rasa-pro 3.12.0.dev13__tar.gz → 3.12.0rc2__tar.gz

{rasa_pro-3.12.0.dev13 → rasa_pro-3.12.0rc2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: rasa-pro
-Version: 3.12.0.dev13
+Version: 3.12.0rc2
 Summary: State-of-the-art open-core Conversational AI framework for Enterprises that natively leverages generative AI for effortless assistant development.
 Home-page: https://rasa.com
 Keywords: nlp,machine-learning,machine-learning-library,bot,bots,botkit,rasa conversational-agents,conversational-ai,chatbot,chatbot-framework,bot-framework
@@ -31,6 +31,7 @@ Requires-Dist: aiogram (>=3.15,<3.16)
 Requires-Dist: aiohttp (>=3.9.4,<3.10)
 Requires-Dist: apscheduler (>=3.10,<3.11)
 Requires-Dist: attrs (>=23.1,<23.2)
+Requires-Dist: azure-identity (>=1.19.0,<1.20.0)
 Requires-Dist: azure-storage-blob (>=12.16.0,<12.17.0)
 Requires-Dist: boto3 (>=1.35.15,<1.36.0)
 Requires-Dist: certifi (>=2024.07.04)
@@ -63,6 +64,7 @@ Requires-Dist: jsonschema (>=4.22)
 Requires-Dist: keras (==2.14.0)
 Requires-Dist: langchain (>=0.2.0,<0.3.0)
 Requires-Dist: langchain-community (>=0.2.0,<0.3.0)
+Requires-Dist: langcodes (>=3.5.0,<4.0.0)
 Requires-Dist: litellm (>=1.52.6,<1.53.0)
 Requires-Dist: matplotlib (>=3.7,<3.8)
 Requires-Dist: mattermostwrapper (>=2.2,<2.3)
@@ -156,15 +158,14 @@ Description-Content-Type: text/markdown
 
 <div align="center">
 
-[![Build Status](https://github.com/RasaHQ/rasa-private/workflows/Continuous%20Integration/badge.svg)](https://github.com/RasaHQ/rasa-private/actions)
 [![Quality Gate Status](https://sonarcloud.io/api/project_badges/measure?project=RasaHQ_rasa&metric=alert_status)](https://sonarcloud.io/summary/new_code?id=RasaHQ_rasa)
-[![Documentation Status](https://img.shields.io/badge/docs-stable-brightgreen.svg)](https://rasa.com/docs/rasa-pro/)
+[![Documentation Status](https://img.shields.io/badge/docs-stable-brightgreen.svg)](https://rasa.com/docs/docs/pro/intro)
+![Python version support](https://img.shields.io/pypi/pyversions/rasa-pro)
 
 </div>
 
 <hr />
 
-
 Rasa Pro is a framework for building scalable, dynamic conversational AI assistants that integrate large language models (LLMs) to enable more contextually aware and agentic interactions. Whether you’re new to conversational AI or an experienced developer, Rasa Pro offers enhanced flexibility, control, and performance for mission-critical applications.
 
 Building on the foundation of Rasa Open Source, Rasa Pro adds advanced features like CALM (Conversational AI with Language Models) and Dialogue Understanding (DU), which enable developers to shift from traditional intent-driven systems to LLM-based agents. This allows for more robust, responsive interactions that adhere strictly to business logic, while reducing risks like prompt injection and minimizing hallucinations.
@@ -177,20 +178,18 @@ Building on the foundation of Rasa Open Source, Rasa Pro adds advanced features
 - **Robustness and Control:** Maintain strict adherence to business logic, preventing unwanted behaviors like prompt injection and hallucinations, leading to more reliable responses and secure interactions.
 - **Built-in Security:** Safeguard sensitive data, control access, and ensure secure deployment, essential for production environments that demand high levels of security and compliance.
 
+A [free developer license](https://rasa.com/docs/pro/intro/#who-rasa-pro-is-for) is available so you can explore and get to know Rasa Pro. It allows you to take your assistant live in production a limited capacity. A paid license is required for larger-scale production use, but all code is visible and can be customized as needed.
 
+To get started right now, you can
 
-A [free developer license](https://rasa.com/docs/rasa-pro/developer-edition/) is available so you can explore and get to know Rasa Pro. For small production deployments, the Extended Developer License allows you to take your assistant live in a limited capacity. A paid license is required for larger-scale production use, but all code is visible and can be customized as needed.
-
-To get started right now, you can 
-
-`pip install rasa-pro` 
+`pip install rasa-pro`
 
-Check out our 
+Check out our
 
-- [Rasa-pro Quickstart](https://rasa.com/docs/rasa-pro/installation/quickstart/),
-- [Conversational AI with Language Models (CALM) conceptual rundown](https://rasa.com/docs/rasa-pro/calm/),
-- [Rasa Pro / CALM tutorial](https://rasa.com/docs/rasa-pro/tutorial), and
-- [Rasa pro changelog](https://rasa.com/docs/rasa/rasa-pro-changelog/)
+- [Rasa-pro Quickstart](https://rasa.com/docs/learn/quickstart/pro),
+- [Conversational AI with Language Models (CALM) conceptual rundown](https://rasa.com/docs/learn/concepts/calm),
+- [Rasa Pro / CALM tutorial](https://rasa.com/docs/pro/tutorial), and
+- [Rasa pro changelog](https://rasa.com/docs/reference/changelogs/rasa-pro-changelog)
 
 for more. Also feel free to reach out to us on the [Rasa forum](https://forum.rasa.com/).
 

{rasa_pro-3.12.0.dev13 → rasa_pro-3.12.0rc2}/README.md

@@ -2,15 +2,14 @@
 
 <div align="center">
 
-[![Build Status](https://github.com/RasaHQ/rasa-private/workflows/Continuous%20Integration/badge.svg)](https://github.com/RasaHQ/rasa-private/actions)
 [![Quality Gate Status](https://sonarcloud.io/api/project_badges/measure?project=RasaHQ_rasa&metric=alert_status)](https://sonarcloud.io/summary/new_code?id=RasaHQ_rasa)
-[![Documentation Status](https://img.shields.io/badge/docs-stable-brightgreen.svg)](https://rasa.com/docs/rasa-pro/)
+[![Documentation Status](https://img.shields.io/badge/docs-stable-brightgreen.svg)](https://rasa.com/docs/docs/pro/intro)
+![Python version support](https://img.shields.io/pypi/pyversions/rasa-pro)
 
 </div>
 
 <hr />
 
-
 Rasa Pro is a framework for building scalable, dynamic conversational AI assistants that integrate large language models (LLMs) to enable more contextually aware and agentic interactions. Whether you’re new to conversational AI or an experienced developer, Rasa Pro offers enhanced flexibility, control, and performance for mission-critical applications.
 
 Building on the foundation of Rasa Open Source, Rasa Pro adds advanced features like CALM (Conversational AI with Language Models) and Dialogue Understanding (DU), which enable developers to shift from traditional intent-driven systems to LLM-based agents. This allows for more robust, responsive interactions that adhere strictly to business logic, while reducing risks like prompt injection and minimizing hallucinations.
@@ -23,19 +22,17 @@ Building on the foundation of Rasa Open Source, Rasa Pro adds advanced features
 - **Robustness and Control:** Maintain strict adherence to business logic, preventing unwanted behaviors like prompt injection and hallucinations, leading to more reliable responses and secure interactions.
 - **Built-in Security:** Safeguard sensitive data, control access, and ensure secure deployment, essential for production environments that demand high levels of security and compliance.
 
+A [free developer license](https://rasa.com/docs/pro/intro/#who-rasa-pro-is-for) is available so you can explore and get to know Rasa Pro. It allows you to take your assistant live in production a limited capacity. A paid license is required for larger-scale production use, but all code is visible and can be customized as needed.
 
+To get started right now, you can
 
-A [free developer license](https://rasa.com/docs/rasa-pro/developer-edition/) is available so you can explore and get to know Rasa Pro. For small production deployments, the Extended Developer License allows you to take your assistant live in a limited capacity. A paid license is required for larger-scale production use, but all code is visible and can be customized as needed.
-
-To get started right now, you can 
-
-`pip install rasa-pro` 
+`pip install rasa-pro`
 
-Check out our 
+Check out our
 
-- [Rasa-pro Quickstart](https://rasa.com/docs/rasa-pro/installation/quickstart/),
-- [Conversational AI with Language Models (CALM) conceptual rundown](https://rasa.com/docs/rasa-pro/calm/),
-- [Rasa Pro / CALM tutorial](https://rasa.com/docs/rasa-pro/tutorial), and
-- [Rasa pro changelog](https://rasa.com/docs/rasa/rasa-pro-changelog/)
+- [Rasa-pro Quickstart](https://rasa.com/docs/learn/quickstart/pro),
+- [Conversational AI with Language Models (CALM) conceptual rundown](https://rasa.com/docs/learn/concepts/calm),
+- [Rasa Pro / CALM tutorial](https://rasa.com/docs/pro/tutorial), and
+- [Rasa pro changelog](https://rasa.com/docs/reference/changelogs/rasa-pro-changelog)
 
 for more. Also feel free to reach out to us on the [Rasa forum](https://forum.rasa.com/).

{rasa_pro-3.12.0.dev13 → rasa_pro-3.12.0rc2}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "poetry.core.masonry.api"
 
 [tool.poetry]
 name = "rasa-pro"
-version = "3.12.0.dev13"
+version = "3.12.0rc2"
 description = "State-of-the-art open-core Conversational AI framework for Enterprises that natively leverages generative AI for effortless assistant development."
 authors = [ "Rasa Technologies GmbH <hi@rasa.com>",]
 maintainers = [ "Tom Bocklisch <tom@rasa.com>",]
@@ -193,6 +193,8 @@ numpy = "~1.26.4"
 scipy = "~1.13.1"
 scikit-learn = "~1.5.1"
 demoji = "^1.1.0"
+azure-identity = "~1.19.0"
+langcodes = "^3.5.0"
 
 [[tool.poetry.dependencies.tensorflow-io-gcs-filesystem]]
 version = "==0.31" # newer versions don't provide wheels for Windows

{rasa_pro-3.12.0.dev13 → rasa_pro-3.12.0rc2}/rasa/anonymization/anonymization_rule_executor.py

@@ -4,20 +4,14 @@ from dataclasses import dataclass
 from typing import Any, Dict, List, Optional, Text, Union
 
 from faker import Faker
-from presidio_analyzer import AnalyzerEngine
-from presidio_analyzer.nlp_engine import (
-    SpacyNlpEngine,
-    StanzaNlpEngine,
-    TransformersNlpEngine,
-)
-from presidio_anonymizer import AnonymizerEngine
-from presidio_anonymizer.entities import OperatorConfig
 
 from rasa.shared.exceptions import RasaException
 from rasa.utils.singleton import Singleton
 
 if typing.TYPE_CHECKING:
+    from presidio_analyzer import AnalyzerEngine
     from presidio_analyzer.nlp_engine.nlp_engine import NlpEngine
+    from presidio_anonymizer.entities import OperatorConfig
 
 DEFAULT_PRESIDIO_LANG_CODE = "en"
 DEFAULT_PRESIDIO_MODEL_NAME = "en_core_web_lg"
@@ -61,8 +55,10 @@ class AnonymizationAnalyzer(metaclass=Singleton):
     @staticmethod
     def _get_analyzer_engine(
         anonymization_rule_list: AnonymizationRuleList,
-    ) -> AnalyzerEngine:
+    ) -> "AnalyzerEngine":
         """Returns an analyzer engine for all the anonymization rule lists."""
+        from presidio_analyzer import AnalyzerEngine
+
         try:
             nlp_engine = AnonymizationAnalyzer._build_presidio_nlp_engine(
                 anonymization_rule_list
@@ -84,6 +80,12 @@ class AnonymizationAnalyzer(metaclass=Singleton):
         anonymization_rule_list: AnonymizationRuleList,
     ) -> "NlpEngine":
         """Creates an instance of the Presidio nlp engine."""
+        from presidio_analyzer.nlp_engine import (
+            SpacyNlpEngine,
+            StanzaNlpEngine,
+            TransformersNlpEngine,
+        )
+
         if anonymization_rule_list.model_provider == "transformers":
             nlp_engine = TransformersNlpEngine(
                 models={
@@ -111,6 +113,8 @@ class AnonymizationRuleExecutor:
 
     def __init__(self, anonymization_rule_list: AnonymizationRuleList):
         """Initialize the anonymization rule executor."""
+        from presidio_anonymizer import AnonymizerEngine
+
         self.anonymization_rule_list = anonymization_rule_list
 
         is_valid_rule_list = self._validate_anonymization_rule_list(
@@ -243,8 +247,10 @@ class AnonymizationRuleExecutor:
 
         return func
 
-    def get_operators(self) -> Dict[Text, OperatorConfig]:
+    def get_operators(self) -> Dict[Text, "OperatorConfig"]:
         """Returns a dictionary of operators for the given anonymization rule list."""
+        from presidio_anonymizer.entities import OperatorConfig
+
         operators = {}
 
         for rule in self.anonymization_rule_list.rule_list:

{rasa_pro-3.12.0.dev13 → rasa_pro-3.12.0rc2}/rasa/cli/data.py

@@ -178,6 +178,22 @@ def _add_data_validate_parsers(
     )
     arguments.set_validator_arguments(flows_structure_parser)
 
+    translations_structure_parser = validate_subparsers.add_parser(
+        "translations",
+        formatter_class=argparse.ArgumentDefaultsHelpFormatter,
+        parents=parents,
+        help="Checks for inconsistencies of the flow and response translation.",
+    )
+    translations_structure_parser.set_defaults(
+        func=lambda args: rasa.cli.utils.validate_files(
+            args.fail_on_warnings,
+            args.max_history,
+            _build_training_data_importer(args),
+            translations_only=True,
+        )
+    )
+    arguments.set_validator_arguments(translations_structure_parser)
+
 
 def _build_training_data_importer(args: argparse.Namespace) -> "TrainingDataImporter":
     config = rasa.cli.utils.get_validated_path(

{rasa_pro-3.12.0.dev13 → rasa_pro-3.12.0rc2}/rasa/cli/project_templates/calm/config.yml

@@ -1,9 +1,9 @@
 recipe: default.v1
 language: en
 pipeline:
-  - name: SingleStepLLMCommandGenerator
+  - name: CompactLLMCommandGenerator
     llm:
-      model_group: openai-gpt-4
+      model_group: openai-gpt-4o
 
 policies:
   - name: FlowPolicy

{rasa_pro-3.12.0.dev13 → rasa_pro-3.12.0rc2}/rasa/cli/project_templates/calm/domain/list_contacts.yml

@@ -7,8 +7,7 @@ slots:
   contacts_list:
     type: text
     mappings:
-      - type: custom
-        action: list_contacts
+      - type: controlled
 
 responses:
   utter_no_contacts:

{rasa_pro-3.12.0.dev13 → rasa_pro-3.12.0rc2}/rasa/cli/project_templates/calm/domain/remove_contact.yml

@@ -7,8 +7,7 @@ slots:
   remove_contact_name:
     type: text
     mappings:
-      - type: custom
-        action: remove_contact
+      - type: controlled
   remove_contact_handle:
     type: text
     mappings:

rasa_pro-3.12.0rc2/rasa/cli/project_templates/calm/domain/shared.yml

@@ -0,0 +1,7 @@
+version: "3.1"
+
+slots:
+  return_value:
+    type: any
+    mappings:
+      - type: controlled

{rasa_pro-3.12.0.dev13 → rasa_pro-3.12.0rc2}/rasa/cli/project_templates/calm/endpoints.yml

@@ -50,9 +50,9 @@ action_endpoint:
 #   type: rephrase
 
 model_groups:
-  - id: openai-gpt-4
+  - id: openai-gpt-4o
     models:
       - provider: openai
-        model: openai/gpt-4
+        model: gpt-4o-2024-11-20
         request_timeout: 7
         max_tokens: 256

{rasa_pro-3.12.0.dev13 → rasa_pro-3.12.0rc2}/rasa/cli/utils.py

@@ -248,6 +248,7 @@ def validate_files(
     importer: TrainingDataImporter,
     stories_only: bool = False,
     flows_only: bool = False,
+    translations_only: bool = False,
 ) -> None:
     """Validates either the story structure or the entire project.
 
@@ -257,6 +258,7 @@ def validate_files(
         importer: The `TrainingDataImporter` to use to load the training data.
         stories_only: If `True`, only the story structure is validated.
         flows_only: If `True`, only the flows are validated.
+        translations_only: If `True`, only the translations data is validated.
     """
     from rasa.validator import Validator
 
@@ -266,6 +268,8 @@ def validate_files(
         all_good = _validate_story_structure(validator, max_history, fail_on_warnings)
     elif flows_only:
         all_good = validator.verify_flows()
+    elif translations_only:
+        all_good = validator.verify_translations()
     else:
         if importer.get_domain().is_empty():
             structlogger.error(
@@ -280,6 +284,7 @@ def validate_files(
             validator, max_history, fail_on_warnings
         )
         valid_flows = validator.verify_flows()
+        valid_translations = validator.verify_translations(summary_mode=True)
         valid_CALM_slot_mappings = validator.validate_CALM_slot_mappings()
 
         all_good = (
@@ -287,6 +292,7 @@ def validate_files(
             and valid_nlu
             and valid_stories
             and valid_flows
+            and valid_translations
             and valid_CALM_slot_mappings
         )
 
@@ -309,6 +315,10 @@ def _validate_domain(validator: "Validator") -> bool:
     valid_slot_mappings = validator.verify_slot_mappings()
     valid_responses = validator.check_for_no_empty_parenthesis_in_responses()
     valid_buttons = validator.validate_button_payloads()
+    valid_slot_validation = validator.verify_slot_validation()
+    valid_conditional_responses = (
+        validator.validate_conditional_response_variation_predicates()
+    )
     return (
         valid_domain_validity
         and valid_actions_in_stories_rules
@@ -317,6 +327,8 @@ def _validate_domain(validator: "Validator") -> bool:
         and valid_slot_mappings
         and valid_responses
         and valid_buttons
+        and valid_slot_validation
+        and valid_conditional_responses
     )