feat: support single file input (input_file) alongside directory mode

src/procleaning_mcp/server.py

@@ -53,43 +53,79 @@ async def _poll_task_result(api_url: str, task_id: str, api_key: str) -> dict:
 
 @mcp.tool()
 def batch_ocr_pdf(
-    input_dir: str,
-    output_dir: str,
+    input_file: str = None,
+    input_dir: str = None,
+    output_dir: str = None,
     api_key: str = None,
     base_url: str = "https://agent.imqimacau.com",
 ) -> str:
     """
-    Batch OCR process all PDF files in the input directory.
-    Submits each PDF to the OCR API and saves results as Markdown files.
+    OCR process PDF file(s) — either a single file or an entire directory.
+    Submits PDF to the OCR API and saves results as Markdown files.
 
     Args:
-        input_dir: Directory containing PDF files to process
-        output_dir: Directory to save OCR results (Markdown files)
+        input_file: Path to a single PDF file to process (mutually exclusive with input_dir)
+        input_dir: Directory containing PDF files to process (mutually exclusive with input_file)
+        output_dir: Directory to save OCR results (Markdown files). Auto-created if not exists.
         api_key: API key for the OCR service (optional, uses environment variable if not provided)
         base_url: Base URL of the OCR API (default: https://agent.imqimacau.com)
 
     Returns:
-        JSON string with batch processing results (success/fail counts and per-file status)
+        JSON string with processing results (success/fail counts and per-file status)
+
+    Note:
+        input_file and input_dir are mutually exclusive. If neither is provided, defaults to input_dir.
     """
     # Use environment variable if api_key not provided
     if not api_key:
         api_key = os.environ.get("OCR_API_KEY", "")
 
-    if not os.path.exists(input_dir):
+    # Validate mutual exclusivity
+    if input_file and input_dir:
         return json.dumps(
-            {"error": f"Input directory does not exist: {input_dir}"},
+            {"error": "input_file and input_dir are mutually exclusive. Please provide only one."},
             ensure_ascii=False,
             indent=2,
         )
 
-    os.makedirs(output_dir, exist_ok=True)
+    # If neither provided, default to input_dir behavior with an error message
+    if not input_file and not input_dir:
+        return json.dumps(
+            {"error": "Either input_file or input_dir must be provided."},
+            ensure_ascii=False,
+            indent=2,
+        )
 
-    # Find all PDF files
-    pdf_files = [
-        os.path.join(input_dir, f)
-        for f in os.listdir(input_dir)
-        if f.lower().endswith(".pdf")
-    ]
+    # Determine file list based on input mode
+    if input_file:
+        # Single file mode
+        if not os.path.isfile(input_file):
+            return json.dumps(
+                {"error": f"Input file does not exist: {input_file}"},
+                ensure_ascii=False,
+                indent=2,
+            )
+        # Auto-create output_dir based on input file location if not provided
+        if not output_dir:
+            output_dir = os.path.join(os.path.dirname(input_file) or ".", "ocr_output")
+        pdf_files = [input_file]
+    else:
+        # Directory mode
+        if not os.path.isdir(input_dir):
+            return json.dumps(
+                {"error": f"Input directory does not exist: {input_dir}"},
+                ensure_ascii=False,
+                indent=2,
+            )
+        if not output_dir:
+            output_dir = os.path.join(input_dir, "ocr_output")
+        pdf_files = [
+            os.path.join(input_dir, f)
+            for f in os.listdir(input_dir)
+            if f.lower().endswith(".pdf")
+        ]
+
+    os.makedirs(output_dir, exist_ok=True)
 
     if not pdf_files:
         return json.dumps(
@@ -208,8 +244,9 @@ def batch_ocr_pdf(
             fail_count += 1
             results.append({"file": filename, "status": "failed", "error": str(e)})
 
-        # Delay between files
-        time.sleep(2)
+        # Delay between files (only in batch mode)
+        if len(pdf_files) > 1:
+            time.sleep(2)
 
     final_result = {
         "summary": {