iatoolkit 2.50.2__tar.gz → 2.51.0__tar.gz

{iatoolkit-2.50.2/src/iatoolkit.egg-info → iatoolkit-2.51.0}/PKG-INFO

@@ -1,10 +1,10 @@
 Metadata-Version: 2.4
 Name: iatoolkit
-Version: 2.50.2
+Version: 2.51.0
 Summary: IAToolkit
 Author: Fernando Libedinsky
 License-Expression: MIT
-Requires-Python: >=3.12
+Requires-Python: >=3.13
 Description-Content-Type: text/markdown
 License-File: LICENSE
 License-File: LICENSE_COMMUNITY.md

{iatoolkit-2.50.2 → iatoolkit-2.51.0}/pyproject.toml

@@ -4,8 +4,8 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "iatoolkit"
-dynamic = ["version", "dependencies"]
-requires-python = ">=3.12"
+dynamic = ["version"]
+requires-python = ">=3.13"
 description = "IAToolkit"
 readme = "readme.md"
 license = "MIT"
@@ -17,6 +17,90 @@ authors = [{ name = "Fernando Libedinsky" }]
 # the whole suite: nothing declared them, so nothing could install them on
 # purpose. Declared here, `pip install iatoolkit[ocr]` is a decision instead of
 # a memory.
+dependencies = [
+  "bcrypt==4.2.1",
+  "anthropic>=0.83.0",
+  "boto3==1.43.10",
+  "botocore==1.43.10",
+  "braintrust>=0.17.0",
+  "build==1.2.2.post1",
+  "click==8.1.8",
+  "cryptography==44.0.3",
+  "Flask==3.1.0",
+  "Flask-Bcrypt==1.0.1",
+  "flask-cors==6.0.0",
+  "Flask-Injector==0.15.0",
+  "Flask-Session==0.8.0",
+  "flatbuffers==24.3.25",
+  "gevent==25.9.1",
+  "google-ai-generativelanguage==0.6.15",
+  "google-api-core==2.24.1",
+  "google-api-python-client==2.161.0",
+  "google-auth>=2.46.0",
+  "google-auth-httplib2==0.2.0",
+  "google-auth-oauthlib==1.2.1",
+  "google-cloud-core==2.4.1",
+  "google-cloud-storage==3.0.0",
+  # 1.6.0 publishes no cp313 wheel. This was the only pin in the whole
+  # dependency set that Python 3.13 required moving.
+  "google-crc32c==1.8.0",
+  "google-genai==1.57.0",
+  "google-resumable-media==2.7.2",
+  "googleapis-common-protos==1.66.0",
+  "gunicorn==23.0.0",
+  "h11==0.14.0",
+  "httpcore==1.0.7",
+  "httplib2==0.22.0",
+  "httptools==0.6.4",
+  "httpx>=0.28.1,<1.0.0",
+  "httpx-sse==0.4.0",
+  "huggingface-hub>=0.16.4",
+  "humanfriendly==10.0",
+  "idna==3.10",
+  "injector==0.22.0",
+  "Jinja2==3.1.5",
+  "langchain==0.3.19",
+  "langchain-core==0.3.35",
+  "langchain-text-splitters==0.3.6",
+  "markdown2==2.5.3",
+  "nh3==0.3.7",
+  "openai==2.8.1",
+  "openpyxl==3.1.5",
+  "pandas==2.3.1",
+  "pgvector==0.3.6",
+  "pillow==11.0.0",
+  "psutil==7.0.0",
+  "psycogreen==1.0.2",
+  "psycopg2-binary==2.9.10",
+  "PyMySQL==1.1.1",
+  "PyJWT==2.10.1",
+  "PyMuPDF==1.25.0",
+  "python-dotenv==1.0.1",
+  "pytest==8.3.4",
+  "pytest-cov==5.0.0",
+  "pytest-mock==3.14.0",
+  "python-dateutil==2.9.0.post0",
+  "python-docx==1.1.2",
+  "pytz==2025.2",
+  "PyYAML==6.0.2",
+  "redis==5.2.1",
+  "regex==2024.11.6",
+  "requests==2.32.3",
+  "requests-oauthlib==2.0.0",
+  "requests-toolbelt==1.0.0",
+  "s3transfer==0.17.0",
+  "sib-api-v3-sdk==7.6.0",
+  "SQLAlchemy==2.0.49",
+  "sqlalchemy-redshift==1.0.0",
+  "redshift_connector==2.1.14",
+  "tiktoken==0.8.0",
+  "tokenizers>=0.22.0",
+  "websocket-client==1.8.0",
+  "websockets==14.1",
+  "Werkzeug==3.1.3",
+  "pyjwt[crypto]>=2.8.0",
+]
+
 [project.optional-dependencies]
 # Parsing a document with the Docling pipeline rather than the basic provider.
 docling = [
@@ -58,7 +142,6 @@ exclude = ["tests*", "*/tests*"]
 
 [tool.setuptools.dynamic]
 version = { attr = "iatoolkit.__version__" }
-dependencies = { file = ["requirements.txt"] }
 
 [tool.setuptools.package-data]
 iatoolkit = ["templates/**/*", "static/**/*", "locales/*.yaml", "config/*.yaml", "config/system_prompts/*.prompt"]

{iatoolkit-2.50.2 → iatoolkit-2.51.0}/src/iatoolkit/__init__.py

@@ -3,7 +3,7 @@
 #
 # IAToolkit is open source software.
 
-__version__ = "2.50.2"
+__version__ = "2.51.0"
 
 # Expose main classes and functions at the top level of the package
 

{iatoolkit-2.50.2 → iatoolkit-2.51.0}/src/iatoolkit/locales/en.yaml

@@ -1065,6 +1065,17 @@ ui:
     scheduled_for: "Scheduled for"
     duration: "Duration"
     detail: "Detail"
+    view_detail: "View detail"
+    run_detail_title: "Run detail"
+    run_id: "Run ID"
+    attempt_label: "Attempt"
+    trigger: "Trigger"
+    output: "Output"
+    error: "Error"
+    no_output: "No output recorded."
+    copy_output: "Copy"
+    output_copied: "Output copied."
+    copy_failed: "Could not copy output."
     identity: "Identity"
     target: "Target"
     target_type: "What to run"

{iatoolkit-2.50.2 → iatoolkit-2.51.0}/src/iatoolkit/locales/es.yaml

@@ -1080,6 +1080,17 @@ ui:
     scheduled_for: "Programada para"
     duration: "Duración"
     detail: "Detalle"
+    view_detail: "Ver detalle"
+    run_detail_title: "Detalle de ejecución"
+    run_id: "ID de ejecución"
+    attempt_label: "Intento"
+    trigger: "Disparador"
+    output: "Salida"
+    error: "Error"
+    no_output: "No hay salida registrada."
+    copy_output: "Copiar"
+    output_copied: "Salida copiada."
+    copy_failed: "No se pudo copiar la salida."
     identity: "Identidad"
     target: "Destino"
     target_type: "Qué ejecutar"

{iatoolkit-2.50.2 → iatoolkit-2.51.0}/src/iatoolkit/repositories/database_manager.py

@@ -4,6 +4,7 @@
 # IAToolkit is open source software.
 
 # database_manager.py
+import os
 from sqlalchemy import create_engine, event, inspect, text
 from sqlalchemy.dialects import registry
 from sqlalchemy.orm import sessionmaker, scoped_session
@@ -21,6 +22,15 @@ class DatabaseManager(DatabaseProvider):
     )
     _DEFAULT_CONNECT_TIMEOUT = 60
     MAX_RESULT_ROWS = 2000
+    #: Connections each pool may hold. Every process opens one pool per database it
+    #: touches, so the ceiling a deployment must respect is
+    #: processes x databases x (pool_size + max_overflow), against the server's
+    #: max_connections. A worker fleet of 16 processes over two databases on a 103-slot
+    #: instance exhausted it with the previous 10/20, and the papers pipeline lost two of
+    #: every three ingestions to "FATAL: remaining connection slots are reserved".
+    #: Override per service with IAT_DB_POOL_SIZE / IAT_DB_MAX_OVERFLOW.
+    DEFAULT_POOL_SIZE = 3
+    DEFAULT_MAX_OVERFLOW = 5
 
     @inject
     def __init__(self,
@@ -51,8 +61,8 @@ class DatabaseManager(DatabaseProvider):
             raw_engine = create_engine(
                 self.engine_url,
                 echo=False,
-                pool_size=10,  # per worker
-                max_overflow=20,
+                pool_size=self._resolve_pool_setting("IAT_DB_POOL_SIZE", self.DEFAULT_POOL_SIZE),
+                max_overflow=self._resolve_pool_setting("IAT_DB_MAX_OVERFLOW", self.DEFAULT_MAX_OVERFLOW),
                 pool_timeout=30,
                 pool_recycle=1800,
                 pool_pre_ping=True,
@@ -89,6 +99,21 @@ class DatabaseManager(DatabaseProvider):
     def _is_mysql(self) -> bool:
         return self.backend == 'mysql'
 
+    @staticmethod
+    def _resolve_pool_setting(env_name: str, default: int) -> int:
+        raw_value = str(os.getenv(env_name, "") or "").strip()
+        if not raw_value:
+            return default
+        try:
+            value = int(raw_value)
+        except (TypeError, ValueError):
+            logging.warning("%s=%r is not an integer; using %s", env_name, raw_value, default)
+            return default
+        if value < 0:
+            logging.warning("%s=%s is negative; using %s", env_name, value, default)
+            return default
+        return value
+
     @classmethod
     def _normalize_timeout(cls, timeout: int | str | None) -> int:
         if timeout is None:

{iatoolkit-2.50.2 → iatoolkit-2.51.0}/src/iatoolkit/repositories/models.py

@@ -451,7 +451,7 @@ class DocumentImage(Base):
     __tablename__ = 'iat_document_images'
 
     id = Column(Integer, primary_key=True, autoincrement=True)
-    document_id = Column(Integer, ForeignKey(f'{ORM_SCHEMA}.iat_documents.id', ondelete='CASCADE'), nullable=False)
+    document_id = Column(Integer, ForeignKey(f'{ORM_SCHEMA}.iat_documents.id', ondelete='CASCADE'), nullable=False, index=True)
 
     page = Column(Integer, nullable=True)
     image_index = Column(Integer, nullable=True)
@@ -612,7 +612,7 @@ class VSDoc(Base):
     company_id = Column(Integer, ForeignKey(f'{ORM_SCHEMA}.iat_companies.id',
                     ondelete='CASCADE'), nullable=False)
     document_id = Column(Integer, ForeignKey(f'{ORM_SCHEMA}.iat_documents.id',
-                        ondelete='CASCADE'), nullable=False)
+                        ondelete='CASCADE'), nullable=False, index=True)
     text = Column(Text, nullable=False)
     meta = Column(JSON_NATIVE, nullable=True)
 
@@ -632,7 +632,7 @@ class VSImage(Base):
     company_id = Column(Integer, ForeignKey(f'{ORM_SCHEMA}.iat_companies.id',
                                             ondelete='CASCADE'), nullable=False)
     document_image_id = Column(Integer, ForeignKey(f'{ORM_SCHEMA}.iat_document_images.id',
-                                                   ondelete='CASCADE'), nullable=False)
+                                                   ondelete='CASCADE'), nullable=False, index=True)
 
     # Vector dimension depends on the multimodal model (e.g., CLIP uses 512 or 768)
     embedding = Column(Vector, nullable=False)

{iatoolkit-2.50.2 → iatoolkit-2.51.0}/src/iatoolkit/services/configuration_service.py

@@ -13,6 +13,7 @@ from iatoolkit.services.structured_output_service import StructuredOutputService
 from injector import inject
 import logging
 import os
+import re
 from urllib.parse import urlparse
 
 
@@ -1472,30 +1473,23 @@ class ConfigurationService:
         """
         main_config_filename = "company.yaml"
 
+        # Requests carry the tenant in the URL, so anything on the internet can ask this
+        # service to load a configuration for a name of its choosing. A scanner walking a
+        # wordlist of leaked-secret filenames (.env.json, .env.backup1, ...) reached here
+        # through the language before_request hook: every probe logged a warning about a
+        # missing company.yaml and left an entry in _loaded_configs keyed by its own input.
+        if not self._is_well_formed_company_short_name(company_short_name):
+            logging.debug("Ignoring configuration request for malformed tenant name: %r", company_short_name)
+            return self._minimal_configuration(company_short_name)
+
         # verify existence of the main configuration file
         if not self.asset_repo.exists(company_short_name, AssetType.CONFIG, main_config_filename):
-            logging.warning(f"Main configuration file not found: {main_config_filename}")
-
-            # return the minimal configuration needed for starting the IAToolkit
-            # this is a for solving a chicken/egg problem when trying to migrate the configuration
-            # from filesystem to database in enterprise installation
-            # see create_assets cli command in enterprise-iatoolkit)
-            return {
-                'id': company_short_name,
-                'name': company_short_name,
-                'llm': {'model': 'gpt-5', 'provider_api_keys': {'openai': ''}},
-                'data_sources': {'sql': []},
-                'tools': [],
-                'prompts': {'prompt_categories': [], 'prompt_list': []},
-                'branding': {},
-                'onboarding_cards': [],
-                'help_content': {
-                    'example_questions': [],
-                    'data_sources': [],
-                    'best_practices': [],
-                    'capabilities': {'can_do': [], 'cannot_do': []},
-                },
-                }
+            logging.warning(
+                "Main configuration file not found for company '%s': %s",
+                company_short_name,
+                main_config_filename,
+            )
+            return self._minimal_configuration(company_short_name)
 
         # read text and parse
         yaml_content = self.asset_repo.read_text(company_short_name, AssetType.CONFIG, main_config_filename)
@@ -1514,6 +1508,41 @@ class ConfigurationService:
 
         return config
 
+    #: A tenant short name as the platform issues them. Anything else is not a tenant that
+    #: could exist, so it is rejected before it reaches the asset repository or the cache.
+    COMPANY_SHORT_NAME_RE = re.compile(r"^[a-z0-9][a-z0-9_-]{0,62}$")
+
+    @classmethod
+    def _is_well_formed_company_short_name(cls, company_short_name) -> bool:
+        # Normalized the way the platform itself normalizes a tenant before looking it up,
+        # so a caller passing a name with padding or casing is not treated as hostile.
+        normalized = str(company_short_name or "").strip().lower()
+        return bool(cls.COMPANY_SHORT_NAME_RE.match(normalized))
+
+    @staticmethod
+    def _minimal_configuration(company_short_name: str) -> dict:
+        """The smallest configuration that still lets IAToolkit boot.
+
+        Solves a chicken/egg problem when migrating configuration from the filesystem to
+        the database in an enterprise installation - see the create_assets CLI command.
+        """
+        return {
+            'id': company_short_name,
+            'name': company_short_name,
+            'llm': {'model': 'gpt-5', 'provider_api_keys': {'openai': ''}},
+            'data_sources': {'sql': []},
+            'tools': [],
+            'prompts': {'prompt_categories': [], 'prompt_list': []},
+            'branding': {},
+            'onboarding_cards': [],
+            'help_content': {
+                'example_questions': [],
+                'data_sources': [],
+                'best_practices': [],
+                'capabilities': {'can_do': [], 'cannot_do': []},
+            },
+        }
+
     def _get_prompt_config(self, config):
         prompts_config = config.get('prompts', {})
         if isinstance(prompts_config, dict):

{iatoolkit-2.50.2 → iatoolkit-2.51.0/src/iatoolkit.egg-info}/PKG-INFO

@@ -1,10 +1,10 @@
 Metadata-Version: 2.4
 Name: iatoolkit
-Version: 2.50.2
+Version: 2.51.0
 Summary: IAToolkit
 Author: Fernando Libedinsky
 License-Expression: MIT
-Requires-Python: >=3.12
+Requires-Python: >=3.13
 Description-Content-Type: text/markdown
 License-File: LICENSE
 License-File: LICENSE_COMMUNITY.md

{iatoolkit-2.50.2 → iatoolkit-2.51.0}/src/iatoolkit.egg-info/SOURCES.txt

@@ -2,7 +2,6 @@ LICENSE
 LICENSE_COMMUNITY.md
 pyproject.toml
 readme.md
-requirements.txt
 src/iatoolkit/__init__.py
 src/iatoolkit/base_company.py
 src/iatoolkit/cli_commands.py

iatoolkit-2.50.2/requirements.txt

@@ -1,81 +0,0 @@
-bcrypt==4.2.1
-anthropic>=0.83.0
-boto3==1.43.10
-botocore==1.43.10
-braintrust>=0.17.0
-build==1.2.2.post1
-click==8.1.8
-cryptography==44.0.3
-Flask==3.1.0
-Flask-Bcrypt==1.0.1
-flask-cors==6.0.0
-Flask-Injector==0.15.0
-Flask-Session==0.8.0
-flatbuffers==24.3.25
-gevent==25.9.1
-google-ai-generativelanguage==0.6.15
-google-api-core==2.24.1
-google-api-python-client==2.161.0
-google-auth>=2.46.0
-google-auth-httplib2==0.2.0
-google-auth-oauthlib==1.2.1
-google-cloud-core==2.4.1
-google-cloud-storage==3.0.0
-# 1.6.0 publishes no cp313 wheel. This was the only pin in the whole
-# dependency set that Python 3.13 required moving.
-google-crc32c==1.8.0
-google-genai==1.57.0
-google-resumable-media==2.7.2
-googleapis-common-protos==1.66.0
-gunicorn==23.0.0
-h11==0.14.0
-httpcore==1.0.7
-httplib2==0.22.0
-httptools==0.6.4
-httpx>=0.28.1,<1.0.0
-httpx-sse==0.4.0
-huggingface-hub>=0.16.4
-humanfriendly==10.0
-idna==3.10
-injector==0.22.0
-Jinja2==3.1.5
-langchain==0.3.19
-langchain-core==0.3.35
-langchain-text-splitters==0.3.6
-markdown2==2.5.3
-nh3==0.3.7
-openai==2.8.1
-openpyxl==3.1.5
-pandas==2.3.1
-pgvector==0.3.6
-pillow==11.0.0
-psutil==7.0.0
-psycogreen==1.0.2
-psycopg2-binary==2.9.10
-PyMySQL==1.1.1
-PyJWT==2.10.1
-PyMuPDF==1.25.0
-python-dotenv==1.0.1
-pytest==8.3.4
-pytest-cov==5.0.0
-pytest-mock==3.14.0
-python-dateutil==2.9.0.post0
-python-docx==1.1.2
-pytz==2025.2
-PyYAML==6.0.2
-redis==5.2.1
-regex==2024.11.6
-requests==2.32.3
-requests-oauthlib==2.0.0
-requests-toolbelt==1.0.0
-s3transfer==0.17.0
-sib-api-v3-sdk==7.6.0
-SQLAlchemy==2.0.49
-sqlalchemy-redshift==1.0.0
-redshift_connector==2.1.14
-tiktoken==0.8.0
-tokenizers>=0.22.0
-websocket-client==1.8.0
-websockets==14.1
-Werkzeug==3.1.3
-pyjwt[crypto]>=2.8.0