadata.

metadata, metadata_version = client.get_stream_metadata(stream_name=stream_name1)

The returned metadata value is a Python dict. The returned metadata_version value is either an int, or None if the stream does not exist. These values can be passed into set_stream_metadata().

Set stream metadata

requires leader

The method set_stream_metadata() sets the metadata for a stream, along with the version of the stream metadata.

metadata["foo"] = "bar"

client.set_stream_metadata(
    stream_name=stream_name1,
    metadata=metadata,
    expected_position=metadata_version,
)

The expected_position argument should be the current version of the stream metadata.

Please refer to the EventStoreDB documentation for more information about stream metadata.

Delete stream

requires leader

The method delete_stream() can be used to "delete" a stream.

commit_position = client.delete_stream(stream_name=stream_name1, expected_position=2)

After deleting a stream, it's still possible to append new events. Reading from a deleted stream will return only events that have been appended after it was deleted.

Tombstone stream

requires leader

The method tombstone_stream() can be used to "tombstone" a stream.

commit_position = client.tombstone_stream(stream_name=stream_name1, expected_position=2)

After tombstoning a stream, it's not possible to append new events.

Catch-up subscriptions

"Catch-up" subscriptions can be used to receive events that have been recorded in the database, and also events that are recorded after a subscription was started.

The method subscribe_all_events() starts a catch-up subscription to receive all events that have been and will be recorded in the database. The method subscribe_stream_events() starts a catch-up subscription to receive events from a specific stream.

Catch-up subscriptions encapsulate a streaming gRPC call which is kept open by the server, with newly recorded events sent to the client as the client iterates over the subscription response.

Many catch-up subscriptions can be created, concurrently or successively, and all will receive all the recorded events they have been requested to receive.

The received recorded event object are instances of the RecordedEvent class (see below).

How to implement exactly-once event processing

The commit positions of recorded events that are received and processed by a downstream component are usefully recorded by the downstream component so that the commit position of last processed event can be determined.

The last recorded commit position can be used to specify the commit position from which to subscribe when processing is resumed. Since this commit position will represent the position of the last successfully processed event in a downstream component, so it will be usual to want the next event after this position, because that is the next event that has not yet been processed. For this reason, when subscribing for events from a specific commit position using a catch-up subscription in EventStoreDB, the recorded event at the specified commit position will NOT be included in the sequence of recorded events that are received.

To accomplish "exactly-once" processing of recorded events in a downstream component when using a catch-up subscription, the commit position of a recorded event should be recorded atomically and uniquely along with the result of processing recorded events, for example in the same database as materialised views when implementing eventually-consistent CQRS, or in the same database as a downstream analytics or reporting or archiving application. By recording the commit position of recorded events atomically with the new state that results from processing recorded events, "dual writing" in the consumption of recorded events can be avoided. By also recording the commit position uniquely, the new state cannot be recorded twice, and hence the recorded state of the downstream component will be updated only once for any recorded event. By using the greatest recorded commit position to resume a catch-up subscription, all recorded events will eventually be processed. The combination of the "at-most-once" condition and the "at-least-once" condition gives the "exactly-once" condition.

The danger with "dual writing" in the consumption of recorded events is that if a recorded event is successfully processed and new state recorded atomically in one transaction with the commit position recorded in a separate transaction, one may happen and not the other. If the new state is recorded but the position is lost, and then the processing is stopped and resumed, the recorded event may be processed twice. On the other hand, if the commit position is recorded but the new state is lost, the recorded event may effectively not be processed at all. By either processing an event more than once, or by failing to process an event, the recorded state of the downstream component might be inaccurate, or possibly inconsistent, and perhaps catastrophically so. Such consequences may or may not matter in your situation. But sometimes inconsistencies may halt processing until the issue is resolved. You can avoid "dual writing" in the consumption of events by atomically recording the commit position of a recorded event along with the new state that results from processing that event in the same atomic transaction. By making the recording of the commit positions unique, so that transactions will be rolled back when there is a conflict, you will prevent the results of any duplicate processing of a recorded event being committed.

Recorded events received from a catch-up subscription cannot be acknowledged back to the EventStoreDB server. Acknowledging events, however, is an aspect of "persistent subscriptions" (see below). Hoping to rely on acknowledging events to an upstream component is an example of dual writing.

Subscribe all events

Thesubscribe_all_events() method can be used to start a "catch-up" subscription that can return events from all streams in the database.

This method can be used by a downstream component to process recorded events with exactly-once semantics.

This method takes an optional commit_position argument, which can be used to specify a commit position from which to subscribe. The default value is None, which means the subscription will operate from the first recorded event in the database. If a commit position is given, it must match an actually existing commit position in the database. The recoded events that are obtained will not include the event recorded at that commit position.

This method also takes four other optional arguments, filter_exclude, filter_include, filter_by_stream_name, and timeout.

The optional argument filter_exclude is a sequence of regular expressions that match recorded events that should not be included. This argument is ignored if filter_include is set to a non-empty sequence. By default, this argument is set to match the event types of "system events", so that EventStoreDB system events will not normally be included. See the Notes section below for more information about filter expressions.

The optional argument filter_include is a sequence of regular expressions that match recorded events that should be included. By default, this argument is an empty tuple. If this argument is set to a non-empty sequence, the filter_exclude argument is ignored.

The optional argument filter_by_stream_name is a boolean value that indicates whether the filter will apply to event types or stream names. By default, this value is False and so the filtering will apply to the event type strings of recorded events.

Please note, the filtering happens on the EventStoreDB server, and the limit argument is applied on the server after filtering. See below for more information about filter regular expressions.

The optional timeout argument is a Python float which sets a deadline for the completion of the gRPC operation.

This method returns a Python iterator that yields recorded events, including events that are recorded after the subscription was created. Iterating over this object will therefore not stop, unless the connection to the database is lost. The call will be ended when the iterator object is deleted from memory, which will happen when the iterator object goes out of scope or is explicitly deleted (see below). The call may also be closed by the server.

The subscription object can be used directly, but it might be used within a threaded loop dedicated to receiving events that can be stopped in a controlled way, with recorded events put on a queue for processing in a different thread. This package doesn't provide such a threaded or queuing object class. Just make sure to reconstruct the subscription (and the queue) using the last recorded commit position when resuming the subscription after an error, to be sure all events are processed once.

The example below shows how to subscribe to receive all recorded events from the start, and then resuming from a specific commit position. Three already-recorded events are received, and then three new events are recorded, which are then received via the subscription.

# Append an event to a new stream.
stream_name2 = str(uuid4())
event4 = NewEvent(type='OrderCreated', data=b'data4')
client.append_events(
    stream_name=stream_name2,
    expected_position=None,
    events=[event4],
)

# Subscribe from the first recorded event in the database.
subscription = client.subscribe_all_events()
received_events = []

# Process events received from the catch-up subscription.
for event in subscription:
    last_commit_position = event.commit_position
    received_events.append(event)
    if event.id == event4.id:
        break

assert received_events[-4].id == event1.id
assert received_events[-3].id == event2.id
assert received_events[-2].id == event3.id
assert received_events[-1].id == event4.id

# Append subsequent events to the stream.
event5 = NewEvent(type='OrderUpdated', data=b'data5')
client.append_events(
    stream_name=stream_name2,
    expected_position=0,
    events=[event5],
)

# Receive subsequent events from the subscription.
for event in subscription:
    last_commit_position = event.commit_position
    received_events.append(event)
    if event.id == event5.id:
        break


assert received_events[-5].id == event1.id
assert received_events[-4].id == event2.id
assert received_events[-3].id == event3.id
assert received_events[-2].id == event4.id
assert received_events[-1].id == event5.id


# Append more events to the stream.
event6 = NewEvent(type='OrderDeleted', data=b'data6')
client.append_events(
    stream_name=stream_name2,
    expected_position=1,
    events=[event6],
)


# Resume subscribing from the last commit position.
subscription = client.subscribe_all_events(
    commit_position=last_commit_position
)


# Catch up by receiving the new event from the subscription.
for event in subscription:
    received_events.append(event)
    if event.id == event6.id:
        break

assert received_events[-6].id == event1.id
assert received_events[-5].id == event2.id
assert received_events[-4].id == event3.id
assert received_events[-3].id == event4.id
assert received_events[-2].id == event5.id
assert received_events[-1].id == event6.id


# Append three more events to a new stream.
stream_name3 = str(uuid4())
event7 = NewEvent(type='OrderCreated', data=b'data7')
event8 = NewEvent(type='OrderUpdated', data=b'data8')
event9 = NewEvent(type='OrderDeleted', data=b'data9')

client.append_events(
    stream_name=stream_name3,
    expected_position=None,
    events=[event7, event8, event9],
)

# Receive the three new events from the resumed subscription.
for event in subscription:
    received_events.append(event)
    if event.id == event9.id:
        break

assert received_events[-9].id == event1.id
assert received_events[-8].id == event2.id
assert received_events[-7].id == event3.id
assert received_events[-6].id == event4.id
assert received_events[-5].id == event5.id
assert received_events[-4].id == event6.id
assert received_events[-3].id == event7.id
assert received_events[-2].id == event8.id
assert received_events[-1].id == event9.id

The catch-up subscription call is ended as soon as the subscription object goes out of scope or is explicitly deleted from memory.

# End the subscription.
del subscription

Subscribe stream events

Thesubscribe_stream_events() method can be used to start a "catch-up" subscription that can return events that are recorded in a single stream.

This method takes a required stream_name argument, which specifies the name of the stream from which recorded events will be received.

This method also takes two optional arguments, stream_position, and timeout.

The optional stream_position argument specifies a position in the stream from which recorded events will be received. The event at the specified stream position will not be included.

The optional timeout argument is a Python float that sets a deadline for the completion of the gRPC operation.

The example below shows how to start a catch-up subscription to a stream.

# Subscribe to events from stream2, from the start.
subscription = client.subscribe_stream_events(stream_name=stream_name2)

# Read from the subscription.
events = []
for event in subscription:
    events.append(event)
    if event.id == event6.id:
        break

# Check we got events only from stream2.
assert len(events) == 3
events[0].stream_name == stream_name2
events[1].stream_name == stream_name2
events[2].stream_name == stream_name2

# Append another event to stream3.
event10 = NewEvent(type="OrderUndeleted", data=b'data10')
client.append_events(
    stream_name=stream_name3,
    expected_position=2,
    events=[event10],
)

# Append another event to stream2.
event11 = NewEvent(type="OrderUndeleted", data=b'data11')
client.append_events(
    stream_name=stream_name2,
    expected_position=2,
    events=[event11]
)

# Continue reading from the subscription.
for event in subscription:
    events.append(event)
    if event.id == event11.id:
        break

# Check we got events only from stream2.
assert len(events) == 4
events[0].stream_name == stream_name2
events[1].stream_name == stream_name2
events[2].stream_name == stream_name2
events[3].stream_name == stream_name2

The example below shows how to start a catch-up subscription to a stream from a specific stream position.

# Subscribe to events from stream2, from the start.
subscription = client.subscribe_stream_events(
    stream_name=stream_name2,
    stream_position=1,
)

# Read event from the subscription.
events = []
for event in subscription:
    events.append(event)
    if event.id == event11.id:
        break

# Check we got events only after position 1.
assert len(events) == 2
events[0].id == event6.id
events[0].stream_position == 2
events[0].stream_name == stream_name2
events[1].id == event11.id
events[1].stream_position == 3
events[1].stream_name == stream_name2

Persistent subscriptions

Create subscription

requires leader

The method create_subscription() can be used to create a "persistent subscription" to EventStoreDB.

This method takes a required group_name argument, which is the name of a "group" of consumers of the subscription.

This method also takes seven optional arguments, from_end, commit_position, filter_exclude, filter_include, filter_by_stream_name, consumer_strategy, and timeout.

The optional from_end argument can be used to specify that the group of consumers of the subscription should only receive events that were recorded after the subscription was created.

Alternatively, the optional commit_position argument can be used to specify a commit position from which commit position the group of consumers of the subscription should receive events. Please note, the recorded event at the specified commit position might be included in the recorded events received by the group of consumers.

If neither from_end or commit_position are specified, the group of consumers of the subscription will potentially receive all recorded events in the database.

The optional filter_exclude argument is a sequence of regular expressions that match recorded events that should not be included. This argument is ignored if filter_include is set to a non-empty sequence. By default, this argument is set to match the event types of "system events", so that EventStoreDB system events will not normally be received. See the Notes section below for more information about filter expressions.

The optional filter_include argument is a sequence of regular expressions that match recorded events that should be received. By default, this argument is an empty tuple. If this argument is set to a non-empty sequence, the filter_exclude argument is ignored.

The optional filter_by_stream_name argument is a boolean value that indicates whether the filter will apply to event types or stream names. By default, this value is False and so the filtering will apply to the event type strings of recorded events.

The optional consumer_strategy argument is a Python str that defines the consumer strategy for this persistent subscription. The value of this argument can be 'DispatchToSingle', 'RoundRobin', 'Pinned', or 'PinnedByCorrelation'. The default value is 'DispatchToSingle'.

The optional timeout argument is a Python float which sets a deadline for the completion of the gRPC operation.

The method create_subscription() does not return a value, because recorded events are obtained by the group of consumers of the subscription using the read_subscription() method.

In the example below, a persistent subscription is created.

# Create a persistent subscription.
group_name = f"group-{uuid4()}"
client.create_subscription(group_name=group_name)

Read subscription

requires leader

The method read_subscription() can be used by a group of consumers to receive recorded events from a persistent subscription created using create_subscription.

This method takes a required group_name argument, which is the name of a "group" of consumers of the subscription specified when create_subscription() was called.

This method also takes an optional timeout argument, that is expected to be a Python float, which sets a deadline for the completion of the gRPC operation.

This method returns a 2-tuple: a "read request" object and a "read response" object.

read_req, read_resp = client.read_subscription(group_name=group_name)

The "read response" object is an iterator that yields recorded events from the specified commit position.

The "read request" object has an ack() method that should be used by a consumer in a group to acknowledge to the server that it has received and successfully processed a recorded event. This will prevent that recorded event being received by another consumer in the same group. The ack() method takes an event_id argument, which is the ID of the recorded event that has been received.

The "read request" object also has an nack() method that should be used by a consumer in a group to negatively acknowledge to the server that it has received but not successfully processed a recorded event. The nack() method takes an event_id argument, which is the ID of the recorded event that has been received, and an action argument, which should be a Python str, either 'unknown', 'park', 'retry', 'skip', or 'stop'.

The example below iterates over the "read response" object, and calls ack() on the "read response" object. The for loop breaks when we have received the last event, so that we can continue with the examples below.

events = []
for event in read_resp:
    events.append(event)

    # Acknowledge the received event.
    read_req.ack(event_id=event.id)

    # Break when the last event has been received.
    if event.id == event11.id:
        break

The received events are the events we appended above.

assert events[-11].id == event1.id
assert events[-10].id == event2.id
assert events[-9].id == event3.id
assert events[-8].id == event4.id
assert events[-7].id == event5.id
assert events[-6].id == event6.id
assert events[-5].id == event7.id
assert events[-4].id == event8.id
assert events[-3].id == event9.id
assert events[-2].id == event10.id
assert events[-1].id == event11.id

The "read request" object also has an nack() method that can be used by a consumer in a group to acknowledge to the server that it has failed successfully to process a recorded event. This will allow that recorded event to be received by this or another consumer in the same group.

It might be more useful to encapsulate the request and response objects and to iterate over the "read response" in a separate thread, to call back to a handler function when a recorded event is received, and call ack() if the handler does not raise an exception, and to call nack() if an exception is raised. The example below shows how this might be done.

from threading import Thread


class SubscriptionReader:
    def __init__(self, client, group_name, callback):
        self.client = client
        self.group_name = group_name
        self.callback = callback
        self.thread = Thread(target=self.read_subscription, daemon=True)
        self.error = None

    def start(self):
        self.thread.start()

    def join(self):
        self.thread.join()

    def read_subscription(self):
        req, resp = self.client.read_subscription(group_name=self.group_name)
        for event in resp:
            try:
                self.callback(event)
            except Exception as e:
                # req.nack(event.id)  # not yet implemented....
                self.error = e
                break
            else:
                req.ack(event.id)


# Create another persistent subscription.
group_name = f"group-{uuid4()}"
client.create_subscription(group_name=group_name)

events = []

def handle_event(event):
    events.append(event)
    if event.id == event11.id:
        raise Exception()


reader = SubscriptionReader(
    client=client,
    group_name=group_name,
    callback=handle_event
)

reader.start()
reader.join()

assert events[-1].id == event11.id

Please note, when processing events in a downstream component, the commit position of the last successfully processed event is usefully recorded by the downstream component so that the commit position can be determined by the downstream component from its own recorded when it is restarted. This commit position can be used to specify the commit position from which to subscribe. Since this commit position represents the position of the last successfully processed event in a downstream component, so it will be usual to want to read from the next event after this position, because that is the next event that needs to be processed. However, when subscribing for events using a persistent subscription in EventStoreDB, the event at the specified commit position MAY be returned as the first event in the received sequence of recorded events, and so it may be necessary to check the commit position of the received events and to discard any recorded event object that has a commit position equal to the commit position specified in the request.

Whilst there are some advantages of persistent subscriptions, in particular the processing of recorded events by a group of consumers, by tracking in the upstream server the position in the commit sequence of events that have been processed, there is a danger of "dual writing" in the consumption of events. Reliability in processing of recorded events by a group of consumers will rely instead on idempotent handling of duplicate messages, and resilience to out-of-order delivery.

Get subscription info

requires leader

The get_subscription_info() method can be used to get information for a persistent subscription.

This method has one required argument, group_name, which should match the value of the argument used when calling create_subscription().

This method also takes an optional timeout argument, that is expected to be a Python float, which sets a deadline for the completion of the gRPC operation.

subscription_info = client.get_subscription_info(
    group_name=group_name,
)

The returned value is a SubscriptionInfo object.

List subscriptions

requires leader

The list_subscriptions() method can be used to get information for all existing persistent subscriptions.

This method takes an optional timeout argument, that is expected to be a Python float, which sets a deadline for the completion of the gRPC operation.

subscriptions = client.list_subscriptions()

The returned value is a list of SubscriptionInfo objects.

Delete subscription

requires leader

The delete_subscription() method can be used to delete a persistent subscription.

This method has one required argument, group_name, which should match the value of argument used when calling create_subscription().

This method also takes an optional timeout argument, that is expected to be a Python float, which sets a deadline for the completion of the gRPC operation.

client.delete_subscription(
    group_name=group_name,
)

Create stream subscription

requires leader

The create_stream_subscription() method can be used to create a persistent subscription for a stream.

This method has two required arguments, group_name and stream_name. The group_name argument names the group of consumers that will receive events from this subscription. The stream_name argument specifies which stream the subscription will follow. The values of both these arguments are expected to be Python str objects.

The method also takes four optional arguments, stream_position, from_end, consumer_strategy, and timeout.

This optional stream_position argument specifies a stream position from which to subscribe. The recorded event at this stream position will be received when reading the subscription.

This optional from_end argument is a Python bool. By default, the value of this argument is False. If this argument is set to a True value, reading from the subscription will receive only events recorded after the subscription was created. That is, it is not inclusive of the current stream position.

The optional consumer_strategy argument is a Python str that defines the consumer strategy for this persistent subscription. The value of this argument can be 'DispatchToSingle', 'RoundRobin', 'Pinned', or 'PinnedByCorrelation'. The default value is 'DispatchToSingle'.

This method also takes an optional timeout argument, that is expected to be a Python float, which sets a deadline for the completion of the gRPC operation.

This method does not return a value. Events can be received by iterating over the value returned by calling read_stream_subscription() (see below).

The example below creates a persistent stream subscription from the start of the stream.

# Create a persistent stream subscription from start of the stream.
group_name1 = f"group-{uuid4()}"
client.create_stream_subscription(
    group_name=group_name1,
    stream_name=stream_name2,
)

The example below creates a persistent stream subscription from a stream position.

# Create a persistent stream subscription from a stream position.
group_name2 = f"group-{uuid4()}"
client.create_stream_subscription(
    group_name=group_name2,
    stream_name=stream_name2,
    stream_position=2
)

The example below creates a persistent stream subscription from the end of the stream.

# Create a persistent stream subscription from end of the stream.
group_name3 = f"group-{uuid4()}"
client.create_stream_subscription(
    group_name=group_name3,
    stream_name=stream_name2,
    from_end=True
)

Read stream subscription

requires leader

The read_stream_subscription() method can be used to create a persistent subscription for a stream.

This method has two required arguments, group_name and stream_name, which should match the values of arguments used when calling create_stream_subscription().

This method also takes an optional timeout argument, that is expected to be a Python float, which sets a deadline for the completion of the gRPC operation.

Just like read_subscription, this method returns a 2-tuple: a "read request" object and a "read response" object.

read_req, read_resp = client.read_stream_subscription(
    group_name=group_name1,
    stream_name=stream_name2,
)

The example below iterates over the "read response" object, and calls ack() on the "read response" object. The for loop breaks when we have received the last event in the stream, so that we can finish the examples in this documentation.

events = []
for event in read_resp:
    events.append(event)

    # Acknowledge the received event.
    read_req.ack(event_id=event.id)

    # Break when the last event has been received.
    if event.id == event11.id:
        break

We can check we received all the events that were appended to stream_name2 in the examples above.

assert len(events) == 4
assert events[0].stream_name == stream_name2
assert events[0].id == event4.id
assert events[1].stream_name == stream_name2
assert events[1].id == event5.id
assert events[2].stream_name == stream_name2
assert events[2].id == event6.id
assert events[3].stream_name == stream_name2
assert events[3].id == event11.id

Get stream subscription info

requires leader

The get_stream_subscription_info() method can be used to get information for a persistent subscription for a stream.

This method has two required arguments, group_name and stream_name, which should match the values of arguments used when calling create_stream_subscription().

This method also takes an optional timeout argument, that is expected to be a Python float, which sets a deadline for the completion of the gRPC operation.

subscription_info = client.get_stream_subscription_info(
    group_name=group_name1,
    stream_name=stream_name2,
)

The returned value is a SubscriptionInfo object.

List stream subscriptions

requires leader

The list_stream_subscriptions() method can be used to get information for all the persistent subscriptions for a stream.

This method has one required argument, stream_name.

This method also takes an optional timeout argument, that is expected to be a Python float, which sets a deadline for the completion of the gRPC operation.

subscriptions = client.list_stream_subscriptions(
    stream_name=stream_name2,
)

The returned value is a list of SubscriptionInfo objects.

Delete stream subscription

requires leader

The delete_stream_subscription() method can be used to delete a persistent subscription for a stream.

This method has two required arguments, group_name and stream_name, which should match the values of arguments used when calling create_stream_subscription().

This method also takes an optional timeout argument, that is expected to be a Python float, which sets a deadline for the completion of the gRPC operation.

client.delete_stream_subscription(
    group_name=group_name1,
    stream_name=stream_name2,
)

Connection

Reconnect

The reconnect() method can be used to manually reconnect the client to a suitable EventStoreDB node. This method uses the same routine for reading the cluster node states and then connecting to a suitable node according to the client's node preference that is specified in the connection string URI when the client is constructed. This method is thread-safe, and it is "conservative" in that, when it is called by several threads at the same time, only one reconnection will occur. Concurrent attempts to reconnect will block until the client has reconnected successfully, and then they will all return normally.

client.reconnect()

An example of when it might be desirable to reconnect manually is when (for performance reasons) the client's node preference is to be connected to a follower node in the cluster, and, after a cluster leader election, the follower becomes the leader. Reconnecting to a follower node in this case is currently beyond the capabilities of this client, but this behavior might be implemented in a future release.

Please note, nearly all the client methods are decorated with @autoreconnect (which calls the reconnect() method when the client detects that reconnecting is required) and a @retry decorator that more generally will retry operations that fail due to connection issues.

The @autoreconnect decorator will reconnect to a suitable node in the cluster when the server to which the client has been connected has become unavailable, or when the client's gRPC channel happens to have been closed. The client will also reconnect when a method is called that requires a leader, and the client's node preference is to be connected to a leader, but the node that the client has been connected to stops being the leader. In this case, the client will reconnect to the current leader. After reconnecting, the failed operation will be retried.

The @retry decorator retries operations that have failed due to more general connection issues, such as a deadline being reached (so that the operation times out), or in case the server throws an exception when handling a client request.

Please also note, the aspects not covered by the reconnect and retry decorator behaviours have to do with methods that return iterators. For example, consider the read_all_events() method, which returns an iterator of RecordedEvent objects. The method has returned and the method decorators have therefore exited before the iterating of the response begins. Therefore, it isn't possible for this method to trigger a reconnection or for it to retry, when the streaming response somehow fails.

Another example is the "catch-up" and persistent subscription read methods, although with these methods there is an initial "confirmation" response from the server which is received and checked by the client before the method returns, so if the server is unavailable when the call is made, or if the channel is somehow closed, or if the server throws an error for some reason, then the client will either reconnect and retry, or more simply just retry the operation, according the type of error that was encountered. However, an event-processing component that iterates over a successfully returned "catch-up" subscription response will need to be monitored for errors, and, if it fails after it has started iterating over the response, the catch-up subscription will need to be restarted from the event-processing component's "last saved commit position". If the event-processing component is recording the commit position of each recorded event atomically along with any new state that results from processing the recorded events, and the commit positions are recorded with a uniqueness constraint, then there will no danger to the projected state becoming inconsistent due to these connection issues. In this case, the client will automatically reconnect to a node in the cluster when the subsequent call to a catch-up subscription method is made. You just need to handle the exception that occurs from the iterator, then read your last saved commit position, and then restart your event-processing component, using the same client object. The client will reconnect if there is a need to do so when the subsequent call is made. Similarly, when reading persistent subscriptions, if there are connectivity issues after you have started iterating over a successfully received response, then the server call will need to be restarted. In this case, the received events will start with the last saved checkpoint on the server (potentially returning events that were in fact successfully processed).

Close

The close() method can be used to cleanly close the gRPC connection.

client.close()

Notes

Connection strings

The EventStoreDB connection string is a URI that conforms with one of two possible schemes, either the "esdb" scheme or the "esdb+discover" scheme. The syntax and semantics of the EventStoreDB URI schemes are explained below. The syntax is defined using EBNF.

The "esdb" URI scheme can be defined in the following way.

esdb-uri = "esdb://" , [ user-info , "@" ] , grpc-target, { "," , grpc-target } , [ "?" , query-string ] ;

In the "esdb" URI scheme, after the optional user info string, there must be at least one gRPC target. If there are several gRPC targets, they must be separated from each other with the "," character. Each gRPC target should indicate an EventStoreDB gRPC server socket, by specifying a host and a port number separated with the ":" character. The host may be a hostname that can be resolved to an IP address, or an IP address.

grpc-target = ( hostname | ip-address ) , ":" , port-number ;

The "esdb+discover" URI scheme can be defined in the following way.

esdb-discover-uri = "esdb+discover://" , [ user-info, "@" ] , cluster-domainname , [ "?" , query-string ] ;

In the "esdb+discover" URI scheme, after the user info string, there must be a domain name which should identify a cluster of EventStoreDB servers. The client will use a DNS server to resolve the domain name to a list of addresses of EventStoreDB servers, by querying for 'A' records. In this case, the port number "2113" will be used to construct gRPC targets from the addresses obtained from 'A' records provided by the DNS server. Therefore, if you want to use the "esdb+discover" URI scheme, you will need to configure DNS when setting up your EventStoreDB cluster.

With both the "esdb" and "esdb+disocver" URI schemes, the client firstly obtains a list of gRPC targets: either directly from "esdb" connection strings; or indirectly from "esdb+discover" connection strings via DNS. This list of targets is known as the "gossip seed". The client will then attempt to connect to each gRPC target in turn, attempting to call the EventStoreDB Gossip API to obtain information about the EventStoreDB cluster. A member of the cluster is selected by the client, according to the "node preference" option (see below). The client may then need to close its connection and reconnect to the selected server.

In both the "esdb" and "esdb+discover" schemes, the URI may include a user info string. If it exists in the URI, the user info string must be separated from the rest of the URI with the "@" character. The user info string must include a username and a password, separated with the ":" character.

user-info = username , ":" , password ;

The user info is sent by the client as "call credentials" in each call to a "secure" server, in a "basic auth" authorization header. This authorization header is used by the server to authenticate the client. The authorization header is not sent to "insecure" servers.

In both the "esdb" and "esdb+discover" schemes, the optional query string must be one or many field-value arguments, separated from each other with the "&" character.

query-string = field-value, { "&", field-value } ;

Each field-value argument must be one of the supported fields, and an appropriate value, separated with the "=" character.

field-value = ( "Tls", "=" , "true" | "false" )
            | ( "TlsVerifyCert", "=" , "true" | "false" )
            | ( "ConnectionName", "=" , string )
            | ( "NodePreference", "=" , "leader" | "follower" | "readonlyreplica" | "random" )
            | ( "DefaultDeadline", "=" , integer )
            | ( "GossipTimeout", "=" , integer )
            | ( "MaxDiscoverAttempts", "=" , integer )
            | ( "DiscoveryInterval", "=" , integer )
            | ( "MaxDiscoverAttempts", "=" , integer )
            | ( "KeepAliveInterval", "=" , integer )
            | ( "KeepAliveInterval", "=" , integer ) ;

The table below describes the query field-values supported by this client.

Here are some examples of EventStoreDB connection string URIs.

# Get cluster info from secure server socket localhost:2113,
# and use "admin" and "changeit" as username and password
# when making calls to EventStoreDB API methods.

esdb://admin:changeit@localhost:2113


# Get cluster info from insecure server socket 127.0.0.1:2114

esdb://127.0.0.1:2114?Tls=false


# Get cluster info from addresses in 'A' records for cluster1.example.com,
# and use a default deadline for making calls to EventStore API method.

esdb+discover://admin:changeit@cluster1.example.com?DefaultDeadline=5


# Get cluster info from either localhost:2111 or localhost:2112 or
# localhost:2113, and then connect to a follower node in the cluster.

esdb://admin:changeit@localhost:2111,localhost:2112,localhost:2113?NodePreference=follower


# Get cluster info from addresses in 'A' records for cluster1.example.com,
# and configure "keep alive" timeout and interval in the gRPC channel.

esdb+discover://admin:changeit@cluster1.example.com?KeepAliveInterval=10000&KeepAliveTimeout=10000

Please note, the client is insensitive to the case of fields and values. If fields are repeated in the query string, the query string will be parsed without error. However, the connection options used by the client will use the value of the first field. All the other field-values in the query string with the same field name will be ignored. Fields without values will also be ignored.

If the client's node preference is "leader" and the node becomes a "follower", the client will attempt to reconnect to the current leader when a method is called that expects to call a leader. Methods which mutate the state of the database expect to call a leader. For such methods, the HTTP header"requires-leader" is set to "true", and this header is observed by the server, and so a node which is not a leader that receives such a request will return an error. This error is detected by the client, which will then close the current gRPC connection and create a new connection to the leader. The request will then be retried with the leader.

If the client's node preference is "follower" and there are no follower nodes in the cluster, then the client will raise an exception. Similarly, if the client's node preference is "readonlyreplica" and there are no read-only replica nodes in the cluster, then the client will also raise an exception.

The gRPC channel option "grpc.max_receive_message_length" is automatically configured to the value 17 * 1024 * 1024. This value cannot be changed.

Regular expression filters

The filter arguments in read_all_events(), subscribe_all_events(), create_subscription() and get_commit_position() are applied to the type attribute of recorded events.

The default value of the filter_exclude arguments is designed to exclude EventStoreDB "system" and "persistence subscription config" events, which otherwise would be included. System events generated by EventStoreDB all have type strings that start with the $ sign. Persistence subscription events generated when manipulating persistence subscriptions all have type strings that start with PersistentConfig.

For example, to match the type of EventStoreDB system events, use the regular expression r'\$.+'. Please note, the constant ESDB_SYSTEM_EVENTS_REGEX is set to r'\$.+'. You can import this value (from esdbclient import ESDB_SYSTEM_EVENTS_REGEX) and use it when building longer sequences of regular expressions.

Similarly, to match the type of EventStoreDB persistence subscription events, use the regular expression r'PersistentConfig\d+'. The constant ESDB_PERSISTENT_CONFIG_EVENTS_REGEX is set to r'PersistentConfig\d+'. You can also import this value (from esdbclient import ESDB_PERSISTENT_CONFIG_EVENTS_REGEX) and use it when building longer sequences of regular expressions.

The constant DEFAULT_EXCLUDE_FILTER is a sequence of regular expressions that match the events that EventStoreDB generates internally, events that are extraneous to those which you append using the append_events() method.

For example, to exclude system events and persistence subscription configuration events, and snapshots, you might use the sequence DEFAULT_EXCLUDE_FILTER + ['.*Snapshot'] as the value of the filter_exclude argument when calling read_all_events(), subscribe_all_events(), create_subscription() or get_commit_position().

New event objects

The NewEvent class is used when appending events.

The required argument type is a Python str, used to indicate the type of the event that will be recorded.

The required argument data is a Python bytes object, used to indicate the data of the event that will be recorded.

The optional argument metadata is a Python bytes object, used to indicate any metadata of the event that will be recorded. The default value is an empty bytes object.

The optional argument content_type is a Python str, used to indicate the type of the data that will be recorded for this event. The default value is application/json, which indicates that the data was serialised using JSON. An alternative value for this argument is application/octet-stream.

The optional argument id is a Python UUID object, used to specify the unique ID of the event that will be recorded. This value will default to a new version-4 UUID.

new_event1 = NewEvent(
    type='OrderCreated',
    data=b'{"name": "Greg"}',
)
assert new_event1.type == 'OrderCreated'
assert new_event1.data == b'{"name": "Greg"}'
assert new_event1.metadata == b''
assert new_event1.content_type == 'application/json'
assert isinstance(new_event1.id, UUID)

event_id = uuid4()
new_event2 = NewEvent(
    type='ImageCreated',
    data=b'01010101010101',
    metadata=b'{"a": 1}',
    content_type='application/octet-stream',
    id=event_id,
)
assert new_event2.type == 'ImageCreated'
assert new_event2.data == b'01010101010101'
assert new_event2.metadata == b'{"a": 1}'
assert new_event2.content_type == 'application/octet-stream'
assert new_event2.id == event_id

Recorded event objects

The RecordedEvent class is used when reading events.

The attribute type is a Python str, used to indicate the type of event that was recorded.

The attribute data is a Python bytes object, used to indicate the data of the event that was recorded.

The attribute metadata is a Python bytes object, used to indicate the metadata of the event that was recorded.

The attribute content_type is a Python str, used to indicate the type of data that was recorded for this event (usually application/json to indicate that this data can be parsed as JSON, but alternatively application/octet-stream to indicate that it is something else).

The attribute id is a Python UUID object, used to indicate the unique ID of the event that was recorded. Please note, when recorded events are returned from a call to read_stream_events() in EventStoreDB v21.10, the commit position is not actually set in the response. This attribute is typed as an optional value (Optional[UUID]), and in the case of using EventStoreDB v21.10 the value of this attribute will be None when reading recorded events from a stream. Recorded events will however have this values set when reading recorded events from read_all_events() and from both catch-up and persistent subscriptions.

The attribute stream_name is a Python str, used to indicate the name of the stream in which the event was recorded.

The attribute stream_position is a Python int, used to indicate the position in the stream at which the event was recorded.

The attribute commit_position is a Python int, used to indicate the commit position at which the event was recorded.

from esdbclient.events import RecordedEvent

recorded_event = RecordedEvent(
    type='OrderCreated',
    data=b'{}',
    metadata=b'',
    content_type='application/json',
    id=uuid4(),
    stream_name='stream1',
    stream_position=0,
    commit_position=512,
)

Asyncio client

The esdbclient package also includes an early version of an asynchronous I/O gRPC Python client. It follows exactly the same behaviors as the multithreaded ESDBClient, but uses the grpc.aio package and the asyncio module, instead of grpc and threading.

The async function AsyncioESDBClient constructs the client, and connects to a server. It can be imported from esdbclient, and can be called with the same arguments as ESDBClient. It supports both the "esdb" and the "esdb+discover" connection string URI schemes, and reconnects and retries methods when connection issues are encountered, just like ESDBClient.

from esdbclient import AsyncioESDBClient

The asynchronous I/O client has async methods append_events(), read_stream_events(), read_all_events(), subscribe_all_events(), delete_stream(), tombstone_stream(), and reconnect().

These methods are equivalent to the methods on ESDBClient. They have the same method signatures, and can be called with the same arguments, to the same effect. The methods which appear on ESDBClient but not on AsyncioESDBClient will be added soon.

The example below demonstrates the append_events(), read_stream_events() and subscribe_all_events() methods. These are the most useful methods for writing an event-sourced application, allowing new aggregate events to be recorded, the recorded events of an aggregate to be obtained so aggregates can be reconstructed, and the state of an application to propagated and processed with "exactly-once" semantics.

import asyncio

async def demonstrate_asyncio_client():

    # Construct client.
    client = await AsyncioESDBClient(
        uri=os.getenv("ESDB_URI"),
        root_certificates=os.getenv("ESDB_ROOT_CERTIFICATES"),
    )

    # Append events.
    stream_name = str(uuid4())
    event1 = NewEvent("OrderCreated", data=b'{}')
    event2 = NewEvent("OrderUpdated", data=b'{}')
    event3 = NewEvent("OrderDeleted", data=b'{}')

    await client.append_events(
        stream_name=stream_name,
        expected_position=None,
        events=[event1, event2, event3]
    )

    # Read stream events.
    recorded = await client.read_stream_events(stream_name)
    assert len(recorded) == 3
    assert recorded[0].id == event1.id
    assert recorded[1].id == event2.id
    assert recorded[2].id == event3.id


    # Subscribe all events.
    received = []
    async for event in await client.subscribe_all_events():
        received.append(event)
        if event.id == event3.id:
            break
    assert received[-3].id == event1.id
    assert received[-2].id == event2.id
    assert received[-1].id == event3.id


    # Close the client.
    await client.close()


# Run the demo.
asyncio.get_event_loop().run_until_complete(
    demonstrate_asyncio_client()
)

Contributors

Install Poetry

The first thing is to check you have Poetry installed.

$ poetry --version

If you don't, then please install Poetry.

$ curl -sSL https://install.python-poetry.org | python3 -

It will help to make sure Poetry's bin directory is in your PATH environment variable.

But in any case, make sure you know the path to the poetry executable. The Poetry installer tells you where it has been installed, and how to configure your shell.

Please refer to the Poetry docs for guidance on using Poetry.

Setup for PyCharm users

You can easily obtain the project files using PyCharm (menu "Git > Clone..."). PyCharm will then usually prompt you to open the project.

Open the project in a new window. PyCharm will then usually prompt you to create a new virtual environment.

Create a new Poetry virtual environment for the project. If PyCharm doesn't already know where your poetry executable is, then set the path to your poetry executable in the "New Poetry Environment" form input field labelled "Poetry executable". In the "New Poetry Environment" form, you will also have the opportunity to select which Python executable will be used by the virtual environment.

PyCharm will then create a new Poetry virtual environment for your project, using a particular version of Python, and also install into this virtual environment the project's package dependencies according to the project's poetry.lock file.

You can add different Poetry environments for different Python versions, and switch between them using the "Python Interpreter" settings of PyCharm. If you want to use a version of Python that isn't installed, either use your favourite package manager, or install Python by downloading an installer for recent versions of Python directly from the Python website.

Once project dependencies have been installed, you should be able to run tests from within PyCharm (right-click on the tests folder and select the 'Run' option).

Because of a conflict between pytest and PyCharm's debugger and the coverage tool, you may need to add --no-cov as an option to the test runner template. Alternatively, just use the Python Standard Library's unittest module.

You should also be able to open a terminal window in PyCharm, and run the project's Makefile commands from the command line (see below).

Setup from command line

Obtain the project files, using Git or suitable alternative.

In a terminal application, change your current working directory to the root folder of the project files. There should be a Makefile in this folder.

Use the Makefile to create a new Poetry virtual environment for the project and install the project's package dependencies into it, using the following command.

$ make install-packages

It's also possible to also install the project in 'editable mode'.

$ make install

Please note, if you create the virtual environment in this way, and then try to open the project in PyCharm and configure the project to use this virtual environment as an "Existing Poetry Environment", PyCharm sometimes has some issues (don't know why) which might be problematic. If you encounter such issues, you can resolve these issues by deleting the virtual environment and creating the Poetry virtual environment using PyCharm (see above).

Project Makefile commands

You can start EventStoreDB using the following command.

$ make start-eventstoredb

You can run tests using the following command (needs EventStoreDB to be running).

$ make test

You can stop EventStoreDB using the following command.

$ make stop-eventstoredb

You can check the formatting of the code using the following command.

$ make lint

You can reformat the code using the following command.

$ make fmt

Tests belong in ./tests. Code-under-test belongs in ./esdbclient.

Edit package dependencies in pyproject.toml. Update installed packages (and the poetry.lock file) using the following command.

$ make update-packages