rasa-pro 3.13.0.dev7__py3-none-any.whl → 3.13.0.dev9__py3-none-any.whl

rasa/tracing/metric_instrument_provider.py

@@ -10,7 +10,10 @@ from rasa.tracing.constants import (
     COMPACT_LLM_COMMAND_GENERATOR_PROMPT_TOKEN_USAGE_METRIC_NAME,
     CONTEXTUAL_RESPONSE_REPHRASER_LLM_RESPONSE_DURATION_METRIC_NAME,
     DURATION_UNIT_NAME,
+    ENTERPRISE_SEARCH_POLICY_CPU_USAGE_METRIC_NAME,
     ENTERPRISE_SEARCH_POLICY_LLM_RESPONSE_DURATION_METRIC_NAME,
+    ENTERPRISE_SEARCH_POLICY_MEMORY_USAGE_METRIC_NAME,
+    ENTERPRISE_SEARCH_POLICY_PROMPT_TOKEN_USAGE_METRIC_NAME,
     INTENTLESS_POLICY_LLM_RESPONSE_DURATION_METRIC_NAME,
     LLM_BASED_COMMAND_GENERATOR_CPU_MEMORY_USAGE_UNIT_NAME,
     LLM_COMMAND_GENERATOR_CPU_USAGE_METRIC_NAME,
@@ -23,6 +26,10 @@ from rasa.tracing.constants import (
     MULTI_STEP_LLM_COMMAND_GENERATOR_PROMPT_TOKEN_USAGE_METRIC_NAME,
     RASA_CLIENT_REQUEST_BODY_SIZE_METRIC_NAME,
     RASA_CLIENT_REQUEST_DURATION_METRIC_NAME,
+    SEARCH_READY_LLM_COMMAND_GENERATOR_CPU_USAGE_METRIC_NAME,
+    SEARCH_READY_LLM_COMMAND_GENERATOR_LLM_RESPONSE_DURATION_METRIC_NAME,
+    SEARCH_READY_LLM_COMMAND_GENERATOR_MEMORY_USAGE_METRIC_NAME,
+    SEARCH_READY_LLM_COMMAND_GENERATOR_PROMPT_TOKEN_USAGE_METRIC_NAME,
     SINGLE_STEP_LLM_COMMAND_GENERATOR_CPU_USAGE_METRIC_NAME,
     SINGLE_STEP_LLM_COMMAND_GENERATOR_LLM_RESPONSE_DURATION_METRIC_NAME,
     SINGLE_STEP_LLM_COMMAND_GENERATOR_MEMORY_USAGE_METRIC_NAME,
@@ -48,7 +55,9 @@ class MetricInstrumentProvider(metaclass=Singleton):
             **self._create_llm_command_generator_instruments(meter),
             **self._create_single_step_llm_command_generator_instruments(meter),
             **self._create_compact_llm_command_generator_instruments(meter),
+            **self._create_search_ready_llm_command_generator_instruments(meter),
             **self._create_multi_step_llm_command_generator_instruments(meter),
+            **self._create_enterprise_search_policy_instruments(meter),
             **self._create_llm_response_duration_instruments(meter),
             **self._create_client_request_instruments(meter),
         }
@@ -162,6 +171,41 @@ class MetricInstrumentProvider(metaclass=Singleton):
             COMPACT_LLM_COMMAND_GENERATOR_LLM_RESPONSE_DURATION_METRIC_NAME: compact_llm_response_duration_llm_command_generator,  # noqa: E501
         }
 
+    @staticmethod
+    def _create_search_ready_llm_command_generator_instruments(
+        meter: Meter,
+    ) -> Dict[str, Any]:
+        search_ready_llm_command_generator_cpu_usage = meter.create_histogram(
+            name=SEARCH_READY_LLM_COMMAND_GENERATOR_CPU_USAGE_METRIC_NAME,
+            description="CPU percentage for SearchReadyLLMCommandGenerator",
+            unit=LLM_BASED_COMMAND_GENERATOR_CPU_MEMORY_USAGE_UNIT_NAME,
+        )
+
+        search_ready_llm_command_generator_memory_usage = meter.create_histogram(
+            name=SEARCH_READY_LLM_COMMAND_GENERATOR_MEMORY_USAGE_METRIC_NAME,
+            description="RAM memory usage for SearchReadyLLMCommandGenerator",
+            unit=LLM_BASED_COMMAND_GENERATOR_CPU_MEMORY_USAGE_UNIT_NAME,
+        )
+
+        search_ready_llm_command_generator_prompt_token_usage = meter.create_histogram(
+            name=SEARCH_READY_LLM_COMMAND_GENERATOR_PROMPT_TOKEN_USAGE_METRIC_NAME,
+            description="SearchReadyLLMCommandGenerator prompt token length",
+            unit="1",
+        )
+
+        search_ready_llm_response_duration_command_generator = meter.create_histogram(
+            name=SEARCH_READY_LLM_COMMAND_GENERATOR_LLM_RESPONSE_DURATION_METRIC_NAME,
+            description="The duration of SearchReadyLLMCommandGenerator's LLM call",
+            unit=DURATION_UNIT_NAME,
+        )
+
+        return {
+            SEARCH_READY_LLM_COMMAND_GENERATOR_CPU_USAGE_METRIC_NAME: search_ready_llm_command_generator_cpu_usage,  # noqa: E501
+            SEARCH_READY_LLM_COMMAND_GENERATOR_MEMORY_USAGE_METRIC_NAME: search_ready_llm_command_generator_memory_usage,  # noqa: E501
+            SEARCH_READY_LLM_COMMAND_GENERATOR_PROMPT_TOKEN_USAGE_METRIC_NAME: search_ready_llm_command_generator_prompt_token_usage,  # noqa: E501
+            SEARCH_READY_LLM_COMMAND_GENERATOR_LLM_RESPONSE_DURATION_METRIC_NAME: search_ready_llm_response_duration_command_generator,  # noqa: E501
+        }
+
     @staticmethod
     def _create_multi_step_llm_command_generator_instruments(
         meter: Meter,
@@ -198,13 +242,42 @@ class MetricInstrumentProvider(metaclass=Singleton):
         }
 
     @staticmethod
-    def _create_llm_response_duration_instruments(meter: Meter) -> Dict[str, Any]:
-        llm_response_duration_enterprise_search = meter.create_histogram(
+    def _create_enterprise_search_policy_instruments(
+        meter: Meter,
+    ) -> Dict[str, Any]:
+        enterprise_search_policy_cpu_usage = meter.create_histogram(
+            name=ENTERPRISE_SEARCH_POLICY_CPU_USAGE_METRIC_NAME,
+            description="CPU percentage for EnterpriseSearchPolicy",
+            unit=LLM_BASED_COMMAND_GENERATOR_CPU_MEMORY_USAGE_UNIT_NAME,
+        )
+
+        enterprise_search_policy_memory_usage = meter.create_histogram(
+            name=ENTERPRISE_SEARCH_POLICY_MEMORY_USAGE_METRIC_NAME,
+            description="RAM memory usage for EnterpriseSearchPolicy",
+            unit=LLM_BASED_COMMAND_GENERATOR_CPU_MEMORY_USAGE_UNIT_NAME,
+        )
+
+        enterprise_search_policy_prompt_token_usage = meter.create_histogram(
+            name=ENTERPRISE_SEARCH_POLICY_PROMPT_TOKEN_USAGE_METRIC_NAME,
+            description="EnterpriseSearchPolicy prompt token length",
+            unit="1",
+        )
+
+        enterprise_search_policy_llm_response_duration = meter.create_histogram(
             name=ENTERPRISE_SEARCH_POLICY_LLM_RESPONSE_DURATION_METRIC_NAME,
             description="The duration of EnterpriseSearchPolicy's LLM call",
             unit=DURATION_UNIT_NAME,
         )
 
+        return {
+            ENTERPRISE_SEARCH_POLICY_CPU_USAGE_METRIC_NAME: enterprise_search_policy_cpu_usage,  # noqa: E501
+            ENTERPRISE_SEARCH_POLICY_MEMORY_USAGE_METRIC_NAME: enterprise_search_policy_memory_usage,  # noqa: E501
+            ENTERPRISE_SEARCH_POLICY_PROMPT_TOKEN_USAGE_METRIC_NAME: enterprise_search_policy_prompt_token_usage,  # noqa: E501
+            ENTERPRISE_SEARCH_POLICY_LLM_RESPONSE_DURATION_METRIC_NAME: enterprise_search_policy_llm_response_duration,  # noqa: E501
+        }
+
+    @staticmethod
+    def _create_llm_response_duration_instruments(meter: Meter) -> Dict[str, Any]:
         llm_response_duration_intentless = meter.create_histogram(
             name=INTENTLESS_POLICY_LLM_RESPONSE_DURATION_METRIC_NAME,
             description="The duration of IntentlessPolicy's LLM call",
@@ -218,7 +291,6 @@ class MetricInstrumentProvider(metaclass=Singleton):
         )
 
         return {
-            ENTERPRISE_SEARCH_POLICY_LLM_RESPONSE_DURATION_METRIC_NAME: llm_response_duration_enterprise_search,  # noqa: E501
             INTENTLESS_POLICY_LLM_RESPONSE_DURATION_METRIC_NAME: llm_response_duration_intentless,  # noqa: E501
             CONTEXTUAL_RESPONSE_REPHRASER_LLM_RESPONSE_DURATION_METRIC_NAME: llm_response_duration_contextual_nlg,  # noqa: E501
         }

rasa/utils/common.py

@@ -29,7 +29,6 @@ import numpy as np
 
 import rasa.shared.utils.io
 import rasa.utils.io
-from rasa.anonymization import ENV_LOG_LEVEL_FAKER, ENV_LOG_LEVEL_PRESIDIO
 from rasa.constants import (
     DEFAULT_LOG_LEVEL_LIBRARIES,
     ENV_LOG_LEVEL_KAFKA,
@@ -53,7 +52,7 @@ EXPECTED_WARNINGS: List[Tuple[Type[Warning], str]] = [
         np.VisibleDeprecationWarning,
         "Creating an ndarray from ragged nested sequences.*",
     ),
-    # raised by pycountry (rasa-plus anonymization), magic_filter, google rpc
+    # raised by magic_filter, google rpc
     # and probably other dependencies that use pkg_resources instead of importlib
     (DeprecationWarning, ".*pkg_resources.*"),
     # This warning is triggered by sanic-cors 2.0.0 and by langchain -> faiss.
@@ -281,8 +280,6 @@ def configure_library_logging() -> None:
     update_matplotlib_log_level(library_log_level)
     update_kafka_log_level(library_log_level)
     update_rabbitmq_log_level(library_log_level)
-    update_presidio_log_level(library_log_level)
-    update_faker_log_level(library_log_level)
 
 
 def update_apscheduler_log_level() -> None:
@@ -407,29 +404,6 @@ def update_rabbitmq_log_level(library_log_level: Text) -> None:
     logging.getLogger("aiormq").setLevel(log_level)
 
 
-def update_presidio_log_level(library_log_level: Text) -> None:
-    """Set the log level of presidio.
-
-    Uses the library specific log level or the general libraries log level.
-    """
-    log_level = os.environ.get(ENV_LOG_LEVEL_PRESIDIO, library_log_level)
-    presidio_loggers = ["presidio_analyzer", "presidio_anonymizer"]
-
-    for logger_name in presidio_loggers:
-        logging.getLogger(logger_name).setLevel(log_level)
-        logging.getLogger(logger_name).propagate = False
-
-
-def update_faker_log_level(library_log_level: Text) -> None:
-    """Set the log level of faker.
-
-    Uses the library specific log level or the general libraries log level.
-    """
-    log_level = os.environ.get(ENV_LOG_LEVEL_FAKER, library_log_level)
-    logging.getLogger("faker").setLevel(log_level)
-    logging.getLogger("faker").propagate = False
-
-
 def sort_list_of_dicts_by_first_key(dicts: List[Dict]) -> List[Dict]:
     """Sorts a list of dictionaries by their first key."""
     return sorted(dicts, key=lambda d: next(iter(d.keys())))

rasa/utils/log_utils.py

@@ -3,14 +3,13 @@ from __future__ import annotations
 import logging
 import os
 import sys
-from typing import Any, Dict, Optional
+from typing import Any, Optional
 
 import structlog
 from structlog.dev import ConsoleRenderer
 from structlog.typing import EventDict, WrappedLogger
 from structlog_sentry import SentryProcessor
 
-from rasa.plugin import plugin_manager
 from rasa.shared.constants import (
     DEFAULT_LOG_LEVEL,
     DEFAULT_LOG_LEVEL_LLM,
@@ -35,48 +34,6 @@ class HumanConsoleRenderer(ConsoleRenderer):
         return super().__call__(logger, name, event_dict)
 
 
-def _anonymizer(
-    _: structlog.BoundLogger, __: str, event_dict: Dict[str, Any]
-) -> Dict[str, Any]:
-    """Anonymizes event dict."""
-    anonymizable_keys = [
-        "text",
-        "response_text",
-        "user_text",
-        "slots",
-        "parse_data_text",
-        "parse_data_entities",
-        "prediction_events",
-        "tracker_latest_message",
-        "prefilled_slots",
-        "message",
-        "response",
-        "slot_candidates",
-        "rasa_event",
-        "rasa_events",
-        "tracker_states",
-        "current_states",
-        "old_states",
-        "current_states",
-        "successes",
-        "current_entity",
-        "next_entity",
-        "states",
-        "entity",
-        "token_text",
-        "user_message",
-        "json_message",
-    ]
-    anonymization_pipeline = plugin_manager().hook.get_anonymization_pipeline()
-
-    if anonymization_pipeline:
-        for key in anonymizable_keys:
-            if key in event_dict:
-                anonymized_value = anonymization_pipeline.log_run(event_dict[key])
-                event_dict[key] = anonymized_value
-    return event_dict
-
-
 def configure_structlog(
     log_level: Optional[int] = None,
     include_time: bool = False,
@@ -95,7 +52,6 @@ def configure_structlog(
     )
 
     shared_processors = [
-        _anonymizer,
         # Processors that have nothing to do with output,
         # e.g., add timestamps or log level names.
         # If log level is too low, abort pipeline and throw away log entry.

rasa/validator.py

@@ -1014,11 +1014,6 @@ class Validator:
             for sv in slot_value
         ]
         if not all(slot_values_validity):
-            invalid_slot_values = [
-                sv
-                for (sv, slot_value_valid) in zip(slot_value, slot_values_validity)
-                if not slot_value_valid
-            ]
             structlogger.error(
                 "validator.verify_predicates.link.invalid_condition",
                 step=step_id,
@@ -1027,8 +1022,7 @@ class Validator:
                 event_info=(
                     f"Detected invalid condition '{link_condition}' "
                     f"at step '{step_id}' for flow id '{flow_id}'. "
-                    f"Values {invalid_slot_values} are not valid values "
-                    f"for slot {slot_name}. "
+                    f"The condition contains invalid values for slot {slot_name}. "
                     f"Please make sure that all conditions are valid."
                 ),
             )
@@ -1998,7 +1992,7 @@ class Validator:
                         "validator.validate_conditional_response_variation_predicates.invalid_slot",
                         utter=utter_name,
                         event_info=(
-                            f"Detected invalid slot '{slot_namespace[1]}' in "
+                            f"Detected invalid slot '{slot_namespace[0]}' in "
                             f"condition '{condition}' for response variation "
                             f"'{utter_name}'. Please make sure that all slots "
                             f"are specified in the domain file."

rasa/version.py

@@ -1,3 +1,3 @@
 # this file will automatically be changed,
 # do not add anything but the version number here!
-__version__ = "3.13.0.dev7"
+__version__ = "3.13.0.dev9"

{rasa_pro-3.13.0.dev7.dist-info → rasa_pro-3.13.0.dev9.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: rasa-pro
-Version: 3.13.0.dev7
+Version: 3.13.0.dev9
 Summary: State-of-the-art open-core Conversational AI framework for Enterprises that natively leverages generative AI for effortless assistant development.
 Keywords: nlp,machine-learning,machine-learning-library,bot,bots,botkit,rasa conversational-agents,conversational-ai,chatbot,chatbot-framework,bot-framework
 Author: Rasa Technologies GmbH
@@ -18,11 +18,12 @@ Provides-Extra: full
 Provides-Extra: gh-release-notes
 Provides-Extra: jieba
 Provides-Extra: metal
+Provides-Extra: pii
 Provides-Extra: spacy
 Provides-Extra: transformers
 Requires-Dist: CacheControl (>=0.14.2,<0.15.0)
 Requires-Dist: PyJWT[crypto] (>=2.8.0,<3.0.0)
-Requires-Dist: SQLAlchemy (>=2.0.40,<2.1.0)
+Requires-Dist: SQLAlchemy (>=2.0.41,<2.1.0)
 Requires-Dist: absl-py (>=2.0,<2.1)
 Requires-Dist: aio-pika (>=8.2.3,<9.4.4)
 Requires-Dist: aiogram (>=3.15,<3.16)
@@ -45,10 +46,10 @@ Requires-Dist: demoji (>=1.1.0,<2.0.0)
 Requires-Dist: diskcache (>=5.6.3,<5.7.0)
 Requires-Dist: dnspython (==2.6.1)
 Requires-Dist: faiss-cpu (>=1.7.4,<2.0.0)
-Requires-Dist: faker (>=26.0.0,<27.0.0)
 Requires-Dist: fbmessenger (>=6.0.0,<6.1.0)
 Requires-Dist: github3.py (>=3.2.0,<3.3.0) ; extra == "gh-release-notes"
 Requires-Dist: gitpython (>=3.1.41,<3.2.0) ; extra == "full"
+Requires-Dist: gliner (>=0.2.20,<0.3.0) ; extra == "pii"
 Requires-Dist: google-auth (>=2.23.4,<3)
 Requires-Dist: google-cloud-storage (>=2.14.0,<2.15.0)
 Requires-Dist: hvac (>=1.2.1,<2.0.0)
@@ -68,6 +69,7 @@ Requires-Dist: matplotlib (>=3.7,<3.8)
 Requires-Dist: mattermostwrapper (>=2.2,<2.3)
 Requires-Dist: networkx (>=3.1,<3.2)
 Requires-Dist: numpy (>=1.26.4,<1.27.0)
+Requires-Dist: onnxruntime (==1.19.2) ; extra == "pii"
 Requires-Dist: openai (>=1.68.2,<1.69.0)
 Requires-Dist: openpyxl (>=3.1.5,<4.0.0)
 Requires-Dist: opentelemetry-api (>=1.16.0,<1.17.0)
@@ -78,13 +80,10 @@ Requires-Dist: packaging (>=23.2,<23.3)
 Requires-Dist: pep440-version-utils (>=1.1.0,<1.2.0)
 Requires-Dist: pluggy (>=1.2.0,<2.0.0)
 Requires-Dist: portalocker (>=2.7.0,<3.0.0)
-Requires-Dist: presidio-analyzer (>=2.2.33,<2.2.34)
-Requires-Dist: presidio-anonymizer (>=2.2.354,<3.0.0)
 Requires-Dist: prompt-toolkit (>=3.0.28,<3.0.29)
 Requires-Dist: protobuf (>=4.23.3,<4.25.4)
 Requires-Dist: psutil (>=5.9.5,<6.0.0)
 Requires-Dist: psycopg2-binary (>=2.9.10,<2.10.0)
-Requires-Dist: pycountry (>=22.3.5,<23.0.0)
 Requires-Dist: pydot (>=1.4,<1.5)
 Requires-Dist: pykwalify (>=1.8,<1.9)
 Requires-Dist: pymilvus (>=2.4.1,<2.4.2)
@@ -92,7 +91,7 @@ Requires-Dist: pymongo (>=4.10.1,<4.11.0)
 Requires-Dist: pypred (>=0.4.0,<0.5.0)
 Requires-Dist: python-dateutil (>=2.8.2,<2.9.0)
 Requires-Dist: python-dotenv (>=1.0.1,<2.0.0)
-Requires-Dist: python-engineio (>=4.12.0,<6,!=5.0.0)
+Requires-Dist: python-engineio (>=4.12.2,<4.13.0)
 Requires-Dist: python-keycloak (>=3.12.0,<4.0.0)
 Requires-Dist: python-socketio (>=5.8,<6)
 Requires-Dist: pytz (>=2022.7.1,<2023.0)
@@ -138,7 +137,7 @@ Requires-Dist: tensorflow_hub (>=0.13.0,<0.14.0)
 Requires-Dist: terminaltables (>=3.1.10,<3.2.0)
 Requires-Dist: tiktoken (>=0.7.0,<0.8.0)
 Requires-Dist: tqdm (>=4.66.2,<5.0.0)
-Requires-Dist: transformers (>=4.36.2,<4.37.0) ; extra == "transformers" or extra == "full"
+Requires-Dist: transformers (>=4.38.2,<4.39.0) ; extra == "transformers" or extra == "full"
 Requires-Dist: twilio (>=8.4,<8.5)
 Requires-Dist: types-protobuf (==4.25.0.20240417)
 Requires-Dist: typing-extensions (>=4.7.1,<5.0.0)