"""Push processed data to the Hugging Face Hub."""
from pathlib import Path

import yaml
from datasets import Dataset, DatasetDict


def get_frontmatter(md_content: str) -> dict:
    """Extract frontmatter from Markdown content."""
    if not md_content.startswith("---"):
        return {}

    end_index = md_content.find("---", 3)
    if end_index == -1:
        return {}

    frontmatter_str = md_content[3:end_index].strip()
    try:
        return yaml.safe_load(frontmatter_str)
    except yaml.YAMLError:
        return {}


def process_file(xml_file: Path) -> dict:
    md_file = xml_file.with_suffix(".md")
    txt_file = xml_file.with_suffix(".txt")

    with xml_file.open("r") as file:
        xml_content = file.read()
    with md_file.open("r") as file:
        md_content = file.read()
    with txt_file.open("r") as file:
        txt_content = file.read()

    # extract front
    frontmatter = get_frontmatter(md_content)

    title = frontmatter.get("title", "")
    if isinstance(title, list):
        if len(title) == 1:
            title = title[0]
        else:
            raise ValueError(f"Title is a list with more than one item: {title}")

    return {
        "id": xml_file.stem,
        "title": title,
        "author": frontmatter.get("author", "Nicolai Frederik Severin Grundtvig"),
        "date": str(frontmatter["date"]) if "date" in frontmatter else "",
        "publisher": frontmatter.get("publisher", "Faculty of Arts, Aarhus University"),
        "xml": xml_content,
        "md": md_content,
        "txt": txt_content,
    }


def main():
    source_data = Path(__file__).parent.parent / "source-data"
    xml_files = list(source_data.glob("*.xml"))

    processed_data = [process_file(xml_file) for xml_file in xml_files]

    for p in processed_data:
        assert isinstance(p, dict), f"Processed data item is not a dict: {p}"
        for key, value in p.items():
            assert isinstance(value, str), (
                f"Value for key '{key}' is not a string: {value}"
            )
    ds = Dataset.from_list(processed_data)
    ds_dict = DatasetDict({"train": ds})

    # push to hub
    ds_dict.push_to_hub("chcaa/grundtvigs-works")


if __name__ == "__main__":
    main()