Changelog update for 2.3.8-alfa

- Add possibility to add allowed_languages to tenant make

common/langchain/llm_metrics_handler.py

@@ -3,7 +3,6 @@ from langchain.callbacks.base import BaseCallbackHandler
 from typing import Dict, Any, List
 from langchain.schema import LLMResult
 from common.utils.business_event_context import current_event
-from flask import current_app
 
 
 class LLMMetricsHandler(BaseCallbackHandler):

common/langchain/persistent_llm_metrics_handler.py

@@ -0,0 +1,47 @@
+import time
+from langchain.callbacks.base import BaseCallbackHandler
+from typing import Dict, Any, List
+from langchain.schema import LLMResult
+from common.utils.business_event_context import current_event
+
+
+class PersistentLLMMetricsHandler(BaseCallbackHandler):
+    """Metrics handler that allows metrics to be retrieved from within any call. In case metrics are required for other
+    purposes than business event logging."""
+
+    def __init__(self):
+        self.total_tokens: int = 0
+        self.prompt_tokens: int = 0
+        self.completion_tokens: int = 0
+        self.start_time: float = 0
+        self.end_time: float = 0
+        self.total_time: float = 0
+
+    def reset(self):
+        self.total_tokens = 0
+        self.prompt_tokens = 0
+        self.completion_tokens = 0
+        self.start_time = 0
+        self.end_time = 0
+        self.total_time = 0
+
+    def on_llm_start(self, serialized: Dict[str, Any], prompts: List[str], **kwargs: Any) -> None:
+        self.start_time = time.time()
+
+    def on_llm_end(self, response: LLMResult, **kwargs: Any) -> None:
+        self.end_time = time.time()
+        self.total_time = self.end_time - self.start_time
+
+        usage = response.llm_output.get('token_usage', {})
+        self.prompt_tokens += usage.get('prompt_tokens', 0)
+        self.completion_tokens += usage.get('completion_tokens', 0)
+        self.total_tokens = self.prompt_tokens + self.completion_tokens
+
+    def get_metrics(self) -> Dict[str, int | float]:
+        return {
+            'total_tokens': self.total_tokens,
+            'prompt_tokens': self.prompt_tokens,
+            'completion_tokens': self.completion_tokens,
+            'time_elapsed': self.total_time,
+            'interaction_type': 'LLM',
+        }

common/models/document.py

@@ -34,6 +34,7 @@ class Processor(db.Model):
     catalog_id = db.Column(db.Integer, db.ForeignKey('catalog.id'), nullable=True)
     type = db.Column(db.String(50), nullable=False)
     sub_file_type = db.Column(db.String(50), nullable=True)
+    active = db.Column(db.Boolean, nullable=True, default=True)
 
     # Tuning enablers
     tuning = db.Column(db.Boolean, nullable=True, default=False)

common/models/user.py

@@ -186,6 +186,7 @@ class TenantMake(db.Model):
     active = db.Column(db.Boolean, nullable=False, default=True)
     website = db.Column(db.String(255), nullable=True)
     logo_url = db.Column(db.String(255), nullable=True)
+    allowed_languages = db.Column(ARRAY(sa.String(2)), nullable=True)
 
     # Chat customisation options
     chat_customisation_options = db.Column(JSONB, nullable=True)
@@ -317,3 +318,27 @@ class SpecialistMagicLinkTenant(db.Model):
 
     magic_link_code = db.Column(db.String(55), primary_key=True)
     tenant_id = db.Column(db.Integer, db.ForeignKey('public.tenant.id'), nullable=False)
+
+
+class TranslationCache(db.Model):
+    __bind_key__ = 'public'
+    __table_args__ = {'schema': 'public'}
+
+    cache_key = db.Column(db.String(16), primary_key=True)
+    source_text = db.Column(db.Text, nullable=False)
+    translated_text = db.Column(db.Text, nullable=False)
+    source_language = db.Column(db.String(2), nullable=False)
+    target_language = db.Column(db.String(2), nullable=False)
+    context = db.Column(db.Text, nullable=True)
+
+    # Translation cost
+    prompt_tokens = db.Column(db.Integer, nullable=False)
+    completion_tokens = db.Column(db.Integer, nullable=False)
+
+    # Tracking
+    created_at = db.Column(db.DateTime, nullable=False, server_default=db.func.now())
+    created_by = db.Column(db.Integer, db.ForeignKey('public.user.id'), nullable=True)
+    updated_at = db.Column(db.DateTime, nullable=False, server_default=db.func.now(), onupdate=db.func.now())
+    updated_by = db.Column(db.Integer, db.ForeignKey('public.user.id'), nullable=True)
+
+    last_used_at = db.Column(db.DateTime, nullable=True)

common/services/utils/translation_services.py

@@ -0,0 +1,43 @@
+import xxhash
+import json
+
+from langchain_core.output_parsers import StrOutputParser
+from langchain_core.prompts import ChatPromptTemplate
+from langchain_core.runnables import RunnablePassthrough
+
+from common.langchain.persistent_llm_metrics_handler import PersistentLLMMetricsHandler
+from common.utils.model_utils import get_template, replace_variable_in_template
+
+class TranslationService:
+    def __init__(self, tenant_id):
+        self.tenant_id = tenant_id
+
+    def translate_text(self, text_to_translate: str, target_lang: str, source_lang: str = None, context: str = None) -> tuple[
+        str, dict[str, int | float]]:
+        prompt_params = {
+            "text_to_translate": text_to_translate,
+            "target_lang": target_lang,
+        }
+        if context:
+            template, llm = get_template("translation_with_context")
+            prompt_params["context"] = context
+        else:
+            template, llm = get_template("translation_without_context")
+
+        # Add a metrics handler to capture usage
+
+        metrics_handler = PersistentLLMMetricsHandler()
+        existing_callbacks = llm.callbacks
+        llm.callbacks = existing_callbacks + [metrics_handler]
+
+        translation_prompt = ChatPromptTemplate.from_template(template)
+
+        setup = RunnablePassthrough()
+
+        chain = (setup | translation_prompt | llm | StrOutputParser())
+
+        translation = chain.invoke(prompt_params)
+
+        metrics = metrics_handler.get_metrics()
+
+        return translation, metrics

common/utils/cache/translation_cache.py

@@ -0,0 +1,156 @@
+import json
+from typing import Dict, Any, Optional
+from datetime import datetime as dt, timezone as tz
+
+import xxhash
+from flask import current_app
+from sqlalchemy import and_
+from sqlalchemy.inspection import inspect
+
+from common.utils.cache.base import CacheHandler, T
+from common.extensions import db
+
+from common.models.user import TranslationCache
+from common.services.utils.translation_services import TranslationService
+from flask_security import current_user
+
+
+class TranslationCacheHandler(CacheHandler[TranslationCache]):
+    """Handles caching of translations with fallback to database and external translation service"""
+    handler_name = 'translation_cache'
+
+    def __init__(self, region):
+        super().__init__(region, 'translation')
+        self.configure_keys('hash_key')
+
+    def _to_cache_data(self, instance: TranslationCache) -> Dict[str, Any]:
+        """Convert TranslationCache instance to cache data using SQLAlchemy inspection"""
+        if not instance:
+            return {}
+
+        mapper = inspect(TranslationCache)
+        data = {}
+
+        for column in mapper.columns:
+            value = getattr(instance, column.name)
+
+            # Handle date serialization
+            if isinstance(value, dt):
+                data[column.name] = value.isoformat()
+            else:
+                data[column.name] = value
+
+        return data
+
+    def _from_cache_data(self, data: Dict[str, Any], **kwargs) -> TranslationCache:
+        if not data:
+            return None
+
+        # Create a new TranslationCache instance
+        translation = TranslationCache()
+        mapper = inspect(TranslationCache)
+
+        # Set all attributes dynamically
+        for column in mapper.columns:
+            if column.name in data:
+                value = data[column.name]
+
+                # Handle date deserialization
+                if column.name.endswith('_date') and value:
+                    if isinstance(value, str):
+                        value = dt.fromisoformat(value).date()
+
+                setattr(translation, column.name, value)
+
+        return translation
+
+    def _should_cache(self, value: TranslationCache) -> bool:
+        """Validate if the translation should be cached"""
+        return value is not None and value.cache_key is not None
+
+    def get_translation(self, text: str, target_lang: str, source_lang:str=None, context: str=None) -> Optional[TranslationCache]:
+        """
+        Get the translation for a text in a specific language
+
+        Args:
+            text: The text to be translated
+            target_lang: The target language for the translation
+            source_lang: The source language of the text to be translated
+            context: Optional context for the translation
+
+        Returns:
+            TranslationCache instance if found, None otherwise
+        """
+
+        def creator_func(text: str, target_lang: str, source_lang: str=None, context: str=None) -> Optional[TranslationCache]:
+            # Generate cache key based on inputs
+            cache_key = self._generate_cache_key(text, target_lang, source_lang, context)
+
+            # Check if translation already exists in database
+            existing_translation = db.session.query(TranslationCache).filter_by(cache_key=cache_key).first()
+
+            if existing_translation:
+                # Update last used timestamp
+                existing_translation.last_used_at = dt.now(tz=tz.utc)
+                db.session.commit()
+                return existing_translation
+
+            # Translation not found in DB, need to create it
+            # Initialize translation service
+            translation_service = TranslationService(getattr(current_app, 'tenant_id', None))
+
+            # Get the translation and metrics
+            translated_text, metrics = translation_service.translate_text(
+                text_to_translate=text,
+                target_lang=target_lang,
+                source_lang=source_lang,
+                context=context
+            )
+
+            # Create new translation cache record
+            new_translation = TranslationCache(
+                cache_key=cache_key,
+                source_text=text,
+                translated_text=translated_text,
+                source_language=source_lang or 'auto',
+                target_language=target_lang,
+                context=context,
+                prompt_tokens=metrics.get('prompt_tokens', 0),
+                completion_tokens=metrics.get('completion_tokens', 0),
+                created_at=dt.now(tz=tz.utc),
+                created_by=getattr(current_user, 'id', None) if 'current_user' in globals() else None,
+                updated_at=dt.now(tz=tz.utc),
+                updated_by=getattr(current_user, 'id', None) if 'current_user' in globals() else None,
+                last_used_at=dt.now(tz=tz.utc)
+            )
+
+            # Save to database
+            db.session.add(new_translation)
+            db.session.commit()
+
+            return new_translation
+
+        return self.get(creator_func, text=text, target_lang=target_lang, source_lang=source_lang, context=context)
+
+    def invalidate_tenant_translations(self, tenant_id: int):
+        """Invalidate cached translations for specific tenant"""
+        self.invalidate(tenant_id=tenant_id)
+
+    def _generate_cache_key(self, text: str, target_lang: str, source_lang: str = None, context: str = None) -> str:
+        """Generate cache key for a translation"""
+        cache_data = {
+            "text": text.strip(),
+            "target_lang": target_lang.lower(),
+            "source_lang": source_lang.lower() if source_lang else None,
+            "context": context.strip() if context else None,
+        }
+
+        cache_string = json.dumps(cache_data, sort_keys=True, ensure_ascii=False)
+        return xxhash.xxh64(cache_string.encode('utf-8')).hexdigest()
+
+def register_translation_cache_handlers(cache_manager) -> None:
+    """Register translation cache handlers with cache manager"""
+    cache_manager.register_handler(
+        TranslationCacheHandler,
+        'eveai_model'  # Use existing eveai_model region
+    )

common/utils/chat_utils.py

@@ -1,14 +1,18 @@
+import json
+
 """
 Utility functions for chat customization.
 """
+from flask import current_app
+
 
 def get_default_chat_customisation(tenant_customisation=None):
     """
     Get chat customization options with default values for missing options.
 
     Args:
-        tenant_customization (dict, optional): The tenant's customization options.
-            Defaults to None.
+        tenant_customisation (dict or str, optional): The tenant's customization options.
+            Defaults to None. Can be a dict or a JSON string.
 
     Returns:
         dict: A dictionary containing all customization options with default values
@@ -37,9 +41,20 @@ def get_default_chat_customisation(tenant_customisation=None):
     # Start with the default customization
     customisation = default_customisation.copy()
 
+    # Convert JSON string to dict if needed
+    if isinstance(tenant_customisation, str):
+        try:
+            tenant_customisation = json.loads(tenant_customisation)
+            current_app.logger.debug(f"Converted JSON string to dict: {tenant_customisation}")
+        except json.JSONDecodeError as e:
+            current_app.logger.error(f"Error parsing JSON customisation: {e}")
+            return default_customisation
+
     # Update with tenant customization
-    for key, value in tenant_customisation.items():
-        if key in customisation:
-            customisation[key] = value
+    current_app.logger.debug(f"Tenant customisation - in default creation: {tenant_customisation}")
+    if tenant_customisation:
+        for key, value in tenant_customisation.items():
+            if key in customisation:
+                customisation[key] = value
 
     return customisation

common/utils/document_utils.py

@@ -3,7 +3,7 @@ from datetime import datetime as dt, timezone as tz
 from sqlalchemy import desc
 from sqlalchemy.exc import SQLAlchemyError
 from werkzeug.utils import secure_filename
-from common.models.document import Document, DocumentVersion, Catalog
+from common.models.document import Document, DocumentVersion, Catalog, Processor
 from common.extensions import db, minio_client
 from common.utils.celery_utils import current_celery
 from flask import current_app
@@ -11,6 +11,7 @@ import requests
 from urllib.parse import urlparse, unquote, urlunparse, parse_qs
 import os
 
+from config.type_defs.processor_types import PROCESSOR_TYPES
 from .config_field_types import normalize_json_field
 from .eveai_exceptions import (EveAIInvalidLanguageException, EveAIDoubleURLException, EveAIUnsupportedFileType,
                                EveAIInvalidCatalog, EveAIInvalidDocument, EveAIInvalidDocumentVersion, EveAIException)
@@ -469,3 +470,15 @@ def lookup_document(tenant_id: int, lookup_criteria: dict, metadata_type: str) -
             "Error during document lookup",
             status_code=500
         )
+
+def is_file_type_supported_by_catalog(catalog_id, file_type):
+    processors = Processor.query.filter_by(catalog_id=catalog_id).filter_by(active=True).all()
+
+    supported_file_types = []
+    for processor in processors:
+        processor_file_types = PROCESSOR_TYPES[processor.type]['file_types']
+        file_types = [f.strip() for f in processor_file_types.split(",")]
+        supported_file_types.extend(file_types)
+
+    if file_type not in supported_file_types:
+        raise EveAIUnsupportedFileType()

common/utils/eveai_exceptions.py

@@ -34,7 +34,25 @@ class EveAIDoubleURLException(EveAIException):
 class EveAIUnsupportedFileType(EveAIException):
     """Raised when an invalid file type is provided"""
 
-    def __init__(self, message="Filetype is not supported", status_code=400, payload=None):
+    def __init__(self, message="Filetype is not supported by current active processors", status_code=400, payload=None):
+        super().__init__(message, status_code, payload)
+
+
+class EveAINoProcessorFound(EveAIException):
+    """Raised when no processor is found for a given file type"""
+
+    def __init__(self, catalog_id, file_type, file_subtype, status_code=400, payload=None):
+        message = f"No active processor found for catalog {catalog_id} with file type {file_type} and subtype {file_subtype}"
+        super().__init__(message, status_code, payload)
+
+
+class EveAINoContentFound(EveAIException):
+    """Raised when no content is found for a given document"""
+
+    def __init__(self, document_id, document_version_id, status_code=400, payload=None):
+        self.document_id = document_id
+        self.document_version_id = document_version_id
+        message = f"No content found while processing Document with ID {document_id} and version {document_version_id}."
         super().__init__(message, status_code, payload)
 
 

config/config.py

@@ -148,7 +148,7 @@ class Config(object):
         },
     }
 
-    SUPPORTED_LANGUAGES_Full = list(SUPPORTED_LANGUAGE_DETAILS.keys())
+    SUPPORTED_LANGUAGES_FULL = list(SUPPORTED_LANGUAGE_DETAILS.keys())
 
     # supported currencies
     SUPPORTED_CURRENCIES = ['€', '$']
@@ -156,10 +156,7 @@ class Config(object):
     # supported LLMs
     # SUPPORTED_EMBEDDINGS = ['openai.text-embedding-3-small', 'openai.text-embedding-3-large', 'mistral.mistral-embed']
     SUPPORTED_EMBEDDINGS = ['mistral.mistral-embed']
-    SUPPORTED_LLMS = ['openai.gpt-4o', 'openai.gpt-4o-mini',
-                      'mistral.mistral-large-latest', 'mistral.mistral-medium_latest', 'mistral.mistral-small-latest']
-
-    ANTHROPIC_LLM_VERSIONS = {'claude-3-5-sonnet': 'claude-3-5-sonnet-20240620', }
+    SUPPORTED_LLMS = ['mistral.mistral-large-latest', 'mistral.mistral-medium_latest', 'mistral.mistral-small-latest']
 
     # Annotation text chunk length
     ANNOTATION_TEXT_CHUNK_LENGTH = 10000

config/customisations/globals/CHAT_CLIENT_CUSTOMISATION/1.0.0.yaml

@@ -56,11 +56,6 @@ configuration:
     description: "Sidebar Markdown-formatted Text"
     type: "text"
     required: false
-  "welcome_message":
-    name: "Welcome Message"
-    description: "Text to be shown as Welcome"
-    type: "text"
-    required: false
 metadata:
   author: "Josako"
   date_added: "2024-06-06"

config/processors/globals/AUTOMAGIC_HTML_PROCESSOR/1.0.0.yaml

@@ -0,0 +1,14 @@
+version: "1.0.0"
+name: "HTML Processor"
+file_types: "html"
+description: "A processor for HTML files, driven by AI"
+configuration:
+  custom_instructions:
+    name: "Custom Instructions"
+    description: "Some custom instruction to guide our AI agent in parsing your HTML file"
+    type: "text"
+    required: false
+metadata:
+  author: "Josako"
+  date_added: "2025-06-25"
+  description: "A processor for HTML files, driven by AI"

config/prompts/globals/automagic_html_parse/1.0.0.yaml

@@ -0,0 +1,30 @@
+version: "1.0.0"
+content: |
+  You are a top administrative assistant specialized in transforming given HTML into markdown formatted files. The 
+  generated files will be used to generate embeddings in a RAG-system.
+  
+  # Best practices are:
+  - Respect wordings and language(s) used in the HTML.
+  - The following items need to be considered: headings, paragraphs, listed items (numbered or not) and tables. Images can be neglected.
+  - Sub-headers can be used as lists. This is true when a header is followed by a series of sub-headers without content (paragraphs or listed items). Present those sub-headers as a list.  
+  - Be careful of encoding of the text. Everything needs to be human readable.
+
+  You only return relevant information, and filter out non-relevant information, such as:
+  - information found in menu bars, sidebars, footers or headers
+  - information in forms, buttons
+  
+  Process the file or text carefully, and take a stepped approach. The resulting markdown should be the result of the 
+  processing of the complete input html file. Answer with the pure markdown, without any other text.
+  
+  {custom_instructions}
+  
+  HTML to be processed is in between triple backquotes.
+  
+  ```{html}```
+
+llm_model: "mistral.mistral-small-latest"
+metadata:
+  author: "Josako"
+  date_added: "2025-06-25"
+  description: "An aid in transforming HTML-based inputs to markdown, fully automatic"
+  changes: "Initial version"

config/prompts/globals/translation_with_context/1.0.0.yaml

@@ -0,0 +1,15 @@
+version: "1.0.0"
+content: >
+  You are a top translator. We need you to translate {text_to_translate} into {target_language}, taking into account 
+  this context:
+  
+  {context}
+  
+  I only want you to return the translation. No explanation, no options. I need to be able to directly use your answer
+  without further interpretation. If more than one option is available, present me with the most probable one.
+llm_model: "mistral.ministral-8b-latest"
+metadata:
+  author: "Josako"
+  date_added: "2025-06-23"
+  description: "An assistant to translate given a context."
+  changes: "Initial version"

config/prompts/globals/translation_without_context/1.0.0.yaml

@@ -0,0 +1,12 @@
+version: "1.0.0"
+content: >
+  You are a top translator. We need you to translate {text_to_translate} into {target_language}.
+  
+  I only want you to return the translation. No explanation, no options. I need to be able to directly use your answer
+  without further interpretation. If more than one option is available, present me with the most probable one.
+llm_model: "mistral.ministral-8b-latest"
+metadata:
+  author: "Josako"
+  date_added: "2025-06-23"
+  description: "An assistant to translate without context."
+  changes: "Initial version"

config/type_defs/processor_types.py

@@ -24,5 +24,10 @@ PROCESSOR_TYPES = {
         "name": "DOCX Processor",
         "description": "A processor for DOCX files",
         "file_types": "docx",
-    }
+    },
+    "AUTOMAGIC_HTML_PROCESSOR": {
+        "name": "AutoMagic HTML Processor",
+        "description": "A processor for HTML files, driven by AI",
+        "file_types": "html, htm",
+    },
 }

config/type_defs/prompt_types.py

@@ -28,4 +28,12 @@ PROMPT_TYPES = {
         "name": "transcript",
         "description": "An assistant to transform a transcript to markdown.",
     },
+    "translation_with_context": {
+        "name": "translation_with_context",
+        "description": "An assistant to translate text with context",
+    },
+    "translation_without_context": {
+        "name": "translation_without_context",
+        "description": "An assistant to translate text without context",
+    },
 }

content/changelog/1.0/1.0.0.md

@@ -5,6 +5,36 @@ All notable changes to EveAI will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [2.3.8-alfa]
+
+### Added
+- Translation Service
+- Automagic HTML Processor 
+- Allowed languages defined at level of Tenant Make
+
+### Changed
+- For changes in existing functionality.
+- Allow to activate / de-activate Processors
+- Align all document views with session catalog
+- Allow different processor types to handle the same file types
+- Remove welcome message from tenant_make customisation, add to specialist configuration
+
+### Deprecated
+- For soon-to-be removed features.
+
+### Removed
+- For now removed features.
+
+### Fixed
+- Adapt TRAICIE_ROLE_DEFINITION_SPECIALIST to latest requirements
+- Allow for empty historical messages
+- Ensure client can cope with empty customisation options
+- Ensure only tenant-defined makes are selectable throughout the application
+- Refresh partner info when adding Partner Services
+
+### Security
+- In case of vulnerabilities.
+
 ## [2.3.7-alfa]
 
 ### Added

docker/compose_dev.yaml

@@ -24,7 +24,7 @@ x-common-variables: &common-variables
   FLOWER_PASSWORD: 'Jungles'
   OPENAI_API_KEY: 'sk-proj-8R0jWzwjL7PeoPyMhJTZT3BlbkFJLb6HfRB2Hr9cEVFWEhU7'
   GROQ_API_KEY: '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'
-  MISTRAL_API_KEY: 'jGDc6fkCbt0iOC0jQsbuZhcjLWBPGc2b'
+  MISTRAL_API_KEY: '0f4ZiQ1kIpgIKTHX8d0a8GOD2vAgVqEn'
   ANTHROPIC_API_KEY: 'sk-ant-api03-c2TmkzbReeGhXBO5JxNH6BJNylRDonc9GmZd0eRbrvyekec2'
   JWT_SECRET_KEY: 'bsdMkmQ8ObfMD52yAFg4trrvjgjMhuIqg2fjDpD/JqvgY0ccCcmlsEnVFmR79WPiLKEA3i8a5zmejwLZKl4v9Q=='
   API_ENCRYPTION_KEY: 'xfF5369IsredSrlrYZqkM9ZNrfUASYYS6TCcAR9UKj4='

docker/compose_test.yaml

@@ -26,7 +26,7 @@ x-common-variables: &common-variables
   REDIS_PORT: '6379'
   FLOWER_USER: 'Felucia'
   FLOWER_PASSWORD: 'Jungles'
-  MISTRAL_API_KEY: 'Vkwgr67vUs6ScKmcFF2QVw7uHKgq0WEN'
+  MISTRAL_API_KEY: 'qunKSaeOkFfLteNiUO77RCsXXSLK65Ec'
   JWT_SECRET_KEY: '7e9c8b3a215f4d6e90712c5d8f3b97a60e482c15f39a7d68bcd45910ef23a784'
   API_ENCRYPTION_KEY: 'kJ7N9p3IstyRGkluYTryM8ZMnfUBSXWR3TCfDG9VLc4='
   MINIO_ENDPOINT: minio:9000

eveai_app/__init__.py

@@ -201,8 +201,3 @@ def register_cache_handlers(app):
     register_specialist_cache_handlers(cache_manager)
     from common.utils.cache.license_cache import register_license_cache_handlers
     register_license_cache_handlers(cache_manager)
-
-
-
-
-

eveai_app/templates/document/document_versions.html

@@ -4,13 +4,13 @@
 {% block title %}Document Versions{% endblock %}
 
 {% block content_title %}Document Versions{% endblock %}
-{% block content_description %}View Versions for {{ document }}{% endblock %}
+{% block content_description %}View Versions for Document <b>{{ document }}</b>{% endblock %}
 {% block content_class %}<div class="col-xl-12 col-lg-5 col-md-7 mx-auto"></div>{% endblock %}
 
 {% block content %}
 <div class="container">
     <form method="POST" action="{{ url_for('document_bp.handle_document_version_selection') }}" id="documentVersionsForm">
-        {{ render_selectable_table(headers=["ID", "URL", "Object Name", "File Type", "Process.", "Proces. Start", "Proces. Finish", "Proces. Error"], rows=rows, selectable=True, id="versionsTable") }}
+        {{ render_selectable_table(headers=["ID", "File Type", "File Size", "Process.", "Proces. Start", "Proces. Finish", "Proces. Error"], rows=rows, selectable=True, id="versionsTable") }}
         <div class="form-group mt-3 d-flex justify-content-between">
             <div>
                 <button type="submit" name="action" value="edit_document_version" class="btn btn-primary" onclick="return validateTableSelection('documentVersionsForm')">Edit Document Version</button>

eveai_app/templates/document/documents.html

@@ -4,14 +4,13 @@
 {% block title %}Documents{% endblock %}
 
 {% block content_title %}Documents{% endblock %}
-{% block content_description %}View Documents for Tenant{% endblock %}
+{% block content_description %}View Documents for Catalog <b>{% if session.catalog_name %}{{ session.catalog_name }}{% else %}No Catalog{% endif %}</b>{% endblock %}
 {% block content_class %}<div class="col-xl-12 col-lg-5 col-md-7 mx-auto"></div>{% endblock %}
 
 {% block content %}
     <!-- Filter Form -->
     {% set filter_form %}
         <form method="GET" action="{{ url_for('document_bp.documents') }}">
-            {{ render_filter_field('catalog_id', 'Catalog', filter_options['catalog_id'], filters.get('catalog_id', [])) }}
             {{ render_filter_field('validity', 'Validity', filter_options['validity'], filters.get('validity', [])) }}
 
             <button type="submit" class="btn btn-primary">Apply Filters</button>
@@ -27,7 +26,6 @@
                 headers=[
                     {"text": "ID", "sort": "id"},
                     {"text": "Name", "sort": "name"},
-                    {"text": "Catalog", "sort": "catalog_name"},
                     {"text": "Valid From", "sort": "valid_from"},
                     {"text": "Valid To", "sort": "valid_to"}
                 ],

eveai_app/templates/document/edit_processor.html

@@ -4,7 +4,7 @@
 {% block title %}Edit Processor{% endblock %}
 
 {% block content_title %}Edit Processor{% endblock %}
-{% block content_description %}Edit a Processor (for a Catalog){% endblock %}
+{% block content_description %}Edit Processor for Catalog <b>{% if session.catalog_name %}{{ session.catalog_name }}{% else %}No Catalog{% endif %}</b>{% endblock %}
 
 {% block content %}
     <form method="post">

eveai_app/templates/document/edit_retriever.html

@@ -4,7 +4,7 @@
 {% block title %}Edit Retriever{% endblock %}
 
 {% block content_title %}Edit Retriever{% endblock %}
-{% block content_description %}Edit a Retriever (for a Catalog){% endblock %}
+{% block content_description %}Edit a Retriever for catalog <b>{% if session.catalog_name %}{{ session.catalog_name }}{% else %}No Catalog{% endif %}</b>{% endblock %}
 
 {% block content %}
     <form method="post">

eveai_app/templates/document/processor.html

@@ -4,7 +4,7 @@
 {% block title %}Processor Registration{% endblock %}
 
 {% block content_title %}Register Processor{% endblock %}
-{% block content_description %}Define a new processor (for a catalog){% endblock %}
+{% block content_description %}Define a new processor for Catalog <b>{% if session.catalog_name %}{{ session.catalog_name }}{% else %}No Catalog{% endif %}</b>{% endblock %}
 
 {% block content %}
     <form method="post">

eveai_app/templates/document/processors.html

@@ -4,13 +4,13 @@
 {% block title %}Processors{% endblock %}
 
 {% block content_title %}Processors{% endblock %}
-{% block content_description %}View Processors for Tenant{% endblock %}
+{% block content_description %}View Processors for Catalog <b>{% if session.catalog_name %}{{ session.catalog_name }}{% else %}No Catalog{% endif %}</b>{% endblock %}
 {% block content_class %}<div class="col-xl-12 col-lg-5 col-md-7 mx-auto"></div>{% endblock %}
 
 {% block content %}
 <div class="container">
     <form method="POST" action="{{ url_for('document_bp.handle_processor_selection') }}" id="processorsForm">
-        {{ render_selectable_table(headers=["Processor ID", "Name", "Type", "Catalog ID"], rows=rows, selectable=True, id="retrieversTable") }}
+        {{ render_selectable_table(headers=["Processor ID", "Name", "Type", "Active"], rows=rows, selectable=True, id="retrieversTable") }}
         <div class="form-group mt-3 d-flex justify-content-between">
             <div>
                 <button type="submit" name="action" value="edit_processor" class="btn btn-primary" onclick="return validateTableSelection('processorsForm')">Edit Processor</button>

eveai_app/templates/document/retriever.html

@@ -4,7 +4,7 @@
 {% block title %}Retriever Registration{% endblock %}
 
 {% block content_title %}Register Retriever{% endblock %}
-{% block content_description %}Define a new retriever (for a catalog){% endblock %}
+{% block content_description %}Define a new retriever for Catalog <b>{% if session.catalog_name %}{{ session.catalog_name }}{% else %}No Catalog{% endif %}</b>{% endblock %}
 
 {% block content %}
     <form method="post">

eveai_app/templates/document/retrievers.html

@@ -4,13 +4,13 @@
 {% block title %}Retrievers{% endblock %}
 
 {% block content_title %}Retrievers{% endblock %}
-{% block content_description %}View Retrievers for Tenant{% endblock %}
+{% block content_description %}View Retrievers for Catalog <b>{% if session.catalog_name %}{{ session.catalog_name }}{% else %}No Catalog{% endif %}</b>{% endblock %}
 {% block content_class %}<div class="col-xl-12 col-lg-5 col-md-7 mx-auto"></div>{% endblock %}
 
 {% block content %}
 <div class="container">
     <form method="POST" action="{{ url_for('document_bp.handle_retriever_selection') }}" id="retrieversForm">
-        {{ render_selectable_table(headers=["Retriever ID", "Name", "Type", "Catalog ID"], rows=rows, selectable=True, id="retrieversTable") }}
+        {{ render_selectable_table(headers=["Retriever ID", "Name", "Type"], rows=rows, selectable=True, id="retrieversTable") }}
         <div class="form-group mt-3 d-flex justify-content-between">
             <div>
                 <button type="submit" name="action" value="edit_retriever" class="btn btn-primary" onclick="return validateTableSelection('retrieversForm')">Edit Retriever</button>

eveai_app/views/document_forms.py

@@ -71,15 +71,6 @@ class ProcessorForm(FlaskForm):
     name = StringField('Name', validators=[DataRequired(), Length(max=50)])
     description = TextAreaField('Description', validators=[Optional()])
 
-    # Catalog for the Retriever
-    catalog = QuerySelectField(
-        'Catalog ID',
-        query_factory=lambda: Catalog.query.all(),
-        allow_blank=True,
-        get_label='name',
-        validators=[DataRequired()],
-    )
-
     # Select Field for Catalog Type (Uses the CATALOG_TYPES defined in config)
     type = SelectField('Processor Type', validators=[DataRequired()])
 
@@ -89,6 +80,7 @@ class ProcessorForm(FlaskForm):
                                   default=2000)
     max_chunk_size = IntegerField('Maximum Chunk Size (3000)', validators=[NumberRange(min=0), Optional()],
                                   default=3000)
+    active = BooleanField('Active', default=True)
     tuning = BooleanField('Enable Embedding Tuning', default=False)
 
     # Metadata fields
@@ -108,14 +100,6 @@ class EditProcessorForm(DynamicFormBase):
     name = StringField('Name', validators=[DataRequired(), Length(max=50)])
     description = TextAreaField('Description', validators=[Optional()])
 
-    # Catalog for the Retriever
-    catalog = QuerySelectField(
-        'Catalog ID',
-        query_factory=lambda: Catalog.query.all(),
-        allow_blank=True,
-        get_label='name',
-        validators=[Optional()],
-    )
     type = StringField('Processor Type', validators=[DataRequired()], render_kw={'readonly': True})
 
     sub_file_type = StringField('Sub File Type', validators=[Optional(), Length(max=50)])
@@ -124,6 +108,7 @@ class EditProcessorForm(DynamicFormBase):
                                   default=2000)
     max_chunk_size = IntegerField('Maximum Chunk Size (3000)', validators=[NumberRange(min=0), Optional()],
                                   default=3000)
+    active = BooleanField('Active', default=True)
     tuning = BooleanField('Enable Embedding Tuning', default=False)
 
     # Metadata fields
@@ -134,14 +119,7 @@ class EditProcessorForm(DynamicFormBase):
 class RetrieverForm(FlaskForm):
     name = StringField('Name', validators=[DataRequired(), Length(max=50)])
     description = TextAreaField('Description', validators=[Optional()])
-    # Catalog for the Retriever
-    catalog = QuerySelectField(
-        'Catalog ID',
-        query_factory=lambda: Catalog.query.all(),
-        allow_blank=True,
-        get_label='name',
-        validators=[Optional()],
-    )
+
     # Select Field for Retriever Type (Uses the RETRIEVER_TYPES defined in config)
     type = SelectField('Retriever Type', validators=[DataRequired()])
     tuning = BooleanField('Enable Tuning', default=False)
@@ -160,14 +138,7 @@ class RetrieverForm(FlaskForm):
 class EditRetrieverForm(DynamicFormBase):
     name = StringField('Name', validators=[DataRequired(), Length(max=50)])
     description = TextAreaField('Description', validators=[Optional()])
-    # Catalog for the Retriever
-    catalog = QuerySelectField(
-        'Catalog ID',
-        query_factory=lambda: Catalog.query.all(),
-        allow_blank=True,
-        get_label='name',
-        validators=[Optional()],
-    )
+
     # Select Field for Retriever Type (Uses the RETRIEVER_TYPES defined in config)
     type = StringField('Processor Type', validators=[DataRequired()], render_kw={'readonly': True})
     tuning = BooleanField('Enable Tuning', default=False)

eveai_app/views/document_list_view.py

@@ -1,5 +1,5 @@
-from datetime import datetime
-from flask import request, render_template, session
+from datetime import datetime as dt, timezone as tz
+from flask import request, render_template, session, current_app
 from sqlalchemy import desc, asc, or_, and_, cast, Integer
 from common.models.document import Document, Catalog
 from common.utils.filtered_list_view import FilteredListView
@@ -7,31 +7,19 @@ from common.utils.view_assistants import prepare_table_for_macro
 
 
 class DocumentListView(FilteredListView):
-    allowed_filters = ['catalog_id', 'validity']
-    allowed_sorts = ['id', 'name', 'catalog_name', 'valid_from', 'valid_to']
+    allowed_filters = ['validity']
+    allowed_sorts = ['id', 'name', 'valid_from', 'valid_to']
 
     def get_query(self):
-        return Document.query.join(Catalog).add_columns(
-            Document.id,
-            Document.name,
-            Catalog.name.label('catalog_name'),
-            Document.valid_from,
-            Document.valid_to
-        )
+        catalog_id = session.get('catalog_id')
+        current_app.logger.debug(f"Catalog ID: {catalog_id}")
+        return Document.query.filter_by(catalog_id=catalog_id)
 
     def apply_filters(self, query):
         filters = request.args.to_dict(flat=False)
 
-        if 'catalog_id' in filters:
-            catalog_ids = filters['catalog_id']
-            if catalog_ids:
-                # Convert catalog_ids to a list of integers
-                catalog_ids = [int(cid) for cid in catalog_ids if cid.isdigit()]
-                if catalog_ids:
-                    query = query.filter(Document.catalog_id.in_(catalog_ids))
-
         if 'validity' in filters:
-            now = datetime.utcnow().date()
+            now = dt.now(tz.utc).date()
             if 'valid' in filters['validity']:
                 query = query.filter(
                     and_(
@@ -47,10 +35,7 @@ class DocumentListView(FilteredListView):
         sort_order = request.args.get('sort_order', 'asc')
 
         if sort_by in self.allowed_sorts:
-            if sort_by == 'catalog_name':
-                column = Catalog.name
-            else:
-                column = getattr(Document, sort_by)
+            column = getattr(Document, sort_by)
 
             if sort_order == 'asc':
                 query = query.order_by(asc(column))
@@ -61,42 +46,39 @@ class DocumentListView(FilteredListView):
 
     def get(self):
         query = self.get_query()
-        query = self.apply_filters(query)
-        query = self.apply_sorting(query)
+        # query = self.apply_filters(query)
+        # query = self.apply_sorting(query)
         pagination = self.paginate(query)
 
         def format_date(date):
-            if isinstance(date, datetime):
+            if isinstance(date, dt):
                 return date.strftime('%Y-%m-%d')
             elif isinstance(date, str):
                 return date
             else:
                 return ''
 
+        current_app.logger.debug(f"Items retrieved: {pagination.items}")
         rows = [
             [
                 {'value': item.id, 'class': '', 'type': 'text'},
                 {'value': item.name, 'class': '', 'type': 'text'},
-                {'value': item.catalog_name, 'class': '', 'type': 'text'},
                 {'value': format_date(item.valid_from), 'class': '', 'type': 'text'},
                 {'value': format_date(item.valid_to), 'class': '', 'type': 'text'}
             ] for item in pagination.items
         ]
 
-        catalogs = Catalog.query.all()
-
         context = {
             'rows': rows,
             'pagination': pagination,
             'filters': request.args.to_dict(flat=False),
             'sort_by': request.args.get('sort_by', 'id'),
             'sort_order': request.args.get('sort_order', 'asc'),
-            'filter_options': self.get_filter_options(catalogs)
+            'filter_options': self.get_filter_options()
         }
         return render_template(self.template, **context)
 
-    def get_filter_options(self, catalogs):
+    def get_filter_options(self):
         return {
-            'catalog_id': [(str(cat.id), cat.name) for cat in catalogs],
             'validity': [('valid', 'Valid'), ('all', 'All')]
-        }
+        }

eveai_app/views/document_views.py

@@ -16,7 +16,7 @@ from common.extensions import db, cache_manager, minio_client
 from common.models.interaction import Specialist, SpecialistRetriever
 from common.utils.document_utils import create_document_stack, start_embedding_task, process_url, \
     edit_document, \
-    edit_document_version, refresh_document, clean_url
+    edit_document_version, refresh_document, clean_url, is_file_type_supported_by_catalog
 from common.utils.dynamic_field_utils import create_default_config_from_type_config
 from common.utils.eveai_exceptions import EveAIInvalidLanguageException, EveAIUnsupportedFileType, \
     EveAIDoubleURLException, EveAIException
@@ -110,7 +110,6 @@ def handle_catalog_selection():
         current_app.logger.info(f'Setting session catalog to {catalog.name}')
         session['catalog_id'] = catalog_id
         session['catalog_name'] = catalog.name
-        current_app.logger.info(f'Finished setting session catalog to {catalog.name}')
     elif action == 'edit_catalog':
         return redirect(prefixed_url_for('document_bp.edit_catalog', catalog_id=catalog_id))
 
@@ -157,7 +156,7 @@ def processor():
         tenant_id = session.get('tenant').get('id')
         new_processor = Processor()
         form.populate_obj(new_processor)
-        new_processor.catalog_id = form.catalog.data.id
+        new_processor.catalog_id = session.get('catalog_id')
         processor_config = cache_manager.processors_config_cache.get_config(new_processor.type)
         new_processor.configuration = create_default_config_from_type_config(
             processor_config["configuration"])
@@ -204,9 +203,6 @@ def edit_processor(processor_id):
         form.populate_obj(processor)
         processor.configuration = form.get_dynamic_data('configuration')
 
-        # Update catalog relationship
-        processor.catalog_id = form.catalog.data.id if form.catalog.data else None
-
         # Update logging information
         update_logging_information(processor, dt.now(tz.utc))
 
@@ -235,14 +231,19 @@ def processors():
     page = request.args.get('page', 1, type=int)
     per_page = request.args.get('per_page', 10, type=int)
 
-    query = Processor.query.order_by(Processor.id)
+    catalog_id = session.get('catalog_id', None)
+    if not catalog_id:
+        flash('You need to set a Session Catalog before adding Documents or URLs', 'warning')
+        return redirect(prefixed_url_for('document_bp.catalogs'))
+
+    query = Processor.query.filter_by(catalog_id=catalog_id).order_by(Processor.id)
 
     pagination = query.paginate(page=page, per_page=per_page)
     the_processors = pagination.items
 
     # prepare table data
     rows = prepare_table_for_macro(the_processors,
-                                   [('id', ''), ('name', ''), ('type', ''), ('catalog_id', '')])
+                                   [('id', ''), ('name', ''), ('type', ''), ('active', '')])
 
     # Render the catalogs in a template
     return render_template('document/processors.html', rows=rows, pagination=pagination)
@@ -272,7 +273,7 @@ def retriever():
         tenant_id = session.get('tenant').get('id')
         new_retriever = Retriever()
         form.populate_obj(new_retriever)
-        new_retriever.catalog_id = form.catalog.data.id
+        new_retriever.catalog_id = session.get('catalog_id')
         new_retriever.type_version = cache_manager.retrievers_version_tree_cache.get_latest_version(
             new_retriever.type)
 
@@ -301,12 +302,6 @@ def edit_retriever(retriever_id):
     # Get the retriever or return 404
     retriever = Retriever.query.get_or_404(retriever_id)
 
-    if retriever.catalog_id:
-        # If catalog_id is just an ID, fetch the Catalog object
-        retriever.catalog = Catalog.query.get(retriever.catalog_id)
-    else:
-        retriever.catalog = None
-
     # Create form instance with the retriever
     form = EditRetrieverForm(request.form, obj=retriever)
 
@@ -319,9 +314,6 @@ def edit_retriever(retriever_id):
         form.populate_obj(retriever)
         retriever.configuration = form.get_dynamic_data('configuration')
 
-        # Update catalog relationship
-        retriever.catalog_id = form.catalog.data.id if form.catalog.data else None
-
         # Update logging information
         update_logging_information(retriever, dt.now(tz.utc))
 
@@ -350,14 +342,19 @@ def retrievers():
     page = request.args.get('page', 1, type=int)
     per_page = request.args.get('per_page', 10, type=int)
 
-    query = Retriever.query.order_by(Retriever.id)
+    catalog_id = session.get('catalog_id', None)
+    if not catalog_id:
+        flash('You need to set a Session Catalog before adding Documents or URLs', 'warning')
+        return redirect(prefixed_url_for('document_bp.catalogs'))
+
+    query = Retriever.query.filter_by(catalog_id=catalog_id).order_by(Retriever.id)
 
     pagination = query.paginate(page=page, per_page=per_page)
     the_retrievers = pagination.items
 
     # prepare table data
     rows = prepare_table_for_macro(the_retrievers,
-                                   [('id', ''), ('name', ''), ('type', ''), ('catalog_id', '')])
+                                   [('id', ''), ('name', ''), ('type', '')])
 
     # Render the catalogs in a template
     return render_template('document/retrievers.html', rows=rows, pagination=pagination)
@@ -400,6 +397,8 @@ def add_document():
             filename = secure_filename(file.filename)
             extension = filename.rsplit('.', 1)[1].lower()
 
+            is_file_type_supported_by_catalog(catalog_id, extension)
+
             catalog_properties = form.get_dynamic_data("tagging_fields")
 
             api_input = {
@@ -451,6 +450,8 @@ def add_url():
 
             file_content, filename, extension = process_url(url, tenant_id)
 
+            is_file_type_supported_by_catalog(catalog_id, extension)
+
             catalog_properties = {}
             full_config = cache_manager.catalogs_config_cache.get_config(catalog.type)
             document_version_configurations = full_config['document_version_configurations']
@@ -489,6 +490,11 @@ def add_url():
 @document_bp.route('/documents', methods=['GET', 'POST'])
 @roles_accepted('Super User', 'Partner Admin', 'Tenant Admin')
 def documents():
+    catalog_id = session.get('catalog_id', None)
+    if not catalog_id:
+        flash('You need to set a Session Catalog before adding Documents or URLs', 'warning')
+        return redirect(prefixed_url_for('document_bp.catalogs'))
+
     view = DocumentListView(Document, 'document/documents.html', per_page=10)
     return view.get()
 
@@ -609,7 +615,7 @@ def edit_document_version_view(document_version_id):
 @roles_accepted('Super User', 'Partner Admin', 'Tenant Admin')
 def document_versions(document_id):
     doc = Document.query.get_or_404(document_id)
-    doc_desc = f'Document {doc.name}'
+    doc_desc = f'{doc.name}'
 
     page = request.args.get('page', 1, type=int)
     per_page = request.args.get('per_page', 10, type=int)
@@ -621,8 +627,7 @@ def document_versions(document_id):
     pagination = query.paginate(page=page, per_page=per_page, error_out=False)
     doc_langs = pagination.items
 
-    rows = prepare_table_for_macro(doc_langs, [('id', ''), ('url', ''),
-                                               ('object_name', ''), ('file_type', ''),
+    rows = prepare_table_for_macro(doc_langs, [('id', ''), ('file_type', ''), ('file_size', ''),
                                                ('processing', ''), ('processing_started_at', ''),
                                                ('processing_finished_at', ''), ('processing_error', '')])
 

eveai_app/views/dynamic_form_base.py

@@ -312,7 +312,7 @@ class DynamicFormBase(FlaskForm):
                 field_class = SelectField
                 tenant_id = session.get('tenant').get('id')
                 makes = TenantMake.query.filter_by(tenant_id=tenant_id).all()
-                choices = [(make.name, make.name) for make in makes]
+                choices = [(make.id, make.name) for make in makes]
                 extra_classes = ''
                 field_kwargs = {'choices': choices}
 
@@ -328,6 +328,16 @@ class DynamicFormBase(FlaskForm):
             initial_data: Optional initial data for the fields
         """
         current_app.logger.debug(f"Adding dynamic fields for collection {collection_name} with config: {config}")
+
+        if isinstance(initial_data, str):
+            try:
+                initial_data = json.loads(initial_data)
+            except (json.JSONDecodeError, TypeError):
+                current_app.logger.error(f"Invalid JSON in initial_data: {initial_data}")
+                initial_data = {}
+        elif initial_data is None:
+            initial_data = {}
+
         # Store the full configuration for later use in get_list_type_configs_js
         if not hasattr(self, '_full_configs'):
             self._full_configs = {}
@@ -581,7 +591,10 @@ class DynamicFormBase(FlaskForm):
                 except Exception as e:
                     current_app.logger.error(f"Error converting initial data to patterns: {e}")
             elif isinstance(field, DateField):
-                data[original_field_name] = field.data.isoformat()
+                if field.data:
+                    data[original_field_name] = field.data.isoformat()
+                else:
+                    data[original_field_name] = None
             else:
                 data[original_field_name] = field.data
         return data

eveai_app/views/interaction_forms.py

@@ -1,3 +1,4 @@
+from flask import session
 from flask_wtf import FlaskForm
 from wtforms import (StringField, BooleanField, SelectField, TextAreaField)
 from wtforms.fields.datetime import DateField
@@ -181,7 +182,8 @@ class EditSpecialistMagicLinkForm(DynamicFormBase):
             self.specialist_name.data = ''
 
         # Dynamically populate the tenant_make field with None as first option
-        tenant_makes = TenantMake.query.all()
+        tenant_id = session.get('tenant').get('id')
+        tenant_makes = TenantMake.query.filter_by(tenant_id=tenant_id).all()
         self.tenant_make_id.choices = [(0, 'None')] + [(make.id, make.name) for make in tenant_makes]
 
 

eveai_app/views/interaction_views.py

@@ -702,12 +702,13 @@ def specialist_magic_link():
             new_spec_ml_tenant.tenant_id = tenant_id
 
             # Define the make valid for this magic link
-            make_id = SpecialistServices.get_specialist_system_field(new_specialist_magic_link.specialist_id,
-                                                                     "make", "tenant_make")
+            specialist = Specialist.query.get(new_specialist_magic_link.specialist_id)
+            make_id = specialist.configuration.get('make', None)
+            current_app.logger.debug(f"make_id defined in specialist: {make_id}")
             if make_id:
-                new_spec_ml_tenant.tenant_make_id = make_id
+                new_specialist_magic_link.tenant_make_id = make_id
             elif session.get('tenant').get('default_tenant_make_id'):
-                new_spec_ml_tenant.tenant_make_id = session.get('tenant').get('default_tenant_make_id')
+                new_specialist_magic_link.tenant_make_id = session.get('tenant').get('default_tenant_make_id')
 
             db.session.add(new_specialist_magic_link)
             db.session.add(new_spec_ml_tenant)

eveai_app/views/partner_views.py

@@ -62,6 +62,7 @@ def edit_partner(partner_id):
         update_logging_information(partner, dt.now(tz.utc))
         db.session.commit()
         flash('Partner updated successfully.', 'success')
+        refresh_session_partner(partner.id)
         return redirect(
             prefixed_url_for('partner_bp.edit_partner',
                              partner_id=partner.id))  # Assuming there's a user profile view to redirect to
@@ -197,6 +198,7 @@ def edit_partner_service(partner_service_id):
             db.session.commit()
             flash('Partner Service updated successfully.', 'success')
             current_app.logger.info(f"Partner Service {partner_service.name} updated successfully! ")
+            refresh_session_partner(partner_id)
         except SQLAlchemyError as e:
             db.session.rollback()
             flash(f'Failed to update Partner Service: {str(e)}', 'danger')
@@ -339,4 +341,7 @@ def add_partner_service_for_tenant(partner_service_id):
         return redirect(prefixed_url_for('partner_bp.partner_services'))
 
 
-
+def refresh_session_partner(partner_id):
+    if session.get('partner', None):
+        if partner_id == session['partner']['id']:
+            session['partner'] = Partner.query.get_or_404(partner_id).to_dict()

eveai_app/views/user_forms.py

@@ -196,6 +196,14 @@ class TenantMakeForm(DynamicFormBase):
     active = BooleanField('Active', validators=[Optional()], default=True)
     website = StringField('Website', validators=[DataRequired(), Length(max=255)])
     logo_url = StringField('Logo URL', validators=[Optional(), Length(max=255)])
+    allowed_languages = SelectMultipleField('Allowed Languages', choices=[], validators=[Optional()])
+
+    def __init__(self, *args, **kwargs):
+        super(TenantMakeForm, self).__init__(*args, **kwargs)
+        # Initialiseer de taalopties met taalcodes en vlaggen
+        lang_details = current_app.config['SUPPORTED_LANGUAGE_DETAILS']
+        self.allowed_languages.choices = [(details['iso 639-1'], f"{details['flag']} {details['iso 639-1']}") 
+                                         for name, details in lang_details.items()]
 
 class EditTenantMakeForm(DynamicFormBase):
     id = IntegerField('ID', widget=HiddenInput())
@@ -204,6 +212,14 @@ class EditTenantMakeForm(DynamicFormBase):
     active = BooleanField('Active', validators=[Optional()], default=True)
     website = StringField('Website', validators=[DataRequired(), Length(max=255)])
     logo_url = StringField('Logo URL', validators=[Optional(), Length(max=255)])
+    allowed_languages = SelectMultipleField('Allowed Languages', choices=[], validators=[Optional()])
+
+    def __init__(self, *args, **kwargs):
+        super(EditTenantMakeForm, self).__init__(*args, **kwargs)
+        # Initialiseer de taalopties met taalcodes en vlaggen
+        lang_details = current_app.config['SUPPORTED_LANGUAGE_DETAILS']
+        self.allowed_languages.choices = [(details['iso 639-1'], f"{details['flag']} {details['iso 639-1']}") 
+                                         for name, details in lang_details.items()]
 
 
 

eveai_app/views/user_views.py

@@ -655,6 +655,8 @@ def tenant_make():
         new_tenant_make.tenant_id = tenant_id
         customisation_options = form.get_dynamic_data("configuration")
         new_tenant_make.chat_customisation_options = json.dumps(customisation_options)
+        # Verwerk allowed_languages als array
+        new_tenant_make.allowed_languages = form.allowed_languages.data if form.allowed_languages.data else None
         set_logging_information(new_tenant_make, dt.now(tz.utc))
 
         try:
@@ -703,6 +705,10 @@ def edit_tenant_make(tenant_make_id):
     # Create form instance with the tenant make
     form = EditTenantMakeForm(request.form, obj=tenant_make)
 
+    # Initialiseer de allowed_languages selectie met huidige waarden
+    if tenant_make.allowed_languages:
+        form.allowed_languages.data = tenant_make.allowed_languages
+
     customisation_config = cache_manager.customisations_config_cache.get_config("CHAT_CLIENT_CUSTOMISATION")
     form.add_dynamic_fields("configuration", customisation_config, tenant_make.chat_customisation_options)
 
@@ -710,6 +716,8 @@ def edit_tenant_make(tenant_make_id):
         # Update basic fields
         form.populate_obj(tenant_make)
         tenant_make.chat_customisation_options = form.get_dynamic_data("configuration")
+        # Verwerk allowed_languages als array
+        tenant_make.allowed_languages = form.allowed_languages.data if form.allowed_languages.data else None
 
         # Update logging information
         update_logging_information(tenant_make, dt.now(tz.utc))

eveai_chat_client/static/assets/js/chat-app.js

@@ -131,18 +131,20 @@ export const ChatApp = {
             const historicalMessages = chatConfig.messages || [];
 
             if (historicalMessages.length > 0) {
-                this.allMessages = historicalMessages.map(msg => {
-                    // Zorg voor een correct geformatteerde bericht-object
-                    return {
-                        id: this.messageIdCounter++,
-                        content: typeof msg === 'string' ? msg : msg.content || '',
-                        sender: msg.sender || 'ai',
-                        type: msg.type || 'text',
-                        timestamp: msg.timestamp || new Date().toISOString(),
-                        formData: msg.formData || null,
-                        status: msg.status || 'delivered'
-                    };
-                });
+                this.allMessages = historicalMessages
+                    .filter(msg => msg !== null && msg !== undefined) // Filter null/undefined berichten uit
+                    .map(msg => {
+                        // Zorg voor een correct geformatteerde bericht-object
+                        return {
+                            id: this.messageIdCounter++,
+                            content: typeof msg === 'string' ? msg : (msg.content || ''),
+                            sender: msg.sender || 'ai',
+                            type: msg.type || 'text',
+                            timestamp: msg.timestamp || new Date().toISOString(),
+                            formData: msg.formData || null,
+                            status: msg.status || 'delivered'
+                        };
+                    });
 
                 console.log(`Loaded ${this.allMessages.length} historical messages`);
             }

eveai_chat_client/views/chat_views.py

@@ -86,7 +86,13 @@ def chat(magic_link_code):
         session['chat_session_id'] = SpecialistServices.start_session()
 
         # Get customisation options with defaults
-        customisation = get_default_chat_customisation(tenant_make.chat_customisation_options)
+        current_app.logger.debug(f"Make Customisation Options: {tenant_make.chat_customisation_options}")
+        try:
+            customisation = get_default_chat_customisation(tenant_make.chat_customisation_options)
+        except Exception as e:
+            current_app.logger.error(f"Error processing customisation options: {str(e)}")
+            # Fallback to default customisation
+            customisation = get_default_chat_customisation(None)
 
         # Start a new chat session
         session['chat_session_id'] = SpecialistServices.start_session()

eveai_chat_workers/specialists/traicie/TRAICIE_ROLE_DEFINITION_SPECIALIST/1_3.py

@@ -69,6 +69,9 @@ class SpecialistExecutor(CrewAIBaseSpecialistExecutor):
             self.role_definition_crew
         )
 
+    def _config_state_result_relations(self):
+        pass
+
     def execute(self, arguments: SpecialistArguments, formatted_context, citations) -> SpecialistResult:
         self.log_tuning("Traicie Role Definition Specialist execution started", {})
 

eveai_workers/processors/__init__.py

@@ -1,5 +1,5 @@
 # Import all processor implementations to ensure registration
-from . import audio_processor, html_processor, pdf_processor, markdown_processor, docx_processor
+from . import audio_processor, html_processor, pdf_processor, markdown_processor, docx_processor, automagic_html_processor
 
 # List of all available processor implementations
-__all__ = ['audio_processor', 'html_processor', 'pdf_processor', 'markdown_processor', 'docx_processor']
+__all__ = ['audio_processor', 'html_processor', 'pdf_processor', 'markdown_processor', 'docx_processor', 'automagic_html_processor']

eveai_workers/processors/automagic_html_processor.py

@@ -0,0 +1,65 @@
+import io
+import pdfplumber
+from flask import current_app
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+from langchain_core.output_parsers import StrOutputParser
+from langchain_core.prompts import ChatPromptTemplate
+import re
+from langchain_core.runnables import RunnablePassthrough
+
+from common.eveai_model.tracked_mistral_ocr_client import TrackedMistralOcrClient
+from common.extensions import minio_client
+from common.utils.model_utils import create_language_template, get_embedding_llm, get_template
+from .base_processor import BaseProcessor
+from common.utils.business_event_context import current_event
+from .processor_registry import ProcessorRegistry
+
+
+class AutomagicHTMLProcessor(BaseProcessor):
+    def __init__(self, tenant, document_version, catalog, processor):
+        super().__init__(tenant, document_version, catalog, processor)
+
+        self.chunk_size = catalog.max_chunk_size
+        self.chunk_overlap = 0
+        self.tuning = self.processor.tuning
+
+        self.prompt_params = {
+            "custom_instructions": self.processor.configuration.get("custom_instructions", ""),
+        }
+        template, llm = get_template("automagic_html_parse")
+
+        translation_prompt = ChatPromptTemplate.from_template(template)
+        setup = RunnablePassthrough()
+        output_parser = StrOutputParser()
+        self.chain = (setup | translation_prompt | llm | output_parser)
+
+
+    def process(self):
+        self._log("Starting Automagic HTML processing")
+        try:
+            # Get HTML-file data
+            file_data = minio_client.download_document_file(
+                self.tenant.id,
+                self.document_version.bucket_name,
+                self.document_version.object_name,
+            )
+
+            # Invoke HTML Processing Agent
+            self.prompt_params["html"] = file_data
+            with current_event.create_span("Markdown Generation"):
+                markdown = self.chain.invoke(self.prompt_params)
+            self._save_markdown(markdown)
+
+            # Retrieve Title
+            match = re.search(r'^# (.+)', markdown, re.MULTILINE)
+            title = match.group(1).strip() if match else None
+
+            self._log("Finished Automagic HTML Processing")
+            return markdown, title
+        except Exception as e:
+            self._log(f"Error automagically processing HTML: {str(e)}", level='error')
+            raise
+
+
+# Register the processor
+ProcessorRegistry.register("AUTOMAGIC_HTML_PROCESSOR", AutomagicHTMLProcessor)

eveai_workers/processors/pdf_processor.py

@@ -44,185 +44,6 @@ class PDFProcessor(BaseProcessor):
             self._log(f"Error processing PDF: {str(e)}", level='error')
             raise
 
-    def _extract_content(self, file_data):
-        extracted_content = []
-        with pdfplumber.open(io.BytesIO(file_data)) as pdf:
-            figure_counter = 1
-            for page_num, page in enumerate(pdf.pages):
-                self._log(f"Extracting content from page {page_num + 1}")
-                page_content = {
-                    'text': page.extract_text(),
-                    'figures': self._extract_figures(page, page_num, figure_counter),
-                    'tables': self._extract_tables(page)
-                }
-                self.log_tuning("_extract_content", {"page_num": page_num, "page_content": page_content})
-                figure_counter += len(page_content['figures'])
-                extracted_content.append(page_content)
-
-        return extracted_content
-
-    def _extract_figures(self, page, page_num, figure_counter):
-        figures = []
-        # Omit figure processing for now!
-        # for img in page.images:
-        #     try:
-        #         # Try to get the bbox, use full page dimensions if not available
-        #         bbox = img.get('bbox', (0, 0, page.width, page.height))
-        #
-        #         figure = {
-        #             'figure_number': figure_counter,
-        #             'filename': f"figure_{page_num + 1}_{figure_counter}.png",
-        #             'caption': self._find_figure_caption(page, bbox)
-        #         }
-        #
-        #         # Extract the figure as an image
-        #         figure_image = page.within_bbox(bbox).to_image()
-        #
-        #         # Save the figure using MinIO
-        #         with io.BytesIO() as output:
-        #             figure_image.save(output, format='PNG')
-        #             output.seek(0)
-        #             minio_client.upload_document_file(
-        #                 self.tenant.id,
-        #                 self.document_version.doc_id,
-        #                 self.document_version.language,
-        #                 self.document_version.id,
-        #                 figure['filename'],
-        #                 output.getvalue()
-        #             )
-        #
-        #         figures.append(figure)
-        #         figure_counter += 1
-        #     except Exception as e:
-        #         self._log(f"Error processing figure on page {page_num + 1}: {str(e)}", level='error')
-
-        return figures
-
-    def _find_figure_caption(self, page, bbox):
-        try:
-            # Look for text below the figure
-            caption_bbox = (bbox[0], bbox[3], bbox[2], min(bbox[3] + 50, page.height))
-            caption_text = page.crop(caption_bbox).extract_text()
-            if caption_text and caption_text.lower().startswith('figure'):
-                return caption_text
-        except Exception as e:
-            self._log(f"Error finding figure caption: {str(e)}", level='error')
-        return None
-
-    def _extract_tables(self, page):
-        tables = []
-        try:
-            for table in page.extract_tables():
-                if table:
-                    markdown_table = self._table_to_markdown(table)
-                    if markdown_table:  # Only add non-empty tables
-                        tables.append(markdown_table)
-                        self.log_tuning("_extract_tables", {"markdown_table": markdown_table})
-        except Exception as e:
-            self._log(f"Error extracting tables from page: {str(e)}", level='error')
-        return tables
-
-    def _table_to_markdown(self, table):
-        if not table or not table[0]:  # Check if table is empty or first row is empty
-            return ""  # Return empty string for empty tables
-
-        def clean_cell(cell):
-            if cell is None:
-                return ""  # Convert None to empty string
-            return str(cell).replace("|", "\\|")  # Escape pipe characters and convert to string
-
-        header = [clean_cell(cell) for cell in table[0]]
-        markdown = "| " + " | ".join(header) + " |\n"
-        markdown += "| " + " | ".join(["---"] * len(header)) + " |\n"
-
-        for row in table[1:]:
-            cleaned_row = [clean_cell(cell) for cell in row]
-            markdown += "| " + " | ".join(cleaned_row) + " |\n"
-
-        return markdown
-
-    def _structure_content(self, extracted_content):
-        structured_content = ""
-        title = "Untitled Document"
-        current_heading_level = 0
-        heading_pattern = re.compile(r'^(\d+(\.\d+)*\.?\s*)?(.+)$')
-
-        def identify_heading(text):
-            match = heading_pattern.match(text.strip())
-            if match:
-                numbering, _, content = match.groups()
-                if numbering:
-                    level = numbering.count('.') + 1
-                    return level, f"{numbering}{content}"
-                else:
-                    return 1, content  # Assume it's a top-level heading if no numbering
-            return 0, text  # Not a heading
-
-        for page in extracted_content:
-            # Assume the title is on the first page
-            if page == extracted_content[0]:
-                lines = page.get('text', '').split('\n')
-                if lines:
-                    title = lines[0].strip()  # Use the first non-empty line as the title
-
-            # Process text
-            paragraphs = page['text'].split('\n\n')
-
-            for para in paragraphs:
-                lines = para.strip().split('\n')
-                if len(lines) == 1:  # Potential heading
-                    level, text = identify_heading(lines[0])
-                    if level > 0:
-                        heading_marks = '#' * level
-                        structured_content += f"\n\n{heading_marks} {text}\n\n"
-                        if level == 1 and not title:
-                            title = text  # Use the first top-level heading as the title if not set
-                    else:
-                        structured_content += f"{para}\n\n"  # Treat as normal paragraph
-                else:
-                    structured_content += f"{para}\n\n"  # Multi-line paragraph
-
-            # Process figures
-            for figure in page.get('figures', []):
-                structured_content += f"\n\n![Figure {figure['figure_number']}]({figure['filename']})\n\n"
-                if figure['caption']:
-                    structured_content += f"*Figure {figure['figure_number']}: {figure['caption']}*\n\n"
-
-            # Add tables
-            if 'tables' in page:
-                for table in page['tables']:
-                    structured_content += f"\n{table}\n"
-
-        if self.tuning:
-            self._save_intermediate(structured_content, "structured_content.md")
-
-        return structured_content, title
-
-    def _split_content_for_llm(self, content):
-        text_splitter = RecursiveCharacterTextSplitter(
-            chunk_size=self.chunk_size,
-            chunk_overlap=self.chunk_overlap,
-            length_function=len,
-            separators=["\n\n", "\n", " ", ""]
-        )
-        return text_splitter.split_text(content)
-
-    def _process_chunks_with_llm(self, chunks):
-        template, llm = get_template('pdf_parse')
-        pdf_prompt = ChatPromptTemplate.from_template(template)
-        setup = RunnablePassthrough()
-        output_parser = StrOutputParser()
-        chain = setup | pdf_prompt | llm | output_parser
-
-        markdown_chunks = []
-        for chunk in chunks:
-            input = {"pdf_content": chunk}
-            result = chain.invoke(input)
-            result = self._clean_markdown(result)
-            markdown_chunks.append(result)
-
-        return "\n\n".join(markdown_chunks)
-
 
 # Register the processor
 ProcessorRegistry.register("PDF_PROCESSOR", PDFProcessor)

eveai_workers/tasks.py

@@ -11,6 +11,7 @@ from langchain_core.prompts import ChatPromptTemplate
 from langchain_core.runnables import RunnablePassthrough
 from sqlalchemy import or_
 from sqlalchemy.exc import SQLAlchemyError
+import traceback
 
 from common.extensions import db, cache_manager
 from common.models.document import DocumentVersion, Embedding, Document, Processor, Catalog
@@ -24,7 +25,8 @@ from common.utils.business_event_context import current_event
 from config.type_defs.processor_types import PROCESSOR_TYPES
 from eveai_workers.processors.processor_registry import ProcessorRegistry
 
-from common.utils.eveai_exceptions import EveAIInvalidEmbeddingModel
+from common.utils.eveai_exceptions import EveAIInvalidEmbeddingModel, EveAINoContentFound, EveAIUnsupportedFileType, \
+    EveAINoProcessorFound
 
 from common.utils.config_field_types import json_to_pattern_list
 
@@ -58,8 +60,8 @@ def create_embeddings(tenant_id, document_version_id):
         catalog = Catalog.query.get_or_404(catalog_id)
 
         # Define processor related information
-        processor_type, processor_class = ProcessorRegistry.get_processor_for_file_type(document_version.file_type)
         processor = get_processor_for_document(catalog_id, document_version.file_type, document_version.sub_file_type)
+        processor_class = ProcessorRegistry.get_processor_class(processor.type)
 
     except Exception as e:
         current_app.logger.error(f'Create Embeddings request received '
@@ -95,7 +97,7 @@ def create_embeddings(tenant_id, document_version_id):
         delete_embeddings_for_document_version(document_version)
 
         try:
-            with current_event.create_span(f"{processor_type} Processing"):
+            with current_event.create_span(f"{processor.type} Processing"):
                 document_processor = processor_class(
                     tenant=tenant,
                     document_version=document_version,
@@ -107,6 +109,8 @@ def create_embeddings(tenant_id, document_version_id):
                     'markdown': markdown,
                     'title': title
                 })
+                if not markdown or markdown.strip() == '':
+                    raise EveAINoContentFound(document_version.doc_id, document_version.id)
 
             with current_event.create_span("Embedding"):
                 embed_markdown(tenant, document_version, catalog, document_processor, markdown, title)
@@ -114,9 +118,11 @@ def create_embeddings(tenant_id, document_version_id):
             current_event.log("Finished Embedding Creation Task")
 
         except Exception as e:
+            stacktrace = traceback.format_exc()
             current_app.logger.error(f'Error creating embeddings for tenant {tenant_id} '
-                                     f'on document version {document_version_id} '
-                                     f'error: {e}')
+                                 f'on document version {document_version_id} '
+                                 f'error: {e}\n'
+                                 f'Stacktrace: {stacktrace}')
             document_version.processing = False
             document_version.processing_finished_at = dt.now(tz.utc)
             document_version.processing_error = str(e)[:255]
@@ -624,25 +630,9 @@ def get_processor_for_document(catalog_id: int, file_type: str, sub_file_type: s
         ValueError: If no matching processor is found
     """
     try:
+        current_app.logger.debug(f"Getting processor for catalog {catalog_id}, file type {file_type}, file sub_type {sub_file_type} ")
         # Start with base query for catalog
-        query = Processor.query.filter_by(catalog_id=catalog_id)
-
-        # Find processor type that handles this file type
-        matching_processor_type = None
-        for proc_type, config in PROCESSOR_TYPES.items():
-            supported_types = config['file_types']
-            if isinstance(supported_types, str):
-                supported_types = [t.strip() for t in supported_types.split(',')]
-
-            if file_type in supported_types:
-                matching_processor_type = proc_type
-                break
-
-        if not matching_processor_type:
-            raise ValueError(f"No processor type found for file type: {file_type}")
-
-        # Add processor type condition
-        query = query.filter_by(type=matching_processor_type)
+        query = Processor.query.filter_by(catalog_id=catalog_id).filter_by(active=True)
 
         # If sub_file_type is provided, add that condition
         if sub_file_type:
@@ -651,22 +641,44 @@ def get_processor_for_document(catalog_id: int, file_type: str, sub_file_type: s
             # If no sub_file_type, prefer processors without sub_file_type specification
             query = query.filter(or_(Processor.sub_file_type.is_(None),
                                      Processor.sub_file_type == ''))
+        
+        available_processors = query.all()
 
-        # Get the first matching processor
-        processor = query.first()
+        if not available_processors:
+            raise EveAINoProcessorFound(catalog_id, file_type, sub_file_type)
+        available_processor_types = [processor.type for processor in available_processors]
+        current_app.logger.debug(f"Available processors for catalog {catalog_id}: {available_processor_types}")
+
+        # Find processor type that handles this file type
+        matching_processor_type = None
+        for proc_type, config in PROCESSOR_TYPES.items():
+            # Alleen verwerken als dit type processor beschikbaar is in de database
+            if proc_type in available_processor_types:
+                supported_types = config['file_types']
+                if isinstance(supported_types, str):
+                    supported_types = [t.strip() for t in supported_types.split(',')]
+                    current_app.logger.debug(f"Supported types for processor type {proc_type}: {supported_types}")
+
+                if file_type in supported_types:
+                    matching_processor_type = proc_type
+                    break
+
+        current_app.logger.debug(f"Processor type found for catalog {catalog_id}, file type {file_type}: {matching_processor_type}")
+        if not matching_processor_type:
+            raise EveAINoProcessorFound(catalog_id, file_type, sub_file_type)
+        else:
+            current_app.logger.debug(f"Processor type found for file type: {file_type}: {matching_processor_type}")
+
+        processor = None
+        for proc in available_processors:
+            if proc.type == matching_processor_type:
+                processor = proc
+                break
 
         if not processor:
-            if sub_file_type:
-                raise ValueError(
-                    f"No processor found for catalog {catalog_id} of type {matching_processor_type}, "
-                    f"file type {file_type}, sub-type {sub_file_type}"
-                )
-            else:
-                raise ValueError(
-                    f"No processor found for catalog {catalog_id}, "
-                    f"file type {file_type}"
-                )
+            raise EveAINoProcessorFound(catalog_id, file_type, sub_file_type)
 
+        current_app.logger.debug(f"Processor found for catalog {catalog_id}, file type {file_type}: {processor}")
         return processor
 
     except Exception as e:

migrations/public/versions/e47dc002b678_add_allowed_languages_to_tenantmake_.py

@@ -0,0 +1,53 @@
+"""Add allowed_languages to TenantMake, introduce TranslationCache
+
+Revision ID: e47dc002b678
+Revises: 83d4e90f87c6
+Create Date: 2025-06-26 13:43:43.719865
+
+"""
+from alembic import op
+import sqlalchemy as sa
+from sqlalchemy.dialects import postgresql
+
+# revision identifiers, used by Alembic.
+revision = 'e47dc002b678'
+down_revision = '83d4e90f87c6'
+branch_labels = None
+depends_on = None
+
+
+def upgrade():
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.create_table('translation_cache',
+    sa.Column('cache_key', sa.String(length=16), nullable=False),
+    sa.Column('source_text', sa.Text(), nullable=False),
+    sa.Column('translated_text', sa.Text(), nullable=False),
+    sa.Column('source_language', sa.String(length=2), nullable=False),
+    sa.Column('target_language', sa.String(length=2), nullable=False),
+    sa.Column('context', sa.Text(), nullable=True),
+    sa.Column('prompt_tokens', sa.Integer(), nullable=False),
+    sa.Column('completion_tokens', sa.Integer(), nullable=False),
+    sa.Column('created_at', sa.DateTime(), server_default=sa.text('now()'), nullable=False),
+    sa.Column('created_by', sa.Integer(), nullable=True),
+    sa.Column('updated_at', sa.DateTime(), server_default=sa.text('now()'), nullable=False),
+    sa.Column('updated_by', sa.Integer(), nullable=True),
+    sa.Column('last_used_at', sa.DateTime(), nullable=True),
+    sa.ForeignKeyConstraint(['created_by'], ['public.user.id'], ),
+    sa.ForeignKeyConstraint(['updated_by'], ['public.user.id'], ),
+    sa.PrimaryKeyConstraint('cache_key'),
+    schema='public'
+    )
+
+    with op.batch_alter_table('tenant_make', schema=None) as batch_op:
+        batch_op.add_column(sa.Column('allowed_languages', postgresql.ARRAY(sa.String(length=2)), nullable=True))
+
+    # ### end Alembic commands ###
+
+
+def downgrade():
+    # ### commands auto generated by Alembic - please adjust! ###
+    with op.batch_alter_table('tenant_make', schema=None) as batch_op:
+        batch_op.drop_column('allowed_languages')
+
+        op.drop_table('translation_cache', schema='public')
+    # ### end Alembic commands ###

migrations/tenant/env.py

@@ -72,7 +72,8 @@ def get_public_table_names():
     # TODO: This function should include the necessary functionality to automatically retrieve table names
     return ['role', 'roles_users', 'tenant', 'user', 'tenant_domain','license_tier', 'license', 'license_usage',
             'business_event_log', 'tenant_project', 'partner', 'partner_service', 'invoice', 'license_period',
-            'license_change_log', 'partner_service_license_tier', 'payment', 'partner_tenant']
+            'license_change_log', 'partner_service_license_tier', 'payment', 'partner_tenant', 'tenant_make',
+            'specialist_magic_link_tenant']
 
 PUBLIC_TABLES = get_public_table_names()
 logger.info(f"Public tables: {PUBLIC_TABLES}")

migrations/tenant/versions/b1647f31339a_add_active_flag_to_processor.py

@@ -0,0 +1,30 @@
+"""Add Active Flag to Processor
+
+Revision ID: b1647f31339a
+Revises: 2b6ae6cc923e
+Create Date: 2025-06-25 12:34:35.391516
+
+"""
+from alembic import op
+import sqlalchemy as sa
+import pgvector
+from sqlalchemy.dialects import postgresql
+
+# revision identifiers, used by Alembic.
+revision = 'b1647f31339a'
+down_revision = '2b6ae6cc923e'
+branch_labels = None
+depends_on = None
+
+
+def upgrade():
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.add_column('processor', sa.Column('active', sa.Boolean(), nullable=True))
+    op.execute("UPDATE processor SET active = true")
+    # ### end Alembic commands ###
+
+
+def downgrade():
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.drop_column('processor', 'active')
+    # ### end Alembic commands ###

requirements.txt

@@ -94,4 +94,5 @@ scaleway~=2.9.0
 html2text~=2025.4.15
 markdown~=3.8
 python-json-logger~=2.0.7
-qrcode[pil]==8.2
+qrcode[pil]==8.2
+xxhash~=3.5.0