feat: support and flush chunks in the chat stream helper by zimeg · Pull Request #1809 · slackapi/python-slack-sdk

zimeg · 2025-12-18T00:06:15Z

Summary

This PR updates the chat_stream helper to support and flush "chunks".

Testing

The following snippet might be useful in testing:

import time

from slack_sdk import WebClient
from slack_sdk.models.messages.chunk import MarkdownTextChunk, TaskUpdateChunk

streamer = client.chat_stream(
    channel="C0123456789",
    recipient_team_id="T0123456789",
    recipient_user_id="U0123456789",
    thread_ts="10101010101.010101",
)

streamer.append(
    chunks=[
        MarkdownTextChunk(
            text="Hello.\nI have received the task. ",
        ),
        MarkdownTextChunk(
            text="This task appears manageable.\nThat is good.",
        ),
        TaskUpdateChunk(
            id="001",
            title="Understanding the task...",
            status="in_progress",
            details="- Identifying the goal\n- Identifying constraints",
        ),
        TaskUpdateChunk(
            id="002",
            title="Performing acrobatics...",
            status="pending",
        ),
    ],
)
time.sleep(4)

streamer.append(
    chunks=[
        TaskUpdateChunk(
            id="001",
            title="Understanding the task...",
            status="complete",
            details="\n- Pretending this was obvious",
            output="We'll continue to ramble now",
        ),
        TaskUpdateChunk(
            id="002",
            title="Performing acrobatics...",
            status="in_progress",
        ),
    ],
)
time.sleep(4)

streamer.stop(
    chunks=[
        TaskUpdateChunk(
            id="002",
            title="Performing acrobatics...",
            status="complete",
            details="- Jumped atop ropes\n- Juggled bowling pins\n- Rode a single wheel too",
        ),
        MarkdownTextChunk(
            text="The crowd appears to be astouned and applauds :popcorn:"
        ),
    ],
)

Requirements

I've read and understood the Contributing Guidelines and have done my best effort to follow them.
I've read and agree to the Code of Conduct.
I've run python3 -m venv .venv && source .venv/bin/activate && ./scripts/run_validation.sh after making the changes.

…thods

codecov · 2025-12-18T00:10:56Z

Codecov Report

✅ All modified and coverable lines are covered by tests.
⚠️ Please upload report for BASE (feat-ai-apps-thinking-steps@1c84f7f). Learn more about missing BASE report.
✅ All tests successful. No failed tests found.

Additional details and impacted files

@@                      Coverage Diff                       @@
##             feat-ai-apps-thinking-steps    #1809   +/-   ##
==============================================================
  Coverage                               ?   83.90%           
==============================================================
  Files                                  ?      116           
  Lines                                  ?    13192           
  Branches                               ?        0           
==============================================================
  Hits                                   ?    11069           
  Misses                                 ?     2123           
  Partials                               ?        0

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

srtaalej

Thanks for adding this ⭐ ⭐ ⭐ left some thoughts on the empty append - would love to hear your thoughts

srtaalej · 2026-01-14T20:48:35Z

slack_sdk/web/client.py

        channel: str,
        ts: str,
-        markdown_text: str,
+        markdown_text: Optional[str] = None,


🌟 it is much easier to have both markdown and chunks optional!

srtaalej · 2026-01-14T20:49:57Z

tests/slack_sdk/web/test_chat_stream.py

+            recipient_team_id="T0123456789",
+            recipient_user_id="U0123456789",
+            thread_ts="123.000",
+        )


i tested this with an empty call to .append() and i think we should raise an error/ warning for when append is sent empty 🤔

A warning is a nice idea. I wouldn't consider it a blocker for the PR, if we want to move forward and add the warning later.

@srtaalej @mwbrooks I super appreciate the suggestions! The API would return an error in this case, but the SDK helper won't... We should make that more obvious in a follow up 👁️‍🗨️ ✨

mwbrooks

✅ Nice, thanks for updating the stream helper! 🙇🏻

🧪 Manual testing works! Below is the modified listeners/assistant/message.py that I used based on the PR test steps.

bolt-python-assistant-template/listeners/assistant/message.py:

from logging import Logger
from typing import Dict, List

from slack_bolt import BoltContext, Say, SetStatus
from slack_sdk import WebClient

from ai.llm_caller import call_llm

from ..views.feedback_block import create_feedback_block

import time
from slack_sdk.models.messages.chunk import MarkdownTextChunk, TaskUpdateChunk, URLSource


def message(
    client: WebClient,
    context: BoltContext,
    logger: Logger,
    payload: dict,
    say: Say,
    set_status: SetStatus,
):
    """
    Handles when users send messages or select a prompt in an assistant thread and generate AI responses:

    Args:
        client: Slack WebClient for making API calls
        context: Bolt context containing channel and thread information
        logger: Logger instance for error tracking
        payload: Event payload with message details (channel, user, text, etc.)
        say: Function to send messages to the thread
        set_status: Function to update the assistant's status
    """
    try:
        channel_id = payload["channel"]
        team_id = context.team_id
        thread_ts = payload["thread_ts"]
        user_id = context.user_id

        set_status(
            status="thinking...",
            loading_messages=[
                "Teaching the hamsters to type faster…",
                "Untangling the internet cables…",
                "Consulting the office goldfish…",
                "Polishing up the response just for you…",
                "Convincing the AI to stop overthinking…",
            ],
        )

        replies = client.conversations_replies(
            channel=context.channel_id,
            ts=context.thread_ts,
            oldest=context.thread_ts,
            limit=10,
        )
        messages_in_thread: List[Dict[str, str]] = []
        for message in replies["messages"]:
            role = "user" if message.get("bot_id") is None else "assistant"
            messages_in_thread.append({"role": role, "content": message["text"]})

        streamer = client.chat_stream(
            channel=channel_id,
            recipient_team_id=team_id,
            recipient_user_id=user_id,
            thread_ts=thread_ts,
        )

        streamer.append(
            chunks=[
                MarkdownTextChunk(
                    text="Hello.\nI have received the task. ",
                ),
                MarkdownTextChunk(
                    text="This task appears manageable.\nThat is good.",
                ),
                TaskUpdateChunk(
                    id="001",
                    title="Understanding the task...",
                    status="in_progress",
                    details="- Identifying the goal\n- Identifying constraints",
                ),
                TaskUpdateChunk(
                    id="002",
                    title="Performing acrobatics...",
                    status="pending",
                ),
            ],
        )

        time.sleep(4)

        streamer.append(
            chunks=[
                TaskUpdateChunk(
                    id="001",
                    title="Understanding the task...",
                    status="complete",
                    details="\n- Pretending this was obvious",
                    output="We'll continue to ramble now",
                ),
                TaskUpdateChunk(
                    id="002",
                    title="Performing acrobatics...",
                    status="in_progress",
                ),
            ],
        )

        time.sleep(4)

        streamer.stop(
            chunks=[
                TaskUpdateChunk(
                    id="12",
                    title="solved equation!",
                    status="complete",
                    sources=[
                        URLSource(
                            url="https://oeis.org",
                            text="The On-Line Encyclopedia of Integer Sequences (OEIS)",
                        ),
                    ],
                ),
                MarkdownTextChunk(text="that computes."),
            ],
        )

    except Exception as e:
        logger.exception(f"Failed to handle a user message event: {e}")
        say(f":warning: Something went wrong! ({e})")

mwbrooks · 2026-01-16T03:22:30Z

slack_sdk/web/async_chat_stream.py

                raise e.SlackRequestError("Failed to stop stream: stream not started")
            self._stream_ts = str(response["ts"])
            self._state = "in_progress"
+        flushings: List[Union[Dict, Chunk]] = []


😉 quite the visual name! 💩 Would chunks_to_flush also work? Perhaps too long.

@mwbrooks I do appreciate this suggestion. Let's blame @claude for this one. But in a follow up with warnings appearing with append too, I agree we should change this!

mwbrooks · 2026-01-16T03:26:16Z

tests/slack_sdk/web/test_chat_stream.py

+            recipient_team_id="T0123456789",
+            recipient_user_id="U0123456789",
+            thread_ts="123.000",
+        )


A warning is a nice idea. I wouldn't consider it a blocker for the PR, if we want to move forward and add the warning later.

WilliamBergamin

These changes look good to me, I left a few comments to make sure I understand the behavior properly but they should be non blocking 💯

WilliamBergamin · 2026-01-16T15:13:40Z

slack_sdk/web/chat_stream.py

+        flushings: List[Union[Dict, Chunk]] = []
+        if len(self._buffer) != 0:
+            flushings.append(MarkdownTextChunk(text=self._buffer))
+        if chunks is not None:
+            flushings.extend(chunks)


Nice! This allows the markdown_text and chunks to be accepted and passed to the method using the chunks param 💯

@WilliamBergamin The chunks argument might be preferred IIRC but I think markdown_text is still so useful for streaming LLM responses direct 👾

WilliamBergamin · 2026-01-16T15:15:13Z

slack_sdk/web/chat_stream.py

+        markdown_text: Optional[str] = None,
+        chunks: Optional[Sequence[Union[Dict, Chunk]]] = None,


Since both arguments are now optional what happens if a developer does not pass any of these arguments when using the method?

@WilliamBergamin A markdown_text_or_chunks_required error should appear from the API but I think @srtaalej and @mwbrooks mention that a warning might be useful to include with the SDK helper...

📝 Noting that we should follow up with this change!

WilliamBergamin · 2026-01-16T15:32:48Z

slack_sdk/web/chat_stream.py

+        if markdown_text is not None:
+            self._buffer += markdown_text
+        if len(self._buffer) >= self._buffer_size or chunks is not None:
+            return self._flush_buffer(chunks=chunks, **kwargs)


If I'm understanding this correctly if I do

streamer = client.chat_stream( channel="C0123456789", recipient_team_id="T0123456789", recipient_user_id="U0123456789", thread_ts="10101010101.010101", ) streamer.append(markdown_text="\n") streamer.append( chunks=[ MarkdownTextChunk( text="Hello.\nI have received the task. ", ), ], )

The first append with markdown will not be sent to Slack but rather added to the buffer

The second append with the chunks will be sent to Slack along with what ever may be in the buffer regarless of if the buffer has exceeded the buffer_size?

@WilliamBergamin Correct! Providing chunks is almost like a forced flush of the buffer 🪬

zimeg · 2026-01-16T23:17:40Z

@srtaalej @mwbrooks @WilliamBergamin Oncemore the kind reviews are so appreciated! 🙏 ✨

I'm holding off on more changes to this PR for testing soonest, but will promise these to be included in upcoming commits.

zimeg added 5 commits December 10, 2025 18:13

feat: accept chunks as arguments to chat.{start,append,stop}Stream me…

c3bdb2d

…thods

fix: remove unsupported and unused identifiers for full support

8c56e22

style: remove mypy extra ignore comment for overriden attributes

783ada5

test: confirm chunks parse as expected json values

1fb7355

feat: support and flush chunks in the chat stream helper

5de794b

zimeg requested review from mwbrooks and srtaalej December 18, 2025 00:06

zimeg self-assigned this Dec 18, 2025

zimeg added enhancement M-T: A feature request for new functionality semver:minor web-client labels Dec 18, 2025

zimeg mentioned this pull request Dec 18, 2025

feat: showcase text generation and thinking steps from suggested prompts slack-samples/bolt-python-assistant-template#37

Merged

3 tasks

zimeg added 4 commits December 17, 2025 16:22

test: dump chunks json before comparison for exact parsings

61d6d53

test: update async tests to expect chunks when flushing buffer

85081e1

style: prefer using markdown text chunks in internal calls

a6bb951

fix: support explicit json values as chunk objects

92c93e0

zimeg mentioned this pull request Jan 12, 2026

feat: accept chunks as arguments to chat.{start,append,stop}Stream methods #1806

Merged

7 tasks

fix: mirror dictionary chunk option to async streamer

7c32814

zimeg marked this pull request as ready for review January 14, 2026 20:05

zimeg requested a review from a team as a code owner January 14, 2026 20:05

srtaalej reviewed Jan 14, 2026

View reviewed changes

mwbrooks approved these changes Jan 16, 2026

View reviewed changes

WilliamBergamin approved these changes Jan 16, 2026

View reviewed changes

Base automatically changed from zimeg-feat-ai-apps-chunks to feat-ai-apps-thinking-steps January 16, 2026 23:04

chore: merge w base branch

833f286

zimeg merged commit f4c0182 into feat-ai-apps-thinking-steps Jan 16, 2026
16 checks passed

zimeg deleted the zimeg-feat-ai-apps-chunks-streamer branch January 16, 2026 23:25

zimeg mentioned this pull request Jan 22, 2026

fix: remove icon_url from url source block element and plan_id from plan block and update chat stream variable #1822

Merged

7 tasks

zimeg mentioned this pull request Jan 29, 2026

feat: add thinking steps support as chunks to chat stream with plan and task blocks #1824

Draft

7 tasks

		markdown_text: Optional[str] = None,
		chunks: Optional[Sequence[Union[Dict, Chunk]]] = None,

Conversation

zimeg commented Dec 18, 2025

Summary

Testing

Category

Requirements

Uh oh!

codecov bot commented Dec 18, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

srtaalej left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

mwbrooks left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

WilliamBergamin left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

zimeg commented Jan 16, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

codecov bot commented Dec 18, 2025 •

edited

Loading