PDFMinerToDocument buggy output

[lukebfox]

Describe the bug
Connecting a pdfminer coverter to a document writer i.e `pipeline.connect('converter.document', 'writer.documents') does not work.

The PDFMinerToDocument output_type does not match the returned json key:

haystack/haystack/components/converters/pdfminer.py

Line 119 in d0da31a

@component.output_types(document=List[Document])

haystack/haystack/components/converters/pdfminer.py

Line 164 in d0da31a

return {"documents": documents}

document != documents

To Reproduce

pipeline = Pipeline()
document_store = InMemoryDocumentStore
pipeline.add_component("converter", PDFMinerToDocument())
pipeline.add_component("writer", DocumentWriter(document_store=document_store))
pipeline.connect("converter.document", "writer.documents")
response = pipeline.run({"converter": {"sources": ['some.pdf']}})
print(response['writer']['documents_written'])

Error message
KeyError: 'writer'

Expected behavior
The writer node is able to use documents from the previous converter node.

FAQ Check

• Have you had a look at our new FAQ page?

System:

• OS: NixOS
• Haystack version (commit): d0da31a
• DocumentStore: InMemoryDocumentStore

[anakin87]

Oh, you are right!

Would you like to open a PR to fix this?