ai-pipeline-core 0.1.1__py3-none-any.whl → 0.1.2__py3-none-any.whl

ai_pipeline_core/__init__.py

@@ -16,7 +16,7 @@ from .prompt_manager import PromptManager
 from .settings import settings
 from .tracing import trace
 
-__version__ = "0.1.1"
+__version__ = "0.1.2"
 
 __all__ = [
     "Document",

ai_pipeline_core/documents/document.py

@@ -10,6 +10,7 @@ from typing import Any, ClassVar, Literal, Self
 from pydantic import BaseModel, ConfigDict, field_serializer, field_validator
 from ruamel.yaml import YAML
 
+from ai_pipeline_core.documents.utils import canonical_name_key
 from ai_pipeline_core.exceptions import DocumentNameError, DocumentSizeError
 
 from .mime_type import (
@@ -26,6 +27,12 @@ class Document(BaseModel, ABC):
     MAX_CONTENT_SIZE: ClassVar[int] = 10 * 1024 * 1024  # 10MB default
     DESCRIPTION_EXTENSION: ClassVar[str] = ".description.md"
 
+    def __init__(self, **data: Any) -> None:
+        """Prevent direct instantiation of abstract Document class."""
+        if type(self) is Document:
+            raise TypeError("Cannot instantiate abstract Document class directly")
+        super().__init__(**data)
+
     # Optional enum of allowed file names. Subclasses may set this.
     # This is used to validate the document name.
     FILES: ClassVar[type[StrEnum] | None] = None
@@ -179,10 +186,10 @@ class Document(BaseModel, ABC):
         """Check if document is an image"""
         return is_image_mime_type(self.mime_type)
 
-    @property
-    def should_be_cached(self) -> bool:
-        """Check if document should be cached"""
-        return False
+    @classmethod
+    def canonical_name(cls) -> str:
+        """Get the canonical name of the document"""
+        return canonical_name_key(cls)
 
     def as_text(self) -> str:
         """Parse document as text"""

ai_pipeline_core/documents/flow_document.py

@@ -1,6 +1,6 @@
 """Flow-specific document base class."""
 
-from typing import Literal, final
+from typing import Any, Literal, final
 
 from .document import Document
 
@@ -15,6 +15,12 @@ class FlowDocument(Document):
     Compared to TaskDocument, FlowDocument are persistent across Prefect flow runs.
     """
 
+    def __init__(self, **data: Any) -> None:
+        """Prevent direct instantiation of abstract FlowDocument class."""
+        if type(self) is FlowDocument:
+            raise TypeError("Cannot instantiate abstract FlowDocument class directly")
+        super().__init__(**data)
+
     @final
     def get_base_type(self) -> Literal["flow"]:
         """Get the document type."""

ai_pipeline_core/documents/task_document.py

@@ -1,6 +1,6 @@
 """Task-specific document base class."""
 
-from typing import Literal, final
+from typing import Any, Literal, final
 
 from .document import Document
 
@@ -16,6 +16,12 @@ class TaskDocument(Document):
     They are used for intermediate results that are not needed after the task completes.
     """
 
+    def __init__(self, **data: Any) -> None:
+        """Prevent direct instantiation of abstract TaskDocument class."""
+        if type(self) is TaskDocument:
+            raise TypeError("Cannot instantiate abstract TaskDocument class directly")
+        super().__init__(**data)
+
     @final
     def get_base_type(self) -> Literal["task"]:
         """Get the document type."""

ai_pipeline_core/documents/utils.py

@@ -1,4 +1,5 @@
 import re
+from typing import Any, Iterable, Type
 from urllib.parse import urlparse
 
 
@@ -31,3 +32,54 @@ def sanitize_url(url: str) -> str:
         sanitized = "unnamed"
 
     return sanitized
+
+
+def camel_to_snake(name: str) -> str:
+    """Convert CamelCase (incl. acronyms) to snake_case."""
+    s1 = re.sub(r"(.)([A-Z][a-z0-9]+)", r"\1_\2", name)
+    s2 = re.sub(r"([a-z0-9])([A-Z])", r"\1_\2", s1)
+    return s2.replace("__", "_").strip("_").lower()
+
+
+def canonical_name_key(
+    obj_or_name: Type[Any] | str,
+    *,
+    max_parent_suffixes: int = 3,
+    extra_suffixes: Iterable[str] = (),
+) -> str:
+    """
+    Produce a canonical snake_case key from a class or name by:
+      1) Starting with the class name (or given string),
+      2) Stripping any trailing parent class names (up to `max_parent_suffixes` from the MRO),
+      3) Stripping any `extra_suffixes`,
+      4) Converting to snake_case.
+
+    Examples (given typical MROs):
+      FinalReportDocument(WorkflowDocument -> Document) -> 'final_report'
+      FooWorkflowDocument(WorkflowDocument -> Document) -> 'foo'
+      BarFlow(Config -> Base -> Flow) -> 'bar'
+    """
+    name = obj_or_name.__name__ if isinstance(obj_or_name, type) else str(obj_or_name)
+
+    # From MRO, collect up to N parent names to consider as removable suffixes
+    suffixes: list[str] = []
+    if isinstance(obj_or_name, type):
+        for base in obj_or_name.mro()[1 : 1 + max_parent_suffixes]:
+            if base is object:
+                continue
+            suffixes.append(base.__name__)
+
+    # Add any custom suffixes the caller wants to strip (e.g., 'Config')
+    suffixes.extend(extra_suffixes)
+
+    # Iteratively trim the longest matching suffix first
+    trimmed = True
+    while trimmed and suffixes:
+        trimmed = False
+        for sfx in sorted(set(suffixes), key=len, reverse=True):
+            if sfx and name.endswith(sfx):
+                name = name[: -len(sfx)]
+                trimmed = True
+                break
+
+    return camel_to_snake(name)

{ai_pipeline_core-0.1.1.dist-info → ai_pipeline_core-0.1.2.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ai-pipeline-core
-Version: 0.1.1
+Version: 0.1.2
 Summary: Core utilities for AI-powered processing pipelines using prefect
 Project-URL: Homepage, https://github.com/bbarwik/ai-pipeline-core
 Project-URL: Repository, https://github.com/bbarwik/ai-pipeline-core
@@ -438,9 +438,15 @@ For learning purposes, see [CLAUDE.md](CLAUDE.md) for our comprehensive coding s
 ## Documentation
 
 - [CLAUDE.md](CLAUDE.md) - Detailed coding standards and architecture guide
-- [Prefect Integration](docs/prefect.md) - Prefect patterns and best practices
-- [Deployment Guide](docs/prefect_deployment.md) - Production deployment
-- [Prefect Logging](docs/prefect_logging.md) - Logging configuration guide
+
+### dependencies_docs/ Directory
+> [!NOTE]
+> The `dependencies_docs/` directory contains guides for AI assistants (like Claude Code) on how to interact with the project's external dependencies and tooling, NOT user documentation for ai-pipeline-core itself. These files are excluded from repository listings to avoid confusion.
+
+**AI Assistant Dependency Guides:**
+- [Prefect Integration](dependencies_docs/prefect.md) - Prefect patterns and best practices for AI assistants
+- [Deployment Guide](dependencies_docs/prefect_deployment.md) - Production deployment guide for AI assistants
+- [Prefect Logging](dependencies_docs/prefect_logging.md) - Logging configuration guide for AI assistants
 
 ## License
 
@@ -465,13 +471,9 @@ Built with:
 
 ## Stability Notice
 
-**Current Version**: 0.1.1
+**Current Version**: 0.1.2
 **Status**: Internal Preview
 **API Stability**: Unstable - Breaking changes expected
 **Recommended Use**: Learning and reference only
 
 For production use, please fork this repository and maintain your own stable version.
-
----
-
-**Remember**: The best code is no code. The second best is minimal, typed, async code that does exactly what's needed.

{ai_pipeline_core-0.1.1.dist-info → ai_pipeline_core-0.1.2.dist-info}/RECORD

@@ -1,16 +1,16 @@
-ai_pipeline_core/__init__.py,sha256=pZsQw7615IfTfZ-4owfhB9QhzRXd71fQoUH0hzVnKt0,779
+ai_pipeline_core/__init__.py,sha256=qJfrXgVvGcXERAsm0EB-5MgxlQjnrnirnRqi12O3B80,779
 ai_pipeline_core/exceptions.py,sha256=_vW0Hbw2LGb5tcVvH0YzTKMff7QOPfCRr3w-w_zPyCE,968
 ai_pipeline_core/prompt_manager.py,sha256=XmNUdMIC0WrE9fF0LIcfozAKOGrlYwj8AfXvCndIH-o,4693
 ai_pipeline_core/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 ai_pipeline_core/settings.py,sha256=Zl2BPa6IHzh-B5V7cg5mtySr1dhWZQYYKxXz3BwrHlQ,615
 ai_pipeline_core/tracing.py,sha256=UcQ_z1F8KrBLq5ZJaXyoIZsiMdqjHzBhADV07pYXY-w,7663
 ai_pipeline_core/documents/__init__.py,sha256=rEnKj-sSlZ9WnFlZAmSGVi1P8vnsHmU9O9_YwtP40ms,242
-ai_pipeline_core/documents/document.py,sha256=p2nlnvsjQhmbEzw5lkXXmb26GMpGZ-QICaNnVapQ5JA,8588
+ai_pipeline_core/documents/document.py,sha256=Gj4WR57VW67hhRN1360oqHfVQg6Xxj4jx_XueK4cvl0,8941
 ai_pipeline_core/documents/document_list.py,sha256=HOG_uZDazA9CJB7Lr_tNcDFzb5Ff9RUt0ELWQK_eYNM,4940
-ai_pipeline_core/documents/flow_document.py,sha256=WxdPvavSCWzyTo7r-R5xKy6RzVNjl56705MoiSE76P8,530
+ai_pipeline_core/documents/flow_document.py,sha256=qsV-2JYOMhkvAj7lW54ZNH_4QUclld9h06CoU59tWww,815
 ai_pipeline_core/documents/mime_type.py,sha256=tMWGH9PVmHe6a_IzdaJUqIHf4qnwQOwOCBhsgW2AyTE,2244
-ai_pipeline_core/documents/task_document.py,sha256=gxfXMeh85gwWd17bVTZ9xVWb_K91YbJ9N7w1zfHqZyw,622
-ai_pipeline_core/documents/utils.py,sha256=P46oaTG266YBiq-FnaD0iQ763ReKKPekhlykf9w-fWQ,905
+ai_pipeline_core/documents/task_document.py,sha256=WjHqtl1d60XFBBqewNRdz1OqBErGI0jRx15oQYCTHo8,907
+ai_pipeline_core/documents/utils.py,sha256=BdE4taSl1vrBhxnFbOP5nDA7lXIcvY__AMRTHoaNb5M,2764
 ai_pipeline_core/flow/__init__.py,sha256=_Sji2yY1ICkvVX6QiiGWKzqIXtg9UAiuvhjHSK_gdO8,57
 ai_pipeline_core/flow/config.py,sha256=crbe_OvNE6qulIKv1D8yKoe8xrEsIlvICyxjhqHHBxQ,2266
 ai_pipeline_core/llm/__init__.py,sha256=3XVK-bSJdOe0s6KmmO7PDbsXHfjlcZEG1MVBmaz3EeU,442
@@ -23,7 +23,7 @@ ai_pipeline_core/logging/__init__.py,sha256=DOO6ckgnMVXl29Sy7q6jhO-iW96h54pCHQDz
 ai_pipeline_core/logging/logging.yml,sha256=YTW48keO_K5bkkb-KXGM7ZuaYKiquLsjsURei8Ql0V4,1353
 ai_pipeline_core/logging/logging_config.py,sha256=6MBz9nnVNvqiLDoyy9-R3sWkn6927Re5hdz4hwTptpI,4903
 ai_pipeline_core/logging/logging_mixin.py,sha256=RDaR2ju2-vKTJRzXGa0DquGPT8_UxahWjvKJnaD0IV8,7810
-ai_pipeline_core-0.1.1.dist-info/METADATA,sha256=88CYHbskiMyW1FkjS0OZOloDaHRR04hWl7PpTge2H2s,15529
-ai_pipeline_core-0.1.1.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
-ai_pipeline_core-0.1.1.dist-info/licenses/LICENSE,sha256=kKj8mfbdWwkyG3U6n7ztB3bAZlEwShTkAsvaY657i3I,1074
-ai_pipeline_core-0.1.1.dist-info/RECORD,,
+ai_pipeline_core-0.1.2.dist-info/METADATA,sha256=LO-DGxVRhZPcwRwb2_zibFzp8aRE59STOYp2BxDag8M,15869
+ai_pipeline_core-0.1.2.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
+ai_pipeline_core-0.1.2.dist-info/licenses/LICENSE,sha256=kKj8mfbdWwkyG3U6n7ztB3bAZlEwShTkAsvaY657i3I,1074
+ai_pipeline_core-0.1.2.dist-info/RECORD,,