fix(wecom): bound req_id cache, revert undocumented is_group change, add tests

Follow-up to the cherry-picked contributor fix:

- Extract `_remember_chat_req_id()` and bound it at DEDUP_MAX_SIZE like
  `_reply_req_ids` — the unbounded dict would grow forever on a long-
  running gateway with many chats.
- Move the cache write to AFTER the group/DM policy check so we don't
  cache req_ids from blocked senders.
- Revert the undocumented `is_group` change: the contributor flipped
  `chattype == 'group'` to `bool(chatid)`, which wasn't mentioned in
  the PR description and weakens the signal (chattype is the explicit
  hint; relying on chatid presence assumes DMs never carry it). Keep
  the original check.
- Drop the defensive `getattr(self, '_last_chat_req_ids', {})` reads
  at both send sites — the attribute is initialized in __init__.
- Update `test_send_uses_passive_reply_stream_...` → `_markdown_...`
  to match the new msgtype, and add a new TestWeComZombieSessionFix
  class covering device_id presence in subscribe, per-chat req_id
  caching + bounding, blocked-sender cache exclusion, and the group
  APP_CMD_RESPONSE fallback path.
This commit is contained in:
Teknium
2026-04-17 18:59:43 -07:00
committed by Teknium
parent 2992802b35
commit 9b14b76eb3
2 changed files with 220 additions and 9 deletions

View File

@@ -492,10 +492,8 @@ class WeComAdapter(BasePlatformAdapter):
if not chat_id:
logger.debug("[%s] Missing chat id, skipping message", self.name)
return
self._last_chat_req_ids[chat_id] = self._payload_req_id(payload)
is_group = bool(body.get("chatid"))
is_group = str(body.get("chattype") or "").lower() == "group"
if is_group:
if not self._is_group_allowed(chat_id, sender_id):
logger.debug("[%s] Group %s / sender %s blocked by policy", self.name, chat_id, sender_id)
@@ -504,6 +502,11 @@ class WeComAdapter(BasePlatformAdapter):
logger.debug("[%s] DM sender %s blocked by policy", self.name, sender_id)
return
# Cache the inbound req_id after policy checks so proactive sends to
# this chat can fall back to APP_CMD_RESPONSE (required for groups —
# WeCom AI Bots cannot initiate APP_CMD_SEND in group chats).
self._remember_chat_req_id(chat_id, self._payload_req_id(payload))
text, reply_text = self._extract_text(body)
media_urls, media_types = await self._extract_media(body)
message_type = self._derive_message_type(body, text, media_types)
@@ -855,6 +858,23 @@ class WeComAdapter(BasePlatformAdapter):
while len(self._reply_req_ids) > DEDUP_MAX_SIZE:
self._reply_req_ids.pop(next(iter(self._reply_req_ids)))
def _remember_chat_req_id(self, chat_id: str, req_id: str) -> None:
"""Cache the most recent inbound req_id per chat.
Used as a fallback reply target when we need to send into a group
without an explicit ``reply_to`` — WeCom AI Bots are blocked from
APP_CMD_SEND in groups and must use APP_CMD_RESPONSE bound to some
prior req_id. Bounded like _reply_req_ids so long-running gateways
don't leak memory across many chats.
"""
normalized_chat_id = str(chat_id or "").strip()
normalized_req_id = str(req_id or "").strip()
if not normalized_chat_id or not normalized_req_id:
return
self._last_chat_req_ids[normalized_chat_id] = normalized_req_id
while len(self._last_chat_req_ids) > DEDUP_MAX_SIZE:
self._last_chat_req_ids.pop(next(iter(self._last_chat_req_ids)))
def _reply_req_id_for_message(self, reply_to: Optional[str]) -> Optional[str]:
normalized = str(reply_to or "").strip()
if not normalized or normalized.startswith("quote:"):
@@ -1239,7 +1259,7 @@ class WeComAdapter(BasePlatformAdapter):
return SendResult(success=False, error=prepared["reject_reason"])
reply_req_id = self._reply_req_id_for_message(reply_to)
if not reply_req_id and chat_id in getattr(self, '_last_chat_req_ids', {}):
if not reply_req_id and chat_id in self._last_chat_req_ids:
reply_req_id = self._last_chat_req_ids[chat_id]
try:
@@ -1310,7 +1330,7 @@ class WeComAdapter(BasePlatformAdapter):
try:
reply_req_id = self._reply_req_id_for_message(reply_to)
if not reply_req_id and chat_id in getattr(self, '_last_chat_req_ids', {}):
if not reply_req_id and chat_id in self._last_chat_req_ids:
reply_req_id = self._last_chat_req_ids[chat_id]
if reply_req_id: