Skip to main content

​
Wait until it’s full

Acontext will start to process the messages once the buffer is full (i.e. unprocessed message turns >= PROJECT_SESSION_MESSAGE_BUFFER_MAX_TURNS) So if your PROJECT_SESSION_MESSAGE_BUFFER_MAX_TURNS is 3: 
from acontext import AcontextClient

client = AcontextClient(
    api_key="sk-ac-your-root-api-bearer-token",
    base_url="http://localhost:8029/api/v1"
)

# this storing won't be processed immediately
client.sessions.store_message(
    session_id="session-uuid",
    blob=message1,
    format="openai"
)

# this storing won't be processed immediately
client.sessions.store_message(
    session_id="session-uuid",
    blob=message2,
    format="openai"
)

# will trigger the processing immediately
client.sessions.store_message(
    session_id="session-uuid",
    blob=message3,
    format="openai"
)

​
Wait until it’s IDLE

Acontext will start to process the messages once the buffer is IDLE for a while (i.e. no new messages for PROJECT_SESSION_MESSAGE_BUFFER_TTL_SECONDS seconds) So if your PROJECT_SESSION_MESSAGE_BUFFER_TTL_SECONDS is 3: 
from acontext import AcontextClient
import time

client = AcontextClient(
    api_key="sk-ac-your-root-api-bearer-token",
    base_url="http://localhost:8029/api/v1"
)

# this storing won't be processed immediately
client.sessions.store_message(
    session_id="session-uuid",
    blob=message1,
    format="openai"
)

time.sleep(3)
# Acontext will start to process the messages in background.

​
Force Buffer Processing

You can force the buffer processing by calling the flush method: 
from acontext import AcontextClient

client = AcontextClient(
    api_key="sk-ac-your-root-api-bearer-token",
    base_url="http://localhost:8029/api/v1"
)

client.sessions.flush(session_id="session-uuid")

​
Monitor Message Processing Status

You can check the observing status of messages in your session to understand how many are observed, in process, or pending: 
from acontext import AcontextClient

client = AcontextClient(
    api_key="sk-ac-your-root-api-bearer-token",
    base_url="http://localhost:8029/api/v1"
)

# Get message observing status for a session
status = client.sessions.messages_observing_status(session_id="session-uuid")

print(f"Observed: {status.observed}")
print(f"In Process: {status.in_process}")
print(f"Pending: {status.pending}")
print(f"Last Updated: {status.updated_at}")

​
Status Meanings

Messages that have been fully processed and observed by the system. These messages are complete and available for analysis.
Messages currently being processed by the system. These are actively being analyzed or transformed.
Messages waiting in the buffer to be processed. These will be processed once the buffer conditions are met (full buffer or TTL expired).
Use messages_observing_status() to monitor buffer health and ensure messages are being processed as expected. A high number of pending messages may indicate you need to adjust buffer settings or call flush() manually.

​
Customization

Settings

Customize the behavior of Buffer.