unstructured-platform-plugins 0.0.42__tar.gz → 0.0.45__tar.gz

{unstructured_platform_plugins-0.0.42 → unstructured_platform_plugins-0.0.45}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: unstructured_platform_plugins
-Version: 0.0.42
+Version: 0.0.45
 Summary: Wrapper to convert arbitrary code into a uvicorn/fastapi implementation for Unstructured Platform
 License-File: LICENSE.md
 Classifier: Development Status :: 4 - Beta
@@ -29,6 +29,10 @@ Description-Content-Type: text/markdown
 # Unstructured Platform Plugins
  ![CI](https://github.com/Unstructured-IO/unstructured-platform-plugins/actions/workflows/ci.yml/badge.svg?branch=main)
 
+> **Deprecation notice:** `wrap_in_fastapi` (and the `etl-uvicorn` CLI it backs)
+> is deprecated. New plugins should build a FastAPI app directly with explicit
+> handlers for the plugin contract routes (`/invoke`, `/schema`, `/id`, etc.).
+
 Information about how to build custom plugins to integrate with Unstructured Platform.
 
 ## Plugin Development

{unstructured_platform_plugins-0.0.42 → unstructured_platform_plugins-0.0.45}/README.md

@@ -1,6 +1,10 @@
 # Unstructured Platform Plugins
  ![CI](https://github.com/Unstructured-IO/unstructured-platform-plugins/actions/workflows/ci.yml/badge.svg?branch=main)
 
+> **Deprecation notice:** `wrap_in_fastapi` (and the `etl-uvicorn` CLI it backs)
+> is deprecated. New plugins should build a FastAPI app directly with explicit
+> handlers for the plugin contract routes (`/invoke`, `/schema`, `/id`, etc.).
+
 Information about how to build custom plugins to integrate with Unstructured Platform.
 
 ## Plugin Development

unstructured_platform_plugins-0.0.45/unstructured_platform_plugins/__version__.py

@@ -0,0 +1 @@
+__version__ = "0.0.45"  # pragma: no cover

{unstructured_platform_plugins-0.0.42 → unstructured_platform_plugins-0.0.45}/unstructured_platform_plugins/etl_uvicorn/api_generator.py

@@ -11,6 +11,7 @@ from fastapi.responses import StreamingResponse
 from opentelemetry.instrumentation.fastapi import FastAPIInstrumentor
 from pydantic import BaseModel, Field, create_model
 from starlette.responses import RedirectResponse
+from typing_extensions import deprecated
 from unstructured_ingest.data_types.file_data import BatchFileData, FileData, file_data_from_dict
 from unstructured_ingest.error import UnstructuredIngestError
 from uvicorn.config import LOG_LEVELS
@@ -108,6 +109,10 @@ def update_filedata_model(new_type) -> BaseModel:
     return new_filedata_model
 
 
+@deprecated(
+    "wrap_in_fastapi is deprecated; build a FastAPI app directly with "
+    "explicit handlers for the plugin contract routes."
+)
 def wrap_in_fastapi(
     func: Callable,
     plugin_id: str,
@@ -249,22 +254,48 @@ def _wrap_in_fastapi(
                 file_data=request_dict.get("file_data", None),
             )
 
-    if input_schema_model.model_fields:
+    async def run_job_with_body(request: BaseModel) -> ResponseType:
+        log_func_and_body(func=func, body=request.json())
+        # Create dictionary from pydantic model while preserving underlying types
+        request_dict = {f: getattr(request, f) for f in request.model_fields}
+        # Make sure nested classes get instantiated correctly. `file_data` can legitimately be None
+        # -- a plugin may declare it optional, and then an absent or partial body leaves it unset --
+        # so convert only a real value. Calling `.model_dump()` on None would raise before `wrap_fn`
+        # runs, turning the optional-body contract into a 500.
+        file_data = request_dict.get("file_data")
+        if file_data is not None:
+            request_dict["file_data"] = file_data_from_dict(file_data.model_dump())
+        map_inputs(func=func, raw_inputs=request_dict)
+        if logger.level == LOG_LEVELS.get("trace", logging.NOTSET):
+            logger.log(level=logger.level, msg=f"passing inputs to function: {request_dict}")
+        return await wrap_fn(func=func, kwargs=request_dict)
+
+    # A pydantic body parameter with no default is mandatory even when every field inside the model
+    # is optional. So a plugin whose parameters are ALL optional would demand a body that no caller
+    # has a reason to populate -- and before it grew those parameters that same plugin accepted no
+    # body at all, so adding one flips the HTTP contract while looking backward-compatible. Default
+    # the body in that case: an absent body resolves each field to its own default, which is exactly
+    # what the function signature already promises.
+    #
+    # A plugin with at least one required field keeps a mandatory body, so an indexer that needs
+    # `file_data` still fails validation rather than silently receiving None.
+    body_is_optional = input_schema_model.model_fields and not any(
+        field.is_required() for field in input_schema_model.model_fields.values()
+    )
+
+    if body_is_optional:
+
+        @fastapi_app.post("/invoke", response_model=InvokeResponse)
+        async def run_job(request: Optional[input_schema_model] = None) -> ResponseType:
+            return await run_job_with_body(
+                request if request is not None else input_schema_model()
+            )
+
+    elif input_schema_model.model_fields:
 
         @fastapi_app.post("/invoke", response_model=InvokeResponse)
         async def run_job(request: input_schema_model) -> ResponseType:
-            log_func_and_body(func=func, body=request.json())
-            # Create dictionary from pydantic model while preserving underlying types
-            request_dict = {f: getattr(request, f) for f in request.model_fields}
-            # Make sure nested classes get instantiated correctly
-            if "file_data" in request_dict:
-                request_dict["file_data"] = file_data_from_dict(
-                    request_dict["file_data"].model_dump()
-                )
-            map_inputs(func=func, raw_inputs=request_dict)
-            if logger.level == LOG_LEVELS.get("trace", logging.NOTSET):
-                logger.log(level=logger.level, msg=f"passing inputs to function: {request_dict}")
-            return await wrap_fn(func=func, kwargs=request_dict)
+            return await run_job_with_body(request)
 
     else:
 

{unstructured_platform_plugins-0.0.42 → unstructured_platform_plugins-0.0.45}/unstructured_platform_plugins/etl_uvicorn/main.py

@@ -1,13 +1,34 @@
+import asyncio
+import signal
+import threading
 from dataclasses import dataclass, field
 from typing import IO, Any, Optional
 
 import click
+import uvicorn
 from uvicorn.config import LOGGING_CONFIG, Config, RawConfigParser
 from uvicorn.main import main, run
 
 from unstructured_platform_plugins.etl_uvicorn.api_generator import generate_fast_api
 
 
+def _install_signal_handlers_ignoring_sigterm(self: uvicorn.Server) -> None:
+    # uvicorn's default load-sheds 504 on SIGTERM, which races with controllers
+    # trying to drain in-flight work during pod shutdown. Plugin webservers are
+    # expected to outlive their controller container so SIGKILL — not SIGTERM —
+    # ends the process. SIGINT is preserved so local Ctrl-C still works.
+    if threading.current_thread() is not threading.main_thread():
+        return
+    try:
+        loop = asyncio.get_event_loop()
+        loop.add_signal_handler(signal.SIGINT, self.handle_exit, signal.SIGINT, None)
+    except NotImplementedError:
+        signal.signal(signal.SIGINT, self.handle_exit)
+
+
+uvicorn.Server.install_signal_handlers = _install_signal_handlers_ignoring_sigterm
+
+
 @dataclass
 class CustomConfig:
     log_config: dict[str, Any] | str | RawConfigParser | IO[Any] | None = field(

unstructured_platform_plugins-0.0.42/unstructured_platform_plugins/__version__.py

@@ -1 +0,0 @@
-__version__ = "0.0.42"  # pragma: no cover