moviewyrm/bookwyrm/connectors/openlibrary.py

186 lines
6.3 KiB
Python
Raw Normal View History

2020-03-07 20:22:28 +00:00
''' openlibrary data connector '''
import re
2020-04-22 13:53:22 +00:00
from bookwyrm import models
2020-05-10 23:41:24 +00:00
from .abstract_connector import AbstractConnector, SearchResult, Mapping
2020-12-20 00:14:05 +00:00
from .abstract_connector import ConnectorException, get_data
2020-03-30 20:15:49 +00:00
from .openlibrary_languages import languages
2020-03-07 20:22:28 +00:00
class Connector(AbstractConnector):
2020-03-07 20:22:28 +00:00
''' instantiate a connector for OL '''
def __init__(self, identifier):
2020-05-10 19:56:59 +00:00
super().__init__(identifier)
2020-05-10 23:41:24 +00:00
get_first = lambda a: a[0]
2020-12-20 00:14:05 +00:00
get_remote_id = lambda a: self.base_url + a
self.book_mappings = [
Mapping('title'),
2020-12-20 00:14:05 +00:00
Mapping('id', remote_field='key', formatter=get_remote_id),
Mapping(
'cover', remote_field='covers', formatter=self.get_cover_url),
Mapping('sortTitle', remote_field='sort_title'),
2020-05-10 23:41:24 +00:00
Mapping('subtitle'),
Mapping('description', formatter=get_description),
Mapping('languages', formatter=get_languages),
Mapping('series', formatter=get_first),
Mapping('seriesNumber', remote_field='series_number'),
2020-05-10 23:41:24 +00:00
Mapping('subjects'),
Mapping('subjectPlaces'),
Mapping('isbn13', formatter=get_first),
Mapping('isbn10', formatter=get_first),
Mapping('lccn', formatter=get_first),
2020-05-10 23:41:24 +00:00
Mapping(
'oclcNumber', remote_field='oclc_numbers',
formatter=get_first
2020-05-10 23:41:24 +00:00
),
Mapping(
'openlibraryKey', remote_field='key',
formatter=get_openlibrary_key
2020-05-10 23:41:24 +00:00
),
Mapping('goodreadsKey', remote_field='goodreads_key'),
Mapping('asin'),
2020-05-10 23:41:24 +00:00
Mapping(
'firstPublishedDate', remote_field='first_publish_date',
2020-05-10 23:41:24 +00:00
),
Mapping('publishedDate', remote_field='publish_date'),
Mapping('pages', remote_field='number_of_pages'),
Mapping('physicalFormat', remote_field='physical_format'),
2020-05-10 23:41:24 +00:00
Mapping('publishers'),
]
self.author_mappings = [
2020-12-20 00:14:05 +00:00
Mapping('id', remote_field='key', formatter=get_remote_id),
2020-12-12 17:43:07 +00:00
Mapping('name'),
2020-12-20 00:14:05 +00:00
Mapping(
'openlibraryKey', remote_field='key',
formatter=get_openlibrary_key
),
2020-12-19 23:20:31 +00:00
Mapping('born', remote_field='birth_date'),
Mapping('died', remote_field='death_date'),
2020-05-10 23:41:24 +00:00
Mapping('bio', formatter=get_description),
]
2020-03-07 20:22:28 +00:00
def get_remote_id_from_data(self, data):
2020-12-20 00:14:05 +00:00
''' format a url from an openlibrary id field '''
try:
key = data['key']
except KeyError:
raise ConnectorException('Invalid book data')
return '%s/%s' % (self.books_url, key)
2020-05-10 19:56:59 +00:00
def is_work_data(self, data):
2020-05-10 21:12:03 +00:00
return bool(re.match(r'^[\/\w]+OL\d+W$', data['key']))
2020-03-07 20:22:28 +00:00
2020-05-10 19:56:59 +00:00
def get_edition_from_work_data(self, data):
try:
key = data['key']
except KeyError:
2020-09-30 17:27:40 +00:00
raise ConnectorException('Invalid book data')
2020-05-10 19:56:59 +00:00
url = '%s/%s/editions' % (self.books_url, key)
data = get_data(url)
return pick_default_edition(data['entries'])
2020-05-10 19:56:59 +00:00
def get_work_from_edition_date(self, data):
try:
key = data['works'][0]['key']
except (IndexError, KeyError):
2020-10-01 02:57:25 +00:00
raise ConnectorException('No work found for edition')
2020-05-10 19:56:59 +00:00
url = '%s/%s' % (self.books_url, key)
return get_data(url)
def get_authors_from_data(self, data):
''' parse author json and load or create authors '''
2020-03-14 04:10:53 +00:00
for author_blob in data.get('authors', []):
2020-03-07 20:22:28 +00:00
author_blob = author_blob.get('author', author_blob)
2020-12-19 23:20:31 +00:00
# this id is "/authors/OL1234567A"
2020-12-20 00:14:05 +00:00
author_id = author_blob['key']
2020-12-19 23:20:31 +00:00
url = '%s/%s.json' % (self.base_url, author_id)
yield self.get_or_create_author(url)
2020-03-07 20:22:28 +00:00
2020-12-20 00:14:05 +00:00
def get_cover_url(self, cover_blob):
2020-05-10 19:56:59 +00:00
''' ask openlibrary for the cover '''
2020-12-20 00:14:05 +00:00
cover_id = cover_blob[0]
2020-05-10 19:56:59 +00:00
image_name = '%s-M.jpg' % cover_id
2020-12-20 00:14:05 +00:00
return '%s/b/id/%s' % (self.covers_url, image_name)
2020-05-10 19:56:59 +00:00
def parse_search_data(self, data):
return data.get('docs')
2020-09-21 17:25:26 +00:00
def format_search_result(self, search_result):
# build the remote id from the openlibrary key
2020-09-21 17:25:26 +00:00
key = self.books_url + search_result['key']
author = search_result.get('author_name') or ['Unknown']
2020-05-10 19:56:59 +00:00
return SearchResult(
2020-10-29 22:29:23 +00:00
title=search_result.get('title'),
key=key,
author=', '.join(author),
year=search_result.get('first_publish_year'),
2020-05-10 19:56:59 +00:00
)
def load_edition_data(self, olkey):
''' query openlibrary for editions of a work '''
url = '%s/works/%s/editions.json' % (self.books_url, olkey)
return get_data(url)
2020-03-07 20:22:28 +00:00
2020-04-02 05:11:31 +00:00
def expand_book_data(self, book):
work = book
2020-10-01 02:43:42 +00:00
# go from the edition to the work, if necessary
2020-04-02 05:11:31 +00:00
if isinstance(book, models.Edition):
work = book.parent_work
2020-10-01 02:43:42 +00:00
# we can mass download edition data from OL to avoid repeatedly querying
edition_options = self.load_edition_data(work.openlibrary_key)
for edition_data in edition_options.get('entries'):
2020-12-20 00:14:05 +00:00
self.create_edition_from_data(work, edition_data)
2020-03-30 19:21:04 +00:00
2020-03-28 04:28:52 +00:00
def get_description(description_blob):
''' descriptions can be a string or a dict '''
if isinstance(description_blob, dict):
return description_blob.get('value')
return description_blob
2020-03-30 20:15:49 +00:00
2020-05-10 23:41:24 +00:00
def get_openlibrary_key(key):
''' convert /books/OL27320736M into OL27320736M '''
return key.split('/')[-1]
2020-03-30 20:15:49 +00:00
def get_languages(language_blob):
''' /language/eng -> English '''
langs = []
for lang in language_blob:
langs.append(
languages.get(lang.get('key', ''), None)
)
return langs
def pick_default_edition(options):
''' favor physical copies with covers in english '''
2020-04-22 13:53:22 +00:00
if not options:
return None
if len(options) == 1:
return options[0]
options = [e for e in options if e.get('cover')] or options
options = [e for e in options if \
'/languages/eng' in str(e.get('languages'))] or options
formats = ['paperback', 'hardcover', 'mass market paperback']
options = [e for e in options if \
str(e.get('physical_format')).lower() in formats] or options
options = [e for e in options if e.get('isbn_13')] or options
options = [e for e in options if e.get('ocaid')] or options
return options[0]