src/researchlink/services/bibtex_parser.py

"""BibTeX parsing utilities."""

from __future__ import annotations

try:
    import bibtexparser
    BIBTEXPARSER_AVAILABLE = True
except ImportError:
    BIBTEXPARSER_AVAILABLE = False


PLACEHOLDER_PATTERNS = ["xx", "xxxx", "art. no.", "tbd", "forthcoming", "in press"]


def parse_bibtex(bibtex_str: str) -> list[dict]:
    """Parse BibTeX string into a list of entry dicts."""
    if not BIBTEXPARSER_AVAILABLE:
        return []
    try:
        bib_db = bibtexparser.loads(bibtex_str)
        return bib_db.entries
    except Exception:
        return []


def has_placeholder(text: str) -> tuple[bool, list[str]]:
    """Detect placeholder values in a citation field. Returns (found, patterns)."""
    found = []
    lower = text.lower()
    for pattern in PLACEHOLDER_PATTERNS:
        if pattern in lower:
            found.append(pattern)
    return bool(found), found


def entry_to_bibtex(entry: dict) -> str:
    """Convert a bibtexparser entry dict back to a BibTeX string."""
    etype = entry.get("ENTRYTYPE", "misc")
    key = entry.get("ID", "unknown")
    fields = {k: v for k, v in entry.items() if k not in ("ENTRYTYPE", "ID")}
    lines = [f"@{etype}{{{key},"]
    for k, v in fields.items():
        lines.append(f"  {k} = {{{v}}},")
    lines.append("}")
    return "\n".join(lines)


def generate_bibtex_from_metadata(
    key: str,
    title: str | None,
    authors: list[str],
    year: str | None,
    venue: str | None,
    url: str | None,
    doi: str | None,
) -> str:
    """Generate a minimal BibTeX entry from known metadata."""
    author_str = " and ".join(authors) if authors else "Unknown Author"
    lines = [f"@article{{{key},"]
    if title:
        lines.append(f"  title = {{{title}}},")
    lines.append(f"  author = {{{author_str}}},")
    if year:
        lines.append(f"  year = {{{year}}},")
    if venue:
        lines.append(f"  journal = {{{venue}}},")
    if doi:
        lines.append(f"  doi = {{{doi}}},")
    if url:
        lines.append(f"  url = {{{url}}},")
    lines.append("  note = {Citation status: needs-verification},")
    lines.append("}")
    return "\n".join(lines)