Source code for app.model.lib.crossref_fetcher

import requests
import re




[docs]
class CrossrefFetcher:
    """
    This class encapsulates requests to api.crossref.org to get information
    about a publication based on its DOI. For now, it only fetches the author
    list and the title of a study.
    """
    def __init__(self, doi):


[docs]
        self.doi = doi





[docs]
        self.authors          = []




[docs]
        self.author_cache     = ''




[docs]
        self.license_url      = None




[docs]
        self.publication_date = None




[docs]
        self.publication_type = None



        self._response_json = None



[docs]
    def make_request(self):
        # Author records for linking and searching:
        quoted_doi = requests.utils.quote(self.doi)
        crossref_url = f"https://api.crossref.org/works/{quoted_doi}"
        response = requests.get(crossref_url)

        if response.status_code == 404:
            raise ValueError("Couldn't find publication")
        if response.status_code != 200:
            raise ValueError(f"Couldn't reach Crossref API (Status {response.status_code})")

        self._response_json = response.json()

        if self._response_json["status"] != "ok":
            raise ValueError("The Crossref API didn't return a successful result")

        message_field = self._response_json.get("message", {})
        title_field   = message_field.get("title", [])

        if canonical_doi := message_field.get("DOI"):
            self.doi = f"https://doi.org/{canonical_doi}"

        self.title        = title_field[0] if len(title_field) else ''
        self.authors      = message_field.get("author", [])
        self.author_cache = ', '.join([a['family'].lower() for a in self.authors])

        licenses = message_field.get("license", [])
        if len(licenses):
            self.license_url = licenses[0].get("URL")

        for key in ['published-online', 'published-print', 'published']:
            publication_date_parts = message_field.get(key, {}).get("date-parts", [])
            if len(publication_date_parts):
                self.publication_date = '-'.join([f"{int(str(part)):02}" for part in publication_date_parts[0]])

                break

        if resource_url := message_field.get('resource', {}).get('primary', {}).get('URL'):
            if re.match(r'https?://(www.)?biorxiv.org/', resource_url):
                self.publication_type = 'preprint'
            else:
                self.publication_type = 'publication'
        else:
            self.publication_type = 'dataset'