iatoolkit 0.4.0__py3-none-any.whl → 0.4.2__py3-none-any.whl

iatoolkit/__init__.py

@@ -21,13 +21,15 @@ from services.excel_service import ExcelService
 from services.dispatcher_service import Dispatcher
 from services.document_service import DocumentService
 from services.search_service import SearchService
+from services.load_documents_service import LoadDocumentsService
 from repositories.profile_repo import ProfileRepo
 from repositories.llm_query_repo import LLMQueryRepo
 from services.query_service import QueryService
+from services.prompt_manager_service import PromptService
 from repositories.database_manager import DatabaseManager
 from infra.call_service import CallServiceClient
 from common.util import Utility
-from repositories.models import Base, Company, Function, TaskType
+from repositories.models import Base, Company, Function, TaskType, Prompt, PromptCategory
 
 
 __all__ = [
@@ -42,8 +44,10 @@ __all__ = [
     'DocumentService',
     'SearchService',
     'QueryService',
+    'LoadDocumentsService',
     'ProfileRepo',
     'LLMQueryRepo',
+    'PromptService',
     'DatabaseManager',
     'CallServiceClient',
     'Utility',
@@ -51,4 +55,6 @@ __all__ = [
     'Function',
     'TaskType',
     'Base',
+    'Prompt',
+    'PromptCategory'
 ]

iatoolkit/cli_commands.py

@@ -24,6 +24,8 @@ def register_core_commands(app):
     def setup_company(company_short_name: str):
         """⚙️ Genera una nueva API key para una compañía ya registrada."""
         try:
+            dispatcher = IAToolkit.get_instance().get_injector().get(Dispatcher)
+            dispatcher.setup_all_companies()
             profile_service = IAToolkit.get_instance().get_injector().get(ProfileService)
             click.echo(f"🔑 Generando API key para '{company_short_name}'...")
             result = profile_service.new_api_key(company_short_name)
@@ -64,14 +66,4 @@ def register_core_commands(app):
             logging.exception(e)
             click.echo(f"Error: {str(e)}")
 
-    @app.cli.command("load")
-    def load_documents():
-        from services.load_documents_service import LoadDocumentsService
 
-        load_documents_service = IAToolkit.get_instance().get_injector().get(LoadDocumentsService)
-        try:
-            result = load_documents_service.load()
-            click.echo(result['message'])
-        except Exception as e:
-            logging.exception(e)
-            click.echo(f"Error: {str(e)}")

iatoolkit/company_registry.py

@@ -15,27 +15,21 @@ class CompanyRegistry:
     def __init__(self):
         self._company_classes: Dict[str, Type[BaseCompany]] = {}
         self._company_instances: Dict[str, BaseCompany] = {}
-        self._injector = None
 
-    def set_injector(self, injector) -> None:
-        """Establece el injector para crear instancias con dependencias"""
-        self._injector = injector
 
-    def instantiate_companies(self) -> Dict[str, BaseCompany]:
+    def instantiate_companies(self, injector) -> Dict[str, BaseCompany]:
         """
         Instancia todas las empresas registradas con inyección de dependencias.
 
         Returns:
             Dict con instancias de empresas {name: instance}
         """
-        if not self._injector:
-            raise RuntimeError("Injector no configurado. Llame a set_injector() primero.")
 
         for company_key, company_class in self._company_classes.items():
             if company_key not in self._company_instances:
                 try:
                     # use de injector to create the instance
-                    company_instance = self._injector.get(company_class)
+                    company_instance = injector.get(company_class)
                     self._company_instances[company_key] = company_instance
                     logging.info(f"company '{company_key}' created in dispatcher")
 

{iatoolkit-0.4.0.dist-info → iatoolkit-0.4.2.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: iatoolkit
-Version: 0.4.0
+Version: 0.4.2
 Summary: IAToolkit
 Author: Fernando Libedinsky
 License-Expression: MIT
@@ -207,46 +207,52 @@ Requires-Dist: yarl==1.18.3
 Requires-Dist: zipp==3.21.0
 Requires-Dist: zstandard==0.23.0
 
-# iatoolkit
+
+<div align="center">
+  <h1>IAToolkit</h1>
+  <p><strong>The Open-Source Framework for Building AI Chatbots on Your Private Data.</strong></p>
+</div>
 
 IAToolkit is a comprehensive, open-source framework designed for building enterprise-grade 
 AI chatbots and conversational applications. 
-Built on Flask with dependency injection, it provides a robust foundation for scalable AI solutions.
+With IAToolkit, you can build production-ready, context-aware chatbots and agents that 
+can query relational databases, perform semantic searches on documents, 
+and connect to your internal APIs in minutes.
+
+IAToolkit bridges the gap between powerful LLMs and your company's data. 
+
 
 ## 🚀 Key Features
-- **Universal LLM Integration**: OpenAI GPT, Google Gemini
-- **Template System**: Jinja2-powered prompt templates with variables
-- **Context Management**: Maintain conversation context across sessions
-
-### 🔒 **Enterprise Security**
-- **JWT Authentication**: Secure token-based authentication
-- **Session Management**: Redis-backed secure sessions
-- **CORS Configuration**: Flexible cross-origin resource sharing
-
-### 🛠 **Function Calling & Tools**
-- **Native Function Calls**: Direct integration with LLM function calling
-- **Custom Tools**: Build and register custom tools for your chatbot
-- **SQL Query Generation**: Natural language to SQL conversion
-- **API Integrations**: Connect to external services and APIs
-
-### 🗄 **Database & Storage**
-- **Multi-Database Support**: PostgreSQL, MySQL, SQLite via SQLAlchemy
-- **Vector Store Integration**: Semantic search and retrieval
-- **Document Processing**: PDF, Word, Excel, and text file handling
-
-### 📊 **Analytics & Monitoring**
-- **Query Logging**: Track all LLM interactions
-- **Performance Metrics**: Response times, token usage, costs
-- **Benchmarking**: Compare model performance
-- **Task Management**: Async task processing with status tracking
-
-### 🔧 **Developer Experience**
-- **Dependency Injection**: Clean, testable architecture
-- **CLI Tools**: Command-line interface for common tasks
-- **Hot Reloading**: Development-friendly configuration
-- **Comprehensive Logging**: Debug and monitor easily
-
-## License
-MIT License
 
+*   **🔗 Unified Data Connection**:
+    *   **Natural Language to SQL**: Let your chatbot query relational databases (PostgreSQL, MySQL, SQLite) using everyday language.
+    *   **Semantic Document Search**: Automatically chunk, embed, and search across your private documents (PDFs, Word, etc.) to provide contextually accurate answers.
+
+*   **🏢 Enterprise-Ready Multi-Tenancy**:
+    *   Deploy isolated "Company" modules, each with its own data, tools, and context. Perfect for SaaS products or internal departmental agents.
+
+*   **🧠 LLM Agnostic**:
+    *   Switch between **OpenAI (GPT-*)** and **Google (Gemini-*)** with a single line change in your configuration. No code refactoring needed.
+
+*   **🛠️ Developer-First Experience**:
+    *   Built with a clean, **Dependency Injection** architecture.
+    *   High-quality code base with **90%+ test coverage**.
+    *   Powerful Flask-based **CLI** for database setup, API key generation, and more.
+
+*   **🔒 Security & Observability Built-In**:
+    *   Comes with JWT-based authentication, user management, and secure session handling out of the box.
+    *   Full traceability with detailed logging of all queries, function calls, token usage, and costs.
+
+## ⚡ Quick Start: Create a Custom Tool in 30 Seconds
+
+See how easy it is to give your AI a new skill. Just define a method inside your Company class and describe it. 
+IAToolkit handles the rest.
+
+## 🤝 Contributing
+
+We welcome contributions! Whether it's adding a new feature, improving documentation, or fixing a bug, 
+please feel free to open a pull request.
+
+## 📄 License
 
+IAToolkit is open-source and licensed under the [MIT License](LICENSE).

{iatoolkit-0.4.0.dist-info → iatoolkit-0.4.2.dist-info}/RECORD

@@ -1,7 +1,7 @@
-iatoolkit/__init__.py,sha256=WImnnjDcaWy8_CoIS4UA2PwqYcMb0dDNUhg-OlkZ2X8,1512
+iatoolkit/__init__.py,sha256=GkFxAQHKPifz4Kd8M73Rc8TWRVIxjxkl1N0nsPvb_sU,1743
 iatoolkit/base_company.py,sha256=FlB-HFYH8FoTl4nbtsYgfKjkdZtizJbKwXqaosxmRqc,2009
-iatoolkit/cli_commands.py,sha256=oWd5kwDYd0W1Lcpuk3N2cEnusPAVefaCrYveMQ1zDvY,3223
-iatoolkit/company_registry.py,sha256=HnDpVyCc41OAn-exVF53b_HMES7GelWZcvxR39S_nI4,2900
+iatoolkit/cli_commands.py,sha256=CyaabHA3HdKd-eIqrJD8IQFT7Tqn_BEdi4jb1utisMo,2909
+iatoolkit/company_registry.py,sha256=KOUzJHLYgzMAV6BxkTiDPlN_ME4fktp7yRzKLlXZ5-w,2597
 iatoolkit/iatoolkit.py,sha256=OwlGujwtNLBYtfZuCpcX_yzrgB8BVo9Jfh72owM8FFc,15651
 iatoolkit/system_prompts/arquitectura.prompt,sha256=2W-7NWy6P6y1Gh5_-zD1iK-BWq1Siu8TuvGCouP67bQ,1267
 iatoolkit/system_prompts/format_styles.prompt,sha256=MSMe1qvR3cF_0IbFshn8R0z6Wx6VCHQq1p37rpu5wwk,3576
@@ -9,14 +9,14 @@ iatoolkit/system_prompts/query_main.prompt,sha256=Eu5VOQzUygJ45Ct1WKYGbi0JMltgI6
 iatoolkit/system_prompts/sql_rules.prompt,sha256=y4nURVnb9AyFwt-lrbMNBHHtZlhk6kC9grYoOhRnrJo,59174
 services/__init__.py,sha256=fSvSfIcPW1dHwTBY1hQ5dBEhaoorzk_GzR4G46gD8tY,173
 services/api_service.py,sha256=InIKTc64BWcp4U4tYKHz28x4ErPxIfvR9x3ZlxJZlXs,2911
-services/benchmark_service.py,sha256=g9JVrmAqIe_iI0D1DwdQ6DJ2_FJRCTndarESNSVfhbw,5907
-services/dispatcher_service.py,sha256=jGixvvQ4DTQGZye8aa05q56B6U-s6NaDb6he6UTXmQc,15534
-services/document_service.py,sha256=sm5QtbrKs2dF9hpLuSLMB-IMWYNBD7yWHv3rd80aD0o,5960
+services/benchmark_service.py,sha256=0Vgsx_FaUZL7igoBYbe1AZkIWOiEUx1FSCV_0Ut0mtk,5921
+services/dispatcher_service.py,sha256=j3Vm3vgDIgwMn9tF1BBHN3sY-V30XIkbHNcXVR0u-kY,15491
+services/document_service.py,sha256=np8wjaFpS8kVgAeVr8JWzGHcdRl1S4vsOX-dxyaLP8E,5961
 services/excel_service.py,sha256=wE9Udbyb96kGRSnZZ6KM2mbE484rKjTEhta9GKKpy-8,3630
-services/file_processor_service.py,sha256=82UArWtwpr94CAMkkoRP0_nPtoqItymdKSIABS0Xkxw,2943
+services/file_processor_service.py,sha256=0CM4CQu6KKfcLVGkxs4hYxgdz8kKRWfkV5rDH9UoccM,4173
 services/history_service.py,sha256=6fGSSWxy60nxtkwp_fodwDHoVKhpIUbHnzAzUSiNi-Y,1657
 services/jwt_service.py,sha256=dC45Sn6FyzdzRiQJnzgkjN3Hy21V1imRxB0hTyWRvlA,3979
-services/load_documents_service.py,sha256=_-OTUih8Zk0m4dHqAhkE7kAwU2mbz_QoMrOKnrq7ZWs,8821
+services/load_documents_service.py,sha256=eDqi4Nr2K0BvHS4om07LL_wbFcyfJ4qIQiMULviZWsE,7098
 services/mail_service.py,sha256=ystFit1LuYUC4ekYYebyiy1rqYQmxeL6K8h58MxEkOY,2233
 services/profile_service.py,sha256=vZV0cregZQiPKYcNLaD7xjez2y6-3Mq97cDndC8NL8w,17922
 services/prompt_manager_service.py,sha256=bWG4SIgt0u45PVUfm0xRLbLfKC7bk6uozVHRdkdgCmc,7761
@@ -26,7 +26,7 @@ services/sql_service.py,sha256=H7CIPpXTcxLXLojD2fBFr_mIAD0PW1vEJhKHLfJi4Hk,1418
 services/tasks_service.py,sha256=hHJDlcsSOPtEleD6_Vv3pocfxWNmthIhmZSdnoWFpEM,6861
 services/user_feedback_service.py,sha256=YtCndRBekDEWYEbac431Ksn2gMO5iBrI3WqKK0xtShE,2513
 services/user_session_context_service.py,sha256=5qn7fqpuiU8KgMpU4M5-iRUsETumz1raBw-EeZLuE1A,3868
-iatoolkit-0.4.0.dist-info/METADATA,sha256=1muE8emXWndqmPwC3xhtpVYjrFiktvWdWdD_UckbW10,8801
-iatoolkit-0.4.0.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-iatoolkit-0.4.0.dist-info/top_level.txt,sha256=dqlBbmgo9okD9d_WMR9uYzdup7Rxgj26yFF85jRGeu4,19
-iatoolkit-0.4.0.dist-info/RECORD,,
+iatoolkit-0.4.2.dist-info/METADATA,sha256=F2KitiTXfL4FN4nIp7ebGr36828ZVRO5QFT5Bvxmfg8,9300
+iatoolkit-0.4.2.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+iatoolkit-0.4.2.dist-info/top_level.txt,sha256=dqlBbmgo9okD9d_WMR9uYzdup7Rxgj26yFF85jRGeu4,19
+iatoolkit-0.4.2.dist-info/RECORD,,

services/benchmark_service.py

@@ -65,7 +65,7 @@ class BenchmarkService:
 
         company = self.profile_repo.get_company_by_short_name(company_short_name)
         if not company:
-            raise IAToolkitException(IAToolkitException.ErrorType.CONFIG_ERROR, "Compañía 'maxxa' no encontrada.")
+            raise IAToolkitException(IAToolkitException.ErrorType.CONFIG_ERROR, f"Compañía {company_short_name} no encontrada.")
 
         total_rows = len(df)
         logging.info(f"Iniciando benchmark para {total_rows} casos de prueba desde el archivo: {file_path}")

services/dispatcher_service.py

@@ -78,8 +78,7 @@ class Dispatcher:
 
         # ✅ NOW it is safe to get the injector and instantiate companies.
         injector = current_iatoolkit().get_injector()
-        self.company_registry.set_injector(injector)
-        self.company_registry.instantiate_companies()
+        self.company_registry.instantiate_companies(injector)
 
 
     def start_execution(self):
@@ -110,7 +109,7 @@ class Dispatcher:
                     prompt_name=prompt['name'],
                     description=prompt['description'],
                     order=1,
-                    is_system_prompt=True
+                    is_system_prompt=True,
                 )
                 i += 1
 
@@ -118,7 +117,7 @@ class Dispatcher:
         for company in self.company_instances.values():
             company.register_company()
 
-    def dispatch(self, company_name: str, action: str, **kwargs) -> str:
+    def dispatch(self, company_name: str, action: str, **kwargs) -> dict:
         company_key = company_name.lower()
 
         if company_key not in self.company_instances:

services/document_service.py

@@ -16,7 +16,7 @@ class DocumentService:
     @inject
     def __init__(self):
         # max number of pages to load
-        self.max_doc_pages = int(os.getenv("MAX_DOC_PAGES", "10"))
+        self.max_doc_pages = int(os.getenv("MAX_DOC_PAGES", "200"))
 
     def file_to_txt(self, filename, file_content):
         try:

services/file_processor_service.py

@@ -7,26 +7,45 @@ from infra.connectors.file_connector import FileConnector
 import logging
 import os
 from typing import Optional, Callable, Dict
+from repositories.models import Company
 
 
 class FileProcessorConfig:
+    """Configuration class for the FileProcessor."""
     def __init__(
         self,
         filters: Dict,
-        action: Callable[[str, bytes], None],
+        callback: Callable[[Company, str, bytes, dict], None],
         continue_on_error: bool = True,
         log_file: str = 'file_processor.log',
         echo: bool = False,
-        context: dict = None,
+        context: dict = None
     ):
+        """
+        Initializes the FileProcessor configuration.
+
+        Args:
+            filters (Dict): A dictionary of filters to apply to file names.
+                Example: {'filename_contains': '.pdf'}
+            action (Callable): The function to execute for each processed file.
+                It receives filename (str), content (bytes), and context (dict).
+            continue_on_error (bool): If True, continues processing other files upon an error.
+            log_file (str): The path to the log file.
+            echo (bool): If True, prints progress to the console.
+            context (dict): A context dictionary passed to the action function.
+        """
         self.filters = filters
-        self.action = action
+        self.callback = callback
         self.continue_on_error = continue_on_error
         self.log_file = log_file
         self.echo = echo
         self.context = context or {}
 
 class FileProcessor:
+    """
+    A generic service to process files from a given data source (connector).
+    It lists files, applies filters, and executes a specific action for each one.
+    """
     def __init__(self,
                  connector: FileConnector,
                  config: FileProcessorConfig,
@@ -45,6 +64,7 @@ class FileProcessor:
         return logging.getLogger(__name__)
 
     def process_files(self):
+        # Fetches files from the connector, filters them, and processes them.
         try:
             files = self.connector.list_files()
         except Exception as e:
@@ -67,9 +87,12 @@ class FileProcessor:
 
                 content = self.connector.get_file_content(file_path)
 
-                # execute the action defined
+                # execute the callback function
                 filename = os.path.basename(file_name)
-                self.config.action(filename, content, self.config.context)
+                self.config.callback(company=self.config.context.get('company'),
+                                     filename=filename,
+                                     content=content,
+                                     context=self.config.context)
                 self.processed_files += 1
 
                 self.logger.info(f"Successfully processed file: {file_path}")

services/load_documents_service.py

@@ -21,6 +21,10 @@ from typing import Dict
 
 
 class LoadDocumentsService:
+    """
+    Orchestrates the process of loading, processing, and storing documents
+    from various sources for different companies.
+    """
     @inject
     def __init__(self,
                  file_connector_factory: FileConnectorFactory,
@@ -38,7 +42,6 @@ class LoadDocumentsService:
         self.vector_store = vector_store
         self.file_connector_factory = file_connector_factory
         self.dispatcher = dispatcher
-        self.company = None
 
         # lower warnings
         logging.getLogger().setLevel(logging.ERROR)
@@ -49,86 +52,46 @@ class LoadDocumentsService:
             separators=["\n\n", "\n", "."]
         )
 
-    # load the files for all of the companies.
-    def load(self, doc_type: str = None):
-        # doc_type: an optional document_type for loading
-        files_loaded = 0
-        companies = self.profile_repo.get_companies()
-
-        for company in companies:
-            load_config = company.parameters.get('load', {})
-            if not load_config:
-                continue
-
-            print(f"Cargando datos de ** {company.short_name} **")
-            self.company = company
-
-            # Si hay configuraciones de tipos de documento específicos
-            doc_types_config = load_config.get('document_types', {})
-
-            if doc_types_config and len(doc_types_config) > 0:
-                # Si se especificó un tipo de documento, cargar solo ese tipo
-                if doc_type and doc_type in doc_types_config:
-                    files_loaded += self._load_document_type(company, doc_type, doc_types_config[doc_type])
-                # Si no se especificó, cargar todos los tipos configurados
-                elif not doc_type:
-                    for type_name, type_config in doc_types_config.items():
-                        files_loaded += self._load_document_type(company, type_name, type_config)
-            else:
-                # Comportamiento anterior: usar la configuración general
-                connector = load_config.get('connector', {})
-                if not connector:
-                    raise IAToolkitException(IAToolkitException.ErrorType.MISSING_PARAMETER,
-                                       f"Falta configurar conector en empresa {company.short_name}")
-
-                files_loaded += self.load_data_source(connector)
-
-        return {'message': f'{files_loaded} files processed'}
-
-    def _load_document_type(self, company: Company, doc_type_name: str, type_config: Dict) -> int:
-        # load specific document_types for a company
-        connector = type_config.get('connector')
-        if not connector:
-            logging.warning(f"Falta configurar conector para tipo {doc_type_name} en empresa {company.short_name}")
-            raise IAToolkitException(IAToolkitException.ErrorType.MISSING_PARAMETER,
-                               f"Falta configurar conector para tipo {doc_type_name} en empresa {company.short_name}")
-
-        # get the metadata for this connector
-        predefined_metadata = type_config.get('metadata', {})
-
-        # config specific filters
-        filters = type_config.get('filters', {"filename_contains": ".pdf"})
-
-        return self.load_data_source(connector, predefined_metadata, filters)
-
-    def load_data_source(self, connector_config: Dict, predefined_metadata: Dict = None, filters: Dict = None):
+    def load_company_files(self,
+                         company: Company,
+                         connector_config: Dict,
+                         predefined_metadata: Dict = None,
+                         filters: Dict = None):
         """
-        Carga archivos desde una fuente de datos usando un conector.
+        Loads all the company files from a connector
 
         Args:
-            connector_config: Configuración del conector
-            predefined_metadata: Metadatos predefinidos para todos los documentos de esta fuente
-            filters: Filtros específicos para esta carga
+            company (Company): The company to load files for.
+            connector_config (Dict): The configuration for the file connector.
+            predefined_metadata (Dict, optional): Metadata to be added to all documents from this source.
+            filters (Dict, optional): Filters to apply to the files.
 
         Returns:
-            int o dict: Número de archivos procesados o diccionario de error
+            int: The number of processed files.
         """
+        if not connector_config:
+            raise IAToolkitException(IAToolkitException.ErrorType.MISSING_PARAMETER,
+                        f"Falta configurar conector")
+
         try:
-            # Si no se proporcionaron filtros, usar el predeterminado
             if not filters:
                 filters = {"filename_contains": ".pdf"}
 
             # Pasar metadata predefinida como parte del contexto al procesador
-            # para que esté disponible en la función load_file
-            extra_context = {}
+            # para que esté disponible en la función load_file_callback
+            context = {
+                'company': company,
+                'metadata': {}
+            }
+
             if predefined_metadata:
-                extra_context['metadata'] = predefined_metadata
+                context['metadata'] = predefined_metadata
 
             # config the processor
             processor_config = FileProcessorConfig(
-                context=extra_context,
+                callback=self.load_file_callback,
+                context=context,
                 filters=filters,
-                action=self.load_file,
                 continue_on_error=True,
                 echo=True
             )
@@ -144,14 +107,21 @@ class LoadDocumentsService:
             logging.exception("Loading files error: %s", str(e))
             return {"error": str(e)}
 
-    # load an individual filename
-    # this method is set up on the FileProcessorConfig object
-    def load_file(self, filename: str, content: bytes, context: dict = {}, company: Company = None):
-        if not company:
-            company = self.company
+    def load_file_callback(self, company: Company, filename: str, content: bytes, context: dict = {}):
+        """
+        Processes a single file: extracts text, generates metadata, and saves it
+        to the relational database and the vector store.
+        This method is intended to be used as the 'action' for FileProcessor.
+
+        Args:
+            company (Company): The company associated with the file.
+            filename (str): The name of the file.
+            content (bytes): The binary content of the file.
+            context (dict, optional): A context dictionary, may contain predefined metadata.
+        """
 
         # check if file exist in repositories
-        if self.doc_repo.get(company=company,filename=filename):
+        if self.doc_repo.get(company_id=company.id,filename=filename):
             return
 
         try: