feat: better progress edits, simpler telegram client (#5)

developing.md

@@ -38,7 +38,7 @@ The orchestrator module containing:
 | `CodexExecRunner` | Spawns `codex exec`, streams JSONL, handles cancellation |
 | `poll_updates()` | Async generator that drains backlog, long-polls updates, filters messages |
 | `_run_main_loop()` | TaskGroup-based main loop that spawns per-message handlers |
-| `_handle_message()` | Per-message handler with progress updates |
+| `handle_message()` | Per-message handler with progress updates |
 | `extract_session_id()` | Parses `resume: <uuid>` from message text |
 | `truncate_for_telegram()` | Smart truncation preserving resume lines |
 
@@ -122,7 +122,7 @@ poll_updates() drains backlog, long-polls, filters chat_id == from_id == cfg.cha
     ↓
 _run_main_loop() spawns tasks in TaskGroup
     ↓
-_handle_message() spawned as task
+handle_message() spawned as task
     ↓
 Send initial progress message (silent)
     ↓

src/takopi/exec_bridge.py

@@ -155,26 +155,17 @@ async def _send_or_edit_markdown(
     reply_to_message_id: int | None = None,
     disable_notification: bool = False,
     limit: int = TELEGRAM_MARKDOWN_LIMIT,
-) -> tuple[dict[str, Any], bool]:
+) -> tuple[dict[str, Any] | None, bool]:
     if edit_message_id is not None:
         rendered, entities = prepare_telegram(text, limit=limit)
-        try:
+        edited = await bot.edit_message_text(
-            return (
+            chat_id=chat_id,
-                await bot.edit_message_text(
+            message_id=edit_message_id,
-                    chat_id=chat_id,
+            text=rendered,
-                    message_id=edit_message_id,
+            entities=entities,
-                    text=rendered,
+        )
-                    entities=entities,
+        if edited is not None:
-                ),
+            return (edited, True)
-                True,
-            )
-        except Exception as e:
-            logger.info(
-                "[tg] edit failed chat_id=%s message_id=%s: %s",
-                chat_id,
-                edit_message_id,
-                e,
-            )
 
     rendered, entities = prepare_telegram(text, limit=limit)
     return (
@@ -192,6 +183,90 @@ async def _send_or_edit_markdown(
 EventCallback = Callable[[dict[str, Any]], Awaitable[None] | None]
 
 
+class ProgressEdits:
+    def __init__(
+        self,
+        *,
+        bot: TelegramClient,
+        chat_id: int,
+        progress_id: int | None,
+        renderer: ExecProgressRenderer,
+        started_at: float,
+        progress_edit_every: float,
+        clock: Callable[[], float],
+        sleep: Callable[[float], Awaitable[None]],
+        limit: int,
+        last_edit_at: float,
+        last_rendered: str | None,
+    ) -> None:
+        self.bot = bot
+        self.chat_id = chat_id
+        self.progress_id = progress_id
+        self.renderer = renderer
+        self.started_at = started_at
+        self.progress_edit_every = progress_edit_every
+        self.clock = clock
+        self.sleep = sleep
+        self.limit = limit
+        self.last_edit_at = last_edit_at
+        self.last_rendered = last_rendered
+        self._event_seq = 0
+        self._published_seq = 0
+        self.wakeup = asyncio.Event()
+        self.task: asyncio.Task[None] | None = (
+            asyncio.create_task(self.run()) if self.progress_id is not None else None
+        )
+
+    async def run(self) -> None:
+        if self.progress_id is None:
+            return
+        while True:
+            await self.wakeup.wait()
+            self.wakeup.clear()
+            while self._published_seq < self._event_seq:
+                await self.sleep(
+                    max(
+                        0.0,
+                        self.last_edit_at + self.progress_edit_every - self.clock(),
+                    )
+                )
+
+                seq_at_render = self._event_seq
+                now = self.clock()
+                md = self.renderer.render_progress(now - self.started_at)
+                rendered, entities = prepare_telegram(md, limit=self.limit)
+                if rendered != self.last_rendered:
+                    logger.debug(
+                        "[progress] edit message_id=%s md=%s", self.progress_id, md
+                    )
+                    self.last_edit_at = now
+                    edited = await self.bot.edit_message_text(
+                        chat_id=self.chat_id,
+                        message_id=self.progress_id,
+                        text=rendered,
+                        entities=entities,
+                    )
+                    if edited is not None:
+                        self.last_rendered = rendered
+
+                self._published_seq = seq_at_render
+            self.wakeup.clear()
+
+    async def on_event(self, evt: dict[str, Any]) -> None:
+        if not self.renderer.note_event(evt):
+            return
+        if self.progress_id is None:
+            return
+        self._event_seq += 1
+        self.wakeup.set()
+
+    async def shutdown(self) -> None:
+        if self.task is None:
+            return
+        self.task.cancel()
+        await asyncio.gather(self.task, return_exceptions=True)
+
+
 class CodexExecRunner:
     def __init__(
         self,
@@ -402,25 +477,20 @@ def _parse_bridge_config(
 
 
 async def _send_startup(cfg: BridgeConfig) -> None:
-    try:
+    logger.debug("[startup] message: %s", cfg.startup_msg)
-        logger.debug("[startup] message: %s", cfg.startup_msg)
+    sent = await cfg.bot.send_message(chat_id=cfg.chat_id, text=cfg.startup_msg)
-        await cfg.bot.send_message(chat_id=cfg.chat_id, text=cfg.startup_msg)
+    if sent is not None:
         logger.info("[startup] sent startup message to chat_id=%s", cfg.chat_id)
-    except Exception as e:
-        logger.info(
-            "[startup] failed to send startup message to chat_id=%s: %s", cfg.chat_id, e
-        )
 
 
 async def _drain_backlog(cfg: BridgeConfig, offset: int | None) -> int | None:
     drained = 0
     while True:
-        try:
+        updates = await cfg.bot.get_updates(
-            updates = await cfg.bot.get_updates(
+            offset=offset, timeout_s=0, allowed_updates=["message"]
-                offset=offset, timeout_s=0, allowed_updates=["message"]
+        )
-            )
+        if updates is None:
-        except Exception as e:
+            logger.info("[startup] backlog drain failed")
-            logger.info("[startup] backlog drain failed: %s", e)
             return offset
         logger.debug("[startup] backlog updates: %s", updates)
         if not updates:
@@ -431,7 +501,7 @@ async def _drain_backlog(cfg: BridgeConfig, offset: int | None) -> int | None:
         drained += len(updates)
 
 
-async def _handle_message(
+async def handle_message(
     cfg: BridgeConfig,
     *,
     chat_id: int,
@@ -440,6 +510,7 @@ async def _handle_message(
     resume_session: str | None,
     running_tasks: dict[str, asyncio.Task[Any]] | None = None,
     clock: Callable[[], float] = time.monotonic,
+    sleep: Callable[[float], Awaitable[None]] = asyncio.sleep,
     progress_edit_every: float = PROGRESS_EDIT_EVERY_S,
 ) -> None:
     logger.debug(
@@ -453,104 +524,57 @@ async def _handle_message(
     progress_renderer = ExecProgressRenderer(max_actions=5)
 
     progress_id: int | None = None
-
     last_edit_at = 0.0
-    edit_task: asyncio.Task[None] | None = None
     last_rendered: str | None = None
-    pending_rendered: str | None = None
 
-    async def _edit_progress(
+    initial_md = progress_renderer.render_progress(0.0)
-        md: str, rendered: str, entities: list[dict[str, Any]] | None
+    initial_rendered, initial_entities = prepare_telegram(
-    ) -> None:
+        initial_md, limit=TELEGRAM_MARKDOWN_LIMIT
-        nonlocal last_rendered, pending_rendered
+    )
-        if progress_id is None:
+    logger.debug(
-            return
+        "[progress] send reply_to=%s md=%s rendered=%s entities=%s",
-        logger.debug(
+        user_msg_id,
-            "[progress] edit message_id=%s md=%s rendered=%s entities=%s",
+        initial_md,
-            progress_id,
+        initial_rendered,
-            md,
+        initial_entities,
-            rendered,
+    )
-            entities,
+    progress_msg = await cfg.bot.send_message(
-        )
+        chat_id=chat_id,
-        try:
+        text=initial_rendered,
-            await cfg.bot.edit_message_text(
+        entities=initial_entities,
-                chat_id=chat_id,
+        reply_to_message_id=user_msg_id,
-                message_id=progress_id,
+        disable_notification=True,
-                text=rendered,
+    )
-                entities=entities,
+    if progress_msg is not None:
-            )
-            last_rendered = rendered
-        except Exception as e:
-            logger.info(
-                "[progress] edit failed chat_id=%s message_id=%s: %s",
-                chat_id,
-                progress_id,
-                e,
-            )
-        finally:
-            if pending_rendered == rendered:
-                pending_rendered = None
-
-    try:
-        initial_md = progress_renderer.render_progress(0.0)
-        initial_rendered, initial_entities = prepare_telegram(
-            initial_md, limit=TELEGRAM_MARKDOWN_LIMIT
-        )
-        logger.debug(
-            "[progress] send reply_to=%s md=%s rendered=%s entities=%s",
-            user_msg_id,
-            initial_md,
-            initial_rendered,
-            initial_entities,
-        )
-        progress_msg = await cfg.bot.send_message(
-            chat_id=chat_id,
-            text=initial_rendered,
-            entities=initial_entities,
-            reply_to_message_id=user_msg_id,
-            disable_notification=True,
-        )
         progress_id = int(progress_msg["message_id"])
         last_edit_at = clock()
         last_rendered = initial_rendered
         logger.debug("[progress] sent chat_id=%s message_id=%s", chat_id, progress_id)
-    except Exception as e:
+
-        logger.info(
+    edits = ProgressEdits(
-            "[handle] failed to send progress message chat_id=%s: %s", chat_id, e
+        bot=cfg.bot,
-        )
+        chat_id=chat_id,
+        progress_id=progress_id,
+        renderer=progress_renderer,
+        started_at=started_at,
+        progress_edit_every=progress_edit_every,
+        clock=clock,
+        sleep=sleep,
+        limit=TELEGRAM_MARKDOWN_LIMIT,
+        last_edit_at=last_edit_at,
+        last_rendered=last_rendered,
+    )
 
     exec_task: asyncio.Task[tuple[str, str, bool]] | None = None
-    tracked_session_id: str | None = None
 
     async def on_event(evt: dict[str, Any]) -> None:
-        nonlocal last_edit_at, edit_task, pending_rendered, tracked_session_id
-        if progress_id is None:
-            return
-        if not progress_renderer.note_event(evt):
-            return
-
         if (
             evt["type"] == "thread.started"
             and running_tasks is not None
             and exec_task is not None
         ):
-            tracked_session_id = progress_renderer.resume_session
+            running_tasks[evt["thread_id"]] = exec_task
-            if tracked_session_id:
+        await edits.on_event(evt)
-                running_tasks[tracked_session_id] = exec_task
-
-        now = clock()
-        if (now - last_edit_at) < progress_edit_every:
-            return
-        if edit_task is not None and not edit_task.done():
-            return
-        elapsed = now - started_at
-        md = progress_renderer.render_progress(elapsed)
-        rendered, entities = prepare_telegram(md, limit=TELEGRAM_MARKDOWN_LIMIT)
-        if rendered == last_rendered or rendered == pending_rendered:
-            return
-        last_edit_at = now
-        pending_rendered = rendered
-        edit_task = asyncio.create_task(_edit_progress(md, rendered, entities))
 
     exec_task = asyncio.create_task(
         cfg.runner.run_serialized(text, resume_session, on_event=on_event)
@@ -561,10 +585,9 @@ async def _handle_message(
         session_id, answer, saw_agent_message = await exec_task
     except asyncio.CancelledError:
         cancelled = True
-        session_id = tracked_session_id or resume_session
+        session_id = progress_renderer.resume_session or resume_session
     except Exception as e:
-        if edit_task is not None:
+        await edits.shutdown()
-            await asyncio.gather(edit_task, return_exceptions=True)
 
         err = _clamp_tg_text(f"Error:\n{e}")
         logger.debug("[error] send reply_to=%s text=%s", user_msg_id, err)
@@ -579,14 +602,12 @@ async def _handle_message(
         )
         return
     finally:
-        if tracked_session_id and running_tasks is not None and exec_task is not None:
+        if running_tasks is not None:
-            # Avoid removing a newer task for the same session_id if another run
+            for sid, task in list(running_tasks.items()):
-            # registered while this one was finishing.
+                if task is exec_task:
-            if running_tasks.get(tracked_session_id) is exec_task:
+                    running_tasks.pop(sid, None)
-                running_tasks.pop(tracked_session_id, None)
 
-    if edit_task is not None:
+    await edits.shutdown()
-        await asyncio.gather(edit_task, return_exceptions=True)
 
     elapsed = clock() - started_at
     if cancelled:
@@ -631,7 +652,7 @@ async def _handle_message(
             final_entities,
         )
 
-    _, edited = await _send_or_edit_markdown(
+    final_msg, edited = await _send_or_edit_markdown(
         cfg.bot,
         chat_id=chat_id,
         text=final_md,
@@ -640,12 +661,11 @@ async def _handle_message(
         disable_notification=False,
         limit=TELEGRAM_MARKDOWN_LIMIT,
     )
+    if final_msg is None:
+        return
     if progress_id is not None and (edit_message_id is None or not edited):
-        try:
+        logger.debug("[final] delete progress message_id=%s", progress_id)
-            logger.debug("[final] delete progress message_id=%s", progress_id)
+        await cfg.bot.delete_message(chat_id=chat_id, message_id=progress_id)
-            await cfg.bot.delete_message(chat_id=chat_id, message_id=progress_id)
-        except Exception:
-            pass
 
 
 async def poll_updates(cfg: BridgeConfig):
@@ -654,12 +674,11 @@ async def poll_updates(cfg: BridgeConfig):
     await _send_startup(cfg)
 
     while True:
-        try:
+        updates = await cfg.bot.get_updates(
-            updates = await cfg.bot.get_updates(
+            offset=offset, timeout_s=50, allowed_updates=["message"]
-                offset=offset, timeout_s=50, allowed_updates=["message"]
+        )
-            )
+        if updates is None:
-        except Exception as e:
+            logger.info("[loop] getUpdates failed")
-            logger.info("[loop] getUpdates failed: %s", e)
             await asyncio.sleep(2)
             continue
         logger.debug("[loop] updates: %s", updates)
@@ -724,7 +743,7 @@ async def _run_main_loop(cfg: BridgeConfig) -> None:
         while True:
             chat_id, user_msg_id, text, resume_session = await queue.get()
             try:
-                await _handle_message(
+                await handle_message(
                     cfg,
                     chat_id=chat_id,
                     user_msg_id=user_msg_id,

src/takopi/telegram.py

@@ -1,8 +1,6 @@
 from __future__ import annotations
 
-import asyncio
 import logging
-from collections.abc import Awaitable, Callable
 from typing import Any
 
 import httpx
@@ -13,67 +11,73 @@ logger = logging.getLogger(__name__)
 logger.addFilter(RedactTokenFilter())
 
 
-class TelegramAPIError(RuntimeError):
-    def __init__(
-        self, method: str, payload: dict[str, Any], status_code: int | None
-    ) -> None:
-        desc = payload.get("description") or str(payload)
-        super().__init__(f"{method} failed: {desc}")
-        self.payload = payload
-        self.status_code = status_code
-
-
 class TelegramClient:
     def __init__(
         self,
         token: str,
         timeout_s: float = 120,
         client: httpx.AsyncClient | None = None,
-        sleep: Callable[[float], Awaitable[None]] = asyncio.sleep,
     ) -> None:
         if not token:
             raise ValueError("Telegram token is empty")
         self._base = f"https://api.telegram.org/bot{token}"
         self._client = client or httpx.AsyncClient(timeout=timeout_s)
         self._owns_client = client is None
-        self._sleep = sleep
 
     async def close(self) -> None:
         if self._owns_client:
             await self._client.aclose()
 
-    async def _post(self, method: str, json_data: dict[str, Any]) -> Any:
+    async def _post(self, method: str, json_data: dict[str, Any]) -> Any | None:
+        logger.debug("[telegram] request %s: %s", method, json_data)
         try:
-            logger.debug("[telegram] request %s: %s", method, json_data)
             resp = await self._client.post(f"{self._base}/{method}", json=json_data)
-            payload: dict[str, Any] | None = None
-            try:
-                payload = resp.json()
-            except Exception:
-                resp.raise_for_status()
-                raise
-            if not payload.get("ok"):
-                params = payload.get("parameters") or {}
-                retry_after = params.get("retry_after")
-                if resp.status_code == 429 and isinstance(retry_after, int):
-                    logger.warning(
-                        "[telegram] 429 retry_after=%s method=%s", retry_after, method
-                    )
-                    await self._sleep(retry_after)
-                    return await self._post(method, json_data)
-                raise TelegramAPIError(method, payload, resp.status_code)
-            logger.debug("[telegram] response %s: %s", method, payload)
-            return payload["result"]
         except httpx.HTTPError as e:
-            logger.error("Telegram network error: %s", e)
+            url = getattr(e.request, "url", None)
-            raise
+            logger.error(
+                "[telegram] network error method=%s url=%s: %s", method, url, e
+            )
+            return None
+
+        try:
+            payload = resp.json()
+        except Exception as e:
+            logger.error(
+                "[telegram] bad response method=%s status=%s url=%s: %s",
+                method,
+                resp.status_code,
+                resp.request.url,
+                e,
+            )
+            return None
+
+        if not isinstance(payload, dict):
+            logger.error(
+                "[telegram] invalid response method=%s url=%s: %r",
+                method,
+                resp.request.url,
+                payload,
+            )
+            return None
+
+        if not payload.get("ok"):
+            logger.error(
+                "[telegram] api error method=%s url=%s: %s",
+                method,
+                resp.request.url,
+                payload,
+            )
+            return None
+
+        logger.debug("[telegram] response %s: %s", method, payload)
+        return payload.get("result")
 
     async def get_updates(
         self,
         offset: int | None,
         timeout_s: int = 50,
         allowed_updates: list[str] | None = None,
-    ) -> list[dict]:
+    ) -> list[dict] | None:
         params: dict[str, Any] = {"timeout": timeout_s}
         if offset is not None:
             params["offset"] = offset
@@ -89,7 +93,7 @@ class TelegramClient:
         disable_notification: bool | None = False,
         entities: list[dict] | None = None,
         parse_mode: str | None = None,
-    ) -> dict:
+    ) -> dict | None:
         params: dict[str, Any] = {
             "chat_id": chat_id,
             "text": text,
@@ -111,7 +115,7 @@ class TelegramClient:
         text: str,
         entities: list[dict] | None = None,
         parse_mode: str | None = None,
-    ) -> dict:
+    ) -> dict | None:
         params: dict[str, Any] = {
             "chat_id": chat_id,
             "message_id": message_id,

tests/test_exec_bridge.py

@@ -186,12 +186,30 @@ class _FakeRunner:
 class _FakeClock:
     def __init__(self, start: float = 0.0) -> None:
         self._now = start
+        self._sleep_until: float | None = None
+        self._sleep_event: asyncio.Event | None = None
+        self.sleep_calls = 0
 
     def __call__(self) -> float:
         return self._now
 
     def set(self, value: float) -> None:
         self._now = value
+        if self._sleep_until is None or self._sleep_event is None:
+            return
+        if self._sleep_until <= self._now:
+            self._sleep_event.set()
+            self._sleep_until = None
+            self._sleep_event = None
+
+    async def sleep(self, delay: float) -> None:
+        self.sleep_calls += 1
+        if delay <= 0:
+            await asyncio.sleep(0)
+            return
+        self._sleep_until = self._now + delay
+        self._sleep_event = asyncio.Event()
+        await self._sleep_event.wait()
 
 
 class _FakeRunnerWithEvents:
@@ -203,12 +221,16 @@ class _FakeRunnerWithEvents:
         clock: _FakeClock,
         answer: str = "ok",
         session_id: str = "019b66fc-64c2-7a71-81cd-081c504cfeb2",
+        advance_after: float | None = None,
+        hold: asyncio.Event | None = None,
     ) -> None:
         self._events = events
         self._times = times
         self._clock = clock
         self._answer = answer
         self._session_id = session_id
+        self._advance_after = advance_after
+        self._hold = hold
 
     async def run_serialized(self, *_args, **kwargs) -> tuple[str, str, bool]:
         on_event = kwargs.get("on_event")
@@ -217,11 +239,16 @@ class _FakeRunnerWithEvents:
                 self._clock.set(when)
                 await on_event(event)
                 await asyncio.sleep(0)
+            if self._advance_after is not None:
+                self._clock.set(self._advance_after)
+                await asyncio.sleep(0)
+        if self._hold is not None:
+            await self._hold.wait()
         return (self._session_id, self._answer, True)
 
 
 def test_final_notify_sends_loud_final_message() -> None:
-    from takopi.exec_bridge import BridgeConfig, _handle_message
+    from takopi.exec_bridge import BridgeConfig, handle_message
 
     bot = _FakeBot()
     runner = _FakeRunner(answer="ok")
@@ -235,7 +262,7 @@ def test_final_notify_sends_loud_final_message() -> None:
     )
 
     asyncio.run(
-        _handle_message(
+        handle_message(
             cfg,
             chat_id=123,
             user_msg_id=10,
@@ -250,7 +277,7 @@ def test_final_notify_sends_loud_final_message() -> None:
 
 
 def test_new_final_message_forces_notification_when_too_long_to_edit() -> None:
-    from takopi.exec_bridge import BridgeConfig, _handle_message
+    from takopi.exec_bridge import BridgeConfig, handle_message
 
     bot = _FakeBot()
     runner = _FakeRunner(answer="x" * 10_000)
@@ -264,7 +291,7 @@ def test_new_final_message_forces_notification_when_too_long_to_edit() -> None:
     )
 
     asyncio.run(
-        _handle_message(
+        handle_message(
             cfg,
             chat_id=123,
             user_msg_id=10,
@@ -279,7 +306,7 @@ def test_new_final_message_forces_notification_when_too_long_to_edit() -> None:
 
 
 def test_progress_edits_are_rate_limited() -> None:
-    from takopi.exec_bridge import BridgeConfig, _handle_message
+    from takopi.exec_bridge import BridgeConfig, handle_message
 
     bot = _FakeBot()
     clock = _FakeClock()
@@ -294,13 +321,61 @@ def test_progress_edits_are_rate_limited() -> None:
             },
         },
         {
-            "type": "item.completed",
+            "type": "item.started",
+            "item": {
+                "id": "item_1",
+                "type": "command_execution",
+                "command": "echo 2",
+                "status": "in_progress",
+            },
+        },
+    ]
+    runner = _FakeRunnerWithEvents(
+        events=events,
+        times=[0.2, 0.4],
+        clock=clock,
+        advance_after=1.0,
+    )
+    cfg = BridgeConfig(
+        bot=bot,  # type: ignore[arg-type]
+        runner=runner,  # type: ignore[arg-type]
+        chat_id=123,
+        final_notify=True,
+        startup_msg="",
+        max_concurrency=1,
+    )
+
+    asyncio.run(
+        handle_message(
+            cfg,
+            chat_id=123,
+            user_msg_id=10,
+            text="hi",
+            resume_session=None,
+            clock=clock,
+            sleep=clock.sleep,
+            progress_edit_every=1.0,
+        )
+    )
+
+    assert len(bot.edit_calls) == 1
+    assert "echo 2" in bot.edit_calls[0]["text"]
+
+
+def test_progress_edits_do_not_sleep_again_without_new_events() -> None:
+    from takopi.exec_bridge import BridgeConfig, handle_message
+
+    bot = _FakeBot()
+    clock = _FakeClock()
+    hold = asyncio.Event()
+    events = [
+        {
+            "type": "item.started",
             "item": {
                 "id": "item_0",
                 "type": "command_execution",
                 "command": "echo 1",
-                "exit_code": 0,
+                "status": "in_progress",
-                "status": "completed",
             },
         },
         {
@@ -315,8 +390,10 @@ def test_progress_edits_are_rate_limited() -> None:
     ]
     runner = _FakeRunnerWithEvents(
         events=events,
-        times=[0.2, 0.4, 1.2],
+        times=[0.2, 0.4],
         clock=clock,
+        advance_after=None,
+        hold=hold,
     )
     cfg = BridgeConfig(
         bot=bot,  # type: ignore[arg-type]
@@ -327,23 +404,50 @@ def test_progress_edits_are_rate_limited() -> None:
         max_concurrency=1,
     )
 
-    asyncio.run(
+    async def run_test() -> None:
-        _handle_message(
+        task = asyncio.create_task(
-            cfg,
+            handle_message(
-            chat_id=123,
+                cfg,
-            user_msg_id=10,
+                chat_id=123,
-            text="hi",
+                user_msg_id=10,
-            resume_session=None,
+                text="hi",
-            clock=clock,
+                resume_session=None,
-            progress_edit_every=1.0,
+                clock=clock,
+                sleep=clock.sleep,
+                progress_edit_every=1.0,
+            )
         )
-    )
 
-    assert len(bot.edit_calls) == 1
+        for _ in range(100):
+            if clock._sleep_until is not None:
+                break
+            await asyncio.sleep(0)
+
+        assert clock._sleep_until == pytest.approx(1.0)
+
+        clock.set(1.0)
+
+        for _ in range(100):
+            if bot.edit_calls:
+                break
+            await asyncio.sleep(0)
+
+        assert len(bot.edit_calls) == 1
+
+        for _ in range(5):
+            await asyncio.sleep(0)
+
+        assert clock.sleep_calls == 1
+        assert clock._sleep_until is None
+
+        hold.set()
+        await task
+
+    asyncio.run(run_test())
 
 
 def test_bridge_flow_sends_progress_edits_and_final_resume() -> None:
-    from takopi.exec_bridge import BridgeConfig, _handle_message
+    from takopi.exec_bridge import BridgeConfig, handle_message
 
     bot = _FakeBot()
     clock = _FakeClock()
@@ -386,13 +490,14 @@ def test_bridge_flow_sends_progress_edits_and_final_resume() -> None:
     )
 
     asyncio.run(
-        _handle_message(
+        handle_message(
             cfg,
             chat_id=123,
             user_msg_id=42,
             text="do it",
             resume_session=None,
             clock=clock,
+            sleep=clock.sleep,
             progress_edit_every=1.0,
         )
     )
@@ -529,7 +634,7 @@ class _FakeRunnerCancellable:
 
 
 def test_handle_message_cancelled_renders_cancelled_state() -> None:
-    from takopi.exec_bridge import BridgeConfig, _handle_message
+    from takopi.exec_bridge import BridgeConfig, handle_message
 
     bot = _FakeBot()
     session_id = "019b66fc-64c2-7a71-81cd-081c504cfeb2"
@@ -546,7 +651,7 @@ def test_handle_message_cancelled_renders_cancelled_state() -> None:
 
     async def run_test():
         task = asyncio.create_task(
-            _handle_message(
+            handle_message(
                 cfg,
                 chat_id=123,
                 user_msg_id=10,

tests/test_telegram_client.py

@@ -8,46 +8,35 @@ from takopi.logging import RedactTokenFilter
 from takopi.telegram import TelegramClient
 
 
-def test_telegram_429_retry_after_calls_sleep() -> None:
+def test_telegram_429_no_retry() -> None:
     calls: list[int] = []
-    sleeps: list[float] = []
-
-    async def fake_sleep(seconds: float) -> None:
-        sleeps.append(seconds)
 
     def handler(request: httpx.Request) -> httpx.Response:
         calls.append(1)
-        if len(calls) == 1:
-            return httpx.Response(
-                429,
-                json={
-                    "ok": False,
-                    "description": "retry",
-                    "parameters": {"retry_after": 3},
-                },
-                request=request,
-            )
         return httpx.Response(
-            200,
+            429,
-            json={"ok": True, "result": {"message_id": 1}},
+            json={
+                "ok": False,
+                "description": "retry",
+                "parameters": {"retry_after": 3},
+            },
             request=request,
         )
 
     transport = httpx.MockTransport(handler)
 
-    async def run() -> dict:
+    async def run() -> dict | None:
         client = httpx.AsyncClient(transport=transport)
         try:
-            tg = TelegramClient("123:abcDEF_ghij", client=client, sleep=fake_sleep)
+            tg = TelegramClient("123:abcDEF_ghij", client=client)
             return await tg._post("sendMessage", {"chat_id": 1, "text": "hi"})
         finally:
             await client.aclose()
 
     result = asyncio.run(run())
 
-    assert result == {"message_id": 1}
+    assert result is None
-    assert sleeps == [3]
+    assert len(calls) == 1
-    assert len(calls) == 2
 
 
 def test_no_token_in_logs_on_http_error(caplog: pytest.LogCaptureFixture) -> None:
@@ -70,8 +59,7 @@ def test_no_token_in_logs_on_http_error(caplog: pytest.LogCaptureFixture) -> Non
             await client.aclose()
 
     caplog.set_level(logging.ERROR)
-    with pytest.raises(httpx.HTTPStatusError):
+    asyncio.run(run())
-        asyncio.run(run())
 
     root_logger.removeFilter(redactor)