propose changes to add input directory checking for sample audio plugin

src/audio-transcription/audio_transcription/main.py

@@ -1,12 +1,14 @@
 """audio transcribe plugin"""
 
 import logging
-from typing import TypedDict
+from typing import List, TypedDict
 
+from pydantic import DirectoryPath
 import typer
 from rb.api.models import (
     BatchTextResponse,
     DirectoryInput,
+    FileFilterDirectory,
     InputSchema,
     InputType,
     ResponseBody,
@@ -19,7 +21,7 @@
 logger = logging.getLogger(__name__)
 
 logging.basicConfig(
-    level=logging.INFO,
+    level=logging.DEBUG,
     format="%(asctime)s [%(levelname)s] %(name)s: %(message)s",
 )
 
@@ -35,9 +37,17 @@
 
 model = AudioTranscriptionModel()
 
+AUDIO_EXTENSIONS = {".mp3", ".wav", ".flac", ".aac"}
+
+
+class AudioDirectory(FileFilterDirectory):
+
+    path: DirectoryPath
+    file_extensions: List[str] = AUDIO_EXTENSIONS
+
 
 class AudioInput(TypedDict):
-    input_dir: DirectoryInput
+    input_dir: AudioDirectory
 
 
 def task_schema() -> TaskSchema:

src/audio-transcription/tests/test_main.py

@@ -1,3 +1,4 @@
+import json
 from pathlib import Path
 
 from rb.api.models import ResponseBody
@@ -55,3 +56,20 @@ def test_api_transcribe_command(self):
         assert response.status_code == 200
         body = ResponseBody(**response.json())
         assert body.root.texts and "Twinkle" in body.root.texts[0].value
+
+    def test_negative_api_transcribe_command(self):
+        """pass in valid directory but no audio files , expect 422 validation error"""
+        transcribe_api = f"/{APP_NAME}/transcribe"
+        full_path = Path.cwd() / "src" / "audio-transcription" / "tests" / "negative"
+        input_json = {
+            "inputs": {
+                "input_dir": {
+                    "path": str(full_path),
+                }
+            }
+        }
+        response = self.client.post(transcribe_api, json=input_json)
+        assert response.status_code == 422
+        resp = json.loads(json.dumps(response.json()))
+        print(f"Response: {resp["detail"]["error"]}")
+        assert resp and "No file extensions matching" in resp["detail"]["error"]

src/rb-api/rb/api/main.py

@@ -1,7 +1,8 @@
 import multiprocessing
 import os
 import sys
-from fastapi import FastAPI
+from fastapi import FastAPI, HTTPException, Request, status
+from fastapi.exceptions import RequestValidationError
 from fastapi.staticfiles import StaticFiles
 from rb.api import routes
 
@@ -21,6 +22,20 @@
     name="static",
 )
 
+
+@app.exception_handler(RequestValidationError)
+async def validation_exception_handler(request: Request, exc: RequestValidationError):  # fmt: skip
+    """response handler for all plugin input validation errors"""
+    error_msg = str(exc)
+    for e in exc.errors():
+        error_msg = e.get("msg")
+
+    raise HTTPException(  # pylint: disable=raise-missing-from
+        status_code=status.HTTP_422_UNPROCESSABLE_ENTITY,
+        detail={"error": f"{error_msg}"},
+    )
+
+
 app.include_router(routes.probes_router, prefix="/probes")
 app.include_router(routes.cli_to_api_router)
 app.include_router(routes.ui_router)

src/rb-api/rb/api/models.py

@@ -5,9 +5,19 @@
 from __future__ import annotations
 
 from enum import Enum
+from pathlib import Path
 from typing import Annotated, Any, Dict, List, Literal, Optional, Union
 
-from pydantic import BaseModel, ConfigDict, DirectoryPath, Field, FilePath, RootModel
+from pydantic import (
+    BaseModel,
+    ConfigDict,
+    DirectoryPath,
+    Field,
+    FilePath,
+    RootModel,
+    field_validator,
+    model_validator,
+)
 
 API_APPMETDATA = "app_metadata"
 API_ROUTES = "routes"
@@ -51,6 +61,42 @@ class DirectoryInput(BaseModel):
     path: DirectoryPath
 
 
+class FileFilterDirectory(DirectoryInput):
+    """Find files with file_extensions in path directory."""
+
+    model_config = ConfigDict(
+        populate_by_name=True,
+    )
+    path: str
+    file_extensions: List[str]
+
+    @field_validator("path")
+    @classmethod
+    def validate_directory(cls, v):
+        path = Path(v)
+
+        if not path.exists():
+            raise ValueError(f"validate directory: '{v}' does not exist.")
+        if not path.is_dir():
+            raise ValueError(f"validate directory: Path '{v}' is not a directory.")
+        return v
+
+    @model_validator(mode="after")
+    def file_filter(self) -> "FileFilterDirectory":
+        path_obj = Path(self.path)
+        files = list(path_obj.glob("*"))
+        if not files:
+            raise ValueError(f"validate directory: Directory {path_obj} is empty.")
+        number_of_matched_files = [
+            f.name for f in files if f.suffix.lower() in self.file_extensions
+        ]
+        if len(number_of_matched_files) < 1:
+            raise ValueError(
+                f"validate directory: No file extensions matching {self.file_extensions} found in directory: {path_obj}"
+            )
+        return self
+
+
 class TextInput(BaseModel):
     model_config = ConfigDict(
         populate_by_name=True,

src/rb-api/rb/api/routes/cli.py

@@ -5,7 +5,7 @@
 from typing import Callable, Generator, Optional
 
 import typer
-from fastapi import APIRouter, HTTPException, Response
+from fastapi import APIRouter, HTTPException, Response, status
 from fastapi.responses import StreamingResponse
 from makefun import with_signature
 from pydantic import BaseModel
@@ -68,10 +68,11 @@ def static_endpoint(callback: Callable, *args, **kwargs) -> ResponseBody:
             # this has an issue of nor sending back details to desktop ui the api caller ?
             raise ValueError(f"Invalid return type from Typer command: {type(result)}")
         except Exception as e:
-            logger.error("Error executing CLI command: %s", e)
+            # response handler for all plugin runtime errors
+            logger.error("Error: %s %s", e, stdout)
             raise HTTPException(  # pylint: disable=raise-missing-from
-                status_code=400,
-                detail={"error": f"Typer CLI aborted {e}", "stdout": stdout[-10:]},
+                status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
+                detail={"error": f"{e}"},
             )
 
 

src/rb-lib/rb/lib/utils.py

@@ -72,8 +72,8 @@ def ensure_ml_func_hinting_and_task_schemas_are_valid(
                     input_type_hint is FileInput
                 ), f"For key {key}, the input type is NewFileInputType, but the TypeDict hint is {input_type_hint}. Change to FileInput."
             case InputType.DIRECTORY:
-                assert (
-                    input_type_hint is DirectoryInput
+                assert issubclass(
+                    input_type_hint, DirectoryInput
                 ), f"For key {key}, the input type is InputType.DIRECTORY, but the TypeDict hint is {input_type_hint}. Change to DirectoryInput."
             case InputType.TEXT:
                 assert (