BibTeX Management Guide

A skill for maintaining clean, consistent, and complete BibTeX bibliography files. Covers formatting standards, deduplication, common errors, and automated cleanup workflows essential for LaTeX-based academic writing.

BibTeX Entry Standards

Required Fields by Entry Type

% Article in a journal
@article{smith2024deep,
  author    = {Smith, John A. and Doe, Jane B.},
  title     = {Deep Learning for Climate Prediction: A Comparative Study},
  journal   = {Nature Machine Intelligence},
  year      = {2024},
  volume    = {6},
  number    = {3},
  pages     = {234--248},
  doi       = {10.1038/s42256-024-00001-1}
}

% Conference proceedings
@inproceedings{lee2024attention,
  author    = {Lee, Wei and Chen, Li},
  title     = {Attention Mechanisms for Scientific Document Understanding},
  booktitle = {Proceedings of the 62nd Annual Meeting of the ACL},
  year      = {2024},
  pages     = {1123--1135},
  publisher = {Association for Computational Linguistics},
  doi       = {10.18653/v1/2024.acl-main.89}
}

% Book
@book{bishop2006pattern,
  author    = {Bishop, Christopher M.},
  title     = {Pattern Recognition and Machine Learning},
  publisher = {Springer},
  year      = {2006},
  isbn      = {978-0387310732}
}

Automated BibTeX Cleanup

Deduplication

import re
from collections import defaultdict

def parse_bibtex_entries(bib_content: str) -> list[dict]:
    """
    Parse a BibTeX file into structured entries.
    """
    entries = []
    pattern = r'@(\w+)\{([^,]+),\s*(.*?)\n\}'
    matches = re.finditer(pattern, bib_content, re.DOTALL)

    for match in matches:
        entry = {
            'type': match.group(1).lower(),
            'key': match.group(2).strip(),
            'raw': match.group(0),
            'fields': {}
        }

        fields_str = match.group(3)
        field_pattern = r'(\w+)\s*=\s*[{\"](.+?)[}\"]'
        for field_match in re.finditer(field_pattern, fields_str, re.DOTALL):
            entry['fields'][field_match.group(1).lower()] = field_match.group(2).strip()

        entries.append(entry)

    return entries


def deduplicate_bibtex(entries: list[dict]) -> dict:
    """
    Find and remove duplicate BibTeX entries.

    Deduplication strategy:
    1. Exact DOI match
    2. Fuzzy title match (normalized)
    3. Author + year + first title word match
    """
    seen_dois = {}
    seen_titles = {}
    duplicates = []
    unique = []

    for entry in entries:
        doi = entry['fields'].get('doi', '').lower().strip()
        title = entry['fields'].get('title', '').lower().strip()
        title_normalized = re.sub(r'[^a-z0-9\s]', '', title)

        is_duplicate = False

        # Check DOI match
        if doi and doi in seen_dois:
            duplicates.append({
                'entry': entry['key'],
                'duplicate_of': seen_dois[doi],
                'reason': 'same DOI'
            })
            is_duplicate = True
        elif doi:
            seen_dois[doi] = entry['key']

        # Check title match
        if not is_duplicate and title_normalized:
            if title_normalized in seen_titles:
                duplicates.append({
                    'entry': entry['key'],
                    'duplicate_of': seen_titles[title_normalized],
                    'reason': 'same title'
                })
                is_duplicate = True
            else:
                seen_titles[title_normalized] = entry['key']

        if not is_duplicate:
            unique.append(entry)

    return {
        'unique_entries': len(unique),
        'duplicates_found': len(duplicates),
        'duplicates': duplicates,
        'entries': unique
    }