mirror of
https://github.com/NousResearch/hermes-agent.git
synced 2026-07-31 19:16:29 +00:00
2008d80a9e
- tests/gateway/test_base_auto_tts_output_format.py (new): the base adapter auto-TTS block passes an explicit .ogg output path on every OPUS_VOICE_PLATFORMS member (parametrized from the tts_tool set — the single source of truth), keeps .mp3 on non-opus platforms, honors the tool's success flag, and stays unique/uuid-based. - tests/gateway/test_auto_voice_reply_format.py: runner _send_voice_reply parametrized across Matrix/Feishu/WhatsApp/Signal (.ogg) alongside the existing Telegram/Slack cases; streamed+global-auto-TTS gate regression. - tests/gateway/test_telegram_voice_caption_markdown.py (new): caption MarkdownV2 formatting, entity-rejection plain fallback, overflow skip, and no-caption passthrough (#32029). - test_voice_command.py filename-uuid contract test updated to point at build_auto_tts_output_path (construction moved there).
130 lines
4.6 KiB
Python
130 lines
4.6 KiB
Python
"""Telegram voice-message captions must render markdown (#32029).
|
|
|
|
``TelegramAdapter.send_voice`` used to pass captions raw with no
|
|
``parse_mode``, so auto-TTS captions carrying the agent's markdown reply
|
|
showed literal *asterisks*, backticks and [links](...). It now formats the
|
|
caption to MarkdownV2 (when the formatted text fits the 1024-char caption
|
|
cap) and falls back to the plain truncated caption when the Bot API rejects
|
|
the entities or formatting overflows.
|
|
"""
|
|
import sys
|
|
from unittest.mock import AsyncMock, MagicMock
|
|
|
|
import pytest
|
|
|
|
from gateway.config import PlatformConfig
|
|
|
|
|
|
def _ensure_telegram_mock():
|
|
if "telegram" in sys.modules and hasattr(sys.modules["telegram"], "__file__"):
|
|
return
|
|
mod = MagicMock()
|
|
mod.error.NetworkError = type("NetworkError", (OSError,), {})
|
|
mod.error.TimedOut = type("TimedOut", (OSError,), {})
|
|
mod.error.BadRequest = type("BadRequest", (Exception,), {})
|
|
for name in ("telegram", "telegram.ext", "telegram.constants", "telegram.request"):
|
|
sys.modules.setdefault(name, mod)
|
|
sys.modules.setdefault("telegram.error", mod.error)
|
|
|
|
|
|
_ensure_telegram_mock()
|
|
|
|
from plugins.platforms.telegram import adapter as telegram_mod # noqa: E402
|
|
from plugins.platforms.telegram.adapter import TelegramAdapter # noqa: E402
|
|
|
|
|
|
def _make_adapter() -> TelegramAdapter:
|
|
adapter = TelegramAdapter(PlatformConfig(enabled=True, token="***"))
|
|
adapter._bot = MagicMock()
|
|
adapter._bot.send_voice = AsyncMock(return_value=MagicMock(message_id=1))
|
|
adapter._bot.send_audio = AsyncMock(return_value=MagicMock(message_id=2))
|
|
return adapter
|
|
|
|
|
|
def _write_ogg(tmp_path):
|
|
audio = tmp_path / "reply.ogg"
|
|
audio.write_bytes(b"\x00" * 16)
|
|
return audio
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_voice_caption_gets_markdown_parse_mode(monkeypatch, tmp_path):
|
|
"""A markdown caption is MarkdownV2-formatted and sent with parse_mode."""
|
|
monkeypatch.setattr(
|
|
telegram_mod, "_probe_voice_duration_seconds", lambda _p: 3
|
|
)
|
|
adapter = _make_adapter()
|
|
|
|
result = await adapter.send_voice(
|
|
"123", str(_write_ogg(tmp_path)), caption="*bold* reply"
|
|
)
|
|
|
|
assert result.success is True
|
|
adapter._bot.send_voice.assert_awaited_once()
|
|
kwargs = adapter._bot.send_voice.await_args.kwargs
|
|
assert kwargs["parse_mode"] is not None
|
|
# format_message converts *bold* to MarkdownV2 bold; the raw caption must
|
|
# have gone through formatting rather than being passed verbatim.
|
|
assert kwargs["caption"] == adapter.format_message("*bold* reply")
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_voice_caption_falls_back_to_plain_on_entity_rejection(
|
|
monkeypatch, tmp_path
|
|
):
|
|
"""A Bot API entity-parse rejection retries with the plain caption."""
|
|
monkeypatch.setattr(
|
|
telegram_mod, "_probe_voice_duration_seconds", lambda _p: 3
|
|
)
|
|
adapter = _make_adapter()
|
|
adapter._bot.send_voice = AsyncMock(
|
|
side_effect=[
|
|
Exception("Bad Request: can't parse entities"),
|
|
MagicMock(message_id=7),
|
|
]
|
|
)
|
|
|
|
result = await adapter.send_voice(
|
|
"123", str(_write_ogg(tmp_path)), caption="*bold* reply"
|
|
)
|
|
|
|
assert result.success is True
|
|
assert adapter._bot.send_voice.await_count == 2
|
|
retry_kwargs = adapter._bot.send_voice.await_args_list[1].kwargs
|
|
assert retry_kwargs["parse_mode"] is None
|
|
assert retry_kwargs["caption"] == "*bold* reply"
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_voice_caption_overflow_skips_formatting(monkeypatch, tmp_path):
|
|
"""When the formatted caption exceeds 1024 UTF-16 units, send plain
|
|
(truncated) — MarkdownV2 escaping inflates length, and a truncated
|
|
formatted caption could cut an entity in half."""
|
|
monkeypatch.setattr(
|
|
telegram_mod, "_probe_voice_duration_seconds", lambda _p: 3
|
|
)
|
|
adapter = _make_adapter()
|
|
# Enough special chars that escaping pushes the formatted text over 1024.
|
|
caption = ("hello. " * 160)[:1020]
|
|
|
|
result = await adapter.send_voice("123", str(_write_ogg(tmp_path)), caption=caption)
|
|
|
|
assert result.success is True
|
|
kwargs = adapter._bot.send_voice.await_args.kwargs
|
|
assert kwargs["parse_mode"] is None
|
|
assert kwargs["caption"] == caption[:1024]
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_voice_without_caption_unchanged(monkeypatch, tmp_path):
|
|
monkeypatch.setattr(
|
|
telegram_mod, "_probe_voice_duration_seconds", lambda _p: 3
|
|
)
|
|
adapter = _make_adapter()
|
|
|
|
result = await adapter.send_voice("123", str(_write_ogg(tmp_path)))
|
|
|
|
assert result.success is True
|
|
kwargs = adapter._bot.send_voice.await_args.kwargs
|
|
assert kwargs["caption"] is None
|
|
assert kwargs["parse_mode"] is None
|