TypeError: 'property' object is not iterable

[HGamalElDin]

Describe the bug
I'm trying to create a Pipeline from 4 Custom Nodes. however when I initialize a pipeline and use add_node function, it thows the below exception.

Error message
TypeError: 'property' object is not iterable

Code Snippet

## FILE TYPE CLASSIFIER
class Type_Classifier:
    """
    Supported Types: [pdf, txt, docx, xlsx, csv, pptx]
    """    
    def __init__(self):
        self.classes = {
            "pdf": ['application/pdf'],
            "docx": ['application/vnd.openxmlformats-officedocument.wordprocessingml.document'],
            "txt": ['text/plain'],
            "pptx": ['application/vnd.openxmlformats-officedocument.presentationml.presentation'],
            "csv": ['text/csv', 'application/csv'],
            "xlsx": ['application/vnd.openxmlformats-officedocument.spreadsheetml.sheet']
            }
    
    def classify(self, file_url: str):
        filename = file_url.split("/")[-1] 
        response = None
        try:
            if 'https' in file_url:
              file_path, _ = urllib.request.urlretrieve(quote(file_url, safe=':/'))
            else:
              file_path = file_url
            file_type = magic.Magic(mime=True).from_file(file_path)
            print(file_type)
            file_classes = list(self.classes.values())
            if file_type in list(itertools.chain(*file_classes)):
                response = [list(self.classes.keys())[indx] for indx in range(len(file_classes)) if file_type in file_classes[indx]][0]
            else:
                response = "[Error] The file is invalid, corrupted or not supported. please make sure you pass a valid and supported file"
            return {"filename": filename, "response": response}
        except FileNotFoundError:
            return {"filename": filename, "response": '[Error] File not found in the path/url specified'}

    def run(self, file_url: str):
        
        return  self.classify(file_url), "output_1"

    def run_batch(self, urls: List[str]):
        print(urls)
        response = {}
        for file_url in tqdm(urls):
            response[file_url] = self.classify(file_url)
        return response, "output_1"

pipeline = Pipeline()
pipeline.add_node(component=Type_Classifier, name="classifier", inputs=["Query"])

System:

• OS: Windows 10
• GPU/CPU: CPU
• Haystack version (commit or version number): 1.18.1

[HGamalElDin]

@julian-risch appreciate your support

[anakin87]

Hello, @HGamalElDin...

Please explain what you are trying to achieve so that we can try to help you better...

• In any case, if you want to create a custom node, you should follow this guide. Remember to create a new class that inherits from BaseComponent.
• Did you check whether the existing FileClassifier works for you?