zsgdp/normalize/normalize_olmocr.py

"""olmOCR Markdown normalization."""

from __future__ import annotations

from zsgdp.normalize.markdown import normalize_markdown_candidate
from zsgdp.schema import DocumentProfile, ParseCandidate


def normalize_olmocr_markdown(*, markdown: str, profile: DocumentProfile, source_path: str) -> ParseCandidate:
    return normalize_markdown_candidate(
        markdown=markdown,
        doc_id=profile.doc_id,
        source_path=source_path,
        file_type=profile.file_type,
        parser_name="olmocr",
        confidence=0.80,
        provenance={"backend": "olmocr"},
    )