fix for doc to image

haystack/components/converters/image/document_to_image.py

@@ -113,23 +113,30 @@ class DocumentToImageContent:
         :returns:
             Dictionary containing one key:
             - "image_contents": ImageContents created from the processed documents. These contain base64-encoded image
-                data and metadata. The order corresponds to order of input documents.
-        :raises ValueError:
-            If any document is missing the required metadata keys, has an invalid file path, or has an unsupported
-            MIME type. The error message will specify which document and what information is missing or incorrect.
+                data and metadata. The order corresponds to order of input documents. A document that is missing the
+                required metadata keys, has an invalid file path, or has an unsupported MIME type gets None in its
+                position and a logged warning with the reason.
         """
         if not documents:
             return {"image_contents": []}
 
-        images_source_info = _extract_image_sources_info(
-            documents=documents, file_path_meta_field=self.file_path_meta_field, root_path=self.root_path
-        )
-
         image_contents: list[ImageContent | None] = [None] * len(documents)
 
         pdf_page_infos: list[_PDFPageInfo] = []
 
-        for doc_idx, image_source_info in enumerate(images_source_info):
+        for doc_idx, document in enumerate(documents):
+            # Validate each document on its own so one invalid document leaves None in its slot
+            # instead of failing the whole batch
+            try:
+                image_source_info = _extract_image_sources_info(
+                    documents=[document], file_path_meta_field=self.file_path_meta_field, root_path=self.root_path
+                )[0]
+            except ValueError as error:
+                logger.warning(
+                    "Skipping document with ID {document_id}: {error}", document_id=document.id, error=str(error)
+                )
+                continue
+
             mime_type = image_source_info["mime_type"]
             path = image_source_info["path"]
             if mime_type == "application/pdf":
@@ -146,7 +153,7 @@ class DocumentToImageContent:
                     base64_image=base64_image,
                     mime_type=mime_type,
                     detail=self.detail,
-                    meta={"file_path": documents[doc_idx].meta[self.file_path_meta_field]},
+                    meta={"file_path": document.meta[self.file_path_meta_field]},
                 )
 
         # efficiently convert PDF pages to images: each PDF is opened and processed only once

releasenotes/notes/document-to-image-partial-failures-462c1534aa227c77.yaml

@@ -0,0 +1,8 @@
+---
+fixes:
+  - |
+    ``DocumentToImageContent`` no longer raises a ``ValueError`` for the whole batch when one document is missing
+    the file path or ``page_number`` metadata, has an invalid file path, or has an unsupported MIME type.
+    That document now gets ``None`` in ``image_contents`` and a warning with the reason is logged, while the other
+    documents are still converted. This lets ``LLMDocumentContentExtractor`` return such documents in
+    ``failed_documents`` instead of failing the entire run.

test/components/converters/image/test_document_to_image_content.py

@@ -53,32 +53,20 @@ class TestDocumentToImageContent:
         results = converter.run(documents=[])
         assert results == {"image_contents": []}
 
-    def test_run_with_missing_file_path_metadata(self) -> None:
+    @pytest.mark.parametrize(
+        "meta, reason",
+        [
+            ({}, "is missing the 'file_path' key"),
+            ({"file_path": "test/test_files/docx/sample_docx.docx"}, "has an unsupported MIME type"),
+            ({"file_path": "wrong_name.jpg"}, "has an invalid file path 'wrong_name.jpg'"),
+            ({"file_path": "test/test_files/pdf/sample_pdf_1.pdf"}, "is missing the 'page_number' key"),
+        ],
+    )
+    def test_run_with_invalid_document(self, meta: dict, reason: str, caplog: pytest.LogCaptureFixture) -> None:
         converter = DocumentToImageContent()
-        # Document without file_path in metadata
-        doc_no_path = Document(content="test", meta={})
-        # Document with file_path but file doesn't exist
-        doc_no_file = Document(content="test", meta={"file_path": "nonexistent.jpg"})
-        with pytest.raises(ValueError, match="is missing the 'file_path' key"):
-            _ = converter.run(documents=[doc_no_path, doc_no_file])
-
-    def test_run_with_non_image_documents(self) -> None:
-        converter = DocumentToImageContent()
-        docx_doc = Document(content="test", meta={"file_path": "test/test_files/docx/sample_docx.docx"})
-        with pytest.raises(ValueError, match="has an unsupported MIME type"):
-            _ = converter.run(documents=[docx_doc])
-
-    def test_run_with_invalid_file_path(self, caplog: pytest.LogCaptureFixture) -> None:
-        converter = DocumentToImageContent()
-        pdf_doc = Document(content="test", meta={"file_path": "wrong_name.jpg"})
-        with pytest.raises(ValueError, match="has an invalid file path 'wrong_name.jpg'"):
-            _ = converter.run(documents=[pdf_doc])
-
-    def test_run_with_pdf_missing_page_number(self, caplog: pytest.LogCaptureFixture) -> None:
-        converter = DocumentToImageContent()
-        pdf_doc = Document(content="test", meta={"file_path": "test/test_files/pdf/sample_pdf_1.pdf"})
-        with pytest.raises(ValueError, match="is missing the 'page_number' key"):
-            _ = converter.run(documents=[pdf_doc])
+        results = converter.run(documents=[Document(content="test", meta=meta)])
+        assert results == {"image_contents": [None]}
+        assert reason in caplog.text
 
     def test_run_with_image_documents(self) -> None:
         converter = DocumentToImageContent(root_path="test/test_files/images")
@@ -105,8 +93,10 @@ class TestDocumentToImageContent:
             Document(content="", meta={"file_path": "pdf/sample_pdf_1.pdf", "page_number": 1}),
             Document(content="text", meta={"file_path": "docx/sample_docx.docx"}),
         ]
-        with pytest.raises(ValueError, match="has an unsupported MIME type"):
-            _ = converter.run(documents=documents)
+        image_contents = converter.run(documents=documents)["image_contents"]
+        assert image_contents[0] is not None
+        assert image_contents[1] is not None
+        assert image_contents[2] is None
 
     @patch("haystack.components.converters.image.document_to_image._extract_image_sources_info")
     @patch("haystack.components.converters.image.document_to_image._batch_convert_pdf_pages_to_images")
@@ -122,9 +112,9 @@ class TestDocumentToImageContent:
     ):
         converter = DocumentToImageContent()
 
-        mocked_extract_image_sources_info.return_value = [
-            {"path": "doc1.pdf", "mime_type": "application/pdf", "page_number": 999},  # Page 999 doesn't exist
-            {"path": "image1.jpg", "mime_type": "image/jpeg"},
+        mocked_extract_image_sources_info.side_effect = [
+            [{"path": "doc1.pdf", "mime_type": "application/pdf", "page_number": 999}],  # Page 999 doesn't exist
+            [{"path": "image1.jpg", "mime_type": "image/jpeg"}],
         ]
         mocked_batch_convert_pdf_pages_to_images.return_value = {}  # Empty dict because page was skipped
         mocked_pil_open.return_value = Image.new("RGB", (100, 100))