- Release script added to tag in both git and docker

- Small bug fixes

.gitignore

@@ -42,3 +42,4 @@ migrations/public/.DS_Store
 scripts/.DS_Store
 scripts/__pycache__/run_eveai_app.cpython-312.pyc
 /eveai_repo.txt
+*repo.txt

.repopackignore_base

@@ -15,7 +15,6 @@ migrations/
 nginx/mime.types
 *.gitignore*
 .python-version
-.repopackignore
+.repopackignore*
 repopack.config.json
-
-
+*repo.txt

.repopackignore_components

@@ -0,0 +1,12 @@
+docker/
+eveai_api/
+eveai_app/
+eveai_beat/
+eveai_chat/
+eveai_chat_workers/
+eveai_entitlements/
+eveai_workers/
+instance/
+integrations/
+nginx/
+scripts/

.repopackignore_docker

@@ -0,0 +1,12 @@
+common/
+config/
+eveai_api/
+eveai_app/
+eveai_beat/
+eveai_chat/
+eveai_chat_workers/
+eveai_entitlements/
+eveai_workers/
+instance/
+integrations/
+nginx/

.repopackignore_eveai_api

@@ -0,0 +1,11 @@
+docker/
+eveai_app/
+eveai_beat/
+eveai_chat/
+eveai_chat_workers/
+eveai_entitlements/
+eveai_workers/
+instance/
+integrations/
+nginx/
+scripts/

.repopackignore_eveai_app

@@ -0,0 +1,11 @@
+docker/
+eveai_api/
+eveai_beat/
+eveai_chat/
+eveai_chat_workers/
+eveai_entitlements/
+eveai_workers/
+instance/
+integrations/
+nginx/
+scripts/

.repopackignore_eveai_beat

@@ -0,0 +1,11 @@
+docker/
+eveai_api/
+eveai_app/
+eveai_chat/
+eveai_chat_workers/
+eveai_entitlements/
+eveai_workers/
+instance/
+integrations/
+nginx/
+scripts/

.repopackignore_eveai_chat

@@ -0,0 +1,11 @@
+docker/
+eveai_api/
+eveai_app/
+eveai_beat/
+eveai_chat_workers/
+eveai_entitlements/
+eveai_workers/
+instance/
+integrations/
+nginx/
+scripts/

.repopackignore_eveai_chat_workers

@@ -0,0 +1,11 @@
+docker/
+eveai_api/
+eveai_app/
+eveai_beat/
+eveai_chat/
+eveai_entitlements/
+eveai_workers/
+instance/
+integrations/
+nginx/
+scripts/

.repopackignore_eveai_entitlements

@@ -0,0 +1,11 @@
+docker/
+eveai_api/
+eveai_app/
+eveai_beat/
+eveai_chat/
+eveai_chat_workers/
+eveai_workers/
+instance/
+integrations/
+nginx/
+scripts/

.repopackignore_eveai_workers

@@ -0,0 +1,11 @@
+docker/
+eveai_api/
+eveai_app/
+eveai_beat/
+eveai_chat/
+eveai_chat_workers/
+eveai_entitlements/
+instance/
+integrations/
+nginx/
+scripts/

.repopackignore_full

@@ -0,0 +1,4 @@
+docker
+integrations
+nginx
+scripts

.repopackignore_integrations

@@ -0,0 +1,13 @@
+common/
+config/
+docker/
+eveai_api/
+eveai_app/
+eveai_beat/
+eveai_chat/
+eveai_chat_workers/
+eveai_entitlements/
+eveai_workers/
+instance/
+nginx/
+scripts/

.repopackignore_nginx

@@ -0,0 +1,11 @@
+docker/
+eveai_api/
+eveai_app/
+eveai_beat/
+eveai_chat/
+eveai_chat_workers/
+eveai_entitlements/
+eveai_workers/
+instance/
+integrations/
+scripts/

CHANGELOG.md

@@ -25,6 +25,54 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 ### Security
 - In case of vulnerabilities.
 
+## [1.0.13-alfa]
+
+### Added
+- Finished Catalog introduction
+- Reinitialization of WordPress site for syncing
+
+### Changed
+- Modification of WordPress Sync Component
+- Cleanup of attributes in Tenant
+
+### Fixed
+- Overall bugfixes as result from the Catalog introduction
+
+## [1.0.12-alfa]
+
+### Added
+- Added Catalog functionality
+
+### Changed
+- For changes in existing functionality.
+
+### Deprecated
+- For soon-to-be removed features.
+
+### Removed
+- For now removed features.
+
+### Fixed
+- Set default language when registering Documents or URLs.
+
+### Security
+- In case of vulnerabilities.
+
+## [1.0.11-alfa]
+
+### Added
+- License Usage Calculation realised
+- View License Usages
+- Celery Beat container added
+- First schedule in Celery Beat for calculating usage (hourly)
+
+### Changed
+- repopack can now split for different components
+
+### Fixed
+- Various fixes as consequence of changing file_location / file_name ==> bucket_name / object_name
+- Celery Routing / Queuing updated
+
 ## [1.0.10-alfa]
 
 ### Added

common/langchain/eveai_retriever.py

@@ -38,7 +38,7 @@ class EveAIRetriever(BaseRetriever, BaseModel):
         similarity_threshold = self.model_variables['similarity_threshold']
         k = self.model_variables['k']
 
-        if self.tenant_info['rag_tuning']:
+        if self.model_variables['rag_tuning']:
             try:
                 current_date = get_date_in_timezone(self.tenant_info['timezone'])
                 current_app.rag_tuning_logger.debug(f'Current date: {current_date}\n')
@@ -73,7 +73,7 @@ class EveAIRetriever(BaseRetriever, BaseModel):
                 current_app.logger.error(f'Error generating overview: {e}')
                 db.session.rollback()
 
-        if self.tenant_info['rag_tuning']:
+        if self.model_variables['rag_tuning']:
             current_app.rag_tuning_logger.debug(f'Parameters for Retrieval of documents: \n')
             current_app.rag_tuning_logger.debug(f'Similarity Threshold: {similarity_threshold}\n')
             current_app.rag_tuning_logger.debug(f'K: {k}\n')
@@ -106,14 +106,14 @@ class EveAIRetriever(BaseRetriever, BaseModel):
                 .limit(k)
             )
 
-            if self.tenant_info['rag_tuning']:
+            if self.model_variables['rag_tuning']:
                 current_app.rag_tuning_logger.debug(f'Query executed for Retrieval of documents: \n')
                 current_app.rag_tuning_logger.debug(f'{query_obj.statement}\n')
                 current_app.rag_tuning_logger.debug(f'---------------------------------------\n')
 
             res = query_obj.all()
 
-            if self.tenant_info['rag_tuning']:
+            if self.model_variables['rag_tuning']:
                 current_app.rag_tuning_logger.debug(f'Retrieved {len(res)} relevant documents \n')
                 current_app.rag_tuning_logger.debug(f'Data retrieved: \n')
                 current_app.rag_tuning_logger.debug(f'{res}\n')
@@ -121,7 +121,7 @@ class EveAIRetriever(BaseRetriever, BaseModel):
 
             result = []
             for doc in res:
-                if self.tenant_info['rag_tuning']:
+                if self.model_variables['rag_tuning']:
                     current_app.rag_tuning_logger.debug(f'Document ID: {doc[0].id} - Distance: {doc[1]}\n')
                     current_app.rag_tuning_logger.debug(f'Chunk: \n {doc[0].chunk}\n\n')
                 result.append(f'SOURCE: {doc[0].id}\n\n{doc[0].chunk}\n\n')

common/models/document.py

@@ -2,12 +2,49 @@ from common.extensions import db
 from .user import User, Tenant
 from pgvector.sqlalchemy import Vector
 from sqlalchemy.dialects.postgresql import JSONB
+from sqlalchemy.dialects.postgresql import ARRAY
+import sqlalchemy as sa
+
+
+class Catalog(db.Model):
+    id = db.Column(db.Integer, primary_key=True)
+    name = db.Column(db.String(50), nullable=False)
+    description = db.Column(db.Text, nullable=True)
+
+    # Embedding variables
+    html_tags = db.Column(ARRAY(sa.String(10)), nullable=True, default=['p', 'h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'li'])
+    html_end_tags = db.Column(ARRAY(sa.String(10)), nullable=True, default=['p', 'li'])
+    html_included_elements = db.Column(ARRAY(sa.String(50)), nullable=True)
+    html_excluded_elements = db.Column(ARRAY(sa.String(50)), nullable=True)
+    html_excluded_classes = db.Column(ARRAY(sa.String(200)), nullable=True)
+
+    min_chunk_size = db.Column(db.Integer, nullable=True, default=2000)
+    max_chunk_size = db.Column(db.Integer, nullable=True, default=3000)
+
+    # Embedding search variables ==> move to specialist?
+    es_k = db.Column(db.Integer, nullable=True, default=8)
+    es_similarity_threshold = db.Column(db.Float, nullable=True, default=0.4)
+
+    # Chat variables ==> Move to Specialist?
+    chat_RAG_temperature = db.Column(db.Float, nullable=True, default=0.3)
+    chat_no_RAG_temperature = db.Column(db.Float, nullable=True, default=0.5)
+
+    # Tuning enablers
+    embed_tuning = db.Column(db.Boolean, nullable=True, default=False)
+    rag_tuning = db.Column(db.Boolean, nullable=True, default=False)    # Move to Specialist?
+
+    # Versioning Information
+    created_at = db.Column(db.DateTime, nullable=False, server_default=db.func.now())
+    created_by = db.Column(db.Integer, db.ForeignKey(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(User.id))
 
 
 class Document(db.Model):
     id = db.Column(db.Integer, primary_key=True)
+    # tenant_id = db.Column(db.Integer, db.ForeignKey(Tenant.id), nullable=False)
+    catalog_id = db.Column(db.Integer, db.ForeignKey(Catalog.id), nullable=True)
     name = db.Column(db.String(100), nullable=False)
-    tenant_id = db.Column(db.Integer, db.ForeignKey(Tenant.id), nullable=False)
     valid_from = db.Column(db.DateTime, nullable=True)
     valid_to = db.Column(db.DateTime, nullable=True)
 
@@ -56,12 +93,6 @@ class DocumentVersion(db.Model):
     def __repr__(self):
         return f"<DocumentVersion {self.document_language.document_id}.{self.document_language.language}>.{self.id}>"
 
-    def calc_file_location(self):
-        return f"{self.document.tenant_id}/{self.document.id}/{self.language}"
-
-    def calc_file_name(self):
-        return f"{self.id}.{self.file_type}"
-
 
 class Embedding(db.Model):
     __tablename__ = 'embeddings'

common/models/entitlements.py

@@ -94,8 +94,8 @@ class LicenseUsage(db.Model):
     id = db.Column(db.Integer, primary_key=True)
     license_id = db.Column(db.Integer, db.ForeignKey('public.license.id'), nullable=False)
     tenant_id = db.Column(db.Integer, db.ForeignKey('public.tenant.id'), nullable=False)
-    storage_mb_used = db.Column(db.Integer, default=0)
-    embedding_mb_used = db.Column(db.Integer, default=0)
+    storage_mb_used = db.Column(db.Float, default=0)
+    embedding_mb_used = db.Column(db.Float, default=0)
     embedding_prompt_tokens_used = db.Column(db.Integer, default=0)
     embedding_completion_tokens_used = db.Column(db.Integer, default=0)
     embedding_total_tokens_used = db.Column(db.Integer, default=0)

common/models/user.py

@@ -34,32 +34,32 @@ class Tenant(db.Model):
     embedding_model = db.Column(db.String(50), nullable=True)
     llm_model = db.Column(db.String(50), nullable=True)
 
-    # Embedding variables
-    html_tags = db.Column(ARRAY(sa.String(10)), nullable=True, default=['p', 'h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'li'])
-    html_end_tags = db.Column(ARRAY(sa.String(10)), nullable=True, default=['p', 'li'])
-    html_included_elements = db.Column(ARRAY(sa.String(50)), nullable=True)
-    html_excluded_elements = db.Column(ARRAY(sa.String(50)), nullable=True)
-    html_excluded_classes = db.Column(ARRAY(sa.String(200)), nullable=True)
-
-    min_chunk_size = db.Column(db.Integer, nullable=True, default=2000)
-    max_chunk_size = db.Column(db.Integer, nullable=True, default=3000)
-
-    # Embedding search variables
-    es_k = db.Column(db.Integer, nullable=True, default=5)
-    es_similarity_threshold = db.Column(db.Float, nullable=True, default=0.7)
-
-    # Chat variables
-    chat_RAG_temperature = db.Column(db.Float, nullable=True, default=0.3)
-    chat_no_RAG_temperature = db.Column(db.Float, nullable=True, default=0.5)
+    # # Embedding variables ==> To be removed once all migrations (dev + prod) have been done
+    # html_tags = db.Column(ARRAY(sa.String(10)), nullable=True, default=['p', 'h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'li'])
+    # html_end_tags = db.Column(ARRAY(sa.String(10)), nullable=True, default=['p', 'li'])
+    # html_included_elements = db.Column(ARRAY(sa.String(50)), nullable=True)
+    # html_excluded_elements = db.Column(ARRAY(sa.String(50)), nullable=True)
+    # html_excluded_classes = db.Column(ARRAY(sa.String(200)), nullable=True)
+    #
+    # min_chunk_size = db.Column(db.Integer, nullable=True, default=2000)
+    # max_chunk_size = db.Column(db.Integer, nullable=True, default=3000)
+    #
+    # # Embedding search variables
+    # es_k = db.Column(db.Integer, nullable=True, default=5)
+    # es_similarity_threshold = db.Column(db.Float, nullable=True, default=0.7)
+    #
+    # # Chat variables
+    # chat_RAG_temperature = db.Column(db.Float, nullable=True, default=0.3)
+    # chat_no_RAG_temperature = db.Column(db.Float, nullable=True, default=0.5)
     fallback_algorithms = db.Column(ARRAY(sa.String(50)), nullable=True)
 
     # Licensing Information
     encrypted_chat_api_key = db.Column(db.String(500), nullable=True)
     encrypted_api_key = db.Column(db.String(500), nullable=True)
 
-    # Tuning enablers
-    embed_tuning = db.Column(db.Boolean, nullable=True, default=False)
-    rag_tuning = db.Column(db.Boolean, nullable=True, default=False)
+    # # Tuning enablers
+    # embed_tuning = db.Column(db.Boolean, nullable=True, default=False)
+    # rag_tuning = db.Column(db.Boolean, nullable=True, default=False)
 
     # Entitlements
     currency = db.Column(db.String(20), nullable=True)
@@ -96,20 +96,7 @@ class Tenant(db.Model):
             'allowed_languages': self.allowed_languages,
             'embedding_model': self.embedding_model,
             'llm_model': self.llm_model,
-            'html_tags': self.html_tags,
-            'html_end_tags': self.html_end_tags,
-            'html_included_elements': self.html_included_elements,
-            'html_excluded_elements': self.html_excluded_elements,
-            'html_excluded_classes': self.html_excluded_classes,
-            'min_chunk_size': self.min_chunk_size,
-            'max_chunk_size': self.max_chunk_size,
-            'es_k': self.es_k,
-            'es_similarity_threshold': self.es_similarity_threshold,
-            'chat_RAG_temperature': self.chat_RAG_temperature,
-            'chat_no_RAG_temperature': self.chat_no_RAG_temperature,
             'fallback_algorithms': self.fallback_algorithms,
-            'embed_tuning': self.embed_tuning,
-            'rag_tuning': self.rag_tuning,
             'currency': self.currency,
             'usage_email': self.usage_email,
         }

common/utils/celery_utils.py

@@ -1,14 +1,16 @@
 from celery import Celery
 from kombu import Queue
 from werkzeug.local import LocalProxy
+from redbeat import RedBeatScheduler
 
 celery_app = Celery()
 
 
-def init_celery(celery, app):
+def init_celery(celery, app, is_beat=False):
     celery_app.main = app.name
     app.logger.debug(f'CELERY_BROKER_URL: {app.config["CELERY_BROKER_URL"]}')
     app.logger.debug(f'CELERY_RESULT_BACKEND: {app.config["CELERY_RESULT_BACKEND"]}')
+
     celery_config = {
         'broker_url': app.config.get('CELERY_BROKER_URL', 'redis://localhost:6379/0'),
         'result_backend': app.config.get('CELERY_RESULT_BACKEND', 'redis://localhost:6379/0'),
@@ -17,19 +19,40 @@ def init_celery(celery, app):
         'accept_content': app.config.get('CELERY_ACCEPT_CONTENT', ['json']),
         'timezone': app.config.get('CELERY_TIMEZONE', 'UTC'),
         'enable_utc': app.config.get('CELERY_ENABLE_UTC', True),
-        'task_routes': {'eveai_worker.tasks.create_embeddings': {'queue': 'embeddings',
-                                                                 'routing_key': 'embeddings.create_embeddings'}},
     }
+
+    if is_beat:
+        # Add configurations specific to Beat scheduler
+        celery_config['beat_scheduler'] = 'redbeat.RedBeatScheduler'
+        celery_config['redbeat_lock_key'] = 'redbeat::lock'
+        celery_config['beat_max_loop_interval'] = 10  # Adjust as needed
+
     celery_app.conf.update(**celery_config)
 
-    # Setting up Celery task queues
-    celery_app.conf.task_queues = (
-        Queue('default', routing_key='task.#'),
-        Queue('embeddings', routing_key='embeddings.#', queue_arguments={'x-max-priority': 10}),
-        Queue('llm_interactions', routing_key='llm_interactions.#', queue_arguments={'x-max-priority': 5}),
-    )
+    # Task queues for workers only
+    if not is_beat:
+        celery_app.conf.task_queues = (
+            Queue('default', routing_key='task.#'),
+            Queue('embeddings', routing_key='embeddings.#', queue_arguments={'x-max-priority': 10}),
+            Queue('llm_interactions', routing_key='llm_interactions.#', queue_arguments={'x-max-priority': 5}),
+            Queue('entitlements', routing_key='entitlements.#', queue_arguments={'x-max-priority': 10}),
+        )
+        celery_app.conf.task_routes = {
+            'eveai_workers.*': {  # All tasks from eveai_workers module
+                'queue': 'embeddings',
+                'routing_key': 'embeddings.#',
+            },
+            'eveai_chat_workers.*': {  # All tasks from eveai_chat_workers module
+                'queue': 'llm_interactions',
+                'routing_key': 'llm_interactions.#',
+            },
+            'eveai_entitlements.*': {  # All tasks from eveai_entitlements module
+                'queue': 'entitlements',
+                'routing_key': 'entitlements.#',
+            }
+        }
 
-    # Ensuring tasks execute with Flask application context
+    # Ensure tasks execute with Flask context
     class ContextTask(celery.Task):
         def __call__(self, *args, **kwargs):
             with app.app_context():
@@ -37,6 +60,39 @@ def init_celery(celery, app):
 
     celery.Task = ContextTask
 
+# Original init_celery before updating for beat
+# def init_celery(celery, app):
+#     celery_app.main = app.name
+#     app.logger.debug(f'CELERY_BROKER_URL: {app.config["CELERY_BROKER_URL"]}')
+#     app.logger.debug(f'CELERY_RESULT_BACKEND: {app.config["CELERY_RESULT_BACKEND"]}')
+#     celery_config = {
+#         'broker_url': app.config.get('CELERY_BROKER_URL', 'redis://localhost:6379/0'),
+#         'result_backend': app.config.get('CELERY_RESULT_BACKEND', 'redis://localhost:6379/0'),
+#         'task_serializer': app.config.get('CELERY_TASK_SERIALIZER', 'json'),
+#         'result_serializer': app.config.get('CELERY_RESULT_SERIALIZER', 'json'),
+#         'accept_content': app.config.get('CELERY_ACCEPT_CONTENT', ['json']),
+#         'timezone': app.config.get('CELERY_TIMEZONE', 'UTC'),
+#         'enable_utc': app.config.get('CELERY_ENABLE_UTC', True),
+#         'task_routes': {'eveai_worker.tasks.create_embeddings': {'queue': 'embeddings',
+#                                                                  'routing_key': 'embeddings.create_embeddings'}},
+#     }
+#     celery_app.conf.update(**celery_config)
+#
+#     # Setting up Celery task queues
+#     celery_app.conf.task_queues = (
+#         Queue('default', routing_key='task.#'),
+#         Queue('embeddings', routing_key='embeddings.#', queue_arguments={'x-max-priority': 10}),
+#         Queue('llm_interactions', routing_key='llm_interactions.#', queue_arguments={'x-max-priority': 5}),
+#     )
+#
+#     # Ensuring tasks execute with Flask application context
+#     class ContextTask(celery.Task):
+#         def __call__(self, *args, **kwargs):
+#             with app.app_context():
+#                 return self.run(*args, **kwargs)
+#
+#     celery.Task = ContextTask
+
 
 def make_celery(app_name, config):
     return celery_app

common/utils/document_utils.py

@@ -17,11 +17,12 @@ from ..models.user import Tenant
 
 def create_document_stack(api_input, file, filename, extension, tenant_id):
     # Create the Document
-    new_doc = create_document(api_input, filename, tenant_id)
+    catalog_id = int(api_input.get('catalog_id'))
+    new_doc = create_document(api_input, filename, catalog_id)
     db.session.add(new_doc)
 
     # Create the DocumentVersion
-    new_doc_vers = create_version_for_document(new_doc,
+    new_doc_vers = create_version_for_document(new_doc, tenant_id,
                                                api_input.get('url', ''),
                                                api_input.get('language', 'en'),
                                                api_input.get('user_context', ''),
@@ -45,7 +46,7 @@ def create_document_stack(api_input, file, filename, extension, tenant_id):
     return new_doc, new_doc_vers
 
 
-def create_document(form, filename, tenant_id):
+def create_document(form, filename, catalog_id):
     new_doc = Document()
     if form['name'] == '':
         new_doc.name = filename.rsplit('.', 1)[0]
@@ -56,13 +57,13 @@ def create_document(form, filename, tenant_id):
         new_doc.valid_from = form['valid_from']
     else:
         new_doc.valid_from = dt.now(tz.utc)
-    new_doc.tenant_id = tenant_id
+    new_doc.catalog_id = catalog_id
     set_logging_information(new_doc, dt.now(tz.utc))
 
     return new_doc
 
 
-def create_version_for_document(document, url, language, user_context, user_metadata):
+def create_version_for_document(document, tenant_id, url, language, user_context, user_metadata):
     new_doc_vers = DocumentVersion()
     if url != '':
         new_doc_vers.url = url
@@ -82,7 +83,7 @@ def create_version_for_document(document, url, language, user_context, user_meta
 
     set_logging_information(new_doc_vers, dt.now(tz.utc))
 
-    mark_tenant_storage_dirty(document.tenant_id)
+    mark_tenant_storage_dirty(tenant_id)
 
     return new_doc_vers
 
@@ -99,12 +100,12 @@ def upload_file_for_version(doc_vers, file, extension, tenant_id):
             doc_vers.doc_id,
             doc_vers.language,
             doc_vers.id,
-            doc_vers.file_name,
+            f"{doc_vers.id}.{extension}",
             file
         )
         doc_vers.bucket_name = bn
         doc_vers.object_name = on
-        doc_vers.file_size_mb = size / 1048576  # Convert bytes to MB
+        doc_vers.file_size = size / 1048576  # Convert bytes to MB
 
         db.session.commit()
         current_app.logger.info(f'Successfully saved document to MinIO for tenant {tenant_id} for '
@@ -222,10 +223,9 @@ def process_multiple_urls(urls, tenant_id, api_input):
 
 
 def start_embedding_task(tenant_id, doc_vers_id):
-    task = current_celery.send_task('create_embeddings', queue='embeddings', args=[
-        tenant_id,
-        doc_vers_id,
-    ])
+    task = current_celery.send_task('create_embeddings',
+                                    args=[tenant_id, doc_vers_id,],
+                                    queue='embeddings')
     current_app.logger.info(f'Embedding creation started for tenant {tenant_id}, '
                             f'Document Version {doc_vers_id}. '
                             f'Embedding creation task: {task.id}')
@@ -287,7 +287,7 @@ def edit_document_version(version_id, user_context):
         return None, str(e)
 
 
-def refresh_document_with_info(doc_id, api_input):
+def refresh_document_with_info(doc_id, tenant_id, api_input):
     doc = Document.query.get_or_404(doc_id)
     old_doc_vers = DocumentVersion.query.filter_by(doc_id=doc_id).order_by(desc(DocumentVersion.id)).first()
 
@@ -295,11 +295,11 @@ def refresh_document_with_info(doc_id, api_input):
         return None, "This document has no URL. Only documents with a URL can be refreshed."
 
     new_doc_vers = create_version_for_document(
-        doc,
+        doc, tenant_id,
         old_doc_vers.url,
         api_input.get('language', old_doc_vers.language),
         api_input.get('user_context', old_doc_vers.user_context),
-        api_input.get('user_metadata', old_doc_vers.user_metadata)
+        api_input.get('user_metadata', old_doc_vers.user_metadata),
     )
 
     set_logging_information(new_doc_vers, dt.now(tz.utc))
@@ -319,18 +319,18 @@ def refresh_document_with_info(doc_id, api_input):
     response.raise_for_status()
     file_content = response.content
 
-    upload_file_for_version(new_doc_vers, file_content, extension, doc.tenant_id)
+    upload_file_for_version(new_doc_vers, file_content, extension, tenant_id)
 
-    task = current_celery.send_task('create_embeddings', queue='embeddings', args=[
-        doc.tenant_id,
-        new_doc_vers.id,
-    ])
+    task = current_celery.send_task('create_embeddings', args=[tenant_id, new_doc_vers.id,], queue='embeddings')
+    current_app.logger.info(f'Embedding creation started for document {doc_id} on version {new_doc_vers.id} '
+                            f'with task id: {task.id}.')
 
     return new_doc_vers, task.id
 
 
 # Update the existing refresh_document function to use the new refresh_document_with_info
-def refresh_document(doc_id):
+def refresh_document(doc_id, tenant_id):
+    current_app.logger.info(f'Refreshing document {doc_id}')
     doc = Document.query.get_or_404(doc_id)
     old_doc_vers = DocumentVersion.query.filter_by(doc_id=doc_id).order_by(desc(DocumentVersion.id)).first()
 
@@ -340,11 +340,11 @@ def refresh_document(doc_id):
         'user_metadata': old_doc_vers.user_metadata
     }
 
-    return refresh_document_with_info(doc_id, api_input)
+    return refresh_document_with_info(doc_id, tenant_id, api_input)
 
 
 # Function triggered when a document_version is created or updated
 def mark_tenant_storage_dirty(tenant_id):
-    tenant = db.session.query(Tenant).filter_by(id=tenant_id).first()
+    tenant = db.session.query(Tenant).filter_by(id=int(tenant_id)).first()
     tenant.storage_dirty = True
     db.session.commit()

common/utils/minio_utils.py

@@ -54,9 +54,7 @@ class MinioClient:
         except S3Error as err:
             raise Exception(f"Error occurred while uploading file: {err}")
 
-    def download_document_file(self, tenant_id, document_id, language, version_id, filename):
-        bucket_name = self.generate_bucket_name(tenant_id)
-        object_name = self.generate_object_name(document_id, language, version_id, filename)
+    def download_document_file(self, tenant_id, bucket_name, object_name):
         try:
             response = self.client.get_object(bucket_name, object_name)
             return response.read()

common/utils/model_utils.py

@@ -14,7 +14,7 @@ from portkey_ai.langchain.portkey_langchain_callback_handler import LangchainCal
 from common.langchain.llm_metrics_handler import LLMMetricsHandler
 from common.langchain.tracked_openai_embeddings import TrackedOpenAIEmbeddings
 from common.langchain.tracked_transcribe import tracked_transcribe
-from common.models.document import EmbeddingSmallOpenAI, EmbeddingLargeOpenAI
+from common.models.document import EmbeddingSmallOpenAI, EmbeddingLargeOpenAI, Catalog
 from common.models.user import Tenant
 from config.model_config import MODEL_CONFIG
 from common.utils.business_event_context import current_event
@@ -42,8 +42,9 @@ def set_language_prompt_template(cls, language_prompt):
 
 
 class ModelVariables(MutableMapping):
-    def __init__(self, tenant: Tenant):
+    def __init__(self, tenant: Tenant, catalog_id=None):
         self.tenant = tenant
+        self.catalog_id = catalog_id
         self._variables = self._initialize_variables()
         self._embedding_model = None
         self._llm = None
@@ -57,25 +58,32 @@ class ModelVariables(MutableMapping):
     def _initialize_variables(self):
         variables = {}
 
-        # We initialize the variables that are available knowing the tenant. For the other, we will apply 'lazy loading'
-        variables['k'] = self.tenant.es_k or 5
-        variables['similarity_threshold'] = self.tenant.es_similarity_threshold or 0.7
-        variables['RAG_temperature'] = self.tenant.chat_RAG_temperature or 0.3
-        variables['no_RAG_temperature'] = self.tenant.chat_no_RAG_temperature or 0.5
-        variables['embed_tuning'] = self.tenant.embed_tuning or False
-        variables['rag_tuning'] = self.tenant.rag_tuning or False
+        # Get the Catalog if catalog_id is passed
+        if self.catalog_id:
+            catalog = Catalog.query.get_or_404(self.catalog_id)
+
+            # We initialize the variables that are available knowing the tenant.
+            variables['embed_tuning'] = catalog.embed_tuning or False
+
+            # Set HTML Chunking Variables
+            variables['html_tags'] = catalog.html_tags
+            variables['html_end_tags'] = catalog.html_end_tags
+            variables['html_included_elements'] = catalog.html_included_elements
+            variables['html_excluded_elements'] = catalog.html_excluded_elements
+            variables['html_excluded_classes'] = catalog.html_excluded_classes
+
+            # Set Chunk Size variables
+            variables['min_chunk_size'] = catalog.min_chunk_size
+            variables['max_chunk_size'] = catalog.max_chunk_size
+
+        # Set the RAG Context (will have to change once specialists are defined
         variables['rag_context'] = self.tenant.rag_context or " "
-
-        # Set HTML Chunking Variables
-        variables['html_tags'] = self.tenant.html_tags
-        variables['html_end_tags'] = self.tenant.html_end_tags
-        variables['html_included_elements'] = self.tenant.html_included_elements
-        variables['html_excluded_elements'] = self.tenant.html_excluded_elements
-        variables['html_excluded_classes'] = self.tenant.html_excluded_classes
-
-        # Set Chunk Size variables
-        variables['min_chunk_size'] = self.tenant.min_chunk_size
-        variables['max_chunk_size'] = self.tenant.max_chunk_size
+        # Temporary setting until we have Specialists
+        variables['rag_tuning'] = False
+        variables['RAG_temperature'] = 0.3
+        variables['no_RAG_temperature'] = 0.5
+        variables['k'] = 8
+        variables['similarity_threshold'] = 0.4
 
         # Set model providers
         variables['embedding_provider'], variables['embedding_model'] = self.tenant.embedding_model.rsplit('.', 1)
@@ -195,7 +203,12 @@ class ModelVariables(MutableMapping):
             return self.transcription_client
         elif key in self._variables.get('prompt_templates', []):
             return self.get_prompt_template(key)
-        return self._variables.get(key)
+        else:
+            value = self._variables.get(key)
+            if value is not None:
+                return value
+            else:
+                raise KeyError(f'Variable {key} does not exist in ModelVariables')
 
     def __setitem__(self, key: str, value: Any) -> None:
         self._variables[key] = value
@@ -225,8 +238,8 @@ class ModelVariables(MutableMapping):
         return self._variables.values()
 
 
-def select_model_variables(tenant):
-    model_variables = ModelVariables(tenant=tenant)
+def select_model_variables(tenant, catalog_id=None):
+    model_variables = ModelVariables(tenant=tenant, catalog_id=catalog_id)
     return model_variables
 
 

common/utils/nginx_utils.py

@@ -6,7 +6,6 @@ def prefixed_url_for(endpoint, **values):
     prefix = request.headers.get('X-Forwarded-Prefix', '')
     scheme = request.headers.get('X-Forwarded-Proto', request.scheme)
     host = request.headers.get('Host', request.host)
-    current_app.logger.debug(f'prefix: {prefix}, scheme: {scheme}, host: {host}')
 
     external = values.pop('_external', False)
     generated_url = url_for(endpoint, **values)

config/logging_config.py

@@ -37,7 +37,7 @@ LOGGING = {
             'level': 'DEBUG',
             'class': 'logging.handlers.RotatingFileHandler',
             'filename': 'logs/eveai_app.log',
-            'maxBytes': 1024 * 1024 * 5,  # 5MB
+            'maxBytes': 1024 * 1024 * 1,  # 1MB
             'backupCount': 10,
             'formatter': 'standard',
         },
@@ -45,7 +45,7 @@ LOGGING = {
             'level': 'DEBUG',
             'class': 'logging.handlers.RotatingFileHandler',
             'filename': 'logs/eveai_workers.log',
-            'maxBytes': 1024 * 1024 * 5,  # 5MB
+            'maxBytes': 1024 * 1024 * 1,  # 1MB
             'backupCount': 10,
             'formatter': 'standard',
         },
@@ -53,7 +53,7 @@ LOGGING = {
             'level': 'DEBUG',
             'class': 'logging.handlers.RotatingFileHandler',
             'filename': 'logs/eveai_chat.log',
-            'maxBytes': 1024 * 1024 * 5,  # 5MB
+            'maxBytes': 1024 * 1024 * 1,  # 1MB
             'backupCount': 10,
             'formatter': 'standard',
         },
@@ -61,7 +61,7 @@ LOGGING = {
             'level': 'DEBUG',
             'class': 'logging.handlers.RotatingFileHandler',
             'filename': 'logs/eveai_chat_workers.log',
-            'maxBytes': 1024 * 1024 * 5,  # 5MB
+            'maxBytes': 1024 * 1024 * 1,  # 1MB
             'backupCount': 10,
             'formatter': 'standard',
         },
@@ -69,7 +69,23 @@ LOGGING = {
             'level': 'DEBUG',
             'class': 'logging.handlers.RotatingFileHandler',
             'filename': 'logs/eveai_api.log',
-            'maxBytes': 1024 * 1024 * 5,  # 5MB
+            'maxBytes': 1024 * 1024 * 1,  # 1MB
+            'backupCount': 10,
+            'formatter': 'standard',
+        },
+        'file_beat': {
+            'level': 'DEBUG',
+            'class': 'logging.handlers.RotatingFileHandler',
+            'filename': 'logs/eveai_beat.log',
+            'maxBytes': 1024 * 1024 * 1,  # 1MB
+            'backupCount': 10,
+            'formatter': 'standard',
+        },
+        'file_entitlements': {
+            'level': 'DEBUG',
+            'class': 'logging.handlers.RotatingFileHandler',
+            'filename': 'logs/eveai_entitlements.log',
+            'maxBytes': 1024 * 1024 * 1,  # 1MB
             'backupCount': 10,
             'formatter': 'standard',
         },
@@ -77,7 +93,7 @@ LOGGING = {
             'level': 'DEBUG',
             'class': 'logging.handlers.RotatingFileHandler',
             'filename': 'logs/sqlalchemy.log',
-            'maxBytes': 1024 * 1024 * 5,  # 5MB
+            'maxBytes': 1024 * 1024 * 1,  # 1MB
             'backupCount': 10,
             'formatter': 'standard',
         },
@@ -85,7 +101,7 @@ LOGGING = {
             'level': 'DEBUG',
             'class': 'logging.handlers.RotatingFileHandler',
             'filename': 'logs/mailman.log',
-            'maxBytes': 1024 * 1024 * 5,  # 5MB
+            'maxBytes': 1024 * 1024 * 1,  # 1MB
             'backupCount': 10,
             'formatter': 'standard',
         },
@@ -93,7 +109,7 @@ LOGGING = {
             'level': 'DEBUG',
             'class': 'logging.handlers.RotatingFileHandler',
             'filename': 'logs/security.log',
-            'maxBytes': 1024 * 1024 * 5,  # 5MB
+            'maxBytes': 1024 * 1024 * 1,  # 1MB
             'backupCount': 10,
             'formatter': 'standard',
         },
@@ -101,7 +117,7 @@ LOGGING = {
             'level': 'DEBUG',
             'class': 'logging.handlers.RotatingFileHandler',
             'filename': 'logs/rag_tuning.log',
-            'maxBytes': 1024 * 1024 * 5,  # 5MB
+            'maxBytes': 1024 * 1024 * 1,  # 1MB
             'backupCount': 10,
             'formatter': 'standard',
         },
@@ -109,7 +125,7 @@ LOGGING = {
             'level': 'DEBUG',
             'class': 'logging.handlers.RotatingFileHandler',
             'filename': 'logs/embed_tuning.log',
-            'maxBytes': 1024 * 1024 * 5,  # 5MB
+            'maxBytes': 1024 * 1024 * 1,  # 1MB
             'backupCount': 10,
             'formatter': 'standard',
         },
@@ -117,7 +133,7 @@ LOGGING = {
             'level': 'INFO',
             'class': 'logging.handlers.RotatingFileHandler',
             'filename': 'logs/business_events.log',
-            'maxBytes': 1024 * 1024 * 5,  # 5MB
+            'maxBytes': 1024 * 1024 * 1,  # 1MB
             'backupCount': 10,
             'formatter': 'standard',
         },
@@ -172,6 +188,16 @@ LOGGING = {
             'level': 'DEBUG',
             'propagate': False
         },
+        'eveai_beat': {  # logger for the eveai_beat
+            'handlers': ['file_beat', 'graylog', ] if env == 'production' else ['file_beat', ],
+            'level': 'DEBUG',
+            'propagate': False
+        },
+        'eveai_entitlements': {  # logger for the eveai_entitlements
+            'handlers': ['file_entitlements', 'graylog', ] if env == 'production' else ['file_entitlements', ],
+            'level': 'DEBUG',
+            'propagate': False
+        },
         'sqlalchemy.engine': {  # logger for the sqlalchemy
             'handlers': ['file_sqlalchemy', 'graylog', ] if env == 'production' else ['file_sqlalchemy', ],
             'level': 'DEBUG',

docker/compose_dev.yaml

@@ -231,6 +231,59 @@ services:
     networks:
       - eveai-network
 
+  eveai_beat:
+    image: josakola/eveai_beat:latest
+    build:
+      context: ..
+      dockerfile: ./docker/eveai_beat/Dockerfile
+      platforms:
+        - linux/amd64
+        - linux/arm64
+    environment:
+      <<: *common-variables
+      COMPONENT_NAME: eveai_beat
+    volumes:
+      - ../eveai_beat:/app/eveai_beat
+      - ../common:/app/common
+      - ../config:/app/config
+      - ../scripts:/app/scripts
+      - ../patched_packages:/app/patched_packages
+      - eveai_logs:/app/logs
+    depends_on:
+      redis:
+        condition: service_healthy
+    networks:
+      - eveai-network
+
+  eveai_entitlements:
+    image: josakola/eveai_entitlements:latest
+    build:
+      context: ..
+      dockerfile: ./docker/eveai_entitlements/Dockerfile
+      platforms:
+        - linux/amd64
+        - linux/arm64
+    environment:
+      <<: *common-variables
+      COMPONENT_NAME: eveai_entitlements
+    volumes:
+      - ../eveai_entitlements:/app/eveai_entitlements
+      - ../common:/app/common
+      - ../config:/app/config
+      - ../scripts:/app/scripts
+      - ../patched_packages:/app/patched_packages
+      - eveai_logs:/app/logs
+    depends_on:
+      db:
+        condition: service_healthy
+      redis:
+        condition: service_healthy
+      minio:
+        condition: service_healthy
+    networks:
+      - eveai-network
+
+
   db:
     hostname: db
     image: ankane/pgvector

docker/compose_stackhero.yaml

@@ -145,6 +145,28 @@ services:
     networks:
       - eveai-network
 
+  eveai_beat:
+    platform: linux/amd64
+    image: josakola/eveai_beat:latest
+    environment:
+      <<: *common-variables
+      COMPONENT_NAME: eveai_beat
+    volumes:
+      - eveai_logs:/app/logs
+    networks:
+      - eveai-network
+
+  eveai_entitlements:
+    platform: linux/amd64
+    image: josakola/eveai_entitlements:latest
+    environment:
+      <<: *common-variables
+      COMPONENT_NAME: eveai_entitlements
+    volumes:
+      - eveai_logs:/app/logs
+    networks:
+      - eveai-network
+
   flower:
     image: josakola/flower:latest
     environment:

docker/eveai_beat/Dockerfile

@@ -0,0 +1,65 @@
+ARG PYTHON_VERSION=3.12.3
+FROM python:${PYTHON_VERSION}-slim as base
+
+# Prevents Python from writing pyc files.
+ENV PYTHONDONTWRITEBYTECODE=1
+
+# Keeps Python from buffering stdout and stderr to avoid situations where
+# the application crashes without emitting any logs due to buffering.
+ENV PYTHONUNBUFFERED=1
+
+# Create directory for patched packages and set permissions
+RUN mkdir -p /app/patched_packages && \
+    chmod 777 /app/patched_packages
+
+# Ensure patches are applied to the application.
+ENV PYTHONPATH=/app/patched_packages:$PYTHONPATH
+
+WORKDIR /app
+
+# Create a non-privileged user that the app will run under.
+# See https://docs.docker.com/go/dockerfile-user-best-practices/
+ARG UID=10001
+RUN adduser \
+    --disabled-password \
+    --gecos "" \
+    --home "/nonexistent" \
+    --shell "/bin/bash" \
+    --no-create-home \
+    --uid "${UID}" \
+    appuser
+
+# Install necessary packages and build tools
+#RUN apt-get update && apt-get install -y \
+#    build-essential \
+#    gcc \
+#    && apt-get clean \
+#    && rm -rf /var/lib/apt/lists/*
+
+# Create logs directory and set permissions
+RUN mkdir -p /app/logs && chown -R appuser:appuser /app/logs
+
+# Install Python dependencies.
+
+# Download dependencies as a separate step to take advantage of Docker's caching.
+# Leverage a cache mount to /root/.cache/pip to speed up subsequent builds.
+# Leverage a bind mount to requirements.txt to avoid having to copy them into
+# into this layer.
+
+COPY requirements.txt /app/
+RUN python -m pip install -r /app/requirements.txt
+
+# Copy the source code into the container.
+COPY eveai_beat /app/eveai_beat
+COPY common /app/common
+COPY config /app/config
+COPY scripts /app/scripts
+COPY patched_packages /app/patched_packages
+COPY --chown=root:root scripts/entrypoint_no_db.sh /app/scripts/
+
+# Set ownership of the application directory to the non-privileged user
+RUN chown -R appuser:appuser /app
+
+# Set entrypoint and command
+ENTRYPOINT ["/app/scripts/entrypoint_no_db.sh"]
+CMD ["/app/scripts/start_eveai_beat.sh"]

docker/eveai_entitlements/Dockerfile

@@ -0,0 +1,69 @@
+ARG PYTHON_VERSION=3.12.3
+FROM python:${PYTHON_VERSION}-slim as base
+
+# Prevents Python from writing pyc files.
+ENV PYTHONDONTWRITEBYTECODE=1
+
+# Keeps Python from buffering stdout and stderr to avoid situations where
+# the application crashes without emitting any logs due to buffering.
+ENV PYTHONUNBUFFERED=1
+
+# Create directory for patched packages and set permissions
+RUN mkdir -p /app/patched_packages && \
+    chmod 777 /app/patched_packages
+
+# Ensure patches are applied to the application.
+ENV PYTHONPATH=/app/patched_packages:$PYTHONPATH
+
+WORKDIR /app
+
+# Create a non-privileged user that the app will run under.
+# See https://docs.docker.com/go/dockerfile-user-best-practices/
+ARG UID=10001
+RUN adduser \
+    --disabled-password \
+    --gecos "" \
+    --home "/nonexistent" \
+    --shell "/bin/bash" \
+    --no-create-home \
+    --uid "${UID}" \
+    appuser
+
+# Install necessary packages and build tools
+RUN apt-get update && apt-get install -y \
+    build-essential \
+    gcc \
+    postgresql-client \
+    && apt-get clean \
+    && rm -rf /var/lib/apt/lists/*
+
+# Create logs directory and set permissions
+RUN mkdir -p /app/logs && chown -R appuser:appuser /app/logs
+
+# Install Python dependencies.
+
+# Download dependencies as a separate step to take advantage of Docker's caching.
+# Leverage a cache mount to /root/.cache/pip to speed up subsequent builds.
+# Leverage a bind mount to requirements.txt to avoid having to copy them into
+# into this layer.
+
+COPY requirements.txt /app/
+RUN python -m pip install -r /app/requirements.txt
+
+# Copy the source code into the container.
+COPY eveai_entitlements /app/eveai_entitlements
+COPY common /app/common
+COPY config /app/config
+COPY scripts /app/scripts
+COPY patched_packages /app/patched_packages
+COPY --chown=root:root scripts/entrypoint.sh /app/scripts/
+
+# Set permissions for entrypoint script
+RUN chmod 777 /app/scripts/entrypoint.sh
+
+# Set ownership of the application directory to the non-privileged user
+RUN chown -R appuser:appuser /app
+
+# Set entrypoint and command
+ENTRYPOINT ["/app/scripts/entrypoint.sh"]
+CMD ["/app/scripts/start_eveai_entitlements.sh"]

docker/release_and_tag_eveai.sh

@@ -0,0 +1,60 @@
+#!/bin/bash
+
+# Initialize variables
+RELEASE_VERSION=""
+RELEASE_MESSAGE=""
+DOCKER_ACCOUNT="josakola"  # Your Docker account name
+
+# Parse input arguments
+while getopts r:m: flag
+do
+    case "${flag}" in
+        r) RELEASE_VERSION=${OPTARG};;
+        m) RELEASE_MESSAGE=${OPTARG};;
+        *)
+          echo "Usage: $0 -r <release_version> -m <release_message>"
+          exit 1 ;;
+    esac
+done
+
+# Ensure both version and message are provided
+if [ -z "$RELEASE_VERSION" ]; then
+  echo "Error: Release version not provided. Use -r <release_version>"
+  exit 1
+fi
+
+if [ -z "$RELEASE_MESSAGE" ]; then
+  echo "Error: Release message not provided. Use -m <release_message>"
+  exit 1
+fi
+
+# Path to your docker-compose file
+DOCKER_COMPOSE_FILE="compose_dev.yaml"
+
+# Get all the services defined in the docker-compose file
+SERVICES=$(docker-compose -f $DOCKER_COMPOSE_FILE config --services)
+
+# Tag and push images for all services that belong to your Docker account
+for SERVICE in $SERVICES; do
+    DOCKER_IMAGE="your-docker-repo/$SERVICE"
+
+    # Check if the image starts with your Docker account name
+    if [[ $DOCKER_IMAGE == $DOCKER_ACCOUNT* ]]; then
+        echo "Tagging Docker image for service: $SERVICE with version: $RELEASE_VERSION"
+
+        # Tag the 'latest' image with the new release version
+        docker tag $DOCKER_IMAGE:latest $DOCKER_IMAGE:$RELEASE_VERSION
+
+        # Push the newly tagged image to Docker Hub
+        docker push $DOCKER_IMAGE:$RELEASE_VERSION
+    else
+        echo "Skipping service: $SERVICE (not part of Docker account $DOCKER_ACCOUNT)"
+    fi
+done
+
+# Step 3: Tag the Git repository with the release version
+echo "Tagging Git repository with version: $RELEASE_VERSION"
+git tag -a v$RELEASE_VERSION -m "Release $RELEASE_VERSION: $RELEASE_MESSAGE"
+git push origin v$RELEASE_VERSION
+
+echo "Release process completed for version: $RELEASE_VERSION"

eveai_api/__init__.py

@@ -56,13 +56,6 @@ def create_app(config_file=None):
         app.logger.debug(f'Request URL: {request.url}')
         app.logger.debug(f'Request headers: {dict(request.headers)}')
 
-        # Log request arguments
-        app.logger.debug(f'Request args: {request.args}')
-
-        # Log form data if it's a POST request
-        if request.method == 'POST':
-            app.logger.debug(f'Form data: {request.form}')
-
         # Log JSON data if the content type is application/json
         if request.is_json:
             app.logger.debug(f'JSON data: {request.json}')
@@ -95,6 +88,10 @@ def create_app(config_file=None):
                 # Don't raise the exception here, let the request continue
                 # The appropriate error handling will be done in the specific endpoints
 
+    @app.route('/api/v1')
+    def swagger():
+        return api_rest.render_doc()
+
     return app
 
 

eveai_api/api/document_api.py

@@ -33,6 +33,7 @@ document_ns = Namespace('documents', description='Document related operations')
 
 # Define models for request parsing and response serialization
 upload_parser = reqparse.RequestParser()
+upload_parser.add_argument('catalog_id', location='form', type=int, required=True, help='The catalog to add the file to')
 upload_parser.add_argument('file', location='files', type=FileStorage, required=True, help='The file to upload')
 upload_parser.add_argument('name', location='form', type=str, required=False, help='Name of the document')
 upload_parser.add_argument('language', location='form', type=str, required=True, help='Language of the document')
@@ -75,6 +76,7 @@ class AddDocument(Resource):
             validate_file_type(extension)
 
             api_input = {
+                'catalog_id': args.get('catalog_id'),
                 'name': args.get('name') or filename,
                 'language': args.get('language'),
                 'user_context': args.get('user_context'),
@@ -102,6 +104,7 @@ class AddDocument(Resource):
 
 # Models for AddURL
 add_url_model = document_ns.model('AddURL', {
+    'catalog_id': fields.Integer(required='True', description='ID of the catalog the URL needs to be added to'),
     'url': fields.String(required=True, description='URL of the document to add'),
     'name': fields.String(required=False, description='Name of the document'),
     'language': fields.String(required=True, description='Language of the document'),
@@ -138,6 +141,7 @@ class AddURL(Resource):
             file_content, filename, extension = process_url(args['url'], tenant_id)
 
             api_input = {
+                'catalog_id': args['catalog_id'],
                 'url': args['url'],
                 'name': args.get('name') or filename,
                 'language': args['language'],
@@ -213,7 +217,8 @@ class DocumentResource(Resource):
     @document_ns.response(200, 'Document refreshed successfully')
     def post(self, document_id):
         """Refresh a document"""
-        new_version, result = refresh_document(document_id)
+        tenant_id = get_jwt_identity()
+        new_version, result = refresh_document(document_id, tenant_id)
         if new_version:
             return {'message': f'Document refreshed. New version: {new_version.id}. Task ID: {result}'}, 200
         else:

eveai_api/views/healthz_views.py

@@ -46,7 +46,7 @@ def check_database():
 def check_celery():
     try:
         # Send a simple task to Celery
-        result = current_celery.send_task('tasks.ping', queue='embeddings')
+        result = current_celery.send_task('ping', queue='eveai_workers.ping')
         response = result.get(timeout=10)  # Wait for up to 10 seconds for a response
         return response == 'pong'
     except CeleryTimeoutError:

eveai_app/__init__.py

@@ -136,6 +136,8 @@ def register_blueprints(app):
     app.register_blueprint(interaction_bp)
     from .views.entitlements_views import entitlements_bp
     app.register_blueprint(entitlements_bp)
+    from .views.administration_views import administration_bp
+    app.register_blueprint(administration_bp)
     from .views.healthz_views import healthz_bp, init_healtz
     app.register_blueprint(healthz_bp)
     init_healtz(app)

eveai_app/templates/administration/trigger_actions.html

@@ -0,0 +1,22 @@
+{% extends 'base.html' %}
+{% from "macros.html" import render_selectable_table, render_pagination, render_field %}
+{% block title %}Trigger Actions{% endblock %}
+{% block content_title %}Trigger Actions{% endblock %}
+{% block content_description %}Manually trigger batch actions{% endblock %}
+{% block content %}
+
+<!-- Trigger action Form -->
+<form method="POST" action="{{ url_for('administration_bp.handle_trigger_action') }}">
+    <div class="form-group mt-3">
+        <button type="submit" name="action" value="update_usages" class="btn btn-secondary">Update Usages</button>
+    </div>
+</form>
+
+{% endblock %}
+
+{% block content_footer %}
+{% endblock %}
+
+{% block scripts %}
+{% endblock %}
+

eveai_app/templates/document/catalog.html

@@ -0,0 +1,23 @@
+{% extends 'base.html' %}
+{% from "macros.html" import render_field %}
+
+{% block title %}Catalog Registration{% endblock %}
+
+{% block content_title %}Register Catalog{% endblock %}
+{% block content_description %}Define a new catalog of documents in Evie's Library{% endblock %}
+
+{% block content %}
+    <form method="post">
+        {{ form.hidden_tag() }}
+        {%  set disabled_fields = [] %}
+        {%  set exclude_fields = [] %}
+        {% for field in form %}
+            {{ render_field(field, disabled_fields, exclude_fields) }}
+        {% endfor %}
+        <button type="submit" class="btn btn-primary">Register Catalog</button>
+    </form>
+{% endblock %}
+
+{% block content_footer %}
+
+{% endblock %}

eveai_app/templates/document/catalogs.html

@@ -0,0 +1,24 @@
+{% extends 'base.html' %}
+{%  from 'macros.html' import render_selectable_table, render_pagination %}
+
+{% block title %}Documents{% endblock %}
+
+{% block content_title %}Catalogs{% endblock %}
+{% block content_description %}View Catalogs for Tenant{% 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_catalog_selection') }}">
+        {{ render_selectable_table(headers=["Catalog ID", "Name"], rows=rows, selectable=True, id="catalogsTable") }}
+        <div class="form-group mt-3">
+            <button type="submit" name="action" value="set_session_catalog" class="btn btn-primary">Set Session Catalog</button>
+            <button type="submit" name="action" value="edit_catalog" class="btn btn-primary">Edit Catalog</button>
+        </div>
+    </form>
+</div>
+{% endblock %}
+
+{% block content_footer %}
+    {{ render_pagination(pagination, 'document_bp.catalogs') }}
+{% endblock %}

eveai_app/templates/document/document_versions.html

@@ -10,7 +10,7 @@
 {% block content %}
 <div class="container">
     <form method="POST" action="{{ url_for('document_bp.handle_document_version_selection') }}">
-        {{ render_selectable_table(headers=["ID", "URL", "File Loc.", "File Name", "File Type", "Process.", "Proces. Start", "Proces. Finish", "Proces. Error"], rows=rows, selectable=True, id="versionsTable") }}
+        {{ render_selectable_table(headers=["ID", "URL", "Object Name", "File Type", "Process.", "Proces. Start", "Proces. Finish", "Proces. Error"], rows=rows, selectable=True, id="versionsTable") }}
         <div class="form-group mt-3">
             <button type="submit" name="action" value="edit_document_version" class="btn btn-primary">Edit Document Version</button>
             <button type="submit" name="action" value="process_document_version" class="btn btn-danger">Process Document Version</button>

eveai_app/templates/document/document_versions_list_view.html

@@ -23,15 +23,23 @@
 
     {{ render_collapsible_section('Filter', 'Filter Options', filter_form) }}
 
-    <!-- Document Versions Table -->
-    {{ render_selectable_sortable_table(
-        headers=["ID", "File Type", "Processing", "Processing Start", "Processing Finish", "Processing Error"],
-        rows=rows,
-        selectable=True,
-        id="documentVersionsTable",
-        sort_by=sort_by,
-        sort_order=sort_order
-    ) }}
+    <div class="form-group mt-3">
+        <form method="POST" action="{{ url_for('document_bp.handle_document_version_selection') }}">
+            <!-- Document Versions Table -->
+            {{ render_selectable_sortable_table(
+                headers=["ID", "File Type", "Processing", "Processing Start", "Processing Finish", "Processing Error"],
+                rows=rows,
+                selectable=True,
+                id="documentVersionsTable",
+                sort_by=sort_by,
+                sort_order=sort_order
+            ) }}
+            <div class="form-group mt-4">
+                <button type="submit" name="action" value="edit_document_version" class="btn btn-primary">Edit Document Version</button>
+                <button type="submit" name="action" value="process_document_version" class="btn btn-danger">Process Document Version</button>
+            </div>
+        </form>
+    </div>
 {% endblock %}
 
 {% block content_footer %}

eveai_app/templates/document/documents.html

@@ -1,5 +1,5 @@
 {% extends 'base.html' %}
-{%  from 'macros.html' import render_selectable_table, render_pagination %}
+{% from 'macros.html' import render_selectable_table, render_pagination, render_filter_field, render_date_filter_field, render_collapsible_section, render_selectable_sortable_table_with_dict_headers %}
 
 {% block title %}Documents{% endblock %}
 
@@ -8,18 +8,88 @@
 {% 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_selection') }}">
-        {{ render_selectable_table(headers=["Document ID", "Name", "Valid From", "Valid To"], rows=rows, selectable=True, id="documentsTable") }}
-        <div class="form-group mt-3">
-            <button type="submit" name="action" value="edit_document" class="btn btn-primary">Edit Document</button>
-            <button type="submit" name="action" value="document_versions" class="btn btn-secondary">Show Document Versions</button>
-            <button type="submit" name="action" value="refresh_document" class="btn btn-secondary">Refresh Document (new version)</button>
-        </div>
-    </form>
-</div>
+    <!-- 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>
+        </form>
+    {% endset %}
+
+    {{ render_collapsible_section('Filter', 'Filter Options', filter_form) }}
+
+    <div class="form-group mt-3">
+        <form method="POST" action="{{ url_for('document_bp.handle_document_selection') }}">
+            <!-- Documents Table -->
+            {{ render_selectable_sortable_table_with_dict_headers(
+                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"}
+                ],
+                rows=rows,
+                selectable=True,
+                id="documentsTable",
+                sort_by=sort_by,
+                sort_order=sort_order
+            ) }}
+            <div class="form-group mt-4">
+                <button type="submit" name="action" value="edit_document" class="btn btn-primary">Edit Document</button>
+                <button type="submit" name="action" value="document_versions" class="btn btn-secondary">Show Document Versions</button>
+                <button type="submit" name="action" value="refresh_document" class="btn btn-secondary">Refresh Document (new version)</button>
+            </div>
+        </form>
+    </div>
 {% endblock %}
 
 {% block content_footer %}
     {{ render_pagination(pagination, 'document_bp.documents') }}
 {% endblock %}
+
+{% block scripts %}
+<script>
+document.addEventListener('DOMContentLoaded', function() {
+    const table = document.getElementById('documentsTable');
+    const headers = table.querySelectorAll('th.sortable');
+
+    headers.forEach(header => {
+        header.addEventListener('click', function() {
+            const sortBy = this.dataset.sort;
+            let sortOrder = 'asc';
+
+            if (this.querySelector('.fa-sort-up')) {
+                sortOrder = 'desc';
+            } else if (this.querySelector('.fa-sort-down')) {
+                sortOrder = 'none';
+            }
+
+            window.location.href = updateQueryStringParameter(window.location.href, 'sort_by', sortBy);
+            window.location.href = updateQueryStringParameter(window.location.href, 'sort_order', sortOrder);
+        });
+    });
+
+    function updateQueryStringParameter(uri, key, value) {
+        var re = new RegExp("([?&])" + key + "=.*?(&|$)", "i");
+        var separator = uri.indexOf('?') !== -1 ? "&" : "?";
+        if (uri.match(re)) {
+            return uri.replace(re, '$1' + key + "=" + value + '$2');
+        }
+        else {
+            return uri + separator + key + "=" + value;
+        }
+    }
+
+    table.addEventListener('change', function(event) {
+        if (event.target.type === 'radio') {
+            var selectedRow = event.target.closest('tr');
+            var documentId = selectedRow.cells[1].textContent;
+            console.log('Selected Document ID:', documentId);
+        }
+    });
+});
+</script>
+{% endblock %}

eveai_app/templates/document/edit_catalog.html

@@ -0,0 +1,25 @@
+{% extends 'base.html' %}
+{% from "macros.html" import render_field %}
+
+{% block title %}Edit Catalog{% endblock %}
+
+{% block content_title %}Edit Catalog{% endblock %}
+{% block content_description %}Edit a catalog of documents in Evie's Library.
+When you change chunking of embedding information, you'll need to manually refresh the library if you want immediate impact.
+{% endblock %}
+
+{% block content %}
+    <form method="post">
+        {{ form.hidden_tag() }}
+        {%  set disabled_fields = [] %}
+        {%  set exclude_fields = [] %}
+        {% for field in form %}
+            {{ render_field(field, disabled_fields, exclude_fields) }}
+        {% endfor %}
+        <button type="submit" class="btn btn-primary">Register Catalog</button>
+    </form>
+{% endblock %}
+
+{% block content_footer %}
+
+{% endblock %}

eveai_app/templates/document/edit_document.html

@@ -8,11 +8,17 @@
 {% block content %}
     <form method="post">
         {{ form.hidden_tag() }}
-        {%  set disabled_fields = [] %}
-        {%  set exclude_fields = [] %}
-        {% for field in form %}
-            {{ render_field(field, disabled_fields, exclude_fields) }}
-        {% endfor %}
+        {% set disabled_fields = [] %}
+        {% set exclude_fields = [] %}
+
+        {{ render_field(form.name, disabled_fields, exclude_fields) }}
+        {{ render_field(form.valid_from, disabled_fields, exclude_fields) }}
+        {{ render_field(form.valid_to, disabled_fields, exclude_fields) }}
+
+        <div class="form-group">
+            <label for="catalog_name">Catalog</label>
+            <input type="text" class="form-control" id="catalog_name" value="{{ catalog_name }}" readonly>
+        </div>
         <button type="submit" class="btn btn-primary">Update Document</button>
     </form>
 {% endblock %}

eveai_app/templates/entitlements/view_usages.html

@@ -0,0 +1,28 @@
+{% extends 'base.html' %}
+{% from "macros.html" import render_selectable_table, render_pagination %}
+
+{% block title %}View License Usage{% endblock %}
+
+{%  block content_title %}View License Usage{% endblock %}
+{% block content_description %}View License Usage{% endblock %}
+
+{% block content %}
+<form action="{{ url_for('user_bp.handle_user_action') }}" method="POST">
+    {{ render_selectable_table(headers=["Usage ID", "Start Date", "End Date", "Storage (MiB)", "Embedding (MiB)", "Interaction (tokens)"], rows=rows, selectable=False, id="usagesTable") }}
+<!--    <div class="form-group mt-3">-->
+<!--        <button type="submit" name="action" value="edit_user" class="btn btn-primary">Edit Selected User</button>-->
+<!--        <button type="submit" name="action" value="resend_confirmation_email" class="btn btn-secondary">Resend Confirmation Email</button>-->
+<!--        <button type="submit" name="action" value="send_password_reset_email" class="btn btn-secondary">Send Password Reset Email</button>-->
+<!--        <button type="submit" name="action" value="reset_uniquifier" class="btn btn-secondary">Reset Uniquifier</button>-->
+<!--        &lt;!&ndash; Additional buttons can be added here for other actions &ndash;&gt;-->
+<!--    </div>-->
+</form>
+{% endblock %}
+
+{% block content_footer %}
+    {{ render_pagination(pagination, 'user_bp.select_tenant') }}
+{% endblock %}
+
+{% block scripts %}
+
+{% endblock %}

eveai_app/templates/header.html

@@ -1,5 +1,5 @@
 <header class="header-2">
-    <div class="page-header min-vh-25" style="background-image: url({{url_for('static', filename='/assets/img/EveAI_bg.jpg')}})" loading="lazy">
+    <div class="page-header min-vh-25" style="background-image: url({{url_for('static', filename='/assets/img/EveAI_bg.jpg')}}); background-position: top left; background-repeat: no-repeat; background-size: cover;" loading="lazy">
         <span class="mask bg-gradient-primary opacity-4"></span>
         <div class="container">
             <div class="row">

eveai_app/templates/interaction/view_chat_session.html

@@ -54,7 +54,7 @@
                             {% if embedding.url %}
                             <a href="{{ embedding.url }}" target="_blank">{{ embedding.url }}</a>
                             {% else %}
-                            {{ embedding.file_name }}
+                            {{ embedding.object_name }}
                             {% endif %}
                         </li>
                         {% endfor %}

eveai_app/templates/macros.html

@@ -177,6 +177,48 @@
 </div>
 {% endmacro %}
 
+{% macro render_selectable_sortable_table_with_dict_headers(headers, rows, selectable, id, sort_by, sort_order) %}
+<div class="card">
+  <div class="table-responsive">
+    <table class="table align-items-center mb-0" id="{{ id }}">
+      <thead>
+        <tr>
+          {% if selectable %}
+            <th class="text-uppercase text-secondary text-xxs font-weight-bolder opacity-7">Select</th>
+          {% endif %}
+          {% for header in headers %}
+            <th class="text-uppercase text-secondary text-xxs font-weight-bolder opacity-7 sortable" data-sort="{{ header['sort'] }}">
+              {{ header['text'] }}
+              {% if sort_by == header['sort'] %}
+                {% if sort_order == 'asc' %}
+                  <i class="fas fa-sort-up"></i>
+                {% elif sort_order == 'desc' %}
+                  <i class="fas fa-sort-down"></i>
+                {% endif %}
+              {% else %}
+                <i class="fas fa-sort"></i>
+              {% endif %}
+            </th>
+          {% endfor %}
+        </tr>
+      </thead>
+      <tbody>
+        {% for row in rows %}
+          <tr>
+            {% if selectable %}
+              <td><input type="radio" name="selected_row" value="{{ row[0].value }}"></td>
+            {% endif %}
+            {% for cell in row %}
+              <td>{{ cell.value }}</td>
+            {% endfor %}
+          </tr>
+        {% endfor %}
+      </tbody>
+    </table>
+  </div>
+</div>
+{% endmacro %}
+
 {% macro render_accordion(accordion_id, accordion_items, header_title, header_description) %}
 <div class="accordion-1">
   <div class="container">

eveai_app/templates/navbar.html

@@ -81,6 +81,8 @@
                             {% endif %}
                             {% if current_user.is_authenticated %}
                                 {{ dropdown('Document Mgmt', 'note_stack', [
+                                    {'name': 'Add Catalog', 'url': '/document/catalog', 'roles': ['Super User', 'Tenant Admin']},
+                                    {'name': 'All Catalogs', 'url': '/document/catalogs', 'roles': ['Super User', 'Tenant Admin']},
                                     {'name': 'Add Document', 'url': '/document/add_document', 'roles': ['Super User', 'Tenant Admin']},
                                     {'name': 'Add URL', 'url': '/document/add_url', 'roles': ['Super User', 'Tenant Admin']},
                                     {'name': 'Add a list of URLs', 'url': '/document/add_urls', 'roles': ['Super User', 'Tenant Admin']},
@@ -98,6 +100,8 @@
                                 {{ dropdown('Administration', 'settings', [
                                     {'name': 'License Tier Registration', 'url': '/entitlements/license_tier', 'roles': ['Super User']},
                                     {'name': 'All License Tiers', 'url': '/entitlements/view_license_tiers', 'roles': ['Super User']},
+                                    {'name': 'Trigger Actions', 'url': '/administration/trigger_actions', 'roles': ['Super User']},
+                                    {'name': 'Usage', 'url': '/entitlements/view_usages', 'roles': ['Super User', 'Tenant Admin']},
                                 ]) }}
                             {% endif %}
                             {% if current_user.is_authenticated %}
@@ -112,6 +116,17 @@
                             {% endif %}
                         </ul>
                         {% if current_user.is_authenticated %}
+                            <ul class="navbar-nav d-lg-block d-none">
+                                <li class="nav-item">
+                                    <a href="/document/catalogs" class="btn btn-sm  bg-gradient-primary  mb-0 me-2">
+                                        {% if 'catalog_name' in session %}
+                                            CATALOG: {{ session['catalog_name'] }}
+                                        {% else %}
+                                            CHOOSE CATALOG
+                                        {% endif %}
+                                    </a>
+                                </li>
+                            </ul>
                             <ul class="navbar-nav d-lg-block d-none">
                                 <li class="nav-item">
                                     <a href="/session_defaults" class="btn btn-sm  bg-gradient-primary  mb-0">

eveai_app/templates/user/tenant.html

@@ -10,7 +10,7 @@
     <form method="post">
         {{ form.hidden_tag() }}
         <!-- Main Tenant Information -->
-        {% set main_fields = ['name', 'website', 'default_language', 'allowed_languages', 'rag_context', 'type'] %}
+        {% set main_fields = ['name', 'website', 'default_language', 'allowed_languages', 'timezone','rag_context', 'type'] %}
         {% for field in form %}
             {{ render_included_field(field, disabled_fields=[], include_fields=main_fields) }}
         {% endfor %}
@@ -30,21 +30,6 @@
                                 License Information
                             </a>
                         </li>
-                        <li class="nav-item">
-                            <a class="nav-link mb-0 px-0 py-1" data-toggle="tab" href="#chunking-tab" role="tab" aria-controls="chunking" aria-selected="false">
-                                Chunking
-                            </a>
-                        </li>
-                        <li class="nav-item">
-                            <a class="nav-link mb-0 px-0 py-1" data-toggle="tab" href="#embedding-search-tab" role="tab" aria-controls="html-chunking" aria-selected="false">
-                                Embedding Search
-                            </a>
-                        </li>
-                        <li class="nav-item">
-                            <a class="nav-link mb-0 px-0 py-1" data-toggle="tab" href="#tuning-tab" role="tab" aria-controls="html-chunking" aria-selected="false">
-                                Tuning
-                            </a>
-                        </li>
                     </ul>
                 </div>
                 <div class="tab-content tab-space">
@@ -62,8 +47,10 @@
                             {{ render_included_field(field, disabled_fields=[], include_fields=license_fields) }}
                         {% endfor %}
                         <!-- Register API Key Button -->
-                        <button type="button" class="btn btn-primary" onclick="generateNewChatApiKey()">Register Chat API Key</button>
-                        <button type="button" class="btn btn-primary" onclick="generateNewApiKey()">Register API Key</button>
+                        <div class="form-group">
+                            <button type="button" class="btn btn-primary" onclick="generateNewChatApiKey()">Register Chat API Key</button>
+                            <button type="button" class="btn btn-primary" onclick="generateNewApiKey()">Register API Key</button>
+                        </div>
                         <!-- API Key Display Field -->
                         <div id="chat-api-key-field" style="display:none;">
                             <label for="chat-api-key">Chat API Key:</label>
@@ -78,27 +65,6 @@
                             <p id="copy-message" style="display:none;color:green;">API key copied to clipboard</p>
                         </div>
                     </div>
-                    <!-- Chunking Settings Tab -->
-                    <div class="tab-pane fade" id="chunking-tab" role="tabpanel">
-                        {% set html_fields = ['html_tags', 'html_end_tags', 'html_included_elements', 'html_excluded_elements', 'html_excluded_classes', 'min_chunk_size', 'max_chunk_size'] %}
-                        {% for field in form %}
-                            {{ render_included_field(field, disabled_fields=[], include_fields=html_fields) }}
-                        {% endfor %}
-                    </div>
-                    <!-- Embedding Search Settings Tab -->
-                    <div class="tab-pane fade" id="embedding-search-tab" role="tabpanel">
-                        {% set es_fields = ['es_k', 'es_similarity_threshold', ] %}
-                        {% for field in form %}
-                            {{ render_included_field(field, disabled_fields=[], include_fields=es_fields) }}
-                        {% endfor %}
-                    </div>
-                    <!-- Tuning Settings Tab -->
-                    <div class="tab-pane fade" id="tuning-tab" role="tabpanel">
-                        {% set tuning_fields = ['embed_tuning', 'rag_tuning', ] %}
-                        {% for field in form %}
-                            {{ render_included_field(field, disabled_fields=[], include_fields=tuning_fields) }}
-                        {% endfor %}
-                    </div>
                 </div>
             </div>
         </div>

eveai_app/templates/user/tenant_overview.html

@@ -30,21 +30,6 @@
                                 License Information
                             </a>
                         </li>
-                        <li class="nav-item">
-                            <a class="nav-link mb-0 px-0 py-1" data-toggle="tab" href="#chunking-tab" role="tab" aria-controls="chunking" aria-selected="false">
-                                Chunking
-                            </a>
-                        </li>
-                        <li class="nav-item">
-                            <a class="nav-link mb-0 px-0 py-1" data-toggle="tab" href="#embedding-search-tab" role="tab" aria-controls="html-chunking" aria-selected="false">
-                                Embedding Search
-                            </a>
-                        </li>
-                        <li class="nav-item">
-                            <a class="nav-link mb-0 px-0 py-1" data-toggle="tab" href="#tuning-tab" role="tab" aria-controls="html-chunking" aria-selected="false">
-                                Tuning
-                            </a>
-                        </li>
                     </ul>
                 </div>
                 <div class="tab-content tab-space">
@@ -78,27 +63,6 @@
                             <p id="copy-message" style="display:none;color:green;">API key copied to clipboard</p>
                         </div>
                     </div>
-                    <!-- Chunking Settings Tab -->
-                    <div class="tab-pane fade" id="chunking-tab" role="tabpanel">
-                        {% set html_fields = ['html_tags', 'html_end_tags', 'html_included_elements', 'html_excluded_elements', 'html_excluded_classes', 'min_chunk_size', 'max_chunk_size'] %}
-                        {% for field in form %}
-                            {{ render_included_field(field, disabled_fields=html_fields, include_fields=html_fields) }}
-                        {% endfor %}
-                    </div>
-                    <!-- Embedding Search Settings Tab -->
-                    <div class="tab-pane fade" id="embedding-search-tab" role="tabpanel">
-                        {% set es_fields = ['es_k', 'es_similarity_threshold', ] %}
-                        {% for field in form %}
-                            {{ render_included_field(field, disabled_fields=es_fields, include_fields=es_fields) }}
-                        {% endfor %}
-                    </div>
-                    <!-- Tuning Settings Tab -->
-                    <div class="tab-pane fade" id="tuning-tab" role="tabpanel">
-                        {% set tuning_fields = ['embed_tuning', 'rag_tuning', ] %}
-                        {% for field in form %}
-                            {{ render_included_field(field, disabled_fields=tuning_fields, include_fields=tuning_fields) }}
-                        {% endfor %}
-                    </div>
                 </div>
             </div>
         </div>

eveai_app/views/administration_forms.py

@@ -0,0 +1,7 @@
+from flask import current_app
+from flask_wtf import FlaskForm
+from wtforms.fields.simple import SubmitField
+
+
+class TriggerActionForm(FlaskForm):
+    submit = SubmitField('Submit')

eveai_app/views/administration_views.py

@@ -0,0 +1,39 @@
+import uuid
+from datetime import datetime as dt, timezone as tz
+from flask import request, redirect, flash, render_template, Blueprint, session, current_app, jsonify
+from flask_security import hash_password, roles_required, roles_accepted, current_user
+from itsdangerous import URLSafeTimedSerializer
+from sqlalchemy.exc import SQLAlchemyError
+
+from common.utils.celery_utils import current_celery
+from common.utils.view_assistants import prepare_table_for_macro, form_validation_failed
+from common.utils.nginx_utils import prefixed_url_for
+from .administration_forms import TriggerActionForm
+
+administration_bp = Blueprint('administration_bp', __name__, url_prefix='/administration')
+
+
+@administration_bp.route('/trigger_actions', methods=['GET'])
+@roles_accepted('Super User')
+def trigger_actions():
+    form = TriggerActionForm()
+    return render_template('administration/trigger_actions.html', form=form)
+
+
+@administration_bp.route('/handle_trigger_action', methods=['POST'])
+@roles_accepted('Super User')
+def handle_trigger_action():
+    action = request.form['action']
+    match action:
+        case 'update_usages':
+            try:
+                # Use send_task to trigger the task since it's part of another component (eveai_entitlements)
+                task = current_celery.send_task('update_usages', queue='entitlements')
+
+                current_app.logger.info(f"Usage update task triggered: {task.id}")
+                flash('Usage update task has been triggered successfully!', 'success')
+            except Exception as e:
+                current_app.logger.error(f"Failed to trigger usage update task: {str(e)}")
+                flash(f'Failed to trigger usage update: {str(e)}', 'danger')
+
+    return redirect(prefixed_url_for('administration_bp.trigger_actions'))

eveai_app/views/document_forms.py

@@ -1,8 +1,8 @@
 from flask import session, current_app
 from flask_wtf import FlaskForm
-from wtforms import (StringField, BooleanField, SubmitField, DateField,
+from wtforms import (StringField, BooleanField, SubmitField, DateField, IntegerField, FloatField, SelectMultipleField,
                      SelectField, FieldList, FormField, TextAreaField, URLField)
-from wtforms.validators import DataRequired, Length, Optional, URL, ValidationError
+from wtforms.validators import DataRequired, Length, Optional, URL, ValidationError, NumberRange
 from flask_wtf.file import FileField, FileAllowed, FileRequired
 import json
 
@@ -23,6 +23,36 @@ def validate_json(form, field):
             raise ValidationError('Invalid JSON format')
 
 
+class CatalogForm(FlaskForm):
+    name = StringField('Name', validators=[DataRequired(), Length(max=50)])
+    description = TextAreaField('Description', validators=[Optional()])
+    # HTML Embedding Variables
+    html_tags = StringField('HTML Tags', validators=[DataRequired()],
+                            default='p, h1, h2, h3, h4, h5, h6, li, , tbody, tr, td')
+    html_end_tags = StringField('HTML End Tags', validators=[DataRequired()],
+                                default='p, li')
+    html_included_elements = StringField('HTML Included Elements', validators=[Optional()])
+    html_excluded_elements = StringField('HTML Excluded Elements', validators=[Optional()])
+    html_excluded_classes = StringField('HTML Excluded Classes', validators=[Optional()])
+    min_chunk_size = IntegerField('Minimum Chunk Size (2000)', validators=[NumberRange(min=0), Optional()],
+                                  default=2000)
+    max_chunk_size = IntegerField('Maximum Chunk Size (3000)', validators=[NumberRange(min=0), Optional()],
+                                  default=3000)
+    # Embedding Search variables
+    es_k = IntegerField('Limit for Searching Embeddings (5)',
+                        default=5,
+                        validators=[NumberRange(min=0)])
+    es_similarity_threshold = FloatField('Similarity Threshold for Searching Embeddings (0.5)',
+                                         default=0.5,
+                                         validators=[NumberRange(min=0, max=1)])
+    # Chat Variables
+    chat_RAG_temperature = FloatField('RAG Temperature', default=0.3, validators=[NumberRange(min=0, max=1)])
+    chat_no_RAG_temperature = FloatField('No RAG Temperature', default=0.5, validators=[NumberRange(min=0, max=1)])
+    # Tuning variables
+    embed_tuning = BooleanField('Enable Embedding Tuning', default=False)
+    rag_tuning = BooleanField('Enable RAG Tuning', default=False)
+
+
 class AddDocumentForm(FlaskForm):
     file = FileField('File', validators=[FileRequired(), allowed_file])
     name = StringField('Name', validators=[Length(max=100)])

eveai_app/views/document_list_view.py

@@ -0,0 +1,102 @@
+from datetime import datetime
+from flask import request, render_template, session
+from sqlalchemy import desc, asc, or_, and_, cast, Integer
+from common.models.document import Document, Catalog
+from common.utils.filtered_list_view import FilteredListView
+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']
+
+    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
+        )
+
+    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()
+            if 'valid' in filters['validity']:
+                query = query.filter(
+                    and_(
+                        or_(Document.valid_from.is_(None), Document.valid_from <= now),
+                        or_(Document.valid_to.is_(None), Document.valid_to >= now)
+                    )
+                )
+
+        return query
+
+    def apply_sorting(self, query):
+        sort_by = request.args.get('sort_by', 'id')
+        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)
+
+            if sort_order == 'asc':
+                query = query.order_by(asc(column))
+            elif sort_order == 'desc':
+                query = query.order_by(desc(column))
+
+        return query
+
+    def get(self):
+        query = self.get_query()
+        query = self.apply_filters(query)
+        query = self.apply_sorting(query)
+        pagination = self.paginate(query)
+
+        def format_date(date):
+            if isinstance(date, datetime):
+                return date.strftime('%Y-%m-%d')
+            elif isinstance(date, str):
+                return date
+            else:
+                return ''
+
+        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)
+        }
+        return render_template(self.template, **context)
+
+    def get_filter_options(self, catalogs):
+        return {
+            'catalog_id': [(str(cat.id), cat.name) for cat in catalogs],
+            'validity': [('valid', 'Valid'), ('all', 'All')]
+        }

eveai_app/views/document_version_list_view.py

@@ -12,7 +12,7 @@ class DocumentVersionListView(FilteredListView):
     allowed_sorts = ['id', 'processing_started_at', 'processing_finished_at', 'processing_error']
 
     def get_query(self):
-        return DocumentVersion.query.join(Document).filter(Document.tenant_id == session.get('tenant', {}).get('id'))
+        return DocumentVersion.query.join(Document)
 
     def apply_filters(self, query):
         filters = request.args.to_dict()

eveai_app/views/document_views.py

@@ -1,9 +1,11 @@
 import ast
 from datetime import datetime as dt, timezone as tz
 
+from babel.messages.setuptools_frontend import update_catalog
 from flask import request, redirect, flash, render_template, Blueprint, session, current_app
 from flask_security import roles_accepted, current_user
 from sqlalchemy import desc
+from sqlalchemy.orm import aliased
 from werkzeug.utils import secure_filename
 from sqlalchemy.exc import SQLAlchemyError
 import requests
@@ -12,18 +14,20 @@ from urllib.parse import urlparse, unquote
 import io
 import json
 
-from common.models.document import Document, DocumentVersion
+from common.models.document import Document, DocumentVersion, Catalog
 from common.extensions import db, minio_client
 from common.utils.document_utils import validate_file_type, create_document_stack, start_embedding_task, process_url, \
     process_multiple_urls, get_documents_list, edit_document, \
     edit_document_version, refresh_document
 from common.utils.eveai_exceptions import EveAIInvalidLanguageException, EveAIUnsupportedFileType, \
     EveAIDoubleURLException
-from .document_forms import AddDocumentForm, AddURLForm, EditDocumentForm, EditDocumentVersionForm, AddURLsForm
+from .document_forms import AddDocumentForm, AddURLForm, EditDocumentForm, EditDocumentVersionForm, AddURLsForm, \
+    CatalogForm
 from common.utils.middleware import mw_before_request
 from common.utils.celery_utils import current_celery
 from common.utils.nginx_utils import prefixed_url_for
 from common.utils.view_assistants import form_validation_failed, prepare_table_for_macro, form_to_dict
+from .document_list_view import DocumentListView
 from .document_version_list_view import DocumentVersionListView
 
 document_bp = Blueprint('document_bp', __name__, url_prefix='/document')
@@ -52,6 +56,123 @@ def before_request():
         raise
 
 
+@document_bp.route('/catalog', methods=['GET', 'POST'])
+@roles_accepted('Super User', 'Tenant Admin')
+def catalog():
+    form = CatalogForm()
+
+    if form.validate_on_submit():
+        tenant_id = session.get('tenant').get('id')
+        new_catalog = Catalog()
+        form.populate_obj(new_catalog)
+        # Handle Embedding Variables
+        new_catalog.html_tags = [tag.strip() for tag in form.html_tags.data.split(',')] if form.html_tags.data else []
+        new_catalog.html_end_tags = [tag.strip() for tag in form.html_end_tags.data.split(',')] \
+            if form.html_end_tags.data else []
+        new_catalog.html_included_elements = [tag.strip() for tag in form.html_included_elements.data.split(',')] \
+            if form.html_included_elements.data else []
+        new_catalog.html_excluded_elements = [tag.strip() for tag in form.html_excluded_elements.data.split(',')] \
+            if form.html_excluded_elements.data else []
+        new_catalog.html_excluded_classes = [cls.strip() for cls in form.html_excluded_classes.data.split(',')] \
+            if form.html_excluded_classes.data else []
+        set_logging_information(new_catalog, dt.now(tz.utc))
+
+        try:
+            db.session.add(new_catalog)
+            db.session.commit()
+            flash('Catalog successfully added!', 'success')
+            current_app.logger.info(f'Catalog {new_catalog.name} successfully added for tenant {tenant_id}!')
+        except SQLAlchemyError as e:
+            db.session.rollback()
+            flash(f'Failed to add catalog. Error: {e}', 'danger')
+            current_app.logger.error(f'Failed to add catalog {new_catalog.name}'
+                                     f'for tenant {tenant_id}. Error: {str(e)}')
+
+    return render_template('document/catalog.html', form=form)
+
+
+@document_bp.route('/catalogs', methods=['GET', 'POST'])
+@roles_accepted('Super User', 'Tenant Admin')
+def catalogs():
+    page = request.args.get('page', 1, type=int)
+    per_page = request.args.get('per_page', 10, type=int)
+
+    query = Catalog.query.order_by(Catalog.id)
+
+    pagination = query.paginate(page=page, per_page=per_page)
+    the_catalogs = pagination.items
+
+    # prepare table data
+    rows = prepare_table_for_macro(the_catalogs, [('id', ''), ('name', '')])
+
+    # Render the catalogs in a template
+    return render_template('document/catalogs.html', rows=rows, pagination=pagination)
+
+
+@document_bp.route('/handle_catalog_selection', methods=['POST'])
+@roles_accepted('Super User', 'Tenant Admin')
+def handle_catalog_selection():
+    catalog_identification = request.form.get('selected_row')
+    catalog_id = ast.literal_eval(catalog_identification).get('value')
+    action = request.form['action']
+    catalog = Catalog.query.get_or_404(catalog_id)
+
+    if action == 'set_session_catalog':
+        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))
+
+    return redirect(prefixed_url_for('document_bp.catalogs'))
+
+
+@document_bp.route('/catalog/<int:catalog_id>', methods=['GET', 'POST'])
+@roles_accepted('Super User', 'Tenant Admin')
+def edit_catalog(catalog_id):
+    catalog = Catalog.query.get_or_404(catalog_id)
+    form = CatalogForm(obj=catalog)
+    tenant_id = session.get('tenant').get('id')
+
+    # Convert arrays to comma-separated strings for display
+    if request.method == 'GET':
+        form.html_tags.data = ', '.join(catalog.html_tags or '')
+        form.html_end_tags.data = ', '.join(catalog.html_end_tags or '')
+        form.html_included_elements.data = ', '.join(catalog.html_included_elements or '')
+        form.html_excluded_elements.data = ', '.join(catalog.html_excluded_elements or '')
+        form.html_excluded_classes.data = ', '.join(catalog.html_excluded_classes or '')
+
+    if request.method == 'POST' and form.validate_on_submit():
+        form.populate_obj(catalog)
+        # Handle Embedding Variables
+        catalog.html_tags = [tag.strip() for tag in form.html_tags.data.split(',')] if form.html_tags.data else []
+        catalog.html_end_tags = [tag.strip() for tag in form.html_end_tags.data.split(',')] \
+            if form.html_end_tags.data else []
+        catalog.html_included_elements = [tag.strip() for tag in form.html_included_elements.data.split(',')] \
+            if form.html_included_elements.data else []
+        catalog.html_excluded_elements = [tag.strip() for tag in form.html_excluded_elements.data.split(',')] \
+            if form.html_excluded_elements.data else []
+        catalog.html_excluded_classes = [cls.strip() for cls in form.html_excluded_classes.data.split(',')] \
+            if form.html_excluded_classes.data else []
+        update_logging_information(catalog, dt.now(tz.utc))
+        try:
+            db.session.add(catalog)
+            db.session.commit()
+            flash('Catalog successfully updated successfully!', 'success')
+            current_app.logger.info(f'Catalog {catalog.name} successfully updated for tenant {tenant_id}')
+        except SQLAlchemyError as e:
+            db.session.rollback()
+            flash(f'Failed to update catalog. Error: {e}', 'danger')
+            current_app.logger.error(f'Failed to update catalog {catalog_id} for tenant {tenant_id}. Error: {str(e)}')
+
+        return redirect(prefixed_url_for('document_bp.catalogs'))
+    else:
+        form_validation_failed(request, form)
+
+    return render_template('document/edit_catalog.html', form=form, catalog_id=catalog_id)
+
+
 @document_bp.route('/add_document', methods=['GET', 'POST'])
 @roles_accepted('Super User', 'Tenant Admin')
 def add_document():
@@ -60,6 +181,7 @@ def add_document():
     if form.validate_on_submit():
         try:
             tenant_id = session['tenant']['id']
+            catalog_id = session['catalog_id']
             file = form.file.data
             filename = secure_filename(file.filename)
             extension = filename.rsplit('.', 1)[1].lower()
@@ -68,6 +190,7 @@ def add_document():
 
             current_app.logger.debug(f'Language on form: {form.language.data}')
             api_input = {
+                'catalog_id': catalog_id,
                 'name': form.name.data,
                 'language': form.language.data,
                 'user_context': form.user_context.data,
@@ -100,11 +223,13 @@ def add_url():
     if form.validate_on_submit():
         try:
             tenant_id = session['tenant']['id']
+            catalog_id = session['catalog_id']
             url = form.url.data
 
             file_content, filename, extension = process_url(url, tenant_id)
 
             api_input = {
+                'catalog_id': catalog_id,
                 'name': form.name.data or filename,
                 'url': url,
                 'language': form.language.data,
@@ -171,22 +296,23 @@ def add_urls():
 @document_bp.route('/documents', methods=['GET', 'POST'])
 @roles_accepted('Super User', 'Tenant Admin')
 def documents():
-    page = request.args.get('page', 1, type=int)
-    per_page = request.args.get('per_page', 10, type=int)
-
-    pagination = get_documents_list(page, per_page)
-    docs = pagination.items
-
-    rows = prepare_table_for_macro(docs, [('id', ''), ('name', ''), ('valid_from', ''), ('valid_to', '')])
-
-    return render_template('document/documents.html', rows=rows, pagination=pagination)
+    view = DocumentListView(Document, 'document/documents.html', per_page=10)
+    return view.get()
 
 
 @document_bp.route('/handle_document_selection', methods=['POST'])
 @roles_accepted('Super User', 'Tenant Admin')
 def handle_document_selection():
     document_identification = request.form['selected_row']
-    doc_id = ast.literal_eval(document_identification).get('value')
+    if isinstance(document_identification, int) or document_identification.isdigit():
+        doc_id = int(document_identification)
+    else:
+        # If it's not an integer, assume it's a string representation of a dictionary
+        try:
+            doc_id = ast.literal_eval(document_identification).get('value')
+        except (ValueError, AttributeError):
+            flash('Invalid document selection.', 'error')
+            return redirect(prefixed_url_for('document_bp.documents'))
 
     action = request.form['action']
 
@@ -208,9 +334,25 @@ def handle_document_selection():
 @document_bp.route('/edit_document/<int:document_id>', methods=['GET', 'POST'])
 @roles_accepted('Super User', 'Tenant Admin')
 def edit_document_view(document_id):
-    doc = Document.query.get_or_404(document_id)
+    # Use an alias for the Catalog to avoid column name conflicts
+    CatalogAlias = aliased(Catalog)
+
+    # Query for the document and its catalog
+    result = db.session.query(Document, CatalogAlias.name.label('catalog_name')) \
+        .join(CatalogAlias, Document.catalog_id == CatalogAlias.id) \
+        .filter(Document.id == document_id) \
+        .first_or_404()
+
+    doc, catalog_name = result
+
     form = EditDocumentForm(obj=doc)
 
+    if request.method == 'GET':
+        # Populate form with current values
+        form.name.data = doc.name
+        form.valid_from.data = doc.valid_from
+        form.valid_to.data = doc.valid_to
+
     if form.validate_on_submit():
         updated_doc, error = edit_document(
             document_id,
@@ -226,7 +368,7 @@ def edit_document_view(document_id):
     else:
         form_validation_failed(request, form)
 
-    return render_template('document/edit_document.html', form=form, document_id=document_id)
+    return render_template('document/edit_document.html', form=form, document_id=document_id, catalog_name=catalog_name)
 
 
 @document_bp.route('/edit_document_version/<int:document_version_id>', methods=['GET', 'POST'])
@@ -268,8 +410,8 @@ 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', ''), ('file_location', ''),
-                                               ('file_name', ''), ('file_type', ''),
+    rows = prepare_table_for_macro(doc_langs, [('id', ''), ('url', ''),
+                                               ('object_name', ''), ('file_type', ''),
                                                ('processing', ''), ('processing_started_at', ''),
                                                ('processing_finished_at', ''), ('processing_error', '')])
 
@@ -280,7 +422,15 @@ def document_versions(document_id):
 @roles_accepted('Super User', 'Tenant Admin')
 def handle_document_version_selection():
     document_version_identification = request.form['selected_row']
-    doc_vers_id = ast.literal_eval(document_version_identification).get('value')
+    if isinstance(document_version_identification, int) or document_version_identification.isdigit():
+        doc_vers_id = int(document_version_identification)
+    else:
+        # If it's not an integer, assume it's a string representation of a dictionary
+        try:
+            doc_vers_id = ast.literal_eval(document_version_identification).get('value')
+        except (ValueError, AttributeError):
+            flash('Invalid document version selection.', 'error')
+            return redirect(prefixed_url_for('document_bp.document_versions_list'))
 
     action = request.form['action']
 
@@ -332,7 +482,7 @@ def refresh_all_documents():
 
 
 def refresh_document_view(document_id):
-    new_version, result = refresh_document(document_id)
+    new_version, result = refresh_document(document_id, session['tenant']['id'])
     if new_version:
         flash(f'Document refreshed. New version: {new_version.id}. Task ID: {result}', 'success')
     else:
@@ -349,10 +499,9 @@ def re_embed_latest_versions():
 
 
 def process_version(version_id):
-    task = current_celery.send_task('create_embeddings', queue='embeddings', args=[
-        session['tenant']['id'],
-        version_id,
-    ])
+    task = current_celery.send_task('create_embeddings',
+                                    args=[session['tenant']['id'], version_id,],
+                                    queue='embeddings')
     current_app.logger.info(f'Embedding creation retriggered by user {current_user.id}, {current_user.email} '
                             f'for tenant {session["tenant"]["id"]}, '
                             f'Document Version {version_id}. '
@@ -398,47 +547,47 @@ def fetch_html(url):
     return response.content
 
 
-def prepare_document_data(docs):
-    rows = []
-    for doc in docs:
-        doc_row = [{'value': doc.name, 'class': '', 'type': 'text'},
-                   {'value': doc.created_at.strftime("%Y-%m-%d %H:%M:%S"), 'class': '', 'type': 'text'}]
-        # Document basic details
-        if doc.valid_from:
-            doc_row.append({'value': doc.valid_from.strftime("%Y-%m-%d"), 'class': '', 'type': 'text'})
-        else:
-            doc_row.append({'value': '', 'class': '', 'type': 'text'})
-
-        # Nested languages and versions
-        languages_rows = []
-        for lang in doc.languages:
-            lang_row = [{'value': lang.language, 'class': '', 'type': 'text'}]
-
-            # Latest version details if available (should be available ;-) )
-            if lang.latest_version:
-                lang_row.append({'value': lang.latest_version.created_at.strftime("%Y-%m-%d %H:%M:%S"),
-                                 'class': '', 'type': 'text'})
-                if lang.latest_version.url:
-                    lang_row.append({'value': lang.latest_version.url,
-                                     'class': '', 'type': 'link', 'href': lang.latest_version.url})
-                else:
-                    lang_row.append({'value': '', 'class': '', 'type': 'text'})
-
-                if lang.latest_version.file_name:
-                    lang_row.append({'value': lang.latest_version.file_name, 'class': '', 'type': 'text'})
-                else:
-                    lang_row.append({'value': '', 'class': '', 'type': 'text'})
-
-                if lang.latest_version.file_type:
-                    lang_row.append({'value': lang.latest_version.file_type, 'class': '', 'type': 'text'})
-                else:
-                    lang_row.append({'value': '', 'class': '', 'type': 'text'})
-                    # Include other details as necessary
-
-            languages_rows.append(lang_row)
-
-        doc_row.append({'is_group': True, 'colspan': '5',
-                        'headers': ['Language', 'Latest Version', 'URL', 'File Name', 'Type'],
-                        'sub_rows': languages_rows})
-        rows.append(doc_row)
-    return rows
+# def prepare_document_data(docs):
+#     rows = []
+#     for doc in docs:
+#         doc_row = [{'value': doc.name, 'class': '', 'type': 'text'},
+#                    {'value': doc.created_at.strftime("%Y-%m-%d %H:%M:%S"), 'class': '', 'type': 'text'}]
+#         # Document basic details
+#         if doc.valid_from:
+#             doc_row.append({'value': doc.valid_from.strftime("%Y-%m-%d"), 'class': '', 'type': 'text'})
+#         else:
+#             doc_row.append({'value': '', 'class': '', 'type': 'text'})
+#
+#         # Nested languages and versions
+#         languages_rows = []
+#         for lang in doc.languages:
+#             lang_row = [{'value': lang.language, 'class': '', 'type': 'text'}]
+#
+#             # Latest version details if available (should be available ;-) )
+#             if lang.latest_version:
+#                 lang_row.append({'value': lang.latest_version.created_at.strftime("%Y-%m-%d %H:%M:%S"),
+#                                  'class': '', 'type': 'text'})
+#                 if lang.latest_version.url:
+#                     lang_row.append({'value': lang.latest_version.url,
+#                                      'class': '', 'type': 'link', 'href': lang.latest_version.url})
+#                 else:
+#                     lang_row.append({'value': '', 'class': '', 'type': 'text'})
+#
+#                 if lang.latest_version.object_name:
+#                     lang_row.append({'value': lang.latest_version.object_name, 'class': '', 'type': 'text'})
+#                 else:
+#                     lang_row.append({'value': '', 'class': '', 'type': 'text'})
+#
+#                 if lang.latest_version.file_type:
+#                     lang_row.append({'value': lang.latest_version.file_type, 'class': '', 'type': 'text'})
+#                 else:
+#                     lang_row.append({'value': '', 'class': '', 'type': 'text'})
+#                     # Include other details as necessary
+#
+#             languages_rows.append(lang_row)
+#
+#         doc_row.append({'is_group': True, 'colspan': '5',
+#                         'headers': ['Language', 'Latest Version', 'URL', 'File Name', 'Type'],
+#                         'sub_rows': languages_rows})
+#         rows.append(doc_row)
+#     return rows

eveai_app/views/entitlements_views.py

@@ -2,18 +2,14 @@ import uuid
 from datetime import datetime as dt, timezone as tz
 from flask import request, redirect, flash, render_template, Blueprint, session, current_app, jsonify
 from flask_security import hash_password, roles_required, roles_accepted, current_user
-from itsdangerous import URLSafeTimedSerializer
 from sqlalchemy.exc import SQLAlchemyError
-from sqlalchemy import or_
+from sqlalchemy import or_, desc
 import ast
 
 from common.models.entitlements import License, LicenseTier, LicenseUsage, BusinessEventLog
 from common.extensions import db, security, minio_client, simple_encryption
-from common.utils.security_utils import send_confirmation_email, send_reset_email
 from .entitlements_forms import LicenseTierForm, LicenseForm
-from common.utils.database import Database
 from common.utils.view_assistants import prepare_table_for_macro, form_validation_failed
-from common.utils.simple_encryption import generate_api_key
 from common.utils.nginx_utils import prefixed_url_for
 
 entitlements_bp = Blueprint('entitlements_bp', __name__, url_prefix='/entitlements')
@@ -174,14 +170,14 @@ def create_license(license_tier_id):
                 db.session.add(new_license)
                 db.session.commit()
                 flash('License created successfully', 'success')
-                return redirect(prefixed_url_for('entitlements_bp/edit_license', license_id=new_license.id))
+                return redirect(prefixed_url_for('entitlements_bp.edit_license', license_id=new_license.id))
             except Exception as e:
                 db.session.rollback()
                 flash(f'Error creating license: {str(e)}', 'error')
         else:
             form_validation_failed(request, form)
 
-    return render_template('entitlements/license.html', form=form)
+    return render_template('entitlements/license.html', form=form, ext_disabled_fields=[])
 
 
 @entitlements_bp.route('/license/<int:license_id>', methods=['GET', 'POST'])
@@ -215,3 +211,25 @@ def edit_license(license_id):
 
     return render_template('entitlements/license.html', form=form, license_tier_id=license_tier.id,
                            ext_disabled_fields=disabled_fields)
+
+
+@entitlements_bp.route('/view_usages')
+@roles_accepted('Super User', 'Tenant Admin')
+def view_usages():
+    page = request.args.get('page', 1, type=int)
+    per_page = request.args.get('per_page', 10, type=int)
+
+    tenant_id = session.get('tenant').get('id')
+    query = LicenseUsage.query.filter_by(tenant_id=tenant_id).order_by(desc(LicenseUsage.id))
+
+    pagination = query.paginate(page=page, per_page=per_page)
+    lus = pagination.items
+
+    # prepare table data
+
+    rows = prepare_table_for_macro(lus, [('id', ''), ('period_start_date', ''), ('period_end_date', ''),
+                                         ('storage_mb_used', ''), ('embedding_mb_used', ''),
+                                         ('interaction_total_tokens_used', '')])
+
+    # Render the users in a template
+    return render_template('entitlements/view_usages.html', rows=rows, pagination=pagination)

eveai_app/views/interaction_views.py

@@ -93,17 +93,17 @@ def view_chat_session(chat_session_id):
     # Fetch all related embeddings for the interactions in this session
     embedding_query = (db.session.query(InteractionEmbedding.interaction_id,
                                         DocumentVersion.url,
-                                        DocumentVersion.file_name)
+                                        DocumentVersion.object_name)
                        .join(Embedding, InteractionEmbedding.embedding_id == Embedding.id)
                        .join(DocumentVersion, Embedding.doc_vers_id == DocumentVersion.id)
                        .filter(InteractionEmbedding.interaction_id.in_([i.id for i in interactions])))
 
     # Create a dictionary to store embeddings for each interaction
     embeddings_dict = {}
-    for interaction_id, url, file_name in embedding_query:
+    for interaction_id, url, object_name in embedding_query:
         if interaction_id not in embeddings_dict:
             embeddings_dict[interaction_id] = []
-        embeddings_dict[interaction_id].append({'url': url, 'file_name': file_name})
+        embeddings_dict[interaction_id].append({'url': url, 'object_name': object_name})
 
     return render_template('interaction/view_chat_session.html',
                            chat_session=chat_session,

eveai_app/views/user_views.py

@@ -48,22 +48,6 @@ def tenant():
         new_tenant = Tenant()
         form.populate_obj(new_tenant)
 
-        # Handle Embedding Variables
-        new_tenant.html_tags = [tag.strip() for tag in form.html_tags.data.split(',')] if form.html_tags.data else []
-        new_tenant.html_end_tags = [tag.strip() for tag in form.html_end_tags.data.split(',')] \
-            if form.html_end_tags.data else []
-        new_tenant.html_included_elements = [tag.strip() for tag in form.html_included_elements.data.split(',')] \
-            if form.html_included_elements.data else []
-        new_tenant.html_excluded_elements = [tag.strip() for tag in form.html_excluded_elements.data.split(',')] \
-            if form.html_excluded_elements.data else []
-        new_tenant.html_excluded_classes = [cls.strip() for cls in form.html_excluded_classes.data.split(',')] \
-            if form.html_excluded_classes.data else []
-
-        current_app.logger.debug(f'html_tags: {new_tenant.html_tags},'
-                                 f'html_end_tags: {new_tenant.html_end_tags},'
-                                 f'html_included_elements: {new_tenant.html_included_elements},'
-                                 f'html_excluded_elements: {new_tenant.html_excluded_elements}')
-
         # Handle Timestamps
         timestamp = dt.now(tz.utc)
         new_tenant.created_at = timestamp
@@ -105,30 +89,11 @@ def edit_tenant(tenant_id):
     if request.method == 'GET':
         # Populate the form with tenant data
         form.populate_obj(tenant)
-        if tenant.html_tags:
-            form.html_tags.data = ', '.join(tenant.html_tags)
-        if tenant.html_end_tags:
-            form.html_end_tags.data = ', '.join(tenant.html_end_tags)
-        if tenant.html_included_elements:
-            form.html_included_elements.data = ', '.join(tenant.html_included_elements)
-        if tenant.html_excluded_elements:
-            form.html_excluded_elements.data = ', '.join(tenant.html_excluded_elements)
-        if tenant.html_excluded_classes:
-            form.html_excluded_classes.data = ', '.join(tenant.html_excluded_classes)
 
     if form.validate_on_submit():
         current_app.logger.debug(f'Updating tenant {tenant_id}')
         # Populate the tenant with form data
         form.populate_obj(tenant)
-        # Then handle the special fields manually
-        tenant.html_tags = [tag.strip() for tag in form.html_tags.data.split(',') if tag.strip()]
-        tenant.html_end_tags = [tag.strip() for tag in form.html_end_tags.data.split(',') if tag.strip()]
-        tenant.html_included_elements = [elem.strip() for elem in form.html_included_elements.data.split(',') if
-                                         elem.strip()]
-        tenant.html_excluded_elements = [elem.strip() for elem in form.html_excluded_elements.data.split(',') if
-                                         elem.strip()]
-        tenant.html_excluded_classes = [elem.strip() for elem in form.html_excluded_classes.data.split(',') if
-                                        elem.strip()]
 
         db.session.commit()
         flash('Tenant updated successfully.', 'success')
@@ -266,10 +231,16 @@ def handle_tenant_selection():
     tenant_identification = request.form['selected_row']
     tenant_id = ast.literal_eval(tenant_identification).get('value')
     the_tenant = Tenant.query.get(tenant_id)
+
+    # set tenant information in the session
     session['tenant'] = the_tenant.to_dict()
     session['default_language'] = the_tenant.default_language
     session['embedding_model'] = the_tenant.embedding_model
     session['llm_model'] = the_tenant.llm_model
+    # remove catalog-related items from the session
+    session.pop('catalog_id', None)
+    session.pop('catalog_name', None)
+
     action = request.form['action']
 
     match action:

eveai_beat/__init__.py

@@ -0,0 +1,44 @@
+import logging
+import logging.config
+from flask import Flask
+import os
+
+from common.utils.celery_utils import make_celery, init_celery
+from config.logging_config import LOGGING
+from config.config import get_config
+
+
+def create_app(config_file=None):
+    app = Flask(__name__)
+
+    environment = os.getenv('FLASK_ENV', 'development')
+
+    match environment:
+        case 'development':
+            app.config.from_object(get_config('dev'))
+        case 'production':
+            app.config.from_object(get_config('prod'))
+        case _:
+            app.config.from_object(get_config('dev'))
+
+    logging.config.dictConfig(LOGGING)
+
+    register_extensions(app)
+
+    celery = make_celery(app.name, app.config)
+    init_celery(celery, app, is_beat=True)
+
+    from . import schedule
+    celery.conf.beat_schedule = schedule.beat_schedule
+
+    app.logger.info("EveAI Beat Scheduler Started Successfully")
+    app.logger.info("-------------------------------------------------------------------------------------------------")
+
+    return app, celery
+
+
+def register_extensions(app):
+    pass
+
+
+app, celery = create_app()

eveai_beat/schedule.py

@@ -0,0 +1,17 @@
+from celery.schedules import crontab
+
+# Define the Celery beat schedule here
+beat_schedule = {
+    'update-tenant-usages-every-hour': {
+        'task': 'update_usages',
+        'schedule': crontab(minute='0'),  # Runs every hour
+        'args': (),
+        'options': {'queue': 'entitlements'}
+    },
+    # 'send-invoices-every-month': {
+    #     'task': 'send_invoices',
+    #     'schedule': crontab(day_of_month=1, hour=0, minute=0),  # Runs on the 1st of every month
+    #     'args': ()
+    # },
+    # Add more schedules as needed
+}

eveai_chat/socket_handlers/chat_handler.py

@@ -109,14 +109,16 @@ def handle_message(data):
         room = session.get('room')
 
         # Offload actual processing of question
-        task = current_celery.send_task('ask_question', queue='llm_interactions', args=[
-            current_tenant_id,
-            data['message'],
-            data['language'],
-            session['session_id'],
-            data['timezone'],
-            room
-        ])
+        task = current_celery.send_task('ask_question',
+                                        queue='llm_interactions',
+                                        args=[
+                                            current_tenant_id,
+                                            data['message'],
+                                            data['language'],
+                                            session['session_id'],
+                                            data['timezone'],
+                                            room
+                                        ])
         current_app.logger.debug(f'SocketIO: Message offloading for tenant {current_tenant_id}, '
                                  f'Question: {task.id}')
         response = {

eveai_chat_workers/tasks.py

@@ -36,7 +36,7 @@ def ping():
 
 def detail_question(question, language, model_variables, session_id):
     current_app.logger.debug(f'Detail question: {question}')
-    current_app.logger.debug(f'model_varialbes: {model_variables}')
+    current_app.logger.debug(f'model_variables: {model_variables}')
     current_app.logger.debug(f'session_id: {session_id}')
     retriever = EveAIHistoryRetriever(model_variables=model_variables, session_id=session_id)
     llm = model_variables['llm']
@@ -93,12 +93,6 @@ def ask_question(tenant_id, question, language, session_id, user_timezone, room)
                     current_app.logger.error(f'ask_question: Error initializing chat session in database: {e}')
                     raise
 
-            if tenant.rag_tuning:
-                current_app.rag_tuning_logger.debug(f'Received question for tenant {tenant_id}:\n{question}. Processing...')
-                current_app.rag_tuning_logger.debug(f'Tenant Information: \n{tenant.to_dict()}')
-                current_app.rag_tuning_logger.debug(f'===================================================================')
-                current_app.rag_tuning_logger.debug(f'===================================================================')
-
             with current_event.create_span("RAG Answer"):
                 result, interaction = answer_using_tenant_rag(question, language, tenant, chat_session)
                 result['algorithm'] = current_app.config['INTERACTION_ALGORITHMS']['RAG_TENANT']['name']
@@ -138,8 +132,7 @@ def answer_using_tenant_rag(question, language, tenant, chat_session):
 
     with current_event.create_span("Detail Question"):
         detailed_question = detail_question(question, language, model_variables, chat_session.session_id)
-        current_app.logger.debug(f'Original question:\n {question}\n\nDetailed question: {detailed_question}')
-        if tenant.rag_tuning:
+        if model_variables['rag_tuning']:
             current_app.rag_tuning_logger.debug(f'Detailed Question for tenant {tenant.id}:\n{question}.')
             current_app.rag_tuning_logger.debug(f'-------------------------------------------------------------------')
         new_interaction.detailed_question = detailed_question
@@ -153,7 +146,7 @@ def answer_using_tenant_rag(question, language, tenant, chat_session):
         full_template = replace_variable_in_template(language_template, "{tenant_context}", model_variables['rag_context'])
         rag_prompt = ChatPromptTemplate.from_template(full_template)
         setup_and_retrieval = RunnableParallel({"context": retriever, "question": RunnablePassthrough()})
-        if tenant.rag_tuning:
+        if model_variables['rag_tuning']:
             current_app.rag_tuning_logger.debug(f'Full prompt for tenant {tenant.id}:\n{full_template}.')
             current_app.rag_tuning_logger.debug(f'-------------------------------------------------------------------')
 
@@ -181,7 +174,7 @@ def answer_using_tenant_rag(question, language, tenant, chat_session):
             current_app.logger.debug(f'ask_question: result answer: {result['answer']}')
             current_app.logger.debug(f'ask_question: result citations: {result["citations"]}')
             current_app.logger.debug(f'ask_question: insufficient information: {result["insufficient_info"]}')
-            if tenant.rag_tuning:
+            if model_variables['rag_tuning']:
                 current_app.rag_tuning_logger.debug(f'ask_question: result answer: {result['answer']}')
                 current_app.rag_tuning_logger.debug(f'ask_question: result citations: {result["citations"]}')
                 current_app.rag_tuning_logger.debug(f'ask_question: insufficient information: {result["insufficient_info"]}')
@@ -197,7 +190,7 @@ def answer_using_tenant_rag(question, language, tenant, chat_session):
             )
             existing_embedding_ids = [emb.id for emb in embeddings]
             urls = list(set(emb.document_version.url for emb in embeddings))
-            if tenant.rag_tuning:
+            if model_variables['rag_tuning']:
                 current_app.rag_tuning_logger.debug(f'Referenced documents for answer for tenant {tenant.id}:\n')
                 current_app.rag_tuning_logger.debug(f'{urls}')
                 current_app.rag_tuning_logger.debug(f'-------------------------------------------------------------------')

eveai_entitlements/tasks.py

@@ -5,13 +5,14 @@ from datetime import datetime as dt, timezone as tz, datetime
 from celery import states
 from dateutil.relativedelta import relativedelta
 from flask import current_app
-from sqlalchemy import or_, and_
+from sqlalchemy import or_, and_, text
 from sqlalchemy.exc import SQLAlchemyError
 from common.extensions import db
 from common.models.user import Tenant
 from common.models.entitlements import BusinessEventLog, LicenseUsage, License
 from common.utils.celery_utils import current_celery
-from common.utils.eveai_exceptions import EveAINoLicenseForTenant
+from common.utils.eveai_exceptions import EveAINoLicenseForTenant, EveAIException
+from common.utils.database import Database
 
 
 # Healthcheck task
@@ -24,32 +25,54 @@ def ping():
 def update_usages():
     current_timestamp = dt.now(tz.utc)
     tenant_ids = get_all_tenant_ids()
+
+    # List to collect all errors
+    error_list = []
+
     for tenant_id in tenant_ids:
-        tenant = Tenant.query.get(tenant_id)
-        if tenant.storage_dirty:
-            recalculate_storage_for_tenant(tenant)
-        check_and_create_license_usage_for_tenant(tenant_id)
-        logs = get_logs_for_processing(tenant_id, current_timestamp)
-        if not logs:
-            continue    # If no logs to be processed, continu to the next tenant
+        try:
+            Database(tenant_id).switch_schema()
+            check_and_create_license_usage_for_tenant(tenant_id)
+            tenant = Tenant.query.get(tenant_id)
+            if tenant.storage_dirty:
+                recalculate_storage_for_tenant(tenant)
+            logs = get_logs_for_processing(tenant_id, current_timestamp)
+            if not logs:
+                continue    # If no logs to be processed, continu to the next tenant
 
-        # Get the min and max timestamp from the logs
-        min_timestamp = min(log.timestamp for log in logs)
-        max_timestamp = max(log.timestamp for log in logs)
+            # Get the min and max timestamp from the logs
+            min_timestamp = min(log.timestamp for log in logs)
+            max_timestamp = max(log.timestamp for log in logs)
 
-        # Retrieve relevant LicenseUsage records
-        license_usages = get_relevant_license_usages(db.session, tenant_id, min_timestamp, max_timestamp)
+            # Retrieve relevant LicenseUsage records
+            current_app.logger.debug(f"Searching relevant usages for tenant {tenant_id}")
+            license_usages = get_relevant_license_usages(db.session, tenant_id, min_timestamp, max_timestamp)
+            current_app.logger.debug(f"Found {license_usages}, end searching relevant usages for tenant {tenant_id}")
 
-        # Split logs based on LicenseUsage periods
-        logs_by_usage = split_logs_by_license_usage(logs, license_usages)
+            # Split logs based on LicenseUsage periods
+            current_app.logger.debug(f"Splitting usages for tenant {tenant_id}")
+            logs_by_usage = split_logs_by_license_usage(logs, license_usages)
+            current_app.logger.debug(f"Found {logs_by_usage}, end splitting logs for tenant {tenant_id}")
 
-        # Now you can process logs for each LicenseUsage
-        for license_usage_id, logs in logs_by_usage.items():
-            process_logs_for_license_usage(tenant_id, license_usage_id, logs)
+            # Now you can process logs for each LicenseUsage
+            for license_usage_id, logs in logs_by_usage.items():
+                current_app.logger.debug(f"Processing logs for usage id {license_usage_id} for tenant {tenant_id}")
+                process_logs_for_license_usage(tenant_id, license_usage_id, logs)
+                current_app.logger.debug(f"Finished processing logs for tenant {tenant_id}")
+        except Exception as e:
+            error = f"Usage Calculation error for Tenant {tenant_id}: {e}"
+            error_list.append(error)
+            current_app.logger.error(error)
+            continue
+
+    if error_list:
+        raise Exception('\n'.join(error_list))
+
+    return "Update Usages taks completed successfully"
 
 
 def get_all_tenant_ids():
-    tenant_ids = db.session.query(Tenant.tenant_id).all()
+    tenant_ids = db.session.query(Tenant.id).all()
     return [tenant_id[0] for tenant_id in tenant_ids]  # Extract tenant_id from tuples
 
 
@@ -57,21 +80,21 @@ def check_and_create_license_usage_for_tenant(tenant_id):
     current_date = dt.now(tz.utc).date()
     license_usages = (db.session.query(LicenseUsage)
                       .filter_by(tenant_id=tenant_id)
-                      .filter_by(and_(LicenseUsage.period_start_date <= current_date,
-                                      LicenseUsage.period_end_date >= current_date))
+                      .filter(and_(LicenseUsage.period_start_date <= current_date,
+                                   LicenseUsage.period_end_date >= current_date))
                       .all())
     if not license_usages:
         active_license = (db.session.query(License).filter_by(tenant_id=tenant_id)
-                          .filter_by(and_(License.start_date <= current_date,
-                                          License.end_date >= current_date))
-                          .one())
+                          .filter(and_(License.start_date <= current_date,
+                                       License.end_date >= current_date))
+                          .one_or_none())
         if not active_license:
             current_app.logger.error(f"No License defined for {tenant_id}. "
                                      f"Impossible to calculate license usage.")
             raise EveAINoLicenseForTenant(message=f"No License defined for {tenant_id}. "
                                                   f"Impossible to calculate license usage.")
 
-        start_date, end_date = calculate_valid_period(current_date, active_license.period_start_date)
+        start_date, end_date = calculate_valid_period(current_date, active_license.start_date)
         new_license_usage = LicenseUsage(period_start_date=start_date,
                                          period_end_date=end_date,
                                          license_id=active_license.id,
@@ -124,8 +147,8 @@ def get_relevant_license_usages(session, tenant_id, min_timestamp, max_timestamp
     # Fetch LicenseUsage records where the log timestamps fall between period_start_date and period_end_date
     return session.query(LicenseUsage).filter(
         LicenseUsage.tenant_id == tenant_id,
-        LicenseUsage.period_start_date <= max_timestamp,
-        LicenseUsage.period_end_date >= min_timestamp
+        LicenseUsage.period_start_date <= max_timestamp.date(),
+        LicenseUsage.period_end_date >= min_timestamp.date()
     ).order_by(LicenseUsage.period_start_date).all()
 
 
@@ -136,7 +159,7 @@ def split_logs_by_license_usage(logs, license_usages):
     for log in logs:
         # Find the corresponding LicenseUsage for each log based on the timestamp
         for license_usage in license_usages:
-            if license_usage.period_start_date <= log.timestamp <= license_usage.period_end_date:
+            if license_usage.period_start_date <= log.timestamp.date() <= license_usage.period_end_date:
                 logs_by_usage[license_usage.id].append(log)
                 break
 
@@ -181,7 +204,7 @@ def process_logs_for_license_usage(tenant_id, license_usage_id, logs):
         log.license_usage_id = license_usage_id
 
     # Update the LicenseUsage record with the accumulated values
-    license_usage.embedding_mb += embedding_mb_used
+    license_usage.embedding_mb_used += embedding_mb_used
     license_usage.embedding_prompt_tokens_used += embedding_prompt_tokens_used
     license_usage.embedding_completion_tokens_used += embedding_completion_tokens_used
     license_usage.embedding_total_tokens_used += embedding_total_tokens_used
@@ -189,27 +212,31 @@ def process_logs_for_license_usage(tenant_id, license_usage_id, logs):
     license_usage.interaction_completion_tokens_used += interaction_completion_tokens_used
     license_usage.interaction_total_tokens_used += interaction_total_tokens_used
 
+    current_app.logger.debug(f"Processed logs for license usage {license_usage.id}:\n{license_usage}")
+
     # Commit the updates to the LicenseUsage and log records
     try:
         db.session.add(license_usage)
-        db.session.add(logs)
+        for log in logs:
+            db.session.add(log)
         db.session.commit()
     except SQLAlchemyError as e:
         db.session.rollback()
-        current_app.logger.error(f"Error trying to update license usage and logs for tenant {tenant_id}. ")
+        current_app.logger.error(f"Error trying to update license usage and logs for tenant {tenant_id}: {e}")
         raise e
 
 
 def recalculate_storage_for_tenant(tenant):
     # Perform a SUM operation to get the total file size from document_versions
-    total_storage = db.session.execute(f"""
+    total_storage = db.session.execute(text(f"""
         SELECT SUM(file_size) 
-        FROM {tenant.id}.document_versions
-    """).scalar()
+        FROM document_version
+    """)).scalar()
+    current_app.logger.debug(f"Recalculating storage for tenant {tenant} - Total storage: {total_storage}")
 
     # Update the LicenseUsage with the recalculated storage
     license_usage = db.session.query(LicenseUsage).filter_by(tenant_id=tenant.id).first()
-    license_usage.storage_mb = total_storage / (1024 * 1024)  # Convert bytes to MB
+    license_usage.storage_mb_used = total_storage
 
     # Reset the dirty flag after recalculating
     tenant.storage_dirty = False

eveai_workers/Processors/audio_processor.py

@@ -27,10 +27,8 @@ class AudioProcessor(TranscriptionProcessor):
     def _get_transcription(self):
         file_data = minio_client.download_document_file(
             self.tenant.id,
-            self.document_version.doc_id,
-            self.document_version.language,
-            self.document_version.id,
-            self.document_version.file_name
+            self.document_version.bucket_name,
+            self.document_version.object_name,
         )
 
         with current_event.create_span("Audio Compression"):

eveai_workers/Processors/html_processor.py

@@ -15,6 +15,7 @@ class HTMLProcessor(Processor):
         self.html_end_tags = model_variables['html_end_tags']
         self.html_included_elements = model_variables['html_included_elements']
         self.html_excluded_elements = model_variables['html_excluded_elements']
+        self.html_excluded_classes = model_variables['html_excluded_classes']
         self.chunk_size = model_variables['processing_chunk_size']  # Adjust this based on your LLM's optimal input size
         self.chunk_overlap = model_variables[
             'processing_chunk_overlap']  # Adjust for context preservation between chunks
@@ -24,10 +25,8 @@ class HTMLProcessor(Processor):
         try:
             file_data = minio_client.download_document_file(
                 self.tenant.id,
-                self.document_version.doc_id,
-                self.document_version.language,
-                self.document_version.id,
-                self.document_version.file_name
+                self.document_version.bucket_name,
+                self.document_version.object_name,
             )
             html_content = file_data.decode('utf-8')
 
@@ -47,7 +46,7 @@ class HTMLProcessor(Processor):
         self._log(f'Parsing HTML for tenant {self.tenant.id}')
         soup = BeautifulSoup(html_content, 'html.parser')
         extracted_html = ''
-        excluded_classes = self._parse_excluded_classes(self.tenant.html_excluded_classes)
+        excluded_classes = self._parse_excluded_classes(self.html_excluded_classes)
 
         if self.html_included_elements:
             elements_to_parse = soup.find_all(self.html_included_elements)

eveai_workers/Processors/pdf_processor.py

@@ -27,10 +27,8 @@ class PDFProcessor(Processor):
         try:
             file_data = minio_client.download_document_file(
                 self.tenant.id,
-                self.document_version.doc_id,
-                self.document_version.language,
-                self.document_version.id,
-                self.document_version.file_name
+                self.document_version.bucket_name,
+                self.document_version.object_name,
             )
 
             with current_event.create_span("PDF Extraction"):

eveai_workers/Processors/srt_processor.py

@@ -7,10 +7,8 @@ class SRTProcessor(TranscriptionProcessor):
     def _get_transcription(self):
         file_data = minio_client.download_document_file(
             self.tenant.id,
-            self.document_version.doc_id,
-            self.document_version.language,
-            self.document_version.id,
-            self.document_version.file_name
+            self.document_version.bucket_name,
+            self.document_version.object_name,
         )
         srt_content = file_data.decode('utf-8')
         return self._clean_srt(srt_content)

eveai_workers/__init__.py

@@ -44,3 +44,4 @@ def register_extensions(app):
 
 
 app, celery = create_app()
+

eveai_workers/tasks.py

@@ -13,7 +13,7 @@ from langchain_core.runnables import RunnablePassthrough
 from sqlalchemy.exc import SQLAlchemyError
 
 from common.extensions import db, minio_client
-from common.models.document import DocumentVersion, Embedding
+from common.models.document import DocumentVersion, Embedding, Document
 from common.models.user import Tenant
 from common.utils.celery_utils import current_celery
 from common.utils.database import Database
@@ -36,34 +36,40 @@ def ping():
 
 @current_celery.task(name='create_embeddings', queue='embeddings')
 def create_embeddings(tenant_id, document_version_id):
-    # Retrieve document version to process
-    document_version = DocumentVersion.query.get(document_version_id)
-    if document_version is None:
-        raise Exception(f'Document version {document_version_id} not found')
+    try:
+        # Retrieve Tenant for which we are processing
+        tenant = Tenant.query.get(tenant_id)
+        if tenant is None:
+            raise Exception(f'Tenant {tenant_id} not found')
+
+        # Ensure we are working in the correct database schema
+        Database(tenant_id).switch_schema()
+
+        # Retrieve document version to process
+        document_version = DocumentVersion.query.get(document_version_id)
+        if document_version is None:
+            raise Exception(f'Document version {document_version_id} not found')
+
+        # Retrieve the Catalog ID
+        doc = Document.query.get_or_404(document_version.doc_id)
+        catalog_id = doc.catalog_id
+
+        # Select variables to work with depending on tenant and model
+        model_variables = select_model_variables(tenant, catalog_id=catalog_id)
+        current_app.logger.debug(f'Model variables: {model_variables}')
+
+    except Exception as e:
+        current_app.logger.error(f'Create Embeddings request received '
+                                 f'for non existing document version {document_version_id} '
+                                 f'for tenant {tenant_id}, '
+                                 f'error: {e}')
+        raise
+
     # BusinessEvent creates a context, which is why we need to use it with a with block
     with BusinessEvent('Create Embeddings', tenant_id,
                        document_version_id=document_version_id,
                        document_version_file_size=document_version.file_size):
         current_app.logger.info(f'Creating embeddings for tenant {tenant_id} on document version {document_version_id}')
-        try:
-            # Retrieve Tenant for which we are processing
-            tenant = Tenant.query.get(tenant_id)
-            if tenant is None:
-                raise Exception(f'Tenant {tenant_id} not found')
-
-            # Ensure we are working in the correct database schema
-            Database(tenant_id).switch_schema()
-
-            # Select variables to work with depending on tenant and model
-            model_variables = select_model_variables(tenant)
-            current_app.logger.debug(f'Model variables: {model_variables}')
-
-        except Exception as e:
-            current_app.logger.error(f'Create Embeddings request received '
-                                     f'for non existing document version {document_version_id} '
-                                     f'for tenant {tenant_id}, '
-                                     f'error: {e}')
-            raise
 
         try:
             db.session.add(document_version)
@@ -204,7 +210,7 @@ def enrich_chunks(tenant, model_variables, document_version, title, chunks):
     if len(chunks) > 1:
         summary = summarize_chunk(tenant, model_variables, document_version, chunks[0])
 
-    chunk_total_context = (f'Filename: {document_version.file_name}\n'
+    chunk_total_context = (f'Filename: {document_version.object_name}\n'
                            f'User Context:\n{document_version.user_context}\n\n'
                            f'User Metadata:\n{document_version.user_metadata}\n\n'
                            f'Title: {title}\n'
@@ -213,7 +219,7 @@ def enrich_chunks(tenant, model_variables, document_version, title, chunks):
                            f'System Metadata:\n{document_version.system_metadata}\n\n'
                            )
     enriched_chunks = []
-    initial_chunk = (f'Filename: {document_version.file_name}\n'
+    initial_chunk = (f'Filename: {document_version.object_name}\n'
                      f'User Context:\n{document_version.user_context}\n\n'
                      f'User Metadata:\n{document_version.user_metadata}\n\n'
                      f'Title: {title}\n'
@@ -304,13 +310,12 @@ def log_parsing_info(tenant, tags, included_elements, excluded_elements, exclude
 def create_potential_chunks_for_markdown(tenant_id, document_version, input_file):
     try:
         current_app.logger.info(f'Creating potential chunks for tenant {tenant_id}')
+        markdown_on = document_version.object_name.rsplit('.', 1)[0] + '.md'
 
         # Download the markdown file from MinIO
         markdown_data = minio_client.download_document_file(tenant_id,
-                                                            document_version.doc_id,
-                                                            document_version.language,
-                                                            document_version.id,
-                                                            input_file
+                                                            document_version.bucket_name,
+                                                            markdown_on,
                                                             )
         markdown = markdown_data.decode('utf-8')
 

integrations/Wordpress/eveai_sync/README.md

@@ -51,6 +51,10 @@ No additional configuration is needed; the plugin will automatically detect the
 
 ## Versions
 
+### 1.1.1 - Add Reinitialisation functionality
+
+### 1.1.0 - Add Catalog Functionality
+
 ### 1.0.x - Bugfixing Releases
 
 ### 1.0.0 - Initial Release

integrations/Wordpress/eveai_sync/eveai_sync.php

@@ -3,7 +3,7 @@
  * Plugin Name: EveAI Sync
  * Plugin URI: https://askeveai.com/
  * Description: Synchronizes WordPress content with EveAI API.
- * Version: 1.0.16
+ * Version: 1.1.1
  * Author: Josako, Pieter Laroy
  * Author URI: https://askeveai.com/about/
  * License: GPL v2 or later
@@ -17,7 +17,7 @@ if (!defined('ABSPATH')) {
 }
 
 // Define plugin constants
-define('EVEAI_SYNC_VERSION', '1.0.0');
+define('EVEAI_SYNC_VERSION', '1.1.1');
 define('EVEAI_SYNC_PLUGIN_DIR', plugin_dir_path(__FILE__));
 define('EVEAI_SYNC_PLUGIN_URL', plugin_dir_url(__FILE__));
 
@@ -50,6 +50,30 @@ function eveai_delete_post_meta($post_id) {
 }
 add_action('before_delete_post', 'eveai_delete_post_meta');
 
+// Clean metadata from Wordpress site
+function eveai_reinitialize_site() {
+    check_ajax_referer('eveai_reinitialize_site', 'nonce');
+
+    if (!current_user_can('manage_options')) {
+        wp_send_json_error('You do not have permission to perform this action.');
+        return;
+    }
+
+    global $wpdb;
+
+    // Remove all EveAI-related post meta
+    $wpdb->query("DELETE FROM $wpdb->postmeta WHERE meta_key LIKE '_eveai_%'");
+
+    // Remove all EveAI-related options
+    delete_option('eveai_last_sync_time');
+    delete_option('eveai_sync_status');
+
+    // Optionally, you might want to clear any custom tables if you have any
+
+    wp_send_json_success('Site reinitialized. All EveAI metadata has been removed.');
+}
+add_action('wp_ajax_eveai_reinitialize_site', 'eveai_reinitialize_site');
+
 // Display sync info in post
 function eveai_display_sync_info($post) {
     $document_id = get_post_meta($post->ID, '_eveai_document_id', true);

integrations/Wordpress/eveai_sync/includes/class-eveai-admin.php

@@ -16,6 +16,7 @@ class EveAI_Admin {
         register_setting('eveai_settings', 'eveai_excluded_categories');
         register_setting('eveai_settings', 'eveai_access_token');
         register_setting('eveai_settings', 'eveai_token_expiry');
+        register_setting('eveai_settings', 'eveai_catalog_id');
     }
 
     public function add_admin_menu() {
@@ -50,6 +51,10 @@ class EveAI_Admin {
                         <th scope="row">API Key</th>
                         <td><input type="text" name="eveai_api_key" value="<?php echo esc_attr(get_option('eveai_api_key')); ?>" style="width: 100%;" /></td>
                     </tr>
+                    <tr valign="top">
+                        <th scope="row">Catalog ID</th>
+                        <td><input type="text" name="eveai_catalog_id" value="<?php echo esc_attr(get_option('eveai_catalog_id')); ?>" style="width: 100%;" /></td>
+                    </tr>
                     <tr valign="top">
                         <th scope="row">Default Language</th>
                         <td><input type="text" name="eveai_default_language" value="<?php echo esc_attr(get_option('eveai_default_language', 'en')); ?>" style="width: 100%;" /></td>
@@ -71,6 +76,11 @@ class EveAI_Admin {
                 <?php wp_nonce_field('eveai_bulk_sync', 'eveai_bulk_sync_nonce'); ?>
                 <input type="submit" name="eveai_bulk_sync" class="button button-primary" value="Start Bulk Sync">
             </form>
+
+            <h2>Reinitialize Site</h2>
+            <p>Click the button below to remove all EveAI metadata from your site. This will reset the sync status for all posts and pages.</p>
+            <button id="eveai-reinitialize" class="button button-secondary">Reinitialize Site</button>
+
             <div id="eveai-sync-results" style="margin-top: 20px;"></div>
         </div>
         <script>
@@ -105,6 +115,29 @@ class EveAI_Admin {
                     });
                 }
             });
+            $('#eveai-reinitialize').on('click', function(e) {
+                e.preventDefault();
+                if (confirm('Are you sure you want to reinitialize? This will remove all EveAI metadata from your site.')) {
+                    $.ajax({
+                        url: ajaxurl,
+                        type: 'POST',
+                        data: {
+                            action: 'eveai_reinitialize_site',
+                            nonce: '<?php echo wp_create_nonce('eveai_reinitialize_site'); ?>'
+                        },
+                        success: function(response) {
+                            if (response.success) {
+                                alert(response.data);
+                            } else {
+                                alert('Error: ' + response.data);
+                            }
+                        },
+                        error: function() {
+                            alert('An error occurred. Please try again.');
+                        }
+                    });
+                }
+            });
         });
         </script>
         <?php

integrations/Wordpress/eveai_sync/includes/class-eveai-api.php

@@ -13,6 +13,7 @@ class EveAI_API {
         $this->api_key = get_option('eveai_api_key');
         $this->access_token = get_option('eveai_access_token');
         $this->token_expiry = get_option('eveai_token_expiry', 0);
+        $this->catalog_id = get_option('eveai_catalog_id');
     }
 
     private function ensure_valid_token() {
@@ -111,6 +112,7 @@ class EveAI_API {
     }
 
     public function add_url($data) {
+        $data['catalog_id'] = get_option('eveai_catalog_id');  // Include catalog_id
         return $this->make_request('POST', '/api/v1/documents/add_url', $data);
     }
 

migrations/public/versions/02debd224316_set_storage_dirty_flag_for_all_tenants.py

@@ -0,0 +1,24 @@
+"""Set storage_dirty flag for all tenants
+
+Revision ID: 02debd224316
+Revises: 8fdd7f2965c1
+Create Date: 2024-10-08 06:53:17.261709
+
+"""
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = '02debd224316'
+down_revision = '8fdd7f2965c1'
+branch_labels = None
+depends_on = None
+
+
+def upgrade():
+    op.execute('UPDATE tenant SET storage_dirty = TRUE')
+
+
+def downgrade():
+    pass

migrations/public/versions/a678c84d5633_licenseusage_correct_mb_fields_to_be_.py

@@ -0,0 +1,46 @@
+"""LicenseUsage: correct mb fields to be floats iso integers
+
+Revision ID: a678c84d5633
+Revises: 02debd224316
+Create Date: 2024-10-11 08:03:22.823327
+
+"""
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = 'a678c84d5633'
+down_revision = '02debd224316'
+branch_labels = None
+depends_on = None
+
+
+def upgrade():
+    # ### commands auto generated by Alembic - please adjust! ###
+    with op.batch_alter_table('license_usage', schema=None) as batch_op:
+        batch_op.alter_column('storage_mb_used',
+               existing_type=sa.INTEGER(),
+               type_=sa.Float(),
+               existing_nullable=True)
+        batch_op.alter_column('embedding_mb_used',
+               existing_type=sa.INTEGER(),
+               type_=sa.Float(),
+               existing_nullable=True)
+
+    # ### end Alembic commands ###
+
+
+def downgrade():
+    # ### commands auto generated by Alembic - please adjust! ###
+    with op.batch_alter_table('license_usage', schema=None) as batch_op:
+        batch_op.alter_column('embedding_mb_used',
+               existing_type=sa.Float(),
+               type_=sa.INTEGER(),
+               existing_nullable=True)
+        batch_op.alter_column('storage_mb_used',
+               existing_type=sa.Float(),
+               type_=sa.INTEGER(),
+               existing_nullable=True)
+
+    # ### end Alembic commands ###

migrations/tenant/env.py

@@ -1,5 +1,7 @@
 import inspect
 import logging
+import sys
+import os
 from logging.config import fileConfig
 
 from flask import current_app
@@ -19,8 +21,26 @@ config = context.config
 
 # Interpret the config file for Python logging.
 # This line sets up loggers basically.
-fileConfig(config.config_file_name)
-logger = logging.getLogger('alembic.env')
+# fileConfig(config.config_file_name)
+# logger = logging.getLogger('alembic.env')
+
+logging.basicConfig(
+    stream=sys.stdout,
+    level=logging.INFO,
+    format='%(asctime)s - %(name)s - %(levelname)s - %(message)s'
+)
+
+logger = logging.getLogger()
+
+# Reset handlers to avoid issues with Alembic overriding them
+for handler in logger.handlers:
+    logger.removeHandler(handler)
+
+# Add a stream handler to output logs to the console
+console_handler = logging.StreamHandler(sys.stdout)
+console_handler.setFormatter(logging.Formatter('%(asctime)s - %(name)s - %(levelname)s - %(message)s'))
+logger.addHandler(console_handler)
+logger.setLevel(logging.INFO)
 
 
 def get_engine():
@@ -124,31 +144,35 @@ def run_migrations_online():
     with connectable.connect() as connection:
         tenants = get_tenant_ids()
         for tenant in tenants:
-            logger.info(f"Migrating tenant: {tenant}")
-            # set search path on the connection, which ensures that
-            # PostgreSQL will emit all CREATE / ALTER / DROP statements
-            # in terms of this schema by default
-            connection.execute(text(f'SET search_path TO "{tenant}", public'))
-            # in SQLAlchemy v2+ the search path change needs to be committed
-            connection.commit()
+            try:
+                os.environ['TENANT_ID'] = str(tenant)
+                logger.info(f"Migrating tenant: {tenant}")
+                # set search path on the connection, which ensures that
+                # PostgreSQL will emit all CREATE / ALTER / DROP statements
+                # in terms of this schema by default
+                connection.execute(text(f'SET search_path TO "{tenant}", public'))
+                # in SQLAlchemy v2+ the search path change needs to be committed
+                connection.commit()
 
-            # make use of non-supported SQLAlchemy attribute to ensure
-            # the dialect reflects tables in terms of the current tenant name
-            connection.dialect.default_schema_name = str(tenant)
+                # make use of non-supported SQLAlchemy attribute to ensure
+                # the dialect reflects tables in terms of the current tenant name
+                connection.dialect.default_schema_name = str(tenant)
 
-            context.configure(
-                connection=connection,
-                target_metadata=get_metadata(),
-                # literal_binds=True,
-                include_object=include_object,
-            )
+                context.configure(
+                    connection=connection,
+                    target_metadata=get_metadata(),
+                    # literal_binds=True,
+                    include_object=include_object,
+                )
 
-            with context.begin_transaction():
-                context.run_migrations()
+                with context.begin_transaction():
+                    context.run_migrations()
 
-            # for checking migrate or upgrade is running
-            if getattr(config.cmd_opts, "autogenerate", False):
-                break
+                # for checking migrate or upgrade is running
+                if getattr(config.cmd_opts, "autogenerate", False):
+                    break
+            except Exception as e:
+                continue
 
 
 if context.is_offline_mode():

migrations/tenant/versions/0f5932ff3051_adding_catalogs_to_document_domain.py

@@ -0,0 +1,56 @@
+"""Adding Catalogs to Document Domain
+
+Revision ID: 0f5932ff3051
+Revises: 5a75fb6da7b8
+Create Date: 2024-10-14 15:20:45.058329
+
+"""
+from alembic import op
+import sqlalchemy as sa
+import pgvector
+from sqlalchemy.dialects import postgresql
+
+# revision identifiers, used by Alembic.
+revision = '0f5932ff3051'
+down_revision = '5a75fb6da7b8'
+branch_labels = None
+depends_on = None
+
+
+def upgrade():
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.create_table('catalog',
+                    sa.Column('id', sa.Integer(), nullable=False),
+                    sa.Column('name', sa.String(length=50), nullable=False),
+                    sa.Column('description', sa.Text(), nullable=True),
+                    sa.Column('html_tags', postgresql.ARRAY(sa.String(length=10)), nullable=True),
+                    sa.Column('html_end_tags', postgresql.ARRAY(sa.String(length=10)), nullable=True),
+                    sa.Column('html_included_elements', postgresql.ARRAY(sa.String(length=50)), nullable=True),
+                    sa.Column('html_excluded_elements', postgresql.ARRAY(sa.String(length=50)), nullable=True),
+                    sa.Column('html_excluded_classes', postgresql.ARRAY(sa.String(length=200)), nullable=True),
+                    sa.Column('min_chunk_size', sa.Integer(), nullable=True),
+                    sa.Column('max_chunk_size', sa.Integer(), nullable=True),
+                    sa.Column('es_k', sa.Integer(), nullable=True),
+                    sa.Column('es_similarity_threshold', sa.Float(), nullable=True),
+                    sa.Column('chat_RAG_temperature', sa.Float(), nullable=True),
+                    sa.Column('chat_no_RAG_temperature', sa.Float(), nullable=True),
+                    sa.Column('embed_tuning', sa.Boolean(), nullable=True),
+                    sa.Column('rag_tuning', sa.Boolean(), nullable=True),
+                    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.ForeignKeyConstraint(['created_by'], ['public.user.id'], ),
+                    sa.ForeignKeyConstraint(['updated_by'], ['public.user.id'], ),
+                    sa.PrimaryKeyConstraint('id')
+                    )
+    op.add_column('document', sa.Column('catalog_id', sa.Integer(), nullable=True))
+    op.create_foreign_key(None, 'document', 'catalog', ['catalog_id'], ['id'])
+    # ### end Alembic commands ###
+
+
+def downgrade():
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.drop_column('document', 'catalog_id')
+    op.drop_table('catalog')
+    # ### end Alembic commands ###

migrations/tenant/versions/28984b05d396_upgrading_existing_documents_to_default_.py

@@ -0,0 +1,99 @@
+"""Upgrading Existing documents to default catalog
+
+Revision ID: 28984b05d396
+Revises: 0f5932ff3051
+Create Date: 2024-10-15 09:02:23.355660
+
+"""
+from alembic import op
+import sqlalchemy as sa
+from sqlalchemy.orm import Session
+import pgvector
+from common.models.document import Catalog, Document
+from common.models.user import Tenant
+from flask import current_app
+import logging
+import os
+
+# Set up logging
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+
+# Create a console handler
+console_handler = logging.StreamHandler()
+console_handler.setLevel(logging.INFO)
+
+# Create a logging format
+formatter = logging.Formatter('%(asctime)s - %(name)s - %(levelname)s - %(message)s')
+console_handler.setFormatter(formatter)
+
+# Add the handler to the logger
+logger.addHandler(console_handler)
+
+logger.info("Starting revision upgrade 28984b05d396")
+
+# revision identifiers, used by Alembic.
+revision = '28984b05d396'
+down_revision = '0f5932ff3051'
+branch_labels = None
+depends_on = None
+
+
+def upgrade():
+    logger.info("Starting migration: Creating default catalog and updating documents.")
+
+    bind = op.get_bind()
+    session = Session(bind=bind)
+
+    try:
+        tenant_id = int(os.getenv('TENANT_ID'))
+        logger.info(f"In migration: tenant_id = {tenant_id}")
+
+        # Step 1: Create a new Default Catalog
+        logger.info("Creating default catalog")
+        default_catalog = Catalog(
+            name='Default Catalog',
+            description=None,
+        )
+        tenant = Tenant.query.get_or_404(tenant_id)
+        default_catalog.html_tags = tenant.html_tags
+        default_catalog.html_end_tags = tenant.html_end_tags
+        default_catalog.html_included_elements = tenant.html_included_elements
+        default_catalog.html_excluded_elements = tenant.html_excluded_elements
+        default_catalog.html_excluded_classes = tenant.html_excluded_classes
+        default_catalog.min_chunk_size = tenant.min_chunk_size
+        default_catalog.max_chunk_size = tenant.max_chunk_size
+        default_catalog.es_k = tenant.es_k
+        default_catalog.es_similarity_threshold = tenant.es_similarity_threshold
+        default_catalog.chat_RAG_temperature = tenant.chat_RAG_temperature
+        default_catalog.chat_no_RAG_temperature = tenant.chat_no_RAG_temperature
+        default_catalog.embed_tuning = tenant.embed_tuning
+        default_catalog.rag_tuning = tenant.rag_tuning
+
+        session.add(default_catalog)
+        session.commit()
+
+        new_catalog_id = default_catalog.id
+        logger.info(f"Default catalog created with ID: {new_catalog_id}")
+
+        # Step 2: Update all documents to use this new catalog
+        logger.info("Updating documents with the new catalog ID.")
+        documents = session.query(Document).all()
+        for document in documents:
+            document.catalog_id = new_catalog_id
+        session.commit()
+
+        logger.info(f"Updated {len(documents)} documents with new catalog ID.")
+
+    except Exception as e:
+        logger.error(f"An error occurred during migration: {e}")
+        session.rollback()
+        raise
+    finally:
+        session.close()
+
+    logger.info("Migration completed successfully.")
+
+
+def downgrade():
+    pass

migrations/tenant/versions/322d3cf1f17b_documentversion_update_to_bucket_name_.py

@@ -26,55 +26,55 @@ def upgrade():
     op.add_column('document_version', sa.Column('object_name', sa.String(length=200), nullable=True))
     op.add_column('document_version', sa.Column('file_size', sa.Float(), nullable=True))
 
-    # ### Upgrade values for bucket_name, object_name and file_size to reflect minio reality ###
-    from common.models.document import DocumentVersion
-    from common.extensions import minio_client
-    from minio.error import S3Error
-
-    # Create a connection
-    connection = op.get_bind()
-    session = Session(bind=connection)
-
-    # Get the current schema name (which should be the tenant ID)
-    current_schema = connection.execute(text("SELECT current_schema()")).scalar()
-    tenant_id = int(current_schema)
-
-    doc_versions = session.query(DocumentVersion).all()
-    for doc_version in doc_versions:
-        try:
-            object_name = minio_client.generate_object_name(doc_version.doc_id,
-                                                            doc_version.language,
-                                                            doc_version.id,
-                                                            doc_version.file_name)
-            bucket_name = minio_client.generate_bucket_name(tenant_id)
-            doc_version.object_name = object_name
-            doc_version.bucket_name = bucket_name
-
-            try:
-                stat = minio_client.client.stat_object(
-                    bucket_name=bucket_name,
-                    object_name=object_name
-                )
-                doc_version.file_size = stat.size / 1048576
-                current_app.logger.info(f"Processed Upgrade for  DocumentVersion {doc_version.id} for Tenant {tenant_id}")
-            except S3Error as e:
-                if e.code == "NoSuchKey":
-                    current_app.logger.warning(
-                        f"Object {doc_version.file_location} not found in bucket {doc_version.bucket_name}. Skipping.")
-                    continue  # Move to the next item
-                else:
-                    raise e  # Handle other types of S3 errors
-        except Exception as e:
-            session.rollback()
-            current_app.logger.error(f"Couldn't process upgrade for DocumentVersion {doc_version.id} for "
-                                     f"Tenant {tenant_id}. Error: {str(e)}")
-
-    try:
-        session.commit()
-        current_app.logger.info(f"Successfully updated file sizes for tenant schema {current_schema}")
-    except Exception as e:
-        session.rollback()
-        current_app.logger.error(f"Error committing changes for tenant schema {current_schema}: {str(e)}")
+    # # ### Upgrade values for bucket_name, object_name and file_size to reflect minio reality ###
+    # from common.models.document import DocumentVersion
+    # from common.extensions import minio_client
+    # from minio.error import S3Error
+    #
+    # # Create a connection
+    # connection = op.get_bind()
+    # session = Session(bind=connection)
+    #
+    # # Get the current schema name (which should be the tenant ID)
+    # current_schema = connection.execute(text("SELECT current_schema()")).scalar()
+    # tenant_id = int(current_schema)
+    #
+    # doc_versions = session.query(DocumentVersion).all()
+    # for doc_version in doc_versions:
+    #     try:
+    #         object_name = minio_client.generate_object_name(doc_version.doc_id,
+    #                                                         doc_version.language,
+    #                                                         doc_version.id,
+    #                                                         doc_version.file_name)
+    #         bucket_name = minio_client.generate_bucket_name(tenant_id)
+    #         doc_version.object_name = object_name
+    #         doc_version.bucket_name = bucket_name
+    #
+    #         try:
+    #             stat = minio_client.client.stat_object(
+    #                 bucket_name=bucket_name,
+    #                 object_name=object_name
+    #             )
+    #             doc_version.file_size = stat.size / 1048576
+    #             current_app.logger.info(f"Processed Upgrade for  DocumentVersion {doc_version.id} for Tenant {tenant_id}")
+    #         except S3Error as e:
+    #             if e.code == "NoSuchKey":
+    #                 current_app.logger.warning(
+    #                     f"Object {doc_version.object_name} not found in bucket {doc_version.bucket_name}. Skipping.")
+    #                 continue  # Move to the next item
+    #             else:
+    #                 raise e  # Handle other types of S3 errors
+    #     except Exception as e:
+    #         session.rollback()
+    #         current_app.logger.error(f"Couldn't process upgrade for DocumentVersion {doc_version.id} for "
+    #                                  f"Tenant {tenant_id}. Error: {str(e)}")
+    #
+    # try:
+    #     session.commit()
+    #     current_app.logger.info(f"Successfully updated file sizes for tenant schema {current_schema}")
+    # except Exception as e:
+    #     session.rollback()
+    #     current_app.logger.error(f"Error committing changes for tenant schema {current_schema}: {str(e)}")
 
     # ### commands auto generated by Alembic - Remove old fields ###
     # op.drop_column('document_version', 'file_location')

nginx/public/chat_evie.html

@@ -17,7 +17,7 @@
     document.addEventListener('DOMContentLoaded', function() {
       const eveAI = new EveAI(
         '2',
-        'EveAI-CHAT-9079-8604-7441-6496-7604',
+        'EveAI-CHAT-6525-5692-6412-5828-7546',
         'http://macstudio.ask-eve-ai-local.com',
         'en',
         'en,fr,nl',

requirements.txt

@@ -81,3 +81,4 @@ anthropic~=0.34.2
 prometheus-client~=0.20.0
 flower~=2.0.1
 psutil~=6.0.0
+celery-redbeat~=2.2.0

scripts/db_squash.sh

@@ -0,0 +1,42 @@
+#!/usr/bin/env bash
+
+# Usage: ./db_migrate.sh -m "Your migration message" -d migrations/public
+
+# Load environment variables
+set -a
+source .env
+set +a
+
+project_dir=$(pwd)
+# Ensure you're in the correct directory
+cd "$project_dir" || exit 1
+
+# Load environment variables (including database connection details)
+source .env
+
+# Set Flask app and Python path
+export FLASK_APP=scripts/run_eveai_app.py
+export PYTHONPATH="$PYTHONPATH:$project_dir"
+
+while getopts m: flag
+do
+    case "${flag}" in
+        m) message=${OPTARG};;
+        *)
+            echo "Invalid option: ${flag}"
+            exit 1
+        ;;
+    esac
+done
+
+# Check if the message and directory are provided
+if [ -z "$message" ]; then
+    echo "Message is required."
+    echo "Usage: ./db_migrate.sh -m \"Your migration message\" -d migrations/public"
+    exit 1
+fi
+
+# Run the alembic commands
+flask db revision --autogenerate -m "Consolidated Public Migrations - $message" -d migrations/public
+flask db revision --autogenerate -m "Consolidated Tenant Migrations - $message" -d migrations/tenant
+

scripts/entrypoint_no_db.sh

@@ -0,0 +1,10 @@
+#!/bin/bash
+set -e
+
+# Ensure the logs directory has the correct permissions
+echo "Changing permissions on logs directory"
+#chown -R appuser:appuser /app/logs
+chmod -R 777 /app/logs
+
+# Switch to appuser and execute the command passed to the script
+exec su -- appuser -c "$@"

scripts/repopack_eveai.sh

@@ -1,17 +1,35 @@
 #!/bin/bash
+# Delete previous repopack files
+rm -f *repo.txt
 
-# Run repopack to generate the file
-repopack
+# Define the list of components
+components=("docker" "eveai_api" "eveai_app" "eveai_beat" "eveai_chat" "eveai_chat_workers" "eveai_entitlements" "eveai_workers" "nginx" "full" "integrations")
 
-# Check if repopack generated the eveai_repo.txt file
-if [[ -f "eveai_repo.txt" ]]; then
-    # Get the current timestamp in the format YYYY-DD-MM_HH:MM:SS
-    timestamp=$(date +"%Y-%d-%m_%H-%M-%S")
+# Get the current date and time in the format YYYY-MM-DD_HH-MM
+timestamp=$(date +"%Y-%m-%d_%H-%M")
 
-    # Rename the file with the timestamp
-    mv eveai_repo.txt "${timestamp}_eveai_repo.txt"
+# Loop through each component and perform the tasks
+for component in "${components[@]}"; do
+    echo "Processing component: $component"
 
-    echo "File renamed to ${timestamp}_eveai_repo.txt"
-else
-    echo "Error: eveai_repo.txt not found. repopack may have failed."
-fi
+    # Merge the .repopackignore_base and .repopackignore_<component> into .repopackignore
+    if [[ -f ".repopackignore_base" && -f ".repopackignore_$component" ]]; then
+        cat .repopackignore_base .repopackignore_$component > .repopackignore
+    else
+        echo "Warning: Missing .repopackignore_base or .repopackignore_$component for $component"
+        continue
+    fi
+
+    # Execute repopack
+    repopack
+
+    # Rename the resulting eveai_repo.txt file to <component>YYYY-MM-DD_HH-MM_repo.txt
+    if [[ -f "eveai_repo.txt" ]]; then
+        mv eveai_repo.txt "${component}_${timestamp}_repo.txt"
+        echo "Renamed eveai_repo.txt to ${component}_${timestamp}_repo.txt"
+    else
+        echo "Error: repopack did not generate eveai_repo.txt for $component"
+    fi
+
+    echo "Finished processing $component"
+done

scripts/start_eveai_app.sh

@@ -32,10 +32,12 @@ export PYTHONPATH="$PROJECT_DIR/patched_packages:$PYTHONPATH:$PROJECT_DIR"  # In
 # Run Alembic upgrade for the public schema
 echo "Applying migrations to the public schema..."
 flask db upgrade -d "${PROJECT_DIR}/migrations/public"
+echo "Finished applying migrations to the public schema..."
 
 # Run Alembic upgrade for the tenant schema
 echo "Applying migrations to the tenant schema..."
 flask db upgrade -d "${PROJECT_DIR}/migrations/tenant"
+echo "Finished applying migrations to the tenant schema..."
 
 # Set flask environment variables
 #export FLASK_ENV=development  # Use 'production' as appropriate

scripts/start_eveai_beat.sh

@@ -0,0 +1,17 @@
+#!/bin/bash
+
+cd "/app/" || exit 1
+export PROJECT_DIR="/app"
+export PYTHONPATH="$PROJECT_DIR/patched_packages:$PYTHONPATH:$PROJECT_DIR"  # Include the app directory in the Python path & patched packages
+
+# Ensure we can write the logs
+chown -R appuser:appuser /app/logs
+
+# Start Celery Beat
+celery -A eveai_beat.celery beat --scheduler=redbeat.RedBeatScheduler --loglevel=debug &
+
+# Start a worker for the 'llm_interactions' queue with auto-scaling - not necessary, in eveai_chat_workers
+# celery -A eveai_workers.celery worker --loglevel=info - Q llm_interactions --autoscale=2,8 --hostname=interactions_worker@%h &
+
+# Wait for all background processes to finish
+wait

scripts/start_eveai_entitlements.sh

@@ -0,0 +1,17 @@
+#!/bin/bash
+
+cd "/app/" || exit 1
+export PROJECT_DIR="/app"
+export PYTHONPATH="$PROJECT_DIR/patched_packages:$PYTHONPATH:$PROJECT_DIR"  # Include the app directory in the Python path & patched packages
+
+# Ensure we can write the logs
+chown -R appuser:appuser /app/logs
+
+# Start a worker for the 'embeddings' queue with higher concurrency
+celery -A eveai_entitlements.celery worker --loglevel=debug -Q entitlements --autoscale=2,8 --hostname=entitlements_worker@%h &
+
+# Start a worker for the 'llm_interactions' queue with auto-scaling - not necessary, in eveai_chat_workers
+# celery -A eveai_workers.celery worker --loglevel=info - Q llm_interactions --autoscale=2,8 --hostname=interactions_worker@%h &
+
+# Wait for all background processes to finish
+wait