Python BibDatabase.entries[0]['arxiv_doi']方法代码示例

# 需要导入模块: from bibtexparser.bibdatabase import BibDatabase [as 别名]
# 或者: from bibtexparser.bibdatabase.BibDatabase import entries[0]['arxiv_doi'] [as 别名]
def arxiv2bibtex(arxiv_id, try_doi=True, ui=None):
    """Return a bibtex string from an arXiv ID

    :param arxiv_id: arXiv id, with or without the `arXiv:` prefix and version
                     suffix (e.g. `v1`). Old an new style are accepted. Here are
                     example of accepted identifiers: `1510.00322`,
                     `arXiv:1510.00322`, `0901.0512`, `arXiv:0901.0512`,
                     `hep-ph/9409201` or `arXiv:hep-ph/9409201`.
                     Note that the `arXiv:` prefix will be automatically
                     removed, and the version suffix automatically added if
                     missing.
    :param try_doi:  if a DOI is referenced in the arXiv metadata,
                     try to download it instead. If that fails for any reason,
                     falls back to the arXiv, with a warning message, if the
                     UI is provided.
    :param ui:       if not None, will display a warning if the doi request
                     fails.
    """
    ## handle errors
    url = 'https://export.arxiv.org/api/query?id_list={}'.format(arxiv_id)
    try:
        r = requests.get(url)
        if r.status_code == 400:  # bad request
            msg = ("the arXiv server returned a bad request error. The "
                   "arXiv id {} is possibly invalid or malformed.".format(arxiv_id))
            raise ReferenceNotFoundError(msg)
        r.raise_for_status()  # raise an exception for HTTP errors:
                              # 401, 404, 400 if `ui` is None, etc.
    except requests.exceptions.RequestException as e:
        msg = ("connection error while retrieving arXiv data for "
               "'{}': {}".format(arxiv_id, e))
        raise ReferenceNotFoundError(msg)

    feed = feedparser.parse(r.text)
    if len(feed.entries) == 0:  # no results.
        msg = "no results for arXiv id {}".format(arxiv_id)
        raise ReferenceNotFoundError(msg)
    if len(feed.entries) > 1:  # I don't know how that could happen, but let's
                               # be ready for it.
        results = '\n'.join('{}. {}'.format(i, entry['title'])
                            for entry in feed.entries)
        msg = ("multiple results for arXiv id {}:\n{}\nThis is unexpected. "
               "Please submit an issue at "
               "https://github.com/pubs/pubs/issues").format(arxiv_id, choices)
        raise ReferenceNotFoundError(msg)

    entry = feed.entries[0]

    ## try to return a doi instead of the arXiv reference
    if try_doi and 'arxiv_doi' in entry:
        try:
            return doi2bibtex(entry['arxiv_doi'])
        except ReferenceNotFoundError as e:
            if ui is not None:
                ui.warning(str(e))

    ## create a bibentry from the arXiv response.
    db = BibDatabase()
    entry_id = _extract_arxiv_id(entry)
    author_str = ' and '.join(
        [author['name'] for author in entry['authors']])
    db.entries = [{
        'ENTRYTYPE': 'article',
        'ID': entry_id,
        'author': author_str,
        'title': entry['title'],
        'year': str(entry['published_parsed'].tm_year),
        'month': _months[entry['published_parsed'].tm_mon-1],
        'eprint': entry_id,
        'eprinttype': 'arxiv',
        'date': entry['published'], # not really standard, but a resolution more
                                    # granular than months is increasinlgy relevant.
        'url': entry['link'],
        'urldate': datetime.datetime.utcnow().isoformat() + 'Z' # can't hurt.
    }]
    # we don't add eprintclass for old-style ids, as it is in the id already.
    if not _is_arxiv_oldstyle(entry_id):
        db.entries[0]['eprintclass'] = entry['arxiv_primary_category']['term']
    if 'arxiv_doi' in entry:
        db.entries[0]['arxiv_doi'] = entry['arxiv_doi']

    bibtex = bibtexparser.dumps(db)
    return bibtex