Adds generalized book data connectors

This commit is contained in:
Mouse Reeve 2020-03-07 12:22:28 -08:00
parent d501e707ee
commit 6f765bd6f4
11 changed files with 249 additions and 144 deletions

View file

@ -0,0 +1,11 @@
''' select and call a connector for whatever book task needs doing '''
from fedireads.connectors import OpenLibraryConnector
openlibrary = OpenLibraryConnector()
def get_or_create_book(key):
''' pull up a book record by whatever means possible '''
return openlibrary.get_or_create_book(key)
def search(query):
''' ya '''
return openlibrary.search(query)

View file

@ -0,0 +1,3 @@
''' bring connectors into the namespace '''
from .settings import CONNECTORS
from .openlibrary import OpenLibraryConnector

View file

@ -0,0 +1,60 @@
''' functionality outline for a book data connector '''
from abc import ABC, abstractmethod
from fedireads.connectors import CONNECTORS
class AbstractConnector(ABC):
''' generic book data connector '''
def __init__(self, connector_name):
# load connector settings
settings = CONNECTORS.get(connector_name)
if not settings:
raise ValueError('No connector with name "%s"' % connector_name)
try:
self.url = settings['BASE_URL']
self.covers_url = settings['COVERS_URL']
self.db_field = settings['DB_KEY_FIELD']
self.key_name = settings['KEY_NAME']
except KeyError:
raise KeyError('Invalid connector settings')
# TODO: politeness settings
@abstractmethod
def search(self, query):
''' free text search '''
# return list of search result objs
pass
@abstractmethod
def get_or_create_book(self, book_id):
''' request and format a book given an identifier '''
# return book model obj
pass
@abstractmethod
def get_or_create_author(self, book_id):
''' request and format a book given an identifier '''
# return book model obj
pass
@abstractmethod
def update_book(self, book_obj):
''' sync a book with the canonical remote copy '''
# return book model obj
pass
class SearchResult(object):
''' standardized search result object '''
def __init__(self, title, key, author, year):
self.title = title
self.key = key
self.author = author
self.year = year

View file

@ -0,0 +1,138 @@
''' openlibrary data connector '''
from django.core.exceptions import ObjectDoesNotExist
from django.core.files.base import ContentFile
import re
import requests
from fedireads import models
from .abstract_connector import AbstractConnector, SearchResult
class OpenLibraryConnector(AbstractConnector):
''' instantiate a connector for OL '''
def __init__(self):
super().__init__('openlibrary')
def search(self, query):
''' query openlibrary search '''
resp = requests.get('%s/search.json' % self.url, params={'q': query})
if not resp.ok:
resp.raise_for_status()
data = resp.json()
results = []
for doc in data['docs'][:5]:
key = doc['key']
key = key.split('/')[-1]
author = doc.get('author_name') or ['Unknown']
results.append(SearchResult(
doc.get('title'),
key,
author[0],
doc.get('first_publish_year'),
))
return results
def get_or_create_book(self, olkey):
''' pull up a book record by whatever means possible '''
if re.match(r'^OL\d+W$', olkey):
model = models.Work
elif re.match(r'^OL\d+M$', olkey):
model = models.Edition
else:
raise ValueError('Invalid OpenLibrary ID')
try:
book = models.Book.objects.get(openlibrary_key=olkey)
return book
except ObjectDoesNotExist:
# no book was found, so we start creating a new one
book = model(openlibrary_key=olkey)
# load the book json from openlibrary.org
response = requests.get('%s/works/%s.json' % (self.url, olkey))
if not response.ok:
response.raise_for_status()
data = response.json()
# great, we can update our book.
book.title = data['title']
description = data.get('description')
if description:
if isinstance(description, dict):
description = description.get('value')
book.description = description
book.pages = data.get('pages')
#book.published_date = data.get('publish_date')
# this book sure as heck better be an edition
if data.get('works'):
key = data.get('works')[0]['key']
key = key.split('/')[-1]
work = self.get_or_create_book(key)
book.parent_work = work
book.save()
# we also need to know the author get the cover
for author_blob in data.get('authors'):
# this id is "/authors/OL1234567A" and we want just "OL1234567A"
author_blob = author_blob.get('author', author_blob)
author_id = author_blob['key']
author_id = author_id.split('/')[-1]
book.authors.add(self.get_or_create_author(author_id))
if data.get('covers') and len(data['covers']):
book.cover.save(*self.get_cover(data['covers'][0]), save=True)
return book
def get_or_create_author(self, olkey):
''' load that author '''
if not re.match(r'^OL\d+A$', olkey):
raise ValueError('Invalid OpenLibrary author ID')
try:
author = models.Author.objects.get(openlibrary_key=olkey)
except ObjectDoesNotExist:
pass
response = requests.get('%s/authors/%s.json' % (self.url, olkey))
if not response.ok:
response.raise_for_status()
data = response.json()
author = models.Author(openlibrary_key=olkey)
bio = data.get('bio')
if bio:
if isinstance(bio, dict):
bio = bio.get('value')
author.bio = bio
name = data['name']
author.name = name
# TODO this is making some BOLD assumption
author.last_name = name.split(' ')[-1]
author.first_name = ' '.join(name.split(' ')[:-1])
#author.born = data.get('birth_date')
#author.died = data.get('death_date')
author.save()
return author
def get_cover(self, cover_id):
''' ask openlibrary for the cover '''
# TODO: get medium and small versions
image_name = '%s-M.jpg' % cover_id
url = '%s/b/id/%s' % (self.covers_url, image_name)
response = requests.get(url)
if not response.ok:
response.raise_for_status()
image_content = ContentFile(requests.get(url).content)
return [image_name, image_content]
def update_book(self, book_obj):
pass

View file

@ -0,0 +1,28 @@
''' settings book data connectors '''
CONNECTORS = {
'openlibrary': {
'KEY_NAME': 'olkey',
'DB_KEY_FIELD': 'openlibrary_key',
'POLITENESS_DELAY': 0,
'MAX_DAILY_QUERIES': -1,
'BASE_URL': 'https://openlibrary.org',
'COVERS_URL': 'https://covers.openlibrary.org',
},
}
''' not implemented yet:
'librarything': {
'KEY_NAME': 'ltkey',
'DB_KEY_FIELD': 'librarything_key',
'POLITENESS_DELAY': 1,
'MAX_DAILY_QUERIES': 1000,
'BASE_URL': 'https://librarything.com',
},
'worldcat': {
'KEY_NAME': 'ocn',
'DB_KEY_FIELD': 'oclc_number',
'POLITENESS_DELAY': 0,
'MAX_DAILY_QUERIES': -1,
'BASE_URL': 'https://worldcat.org',
},
'''

View file

@ -1,135 +0,0 @@
''' activitystream api and books '''
from django.core.exceptions import ObjectDoesNotExist
from django.core.files.base import ContentFile
import re
import requests
from fedireads import models
from fedireads.settings import OL_URL
def book_search(query):
''' look up a book '''
response = requests.get('%s/search.json' % OL_URL, params={'q': query})
if not response.ok:
response.raise_for_status()
data = response.json()
results = []
for doc in data['docs'][:5]:
key = doc['key']
key = key.split('/')[-1]
author = doc.get('author_name') or ['Unknown']
results.append({
'title': doc.get('title'),
'olkey': key,
'year': doc.get('first_publish_year'),
'author': author[0],
})
return results
def get_or_create_book(olkey, update=False):
''' create a book or work '''
# check if this is in the format of an OL book identifier
if re.match(r'^OL\d+W$', olkey):
model = models.Work
elif re.match(r'^OL\d+M$', olkey):
model = models.Edition
else:
raise ValueError('Invalid OpenLibrary ID')
# get the existing entry from our db, if it exists
try:
book = model.objects.get(openlibrary_key=olkey)
if not update:
return book
# we have the book, but still want to update it from OL
except ObjectDoesNotExist:
# no book was found, so we start creating a new one
book = model(openlibrary_key=olkey)
# load the book json from openlibrary.org
response = requests.get('%s/works/%s.json' % (OL_URL, olkey))
if not response.ok:
response.raise_for_status()
data = response.json()
# great, we can update our book.
book.title = data['title']
description = data.get('description')
if description:
if isinstance(description, dict):
description = description.get('value')
book.description = description
book.pages = data.get('pages')
#book.published_date = data.get('publish_date')
# this book sure as heck better be an edition
if data.get('works'):
key = data.get('works')[0]['key']
key = key.split('/')[-1]
work = get_or_create_book(key)
book.parent_work = work
book.save()
# we also need to know the author get the cover
for author_blob in data.get('authors'):
# this id starts as "/authors/OL1234567A" and we want just "OL1234567A"
author_blob = author_blob.get('author', author_blob)
author_id = author_blob['key']
author_id = author_id.split('/')[-1]
book.authors.add(get_or_create_author(author_id))
if data.get('covers') and len(data['covers']):
book.cover.save(*get_cover(data['covers'][0]), save=True)
return book
def get_cover(cover_id):
''' ask openlibrary for the cover '''
# TODO: get medium and small versions
image_name = '%s-M.jpg' % cover_id
url = 'https://covers.openlibrary.org/b/id/%s' % image_name
response = requests.get(url)
if not response.ok:
response.raise_for_status()
image_content = ContentFile(requests.get(url).content)
return [image_name, image_content]
def get_or_create_author(olkey, update=False):
''' load that author '''
if not re.match(r'^OL\d+A$', olkey):
raise ValueError('Invalid OpenLibrary author ID')
try:
author = models.Author.objects.get(openlibrary_key=olkey)
if not update:
return author
except ObjectDoesNotExist:
pass
response = requests.get('%s/authors/%s.json' % (OL_URL, olkey))
if not response.ok:
response.raise_for_status()
data = response.json()
author = models.Author(openlibrary_key=olkey)
bio = data.get('bio')
if bio:
if isinstance(bio, dict):
bio = bio.get('value')
author.bio = bio
name = data['name']
author.name = name
# TODO this is making some BOLD assumption
author.last_name = name.split(' ')[-1]
author.first_name = ' '.join(name.split(' ')[:-1])
#author.born = data.get('birth_date')
#author.died = data.get('death_date')
author.save()
return author

View file

@ -1,6 +1,6 @@
''' Handle user activity '''
from fedireads import models
from fedireads.openlibrary import get_or_create_book
from fedireads.books_manager import get_or_create_book
from fedireads.sanitize_html import InputHtmlParser
from django.db import IntegrityError

View file

@ -5,7 +5,7 @@
<h1>Search results</h1>
{% for result in results %}
<div>
<a href="/book/{{ result.olkey }}">{{ result.title }}</a> by {{ result.author }} ({{ result.year }})
<a href="/book/{{ result.key }}">{{ result.title }}</a> by {{ result.author }} ({{ result.year }})
</div>
{% endfor %}
</div>

View file

@ -5,7 +5,7 @@ from django.shortcuts import redirect
from django.template.response import TemplateResponse
import re
from fedireads import forms, models, openlibrary, outgoing
from fedireads import forms, models, books_manager, outgoing
from fedireads.views import get_user_from_username
@ -150,8 +150,8 @@ def search(request):
results = [outgoing.handle_account_search(query)]
template = 'user_results.html'
else:
# just send the question over to openlibrary for book search
results = openlibrary.book_search(query)
# just send the question over to book search
results = books_manager.search(query)
template = 'book_results.html'
return TemplateResponse(request, template, {'results': results})

View file

@ -6,7 +6,7 @@ from django.http import HttpResponseNotFound
from django.shortcuts import redirect
from django.template.response import TemplateResponse
from fedireads import forms, models, openlibrary, incoming
from fedireads import forms, models, books_manager, incoming
from fedireads.settings import DOMAIN
@ -188,7 +188,7 @@ def edit_profile_page(request, username):
@login_required
def book_page(request, book_identifier, tab='friends'):
''' info about a book '''
book = openlibrary.get_or_create_book(book_identifier)
book = books_manager.get_or_create_book(book_identifier)
if isinstance(book, models.Work):
book_reviews = models.Review.objects.filter(
@ -258,7 +258,7 @@ def book_page(request, book_identifier, tab='friends'):
def author_page(request, author_identifier):
''' landing page for an author '''
try:
author = models.Author.objects.get(openlibrary_key=author_identifier)
author = models.Author.objects.get(books_manager_key=author_identifier)
except ValueError:
return HttpResponseNotFound()

View file

@ -1,5 +1,5 @@
from fedireads.models import User
from fedireads.openlibrary import get_or_create_book
from fedireads.books_manager import get_or_create_book
User.objects.create_user('mouse', 'mouse.reeve@gmail.com', 'password123')
User.objects.create_user('rat', 'rat@rat.com', 'ratword')