mirrors/hermes-agent
1
0
Fork 0
mirror of https://github.com/NousResearch/hermes-agent.git synced 2026-07-31 19:16:29 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 2

fix(telegram): render markdown in voice-message captions

Browse source 
Closes the #32029 gap: TelegramAdapter.send_voice passed captions raw with
no parse_mode, so auto-TTS captions (which carry the agent's markdown
reply) showed literal *asterisks*, backticks and [links](...). Captions
are now formatted to MarkdownV2 via the adapter's format_message when the
formatted text fits Telegram's 1024-char caption cap, with fallback to the
plain truncated caption when formatting overflows or the Bot API rejects
the entities (mirrors the text-send markdown fallback ladder).

Fixes #32029
This commit is contained in:
Teknium 2026-07-28 09:38:45 -07:00
parent 1753369f7c
commit ee15e04803
1 changed files with 66 additions and 16 deletions
Download patch file Download diff file Expand all files Collapse all files

82
plugins/platforms/telegram/adapter.py
View file

@ -6742,6 +6742,29 @@ class TelegramAdapter(BasePlatformAdapter):
_probe_voice_duration_seconds, audio_path
)
# Render caption markdown (#32029): auto-TTS captions carry the
# agent's markdown reply, which showed literal *asterisks* and
# [links](...) without a parse_mode. Format to MarkdownV2 when it
# fits the 1024-char caption cap; fall back to the raw text
# (previous behaviour) when formatting would overflow or the
# Bot API rejects the entities.
_caption_variants: List[tuple] = []
if caption:
try:
_formatted_caption = self.format_message(caption)
if utf16_len(_formatted_caption) <= 1024:
_caption_variants.append(
(_formatted_caption, ParseMode.MARKDOWN_V2)
)
except Exception:
logger.debug(
"[%s] voice caption MarkdownV2 formatting failed; "
"sending plain caption", self.name, exc_info=True,
)
_caption_variants.append((caption[:1024], None))
else:
_caption_variants.append((None, None))
with open(audio_path, "rb") as audio_file:
ext = os.path.splitext(audio_path)[1].lower()
# .ogg / .opus files -> send as voice (round playable bubble)
@ -6755,22 +6778,49 @@ class TelegramAdapter(BasePlatformAdapter):
reply_to_message_id=reply_to_id,
reply_to_mode=self._reply_to_mode
)
msg = await self._send_with_dm_topic_reply_anchor_retry(
self._bot.send_voice,
{
"chat_id": normalize_telegram_chat_id(chat_id),
"voice": audio_file,
"caption": caption[:1024] if caption else None,
"reply_to_message_id": reply_to_id,
"duration": _duration_secs,
**voice_thread_kwargs,
**self._notification_kwargs(metadata),
},
metadata,
reply_to_id,
"voice",
reset_media=lambda: audio_file.seek(0),
)
msg = None
_last_parse_error: Optional[Exception] = None
for _cap_text, _cap_parse_mode in _caption_variants:
try:
msg = await self._send_with_dm_topic_reply_anchor_retry(
self._bot.send_voice,
{
"chat_id": normalize_telegram_chat_id(chat_id),
"voice": audio_file,
"caption": _cap_text,
"parse_mode": _cap_parse_mode,
"reply_to_message_id": reply_to_id,
"duration": _duration_secs,
**voice_thread_kwargs,
**self._notification_kwargs(metadata),
},
metadata,
reply_to_id,
"voice",
reset_media=lambda: audio_file.seek(0),
)
break
except Exception as _cap_error:
# Only retry the next (plain) variant on entity
# parse failures; anything else is a real send
# error for the outer handler.
if (_cap_parse_mode is not None
and ("parse" in str(_cap_error).lower()
or "entit" in str(_cap_error).lower())):
logger.warning(
"[%s] voice caption MarkdownV2 rejected, "
"retrying plain: %s",
self.name,
_redact_telegram_error_text(_cap_error),
)
_last_parse_error = _cap_error
audio_file.seek(0)
continue
raise
if msg is None:
raise _last_parse_error or RuntimeError(
"Telegram send_voice failed for all caption variants"
)
elif ext in {".mp3", ".m4a"}:
# Telegram's Bot API sendAudio only accepts MP3 / M4A.
_audio_thread = self._metadata_thread_id(metadata)