from datasets import Dataset
import json

# Load or define your metadata
sample_data = [
    {
        "filename": "CIA-RDP79-00999A000200010002-3.pdf",
        "document_id": "CIA-RDP79-00999A000200010002-3",
        "page_count": 10,
        "image_count": 10,
        "processed_at": "2025-03-22T22:04:19.172208",
        "ocr_status": "pending",
        "text_extracted": False,
        "source": "CIA Stargate Archive",
        "tags": [],
        "pdf": None  # Add actual file path or binary later
    },
    {
        "filename": "CIA-RDP96-00788R000700020002-2.pdf",
        "document_id": "CIA-RDP96-00788R000700020002-2",
        "page_count": 14,
        "image_count": 14,
        "processed_at": "2025-03-22T22:10:03.984102",
        "ocr_status": "pending",
        "text_extracted": False,
        "source": "CIA Stargate Archive",
        "tags": ["remote_viewing", "tasking_response"],
        "pdf": None
    }
]

# Create the dataset
dataset = Dataset.from_list(sample_data)

# Save it locally as a Hugging Face Arrow dataset
dataset.save_to_disk("stargate_raw_pdfs")

# (Optional) Preview
print(dataset)