rasa-pro 3.14.0.dev4__py3-none-any.whl → 3.14.0.dev5__py3-none-any.whl

rasa/shared/providers/llm/self_hosted_llm_client.py

@@ -237,7 +237,7 @@ class SelfHostedLLMClient(_BaseLiteLLMClient):
             raise ProviderClientAPIException(e)
 
     async def acompletion(
-        self, messages: Union[List[dict], List[str], str], **kwargs: Any
+        self, messages: Union[List[dict], List[str], str]
     ) -> LLMResponse:
         """Asynchronous completion of the model with the given messages.
 
@@ -255,18 +255,15 @@ class SelfHostedLLMClient(_BaseLiteLLMClient):
                 - a list of messages. Each message is a string and will be formatted
                     as a user message.
                 - a single message as a string which will be formatted as user message.
-            **kwargs: Additional parameters to pass to the completion call.
 
         Returns:
             The completion response.
         """
         if self._use_chat_completions_endpoint:
-            return await super().acompletion(messages, **kwargs)
+            return await super().acompletion(messages)
         return await self._atext_completion(messages)
 
-    def completion(
-        self, messages: Union[List[dict], List[str], str], **kwargs: Any
-    ) -> LLMResponse:
+    def completion(self, messages: Union[List[dict], List[str], str]) -> LLMResponse:
         """Completion of the model with the given messages.
 
         Method overrides the base class method to call the appropriate
@@ -276,13 +273,12 @@ class SelfHostedLLMClient(_BaseLiteLLMClient):
 
         Args:
             messages: The messages to be used for completion.
-            **kwargs: Additional parameters to pass to the completion call.
 
         Returns:
             The completion response.
         """
         if self._use_chat_completions_endpoint:
-            return super().completion(messages, **kwargs)
+            return super().completion(messages)
         return self._text_completion(messages)
 
     @staticmethod

rasa/shared/utils/llm.py

@@ -49,15 +49,7 @@ from rasa.shared.constants import (
     RASA_PATTERN_INTERNAL_ERROR_USER_INPUT_TOO_LONG,
     ROUTER_CONFIG_KEY,
 )
-from rasa.shared.core.events import (
-    AgentCancelled,
-    AgentCompleted,
-    AgentInterrupted,
-    AgentResumed,
-    AgentStarted,
-    BotUttered,
-    UserUttered,
-)
+from rasa.shared.core.events import BotUttered, UserUttered
 from rasa.shared.core.slots import BooleanSlot, CategoricalSlot, Slot
 from rasa.shared.engine.caching import get_local_cache_location
 from rasa.shared.exceptions import (
@@ -120,7 +112,7 @@ DEPLOYMENT_CENTRIC_PROVIDERS = [AZURE_OPENAI_PROVIDER]
 
 # Placeholder messages used in the transcript for
 # instances where user input results in an error
-ERROR_PLACEHOLDER: Dict[str, str] = {
+ERROR_PLACEHOLDER = {
     RASA_PATTERN_INTERNAL_ERROR_USER_INPUT_TOO_LONG: "[User sent really long message]",
     RASA_PATTERN_INTERNAL_ERROR_USER_INPUT_EMPTY: "",
     "default": "[User input triggered an error]",
@@ -233,7 +225,6 @@ def tracker_as_readable_transcript(
     ai_prefix: str = AI,
     max_turns: Optional[int] = 20,
     turns_wrapper: Optional[Callable[[List[str]], List[str]]] = None,
-    highlight_agent_turns: bool = False,
 ) -> str:
     """Creates a readable dialogue from a tracker.
 
@@ -243,7 +234,6 @@ def tracker_as_readable_transcript(
         ai_prefix: the prefix to use for ai utterances
         max_turns: the maximum number of turns to include in the transcript
         turns_wrapper: optional function to wrap the turns in a custom way
-        highlight_agent_turns: whether to highlight agent turns in the transcript
 
     Example:
         >>> tracker = Tracker(
@@ -261,9 +251,7 @@ def tracker_as_readable_transcript(
     Returns:
     A string representing the transcript of the tracker
     """
-    transcript: List[str] = []
-
-    current_ai_prefix = ai_prefix
+    transcript = []
 
     # using `applied_events` rather than `events` means that only events after the
     # most recent `Restart` or `SessionStarted` are included in the transcript
@@ -278,20 +266,9 @@ def tracker_as_readable_transcript(
             else:
                 message = sanitize_message_for_prompt(event.text)
             transcript.append(f"{human_prefix}: {message}")
-        elif isinstance(event, BotUttered):
-            transcript.append(
-                f"{current_ai_prefix}: {sanitize_message_for_prompt(event.text)}"
-            )
 
-        if highlight_agent_turns:
-            if isinstance(event, AgentStarted) or isinstance(event, AgentResumed):
-                current_ai_prefix = event.agent_id
-            elif (
-                isinstance(event, AgentCompleted)
-                or isinstance(event, AgentCancelled)
-                or isinstance(event, AgentInterrupted)
-            ):
-                current_ai_prefix = ai_prefix
+        elif isinstance(event, BotUttered):
+            transcript.append(f"{ai_prefix}: {sanitize_message_for_prompt(event.text)}")
 
     # turns_wrapper to count multiple utterances by bot/user as single turn
     if turns_wrapper:

rasa/shared/utils/schemas/events.py

@@ -160,43 +160,6 @@ FLOW_CANCELLED = {
         "step_id": {"type": "string"},
     }
 }
-AGENT_STARTED = {
-    "properties": {
-        "event": {"const": "agent_started"},
-        "agent_id": {"type": "string"},
-        "flow_id": {"type": "string"},
-    }
-}
-AGENT_COMPLETED = {
-    "properties": {
-        "event": {"const": "agent_completed"},
-        "agent_id": {"type": "string"},
-        "flow_id": {"type": "string"},
-        "status": {"type": "string"},
-    }
-}
-AGENT_INTERRUPTED = {
-    "properties": {
-        "event": {"const": "agent_interrupted"},
-        "agent_id": {"type": "string"},
-        "flow_id": {"type": "string"},
-    }
-}
-AGENT_RESUMED = {
-    "properties": {
-        "event": {"const": "agent_resumed"},
-        "agent_id": {"type": "string"},
-        "flow_id": {"type": "string"},
-    }
-}
-AGENT_CANCELLED = {
-    "properties": {
-        "event": {"const": "agent_cancelled"},
-        "agent_id": {"type": "string"},
-        "flow_id": {"type": "string"},
-        "reason": {"type": "string"},
-    }
-}
 DIALOGUE_STACK_UPDATED = {
     "properties": {"event": {"const": "stack"}, "update": {"type": "string"}}
 }
@@ -241,11 +204,6 @@ EVENT_SCHEMA = {
         FLOW_RESUMED,
         FLOW_COMPLETED,
         FLOW_CANCELLED,
-        AGENT_STARTED,
-        AGENT_COMPLETED,
-        AGENT_INTERRUPTED,
-        AGENT_RESUMED,
-        AGENT_CANCELLED,
         DIALOGUE_STACK_UPDATED,
         ROUTING_SESSION_ENDED,
         SESSION_ENDED,

rasa/shared/utils/yaml.py

@@ -335,9 +335,7 @@ class YamlValidationException(YamlException, ValueError):
                     )
                     if found_lc:
                         return line_offset
-                    if this_line is not None:
-                        return this_line + line_offset
-                    return line_offset
+                    return this_line + line_offset
                 return line
             elif isinstance(current, list) and head.isdigit():
                 return (

rasa/studio/upload.py

@@ -7,7 +7,6 @@ from typing import Any, Dict, Iterable, List, Optional, Set, Text, Tuple, Union
 import questionary
 import requests
 import structlog
-from pydantic import BaseModel, Field
 
 import rasa.cli.telemetry
 import rasa.cli.utils
@@ -24,9 +23,13 @@ from rasa.shared.constants import (
     DEFAULT_DOMAIN_PATHS,
 )
 from rasa.shared.core.domain import Domain
-from rasa.shared.core.flows.yaml_flows_io import YAMLFlowsReader, YamlFlowsWriter
+from rasa.shared.core.flows.yaml_flows_io import YAMLFlowsReader
 from rasa.shared.exceptions import RasaException
-from rasa.shared.importers.importer import FlowSyncImporter, TrainingDataImporter
+from rasa.shared.importers.importer import TrainingDataImporter
+from rasa.shared.importers.utils import (
+    CALMUserData,
+    extract_calm_import_parts_from_importer,
+)
 from rasa.shared.nlu.training_data.formats.rasa_yaml import (
     RasaYAMLReader,
     RasaYAMLWriter,
@@ -34,7 +37,6 @@ from rasa.shared.nlu.training_data.formats.rasa_yaml import (
 from rasa.shared.utils.llm import collect_custom_prompts
 from rasa.shared.utils.yaml import (
     dump_obj_as_yaml_to_string,
-    read_yaml,
     read_yaml_file,
 )
 from rasa.studio import results_logger
@@ -43,6 +45,7 @@ from rasa.studio.config import StudioConfig
 from rasa.studio.results_logger import StudioResult, with_studio_error_handler
 from rasa.studio.utils import validate_argument_paths
 from rasa.telemetry import track_upload_to_studio_failed
+from rasa.utils.json_utils import extract_values
 
 structlogger = structlog.get_logger()
 
@@ -68,16 +71,6 @@ DOMAIN_KEYS = [
 ]
 
 
-class CALMImportParts(BaseModel):
-    """All pieces that will be uploaded to Rasa Studio."""
-
-    flows: Dict[str, Any]
-    domain: Dict[str, Any]
-    config: Dict[str, Any]
-    endpoints: Dict[str, Any]
-    nlu: Dict[str, Any] = Field(default_factory=dict)
-
-
 def _get_selected_entities_and_intents(
     args: argparse.Namespace,
     intents_from_files: Set[Text],
@@ -115,9 +108,10 @@ def run_validation(args: argparse.Namespace) -> None:
     """
     from rasa.validator import Validator
 
+    training_data_paths = args.data if isinstance(args.data, list) else [args.data]
     training_data_importer = TrainingDataImporter.load_from_dict(
         domain_path=args.domain,
-        training_data_paths=[args.data],
+        training_data_paths=training_data_paths,
         config_path=args.config,
         expand_env_vars=False,
     )
@@ -194,11 +188,6 @@ config_keys = [
 ]
 
 
-def extract_values(data: Dict, keys: List[Text]) -> Dict:
-    """Extracts values for given keys from a dictionary."""
-    return {key: data.get(key) for key in keys if data.get(key)}
-
-
 def _get_assistant_name(config: Dict[Text, Any]) -> str:
     config_assistant_id = config.get("assistant_id", "")
     assistant_name = questionary.text(
@@ -237,7 +226,7 @@ def build_calm_import_parts(
     config_path: Text,
     endpoints_path: Optional[Text] = None,
     assistant_name: Optional[Text] = None,
-) -> Tuple[str, CALMImportParts]:
+) -> Tuple[str, CALMUserData]:
     """Builds the parts of the assistant to be uploaded to Studio.
 
     Args:
@@ -250,9 +239,11 @@ def build_calm_import_parts(
     Returns:
         The assistant name and the parts to be uploaded
     """
+    training_data_paths = data_path if isinstance(data_path, list) else [str(data_path)]
     importer = TrainingDataImporter.load_from_dict(
         domain_path=domain_path,
         config_path=config_path,
+        training_data_paths=training_data_paths,
         expand_env_vars=False,
     )
 
@@ -260,33 +251,10 @@ def build_calm_import_parts(
     endpoints = read_yaml_file(endpoints_path, expand_env_vars=False)
     assistant_name = assistant_name or _get_assistant_name(config)
 
-    domain_from_files = importer.get_user_domain().as_dict()
-    domain = extract_values(domain_from_files, DOMAIN_KEYS)
-
-    flow_importer = FlowSyncImporter.load_from_dict(
-        training_data_paths=[str(data_path)], expand_env_vars=False
-    )
-
-    flows = list(flow_importer.get_user_flows())
-    flows_yaml = YamlFlowsWriter().dumps(flows)
-    flows = read_yaml(flows_yaml, expand_env_vars=False)
-
-    nlu_importer = TrainingDataImporter.load_from_dict(
-        training_data_paths=[str(data_path)], expand_env_vars=False
-    )
-    nlu_data = nlu_importer.get_nlu_data()
-    nlu_examples = nlu_data.filter_training_examples(
-        lambda ex: ex.get("intent") in nlu_data.intents
-    )
-    nlu_examples_yaml = RasaYAMLWriter().dumps(nlu_examples)
-    nlu = read_yaml(nlu_examples_yaml, expand_env_vars=False)
-
-    parts = CALMImportParts(
-        flows=flows,
-        domain=domain,
+    parts = extract_calm_import_parts_from_importer(
+        importer=importer,
         config=config,
         endpoints=endpoints,
-        nlu=nlu,
     )
 
     return assistant_name, parts
@@ -349,9 +317,10 @@ def upload_nlu_assistant(
         "rasa.studio.upload.nlu_data_read",
         event_info="Found DM1 assistant data, parsing...",
     )
+    training_data_paths = args.data if isinstance(args.data, list) else [args.data]
     importer = TrainingDataImporter.load_from_dict(
         domain_path=args.domain,
-        training_data_paths=[args.data],
+        training_data_paths=training_data_paths,
         config_path=args.config,
         expand_env_vars=False,
     )

rasa/telemetry.py

@@ -3,7 +3,6 @@ import contextlib
 import hashlib
 import inspect
 import json
-import logging
 import multiprocessing
 import os
 import platform
@@ -70,7 +69,7 @@ if typing.TYPE_CHECKING:
     from rasa.shared.importers.importer import TrainingDataImporter
     from rasa.shared.nlu.training_data.training_data import TrainingData
 
-logger = logging.getLogger(__name__)
+structlogger = structlog.get_logger()
 
 SEGMENT_TRACK_ENDPOINT = "https://api.segment.io/v1/track"
 SEGMENT_IDENTIFY_ENDPOINT = "https://api.segment.io/v1/identify"
@@ -197,6 +196,10 @@ TELEMETRY_E2E_TEST_CONVERSION_EVENT = "E2E Test Conversion Completed"
 E2E_TEST_CONVERSION_FILE_TYPE = "file_type"
 E2E_TEST_CONVERSION_TEST_CASE_COUNT = "test_case_count"
 
+# Copilot telemetry
+TELEMETRY_COPILOT_USER_MESSAGE_EVENT = "copilot_user_message"
+TELEMETRY_COPILOT_BOT_MESSAGE_EVENT = "copilot_bot_message"
+
 
 def print_telemetry_reporting_info() -> None:
     """Print telemetry information to std out."""
@@ -255,7 +258,11 @@ def _is_telemetry_enabled_in_configuration() -> bool:
 
         return stored_config[CONFIG_TELEMETRY_ENABLED]
     except ValueError as e:
-        logger.debug(f"Could not read telemetry settings from configuration file: {e}")
+        structlogger.debug(
+            "telemetry.is_telemetry_enabled_in_configuration.error",
+            error=str(e),
+            event_info="Could not read telemetry settings from configuration file",
+        )
 
         # seems like there is no config, we'll create one and enable telemetry
         success = _write_default_telemetry_configuration()
@@ -272,7 +279,10 @@ def is_telemetry_enabled() -> bool:
     from rasa.utils import licensing
 
     if licensing.is_champion_server_license():
-        logger.debug("Telemetry is enabled for developer licenses.")
+        structlogger.debug(
+            "telemetry.enabled.developer_license",
+            event_info="Telemetry is enabled for developer licenses.",
+        )
         return True
 
     telemetry_environ = os.environ.get(TELEMETRY_ENABLED_ENVIRONMENT_VARIABLE)
@@ -308,9 +318,13 @@ def initialize_telemetry() -> bool:
 
         return telemetry_environ.lower() == "true"
     except Exception as e:  # skipcq:PYL-W0703
-        logger.exception(
-            f"Failed to initialize telemetry reporting: {e}."
-            f"Telemetry reporting will be disabled."
+        structlogger.exception(
+            "telemetry.initialize_telemetry.error",
+            error=str(e),
+            event_info=(
+                "Failed to initialize telemetry reporting. "
+                "Telemetry reporting will be disabled."
+            ),
         )
         return False
 
@@ -481,7 +495,10 @@ def print_telemetry_payload(payload: Dict[Text, Any]) -> None:
         payload: payload to be delivered to segment.
     """
     payload_json = json.dumps(payload, indent=2)
-    logger.debug(f"Telemetry payload: {payload_json}")
+    structlogger.debug(
+        "telemetry.print_telemetry_payload.debug",
+        event_info=f"Telemetry payload: {payload_json}",
+    )
 
 
 def _get_telemetry_write_key() -> Optional[Text]:
@@ -535,10 +552,24 @@ def _send_request(url: Text, payload: Dict[Text, Any]) -> None:
     if not write_key:
         # If RASA_TELEMETRY_WRITE_KEY is empty or `None`, telemetry has not
         # been enabled for this build (e.g. because it is running from source)
-        logger.debug("Skipping request to external service: telemetry key not set.")
+        structlogger.debug(
+            "telemetry.send_request.no_telemetry_key",
+            event_info="Skipping request to external service: telemetry key not set.",
+        )
         return
 
-    headers = rasa.telemetry.segment_request_header(write_key)
+    send_segment_request(url, payload, write_key)
+
+
+def send_segment_request(url: Text, payload: Dict[Text, Any], write_key: Text) -> None:
+    """Send a request to the Segment API.
+
+    Args:
+        url: URL of the Segment API endpoint
+        payload: payload to send to the Segment API
+        write_key: write key for the Segment API
+    """
+    headers = segment_request_header(write_key)
 
     resp = requests.post(
         url=url,
@@ -548,15 +579,22 @@ def _send_request(url: Text, payload: Dict[Text, Any]) -> None:
     )
     # handle different failure cases
     if resp.status_code != 200:
-        logger.debug(
-            f"Segment telemetry request returned a {resp.status_code} response. "
-            f"Body: {resp.text}"
+        structlogger.debug(
+            "telemetry.send_segment_request.error_response",
+            event_info=(
+                f"Segment telemetry request returned a {resp.status_code} "
+                f"response. Body: {resp.text}"
+            ),
         )
     else:
         data = resp.json()
         if not data.get("success"):
-            logger.debug(
-                f"Segment telemetry request returned a failure. Response: {data}"
+            structlogger.debug(
+                "telemetry.send_segment_request.failure",
+                event_info=(
+                    f"Segment telemetry request returned a failure. "
+                    f"Response: {data}"
+                ),
             )
 
 
@@ -609,6 +647,15 @@ def with_default_context_fields(
     return {**_default_context_fields(), **context}
 
 
+def get_deployment_stack() -> Text:
+    """Return the deployment stack.
+
+    Returns:
+        The deployment stack.
+    """
+    return os.environ.get("DEPLOYMENT_STACK", "")
+
+
 def _default_context_fields() -> Dict[Text, Any]:
     """Return a dictionary that contains the default context values.
 
@@ -632,6 +679,7 @@ def _default_context_fields() -> Dict[Text, Any]:
             "cpu": multiprocessing.cpu_count(),
             "docker": _is_docker(),
             "license_hash": get_license_hash(),
+            "deployment_stack": get_deployment_stack(),
             "company": property_of_active_license(
                 lambda active_license: active_license.company
             ),
@@ -663,7 +711,10 @@ def _track(
         telemetry_id = get_telemetry_id()
 
         if not telemetry_id:
-            logger.debug("Will not report telemetry events as no ID was found.")
+            structlogger.debug(
+                "telemetry.track.no_id_found",
+                event_info="Will not report telemetry events as no ID was found.",
+            )
             return
 
         if not properties:
@@ -681,7 +732,11 @@ def _track(
                 with_default_context_fields(context),
             )
     except Exception as e:  # skipcq:PYL-W0703
-        logger.debug(f"Skipping telemetry reporting: {e}")
+        structlogger.debug(
+            "telemetry.track.error",
+            error=str(e),
+            event_info="Skipping telemetry reporting",
+        )
 
 
 def _identify(
@@ -702,7 +757,10 @@ def _identify(
         telemetry_id = get_telemetry_id()
 
         if not telemetry_id:
-            logger.debug("Will not report telemetry events as no ID was found.")
+            structlogger.debug(
+                "telemetry.identify.no_id_found",
+                event_info="Will not report telemetry events as no ID was found.",
+            )
             return
 
         if not traits:
@@ -710,7 +768,11 @@ def _identify(
 
         _send_traits(telemetry_id, traits, with_default_context_fields(context))
     except Exception as e:
-        logger.debug(f"Skipping telemetry reporting: {e}")
+        structlogger.debug(
+            "telemetry.identify.error",
+            error=str(e),
+            event_info="Skipping telemetry reporting",
+        )
 
 
 def _send_traits(
@@ -868,13 +930,16 @@ def strip_sensitive_data_from_sentry_event(
 
 
 @ensure_telemetry_enabled
-def initialize_error_reporting() -> None:
+def initialize_error_reporting(private_mode: bool = True) -> None:
     """Sets up automated error reporting.
 
     Exceptions are reported to sentry. We avoid sending any metadata (local
     variables, paths, ...) to make sure we don't compromise any data. Only the
     exception and its stacktrace is logged and only if the exception origins
     from the `rasa` package.
+
+    Args:
+        private_mode: If True, try to send as little data as possible.
     """
     import sentry_sdk
     from sentry_sdk import configure_scope
@@ -892,11 +957,18 @@ def initialize_error_reporting() -> None:
 
     telemetry_id = get_telemetry_id()
 
+    # in hello rasa we use a different project, so we need to be able
+    # to set the whole url. since we can't change the behavior of sentry in pro
+    # we have two kinds of keys, full urls and jsut the key within the fixed rasa
+    # pro project.
+    if not key.startswith("https://"):
+        key = f"https://{key}.ingest.sentry.io/2801673"
+
     # this is a very defensive configuration, avoiding as many integrations as
     # possible. it also submits very little data (exception with error message
     # and line numbers).
     sentry_sdk.init(
-        f"https://{key}.ingest.sentry.io/2801673",
+        key,
         before_send=before_send,
         integrations=[
             ExcepthookIntegration(),
@@ -916,7 +988,7 @@ def initialize_error_reporting() -> None:
             OSError,
         ],
         in_app_include=["rasa"],  # only submit errors in this package
-        include_local_variables=False,  # don't submit local variables
+        include_local_variables=not private_mode,
         release=f"rasa-{rasa.__version__}",
         default_integrations=False,
         environment="development" if in_continuous_integration() else "production",
@@ -937,6 +1009,7 @@ def initialize_error_reporting() -> None:
                 # os is a nested dict, hence we report it separately
                 scope.set_context("Operating System", default_context.pop("os"))
             scope.set_context("Environment", default_context)
+    structlogger.debug("telemetry.sentry.initialized")
 
 
 @contextlib.contextmanager
@@ -1426,6 +1499,7 @@ def track_shell_started(model_type: Text, assistant_id: Text) -> None:
 
     Args:
         model_type: Type of the model, core / nlu or rasa.
+        assistant_id: ID of the assistant being inspected.
     """
     _track(
         TELEMETRY_SHELL_STARTED_EVENT,
@@ -1997,7 +2071,7 @@ def _extract_stream_pii(event_broker: Optional["EventBroker"]) -> bool:
 def track_privacy_enabled(
     privacy_config: "PrivacyConfig", event_broker: Optional["EventBroker"]
 ) -> None:
-    """Track when PII management capability is enabled"""
+    """Track when PII management capability is enabled."""
     stream_pii = _extract_stream_pii(event_broker)
     privacy_properties = _extract_privacy_enabled_event_properties(
         privacy_config, stream_pii