From 8e57c85bf4019937ca867aaf39221e1ed3a953be Mon Sep 17 00:00:00 2001
From: Panos Vagenas <35837085+vagenas@users.noreply.github.com>
Date: Mon, 2 Dec 2024 13:32:05 +0100
Subject: [PATCH] rename new status, populate ConversionResult errors

Signed-off-by: Panos Vagenas <35837085+vagenas@users.noreply.github.com>
---
 docling/datamodel/base_models.py |  3 ++-
 docling/document_converter.py    | 18 +++++++++++++++---
 tests/test_invalid_input.py      |  2 +-
 3 files changed, 18 insertions(+), 5 deletions(-)

diff --git a/docling/datamodel/base_models.py b/docling/datamodel/base_models.py
index 6cf5aa5a..86dc6ee5 100644
--- a/docling/datamodel/base_models.py
+++ b/docling/datamodel/base_models.py
@@ -22,7 +22,7 @@ class ConversionStatus(str, Enum):
     FAILURE = auto()
     SUCCESS = auto()
     PARTIAL_SUCCESS = auto()
-    UNSUPPORTED = auto()
+    SKIPPED = auto()
 
 
 class InputFormat(str, Enum):
@@ -94,6 +94,7 @@ class DoclingComponentType(str, Enum):
     DOCUMENT_BACKEND = auto()
     MODEL = auto()
     DOC_ASSEMBLER = auto()
+    USER_INPUT = auto()
 
 
 class ErrorItem(BaseModel):
diff --git a/docling/document_converter.py b/docling/document_converter.py
index 054e03f2..503a4c5b 100644
--- a/docling/document_converter.py
+++ b/docling/document_converter.py
@@ -15,7 +15,13 @@ from docling.backend.md_backend import MarkdownDocumentBackend
 from docling.backend.msexcel_backend import MsExcelDocumentBackend
 from docling.backend.mspowerpoint_backend import MsPowerpointDocumentBackend
 from docling.backend.msword_backend import MsWordDocumentBackend
-from docling.datamodel.base_models import ConversionStatus, DocumentStream, InputFormat
+from docling.datamodel.base_models import (
+    ConversionStatus,
+    DoclingComponentType,
+    DocumentStream,
+    ErrorItem,
+    InputFormat,
+)
 from docling.datamodel.document import (
     ConversionResult,
     InputDocument,
@@ -262,11 +268,17 @@ class DocumentConverter:
         if valid:
             conv_res = self._execute_pipeline(in_doc, raises_on_error=raises_on_error)
         else:
+            error_message = f"File format not allowed: {in_doc.file}"
             if raises_on_error:
-                raise ConversionError(f"Unsupported format in: {in_doc.file}")
+                raise ConversionError(error_message)
             else:
+                error_item = ErrorItem(
+                    component_type=DoclingComponentType.USER_INPUT,
+                    module_name="",
+                    error_message=error_message,
+                )
                 conv_res = ConversionResult(
-                    input=in_doc, status=ConversionStatus.UNSUPPORTED
+                    input=in_doc, status=ConversionStatus.SKIPPED, errors=[error_item]
                 )
 
         return conv_res
diff --git a/tests/test_invalid_input.py b/tests/test_invalid_input.py
index 6e6fc53f..f40d79e4 100644
--- a/tests/test_invalid_input.py
+++ b/tests/test_invalid_input.py
@@ -24,7 +24,7 @@ def test_convert_unsupported_doc_format_wout_exception(converter: DocumentConver
     result = converter.convert(
         DocumentStream(name="input.xyz", stream=BytesIO(b"xyz")), raises_on_error=False
     )
-    assert result.status == ConversionStatus.UNSUPPORTED
+    assert result.status == ConversionStatus.SKIPPED
 
 
 def test_convert_unsupported_doc_format_with_exception(converter: DocumentConverter):