Skip to main content

​
Overview

The StreamClient provides access to real-time streaming endpoints including filtered streams, sample streams, and compliance streams. Streaming connections automatically handle reconnection with exponential backoff.

​
Initialization

Access the stream client through the main Client instance: 
from xdk import Client

client = Client(bearer_token="your_token")
stream = client.stream

​
Key Methods

Stream real-time posts matching filter rules with automatic pagination and reconnection.
​
backfill_minutes
int
Number of minutes of backfill to retrieve
​
tweet_fields
List[str]
Tweet fields to include in the response
​
expansions
List[str]
Expansions to include in the response
Returns: Generator yielding posts as they arrive Example: 
from xdk.streaming import StreamConfig

# Configure streaming with retry behavior
config = StreamConfig(
    max_retries=10,
    initial_backoff=1.0,
    max_backoff=60.0
)

# Stream filtered posts
for post in client.stream.posts_search(
    tweet_fields=["created_at", "author_id"],
    config=config
):
    print(f"New post: {post.data.text}")

​
posts_sample()

Stream a random sample of public posts in real-time.
​
backfill_minutes
int
Number of minutes of backfill to retrieve
​
tweet_fields
List[str]
Tweet fields to include in the response
Returns: Generator yielding random sample posts Example: 
# Stream a sample of posts
for post in client.stream.posts_sample(
    tweet_fields=["lang", "created_at"]
):
    print(f"Sample post: {post.data.text}")

​
Stream Rules Management

Manage filtering rules for the filtered stream: 
from xdk.stream.models import UpdateRulesRequest

# Add stream rules
rules_request = UpdateRulesRequest(
    add=[
        {"value": "python programming", "tag": "python-tweets"},
        {"value": "from:xdevplatform", "tag": "xdev-tweets"}
    ]
)
response = client.stream.update_rules(body=rules_request)

# Get current rules
rules = client.stream.get_rules()
for rule in rules.data:
    print(f"Rule: {rule.value} (tag: {rule.tag})")

# Get rule counts
counts = client.stream.get_rule_counts()

​
Compliance Streams

​
posts_compliance()

Stream compliance events for posts (deletions, withheld content, etc.) Example: 
for event in client.stream.posts_compliance():
    print(f"Compliance event: {event}")

​
users_compliance()

Stream compliance events for users (suspensions, deactivations, etc.)

​
likes_compliance()

Stream compliance events for likes (unlike events, etc.)

​
Firehose Streams

For Enterprise access, firehose streams are available:
  • posts_firehose() - Full firehose of all public posts
  • posts_firehose_ja() - Japanese language firehose
  • posts_firehose_ko() - Korean language firehose
  • posts_firehose_pt() - Portuguese language firehose
  • likes_firehose() - All public like events

​
Authentication

Streaming endpoints require:
  • Bearer Token (App-only)
  • OAuth 2.0 User Context
Some firehose endpoints require Enterprise API access.

​
See Also

Build docs developers (and LLMs) love

Get started for freeTalk to us