Dev/1.2.5 #35
Dev/1.2.5 #35
Conversation
… поддержку двухфакторной аутентификации и исправления для работы с сокетами.
…ые функции и исправления. Добавлены пометки для методов аутентификации в enum.
📝 WalkthroughWalkthroughAdds full two-factor authentication flows (password challenges, email codes, recovery hints) and account password management, extends WebSocket/protocol handling (ver=11, seq masking, timeout and connection checks), introduces new payloads and Capability enum, and bumps version to v1.2.5. Changes
Sequence Diagram(s)sequenceDiagram
participant User
participant Client as MaxClient
participant Server as AuthServer
rect rgb(220,235,255)
Note over User,Server: Login request returns passwordChallenge
User->>Client: login_with_code(code)
Client->>Server: send login request
Server-->>Client: { ..., passwordChallenge: {...}, tokenAttrs: {...} }
end
rect rgb(235,220,255)
Note over Client,User: Two‑factor interactive flow
Client->>Server: CreateTrackPayload(type=0)
Server-->>Client: { track_id }
Client->>User: prompt password
User-->>Client: password
Client->>Server: CheckPasswordChallengePayload(track_id, password)
Server-->>Client: tokenAttrs or error
alt tokenAttrs present
Client->>User: optional hint/email prompts
User-->>Client: hint / email / verify_code
Client->>Server: SetHintPayload / RequestEmailCodePayload / SendEmailCodePayload
Client->>Server: SetTwoFactorPayload(expected_capabilities, track_id, password, hint?)
Server-->>Client: 2FA setup success
else error
Client-->>User: invalid password / retry or fail
end
end
rect rgb(220,255,230)
Note over Client: Finalize login
Client->>Client: extract LOGIN token from tokenAttrs or proceed with token
Client-->>User: login successful
end
Estimated code review effort🎯 4 (Complex) | ⏱️ ~50 minutes Possibly related PRs
Poem
Pre-merge checks and finishing touches❌ Failed checks (1 warning, 2 inconclusive)
✨ Finishing touches
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 5
🧹 Nitpick comments (8)
src/pymax/mixins/telemetry.py (1)
106-113: Exception handling looks good; consider addingexc_infofor debugging.The
SocketNotConnectedErrorhandling correctly exits the telemetry loop on disconnect. The genericExceptioncatch is appropriate here to prevent telemetry failures from crashing the client.Consider adding
exc_info=Trueto the warning on line 112 for easier debugging:except Exception: - self.logger.warning("Failed to send random navigation") + self.logger.warning("Failed to send random navigation", exc_info=True)src/pymax/static/constant.py (1)
92-96: Sentinel pattern is well-implemented.Using a dedicated
_Unsetclass is the correct approach for distinguishing between "not provided" andNone. Consider adding__repr__for debugging clarity:class _Unset: def __repr__(self) -> str: return "UNSET"src/pymax/mixins/websocket.py (1)
137-144: Fix exception chaining per static analysis hints.Use
raise ... from Noneto suppress exception chaining noise, and consider usinglogger.warninginstead oflogger.exceptionforTimeoutError(no useful traceback).🔎 Proposed fix
except asyncio.TimeoutError: - self.logger.exception("Send and wait failed (opcode=%s, seq=%s)", opcode, msg["seq"]) - raise RuntimeError("Send and wait failed") + self.logger.warning("Send and wait timed out (opcode=%s, seq=%s)", opcode, msg["seq"]) + raise RuntimeError("Send and wait failed") from None except Exception: self.logger.exception("Send and wait failed (opcode=%s, seq=%s)", opcode, msg["seq"]) - raise RuntimeError("Send and wait failed") + raise RuntimeError("Send and wait failed") from None finally: self._pending.pop(seq_key, None)src/pymax/static/enum.py (1)
226-231: Replace Cyrillic characters with ASCII in the comment.The comment on line 227 contains Cyrillic characters that could cause encoding issues or confusion. Static analysis flagged ambiguous characters (
Вandа). Consider rewriting the comment in English or ensuring it uses only ASCII-compatible Cyrillic if Russian is intentional.Suggested fix
class Capability(int, Enum): - DEFAULT = 0 # В душе не чаю что это такое но при первой установке 2фа там 0 3 4 так что пусть будет дефолт + DEFAULT = 0 # Unknown purpose; observed values 0, 3, 4 during initial 2FA setup ESIA_VERIFIED_FLAG = 1 SECOND_FACTOR_PASSWORD_ENABLED = 2 SECOND_FACTOR_HAS_EMAIL = 3 SECOND_FACTOR_HAS_HINT = 4src/pymax/mixins/auth.py (4)
389-400: Consider propagating errors rather than silently returning None.
_check_passwordreturnsNoneon error without logging or re-raising. This makes it harder to distinguish between "wrong password" and other server errors. Consider logging the specific error for debugging purposes.Suggested improvement
async def _check_password(self, password: str, track_id: str) -> dict[str, Any] | None: payload = CheckPasswordChallengePayload( track_id=track_id, password=password, ).model_dump(by_alias=True) data = await self._send_and_wait(opcode=Opcode.AUTH_LOGIN_CHECK_PASSWORD, payload=payload) token_attrs = data.get("payload", {}).get("tokenAttrs", {}) if data.get("payload", {}).get("error"): + self.logger.debug("Password check failed: %s", data.get("payload", {}).get("error")) return None return token_attrs
433-441: Variable shadowing:payloadis reassigned from request to response.The variable
payloadis used for both the request payload (line 434) and reassigned to the response payload (line 440). This reduces readability and can lead to confusion.Suggested fix
async def _set_password(self, password: str, track_id: str) -> bool: payload = SetPasswordPayload( track_id=track_id, password=password, ).model_dump(by_alias=True) data = await self._send_and_wait(opcode=Opcode.AUTH_VALIDATE_PASSWORD, payload=payload) - payload = data.get("payload", {}) - return not payload + response_payload = data.get("payload", {}) + return not response_payload
443-451: Same variable shadowing issue as_set_password.Suggested fix
async def _set_hint(self, hint: str, track_id: str) -> bool: payload = SetHintPayload( track_id=track_id, hint=hint, ).model_dump(by_alias=True) data = await self._send_and_wait(opcode=Opcode.AUTH_VALIDATE_HINT, payload=payload) - payload = data.get("payload", {}) - return not payload + response_payload = data.get("payload", {}) + return not response_payload
575-593: Variable shadowing:payloadreassigned from request to response.Same issue as in other methods. Also, the
expected_capabilitieslist hardcodesSECOND_FACTOR_HAS_HINTandSECOND_FACTOR_HAS_EMAILeven when hint/email might not have been set successfully.Suggested fix
+ capabilities = [Capability.DEFAULT] + if hint and isinstance(hint, str): + capabilities.append(Capability.SECOND_FACTOR_HAS_HINT) + if email: + capabilities.append(Capability.SECOND_FACTOR_HAS_EMAIL) + payload = SetTwoFactorPayload( - expected_capabilities=[ - Capability.DEFAULT, - Capability.SECOND_FACTOR_HAS_HINT, - Capability.SECOND_FACTOR_HAS_EMAIL, - ], + expected_capabilities=capabilities, track_id=track_id, password=password, hint=hint if isinstance(hint, (str, type(None))) else None, ) data = await self._send_and_wait( opcode=Opcode.AUTH_SET_2FA, payload=payload.model_dump(by_alias=True), ) - payload = data.get("payload", {}) - if payload and payload.get("error"): + response_payload = data.get("payload", {}) + if response_payload and response_payload.get("error"): MixinsUtils.handle_error(data) return True
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (11)
redocs/source/release_notes.rstredocs/source/release_notes_archive.rstsrc/pymax/core.pysrc/pymax/interfaces.pysrc/pymax/mixins/auth.pysrc/pymax/mixins/socket.pysrc/pymax/mixins/telemetry.pysrc/pymax/mixins/websocket.pysrc/pymax/payloads.pysrc/pymax/static/constant.pysrc/pymax/static/enum.py
🧰 Additional context used
🧬 Code graph analysis (6)
src/pymax/mixins/telemetry.py (1)
src/pymax/exceptions.py (2)
Error(60-86)SocketNotConnectedError(23-30)
src/pymax/interfaces.py (1)
src/pymax/exceptions.py (2)
SocketNotConnectedError(23-30)WebSocketNotConnectedError(13-20)
src/pymax/mixins/auth.py (4)
src/pymax/static/enum.py (2)
Capability(226-231)Opcode(4-146)src/pymax/payloads.py (7)
CheckPasswordChallengePayload(370-372)CreateTrackPayload(375-376)RequestEmailCodePayload(396-398)SendEmailCodePayload(401-403)SetHintPayload(384-386)SetPasswordPayload(379-381)SetTwoFactorPayload(389-393)src/pymax/protocols.py (2)
ClientProtocol(32-123)_send_and_wait(95-102)src/pymax/static/constant.py (1)
_Unset(92-93)
src/pymax/mixins/socket.py (2)
src/pymax/interfaces.py (1)
_handle_pending(259-266)src/pymax/exceptions.py (1)
SocketSendError(33-39)
src/pymax/core.py (1)
src/pymax/mixins/auth.py (1)
_two_factor_auth(402-431)
src/pymax/payloads.py (1)
src/pymax/static/enum.py (1)
Capability(226-231)
🪛 Ruff (0.14.10)
src/pymax/mixins/telemetry.py
111-111: Do not catch blind exception: Exception
(BLE001)
src/pymax/interfaces.py
213-213: Do not catch blind exception: Exception
(BLE001)
src/pymax/mixins/auth.py
408-408: Avoid specifying long messages outside the exception class
(TRY003)
430-430: Avoid specifying long messages outside the exception class
(TRY003)
527-527: Avoid specifying long messages outside the exception class
(TRY003)
src/pymax/mixins/websocket.py
139-139: Within an except clause, raise exceptions with raise ... from err or raise ... from None to distinguish them from errors in exception handling
(B904)
139-139: Avoid specifying long messages outside the exception class
(TRY003)
142-142: Within an except clause, raise exceptions with raise ... from err or raise ... from None to distinguish them from errors in exception handling
(B904)
142-142: Avoid specifying long messages outside the exception class
(TRY003)
src/pymax/static/enum.py
227-227: Comment contains ambiguous В (CYRILLIC CAPITAL LETTER VE). Did you mean B (LATIN CAPITAL LETTER B)?
(RUF003)
227-227: Comment contains ambiguous а (CYRILLIC SMALL LETTER A). Did you mean a (LATIN SMALL LETTER A)?
(RUF003)
🔇 Additional comments (15)
src/pymax/interfaces.py (3)
191-191: Protocol version bump looks correct.The
ver=11aligns with the WebSocket protocol improvements mentioned in the release notes.
210-214: Socket disconnect handling in ping loop is correct.Breaking out of the loop on
SocketNotConnectedErrorand logging at debug level is appropriate. Removingexc_infofrom the generic exception warning is fine since the message is self-explanatory.
289-296: Potential logic inversion — verify notification behavior.The early return skips sending
NOTIF_MESSAGEwhen the socket is connected. This appears inverted: typically you'd want to send read notifications when connected, not skip them. If intentional (e.g., socket clients handle notifications differently), please add a clarifying comment.async def _send_notification_response(self, chat_id: int, message_id: str) -> None: - if self._socket is not None and self.is_connected: - return + # Socket clients handle notifications via a different path + if self._socket is not None and self.is_connected: + return await self._send_and_wait(src/pymax/core.py (1)
137-137: Sequence initialization at 255 is correct for 1-byte wrapping.With
seq % 256used in packing, starting at 255 means the first message increments to 256, which wraps to 0. This aligns with the 8-bit sequence handling in socket.py.src/pymax/mixins/websocket.py (1)
114-120: Good improvement: seq_key and pending future cancellation.Using
seq_keyfor consistent cleanup and canceling stale pending futures prevents memory leaks when sequence numbers wrap around.src/pymax/mixins/socket.py (4)
87-87: Correct 1-byte sequence packing.Masking
seq % 256ensures the sequence fits in the 1-byte wire format.
220-223: Sequence normalization in recv loop aligns with send path.Using
seq % 256(withNoneguard) ensures pending lookups match the keys used when registering futures.
252-258: Future cancellation prevents stale pending entries.Canceling an existing future before registering a new one for the same
seq_keyis essential for correctness when sequences wrap.
291-299: Timeout and cleanup handling looks correct.Using
SocketSendErrorwith proper exception chaining (from None/from exc) and cleaning up withmsg["seq"] % 256maintains consistency.redocs/source/release_notes.rst (1)
1-61: Release notes accurately document the changes.The documentation covers all key changes: 2FA support,
set_passwordmethod, sequence number fixes, and protocol version bump. The new payload classes andCapabilityenum are properly listed.src/pymax/payloads.py (1)
370-403: New 2FA payload classes are well-structured.The payload models follow the existing
CamelModelpattern consistently.SetTwoFactorPayloadcorrectly useslist[Capability]for type safety.src/pymax/static/enum.py (1)
5-146: LGTM on Opcode additions.The new opcode values and inline checkmark annotations are consistent with the existing pattern. The new opcodes for 2FA (AUTH_VALIDATE_PASSWORD, AUTH_LOGIN_CHECK_PASSWORD, etc.) align with the authentication features being added in this PR.
src/pymax/mixins/auth.py (2)
10-26: LGTM on imports.The new imports align with the 2FA functionality being added and are properly organized.
209-223: LGTM on login flow enhancement.The conditional handling for password challenges vs direct token extraction is clear and properly structured for 2FA support.
redocs/source/release_notes_archive.rst (1)
1-141: Documentation structure is correct—v1.2.5 release notes already exist.The archive file properly contains only previous versions (v1.2.3 and v1.2.4). The v1.2.5 release notes are already documented in
redocs/source/release_notes.rstwith comprehensive coverage of the new 2FA functionality, includingset_password(),Capabilityenum, and all related payload classes. No additional documentation is needed.
| async def _two_factor_auth(self, password_challenge: dict[str, Any]) -> None: | ||
| self.logger.info("Starting two-factor authentication flow") | ||
|
|
||
| track_id = password_challenge.get("trackId") | ||
| if not track_id: | ||
| self.logger.critical("Password challenge missing track ID") | ||
| raise ValueError("Password challenge missing track ID") | ||
|
|
||
| hint = password_challenge.get("hint", "No hint provided") | ||
|
|
||
| while True: | ||
| password = await asyncio.to_thread( | ||
| lambda: input(f"Введите пароль (Подсказка: {hint}): ").strip() | ||
| ) | ||
| if not password: | ||
| self.logger.warning("Password is empty, please try again") | ||
| continue | ||
|
|
||
| token_attrs = await self._check_password(password, track_id) | ||
| if not token_attrs: | ||
| self.logger.error("Incorrect password, please try again") | ||
| continue | ||
|
|
||
| login_attrs = token_attrs.get("LOGIN", {}) | ||
| if login_attrs: | ||
| token = login_attrs.get("token") | ||
| if not token: | ||
| self.logger.critical("Login response did not contain tokenAttrs.LOGIN.token") | ||
| raise ValueError("Login response did not contain tokenAttrs.LOGIN.token") | ||
| return token |
There was a problem hiding this comment.
Return type annotation mismatch: method returns str but is annotated as None.
The method is annotated to return None but actually returns a token string on line 431. This will cause type checking issues.
Proposed fix
- async def _two_factor_auth(self, password_challenge: dict[str, Any]) -> None:
+ async def _two_factor_auth(self, password_challenge: dict[str, Any]) -> str:
self.logger.info("Starting two-factor authentication flow")
# ... rest of method🧰 Tools
🪛 Ruff (0.14.10)
408-408: Avoid specifying long messages outside the exception class
(TRY003)
430-430: Avoid specifying long messages outside the exception class
(TRY003)
🤖 Prompt for AI Agents
In src/pymax/mixins/auth.py around lines 402 to 431, the _two_factor_auth method
is annotated to return None but actually returns a token string; change the
return type annotation to str (or Optional[str] if you prefer a nullable
contract) on the method signature, update any necessary typing imports, and
ensure any callers/types are compatible with the new return type.
| opcode=Opcode.AUTH_CREATE_TRACK, | ||
| payload=payload, | ||
| ) | ||
| print(data) |
There was a problem hiding this comment.
Remove debug print statement.
This print(data) statement appears to be debug code that should be removed before merging.
Proposed fix
- print(data)
if data.get("payload", {}).get("error"):📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
| print(data) | |
| if data.get("payload", {}).get("error"): |
🤖 Prompt for AI Agents
In src/pymax/mixins/auth.py around line 520, remove the leftover debug statement
`print(data)` so no debug output is emitted; simply delete that line (or replace
it with appropriate logging at debug level if needed) and ensure surrounding
code and imports remain unchanged.
| while True: | ||
| if not password: | ||
| password = await asyncio.to_thread(lambda: input("Введите пароль: ").strip()) | ||
| if not password: | ||
| self.logger.warning("Password is empty, please try again") | ||
| continue | ||
|
|
||
| success = await self._set_password(password, track_id) | ||
| if success: | ||
| self.logger.info("Password set successfully") | ||
| break | ||
| else: | ||
| self.logger.error("Failed to set password, please try again") |
There was a problem hiding this comment.
Potential infinite loop if password validation repeatedly fails.
If _set_password keeps returning False (e.g., due to server-side validation rules the user can't satisfy), the loop never exits. Consider adding a retry limit or allowing the user to abort.
Suggested fix with retry limit
+ max_retries = 5
+ attempts = 0
while True:
+ if attempts >= max_retries:
+ self.logger.error("Maximum password set attempts exceeded")
+ raise ValueError("Failed to set password after maximum retries")
if not password:
password = await asyncio.to_thread(lambda: input("Введите пароль: ").strip())
if not password:
self.logger.warning("Password is empty, please try again")
continue
success = await self._set_password(password, track_id)
if success:
self.logger.info("Password set successfully")
break
else:
+ attempts += 1
+ password = "" # Reset to prompt again
self.logger.error("Failed to set password, please try again")📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
| while True: | |
| if not password: | |
| password = await asyncio.to_thread(lambda: input("Введите пароль: ").strip()) | |
| if not password: | |
| self.logger.warning("Password is empty, please try again") | |
| continue | |
| success = await self._set_password(password, track_id) | |
| if success: | |
| self.logger.info("Password set successfully") | |
| break | |
| else: | |
| self.logger.error("Failed to set password, please try again") | |
| max_retries = 5 | |
| attempts = 0 | |
| while True: | |
| if attempts >= max_retries: | |
| self.logger.error("Maximum password set attempts exceeded") | |
| raise ValueError("Failed to set password after maximum retries") | |
| if not password: | |
| password = await asyncio.to_thread(lambda: input("Введите пароль: ").strip()) | |
| if not password: | |
| self.logger.warning("Password is empty, please try again") | |
| continue | |
| success = await self._set_password(password, track_id) | |
| if success: | |
| self.logger.info("Password set successfully") | |
| break | |
| else: | |
| attempts += 1 | |
| password = "" # Reset to prompt again | |
| self.logger.error("Failed to set password, please try again") |
🤖 Prompt for AI Agents
In src/pymax/mixins/auth.py around lines 529 to 541, the password prompt loop
can become infinite if _set_password keeps returning False; add a retry limit
and allow user abort: introduce a max_retries counter (e.g., 3-5), increment it
on each failed attempt from _set_password, and after exceeding the limit log an
error and break/raise so the flow can continue or surface the failure;
additionally, when prompting for the password accept a special abort token
(e.g., empty input interpreted after first prompt or a 'q' input) to let the
user cancel immediately and handle that case by logging/wrapping up instead of
retrying indefinitely.
| while True: | ||
| if not email: | ||
| email = await asyncio.to_thread( | ||
| lambda: input("Введите email для восстановления пароля: ").strip() | ||
| ) | ||
| if not email: | ||
| self.logger.warning("Email is empty, please try again") | ||
| continue | ||
|
|
||
| success = await self._set_email(email, track_id) | ||
| if success: | ||
| self.logger.info("Recovery email set successfully") | ||
| break |
There was a problem hiding this comment.
Potential infinite loop if email validation repeatedly fails.
Similar to the password loop, if _set_email keeps failing, there's no exit condition. Additionally, if the user provides an empty email and email parameter was not provided, the loop continues indefinitely asking for email with no way to skip.
Suggested fix allowing user to skip email
while True:
if not email:
email = await asyncio.to_thread(
- lambda: input("Введите email для восстановления пароля: ").strip()
+ lambda: input("Введите email для восстановления пароля (пустой - пропустить): ").strip()
)
if not email:
- self.logger.warning("Email is empty, please try again")
- continue
+ self.logger.info("Email recovery skipped by user")
+ break
success = await self._set_email(email, track_id)
if success:
self.logger.info("Recovery email set successfully")
break
+ else:
+ email = "" # Reset to prompt again📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
| while True: | |
| if not email: | |
| email = await asyncio.to_thread( | |
| lambda: input("Введите email для восстановления пароля: ").strip() | |
| ) | |
| if not email: | |
| self.logger.warning("Email is empty, please try again") | |
| continue | |
| success = await self._set_email(email, track_id) | |
| if success: | |
| self.logger.info("Recovery email set successfully") | |
| break | |
| while True: | |
| if not email: | |
| email = await asyncio.to_thread( | |
| lambda: input("Введите email для восстановления пароля (пустой - пропустить): ").strip() | |
| ) | |
| if not email: | |
| self.logger.info("Email recovery skipped by user") | |
| break | |
| success = await self._set_email(email, track_id) | |
| if success: | |
| self.logger.info("Recovery email set successfully") | |
| break | |
| else: | |
| email = "" # Reset to prompt again |
🤖 Prompt for AI Agents
In src/pymax/mixins/auth.py around lines 561 to 573, the while loop that prompts
for an email can spin indefinitely if _set_email repeatedly fails or the user
keeps submitting empty input; update the loop to enforce a bounded number of
attempts (e.g., max_retries), accept an explicit skip option (e.g., empty input
or 'skip') to exit gracefully when no email is desired, and on repeated
_set_email failures return or raise a clear error after max_retries instead of
looping forever; also log the reason when aborting so callers can act
accordingly.
There was a problem hiding this comment.
Actionable comments posted: 0
🧹 Nitpick comments (1)
src/pymax/static/enum.py (1)
226-231: Consider revising the comment to use only Latin characters.The new
Capabilityenum is well-structured for 2FA features. However, the comment on line 227 contains ambiguous Cyrillic characters (В, а) that visually resemble Latin characters (B, a), which can cause confusion.🔎 Proposed fix
- DEFAULT = 0 # В душе не чаю что это такое но при первой установке 2фа там 0 3 4 так что пусть будет дефолт + DEFAULT = 0 # Default capability value (observed as 0, 3, 4 during initial 2FA setup)
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (3)
pyproject.tomlsrc/pymax/core.pysrc/pymax/static/enum.py
✅ Files skipped from review due to trivial changes (1)
- pyproject.toml
🧰 Additional context used
🧬 Code graph analysis (1)
src/pymax/core.py (1)
src/pymax/mixins/auth.py (1)
_two_factor_auth(402-431)
🪛 Ruff (0.14.10)
src/pymax/static/enum.py
227-227: Comment contains ambiguous В (CYRILLIC CAPITAL LETTER VE). Did you mean B (LATIN CAPITAL LETTER B)?
(RUF003)
227-227: Comment contains ambiguous а (CYRILLIC SMALL LETTER A). Did you mean a (LATIN SMALL LETTER A)?
(RUF003)
🔇 Additional comments (1)
src/pymax/core.py (1)
255-262: LGTM! The token extraction bug has been fixed.The previous issue where line 262 called
.get("LOGIN", {})again onlogin_attrs(which was already the LOGIN dict) has been correctly resolved. The code now properly extracts the token directly fromlogin_attrs.get("token").The new 2FA flow logic is sound:
- If a
passwordChallengeis present andlogin_attrsis empty, the code triggers two-factor authentication.- Otherwise, it retrieves the token directly from
login_attrs.
Описание
Кратко, что делает этот PR. Например, добавляет новый метод, исправляет баг, улучшает документацию.
Новые методы
Измененные методы
И так же кучу системных методов поменял + фикс сокета частичный
Тип изменений
Связанные задачи / Issue
Ссылка на issue, если есть: #
Тестирование
Покажите пример кода, который проверяет изменения:
Summary by CodeRabbit
New Features
Bug Fixes
Other
✏️ Tip: You can customize this high-level summary in your review settings.