mdify-cli 2.6.0__py3-none-any.whl → 2.8.0__py3-none-any.whl

mdify/__init__.py

@@ -1,3 +1,3 @@
 """mdify - Convert documents to Markdown via Docling container."""
 
-__version__ = "2.6.0"
+__version__ = "2.8.0"

mdify/docling_client.py

@@ -4,6 +4,8 @@ from dataclasses import dataclass
 from pathlib import Path
 from typing import Optional
 
+import mimetypes
+
 import requests
 
 
@@ -40,6 +42,48 @@ class DoclingHTTPError(DoclingClientError):
         super().__init__(f"HTTP {status_code}: {message}")
 
 
+def _get_mime_type(file_path: Path) -> str:
+    """Get MIME type for file, with fallback for unknown types."""
+    mime_type, _ = mimetypes.guess_type(str(file_path))
+    return mime_type or "application/octet-stream"
+
+
+def _extract_content(result_data) -> str:
+    """Extract content from API response, supporting both old and new formats.
+
+    Supports:
+    - New format: {"document": {"md_content": "..."}}
+    - Fallback: {"document": {"content": "..."}}
+    - Old format: {"content": "..."}
+    - List format: [{"document": {...}} or {"content": "..."}]
+
+    Args:
+        result_data: Response data from docling-serve API
+
+    Returns:
+        Extracted content string, or empty string if not found
+    """
+    if isinstance(result_data, dict):
+        # New format with document field
+        if "document" in result_data:
+            doc = result_data["document"]
+            # Try md_content first, then content
+            return doc.get("md_content", "") or doc.get("content", "")
+        # Old format without document field
+        return result_data.get("content", "")
+    elif isinstance(result_data, list) and len(result_data) > 0:
+        # List format - process first item
+        first_result = result_data[0]
+        if isinstance(first_result, dict):
+            if "document" in first_result:
+                doc = first_result["document"]
+                # Try md_content first, then content
+                return doc.get("md_content", "") or doc.get("content", "")
+            # Old format without document field
+            return first_result.get("content", "")
+    return ""
+
+
 def check_health(base_url: str) -> bool:
     """Check if docling-serve is healthy.
 
@@ -77,7 +121,7 @@ def convert_file(
         with open(file_path, "rb") as f:
             response = requests.post(
                 f"{base_url}/v1/convert/file",
-                files={"files": (file_path.name, f, "application/pdf")},
+                files={"files": (file_path.name, f, _get_mime_type(file_path))},
                 data={"to_formats": to_format, "do_ocr": str(do_ocr).lower()},
             )
 
@@ -87,17 +131,10 @@ def convert_file(
             )
 
         result_data = response.json()
+        content = _extract_content(result_data)
 
-        # docling-serve returns results in a list format
-        if isinstance(result_data, list) and len(result_data) > 0:
-            first_result = result_data[0]
-            return ConvertResult(
-                content=first_result.get("content", ""), format=to_format, success=True
-            )
-        elif isinstance(result_data, dict):
-            return ConvertResult(
-                content=result_data.get("content", ""), format=to_format, success=True
-            )
+        if content or isinstance(result_data, (dict, list)):
+            return ConvertResult(content=content, format=to_format, success=True)
         else:
             raise DoclingHTTPError(200, f"Unexpected response format: {result_data}")
 
@@ -126,7 +163,7 @@ def convert_file_async(
         with open(file_path, "rb") as f:
             response = requests.post(
                 f"{base_url}/v1/convert/file/async",
-                files={"files": (file_path.name, f, "application/pdf")},
+                files={"files": (file_path.name, f, _get_mime_type(file_path))},
                 data={"to_formats": to_format, "do_ocr": str(do_ocr).lower()},
             )
 
@@ -202,19 +239,21 @@ def get_result(base_url: str, task_id: str) -> ConvertResult:
             )
 
         result_data = response.json()
+        content = _extract_content(result_data)
 
-        # Similar to sync conversion, handle list or dict format
-        if isinstance(result_data, list) and len(result_data) > 0:
+        # Determine format from response, defaulting to "md"
+        result_format = "md"
+        if isinstance(result_data, dict):
+            result_format = result_data.get("format", "md")
+        elif isinstance(result_data, list) and len(result_data) > 0:
             first_result = result_data[0]
+            if isinstance(first_result, dict):
+                result_format = first_result.get("format", "md")
+
+        if content or isinstance(result_data, (dict, list)):
             return ConvertResult(
-                content=first_result.get("content", ""),
-                format=first_result.get("format", "md"),
-                success=True,
-            )
-        elif isinstance(result_data, dict):
-            return ConvertResult(
-                content=result_data.get("content", ""),
-                format=result_data.get("format", "md"),
+                content=content,
+                format=result_format,
                 success=True,
             )
         else:

{mdify_cli-2.6.0.dist-info → mdify_cli-2.8.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: mdify-cli
-Version: 2.6.0
+Version: 2.8.0
 Summary: Convert PDFs and document images into structured Markdown for LLM workflows
 Author: tiroq
 License-Expression: MIT

mdify_cli-2.8.0.dist-info/RECORD

@@ -0,0 +1,12 @@
+assets/mdify.png,sha256=qUj7WXWqNwpI2KNXOW79XJwqFqa-UI0JEkmt1mmy4Rg,1820418
+mdify/__init__.py,sha256=YBAx8MdkINw38Jx6zeS5ikx8buI1avmzcHF2v23nZQU,90
+mdify/__main__.py,sha256=bhpJ00co6MfaVOdH4XLoW04NtLYDa_oJK7ODzfLrn9M,143
+mdify/cli.py,sha256=LqIibolYSKGCNYqxuIyFnvPkjJyNlXvfWeKaSaoOrqo,28542
+mdify/container.py,sha256=tkk0nv7EquL-rKUY4nkS_yGITb7mqw8B7eEfuqaeVrg,5239
+mdify/docling_client.py,sha256=xuQR6sC1v3EPloOSwExoHCqT4uUxE8myYq-Yeby3C2I,7975
+mdify_cli-2.8.0.dist-info/licenses/LICENSE,sha256=NWM66Uv-XuSMKaU-gaPmvfyk4WgE6zcIPr78wyg6GAo,1065
+mdify_cli-2.8.0.dist-info/METADATA,sha256=LKU3PAHABNp5dT9KJ3hGeCMSXxjDkIFXNveXzRv2fIA,7923
+mdify_cli-2.8.0.dist-info/WHEEL,sha256=wUyA8OaulRlbfwMtmQsvNngGrxQHAvkKcvRmdizlJi0,92
+mdify_cli-2.8.0.dist-info/entry_points.txt,sha256=0Xki8f5lADQUtwdt6Eq_FEaieI6Byhk8UE7BuDhChMg,41
+mdify_cli-2.8.0.dist-info/top_level.txt,sha256=qltzf7h8owHq7dxCdfCkSHY8gT21hn1_E8P-VWS_OKM,6
+mdify_cli-2.8.0.dist-info/RECORD,,

mdify_cli-2.6.0.dist-info/RECORD

@@ -1,12 +0,0 @@
-assets/mdify.png,sha256=qUj7WXWqNwpI2KNXOW79XJwqFqa-UI0JEkmt1mmy4Rg,1820418
-mdify/__init__.py,sha256=4mWutp3KF_BH9sz_oEPFBoN7Ee6vamK3cHDBpUtRQVY,90
-mdify/__main__.py,sha256=bhpJ00co6MfaVOdH4XLoW04NtLYDa_oJK7ODzfLrn9M,143
-mdify/cli.py,sha256=LqIibolYSKGCNYqxuIyFnvPkjJyNlXvfWeKaSaoOrqo,28542
-mdify/container.py,sha256=tkk0nv7EquL-rKUY4nkS_yGITb7mqw8B7eEfuqaeVrg,5239
-mdify/docling_client.py,sha256=_9qjL5yOOeJahOg6an2P6Iii1xkeR6wmNJZG4Q6NRkk,6553
-mdify_cli-2.6.0.dist-info/licenses/LICENSE,sha256=NWM66Uv-XuSMKaU-gaPmvfyk4WgE6zcIPr78wyg6GAo,1065
-mdify_cli-2.6.0.dist-info/METADATA,sha256=NcyfsGSLiSkz0NkRdc6g5pOervCpXJbWEIDSPnYSvFk,7923
-mdify_cli-2.6.0.dist-info/WHEEL,sha256=wUyA8OaulRlbfwMtmQsvNngGrxQHAvkKcvRmdizlJi0,92
-mdify_cli-2.6.0.dist-info/entry_points.txt,sha256=0Xki8f5lADQUtwdt6Eq_FEaieI6Byhk8UE7BuDhChMg,41
-mdify_cli-2.6.0.dist-info/top_level.txt,sha256=qltzf7h8owHq7dxCdfCkSHY8gT21hn1_E8P-VWS_OKM,6
-mdify_cli-2.6.0.dist-info/RECORD,,