Add wikidata links.
[redakcja.git] / src / documents / management / commands / import_wl.py
1 # This file is part of FNP-Redakcja, licensed under GNU Affero GPLv3 or later.
2 # Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
3 #
4 from collections import defaultdict
5 import json
6 from urllib.request import urlopen
7
8 from django.core.management.base import BaseCommand
9 from django.core.management.color import color_style
10 from django.db import transaction
11 from librarian.dcparser import BookInfo
12 from librarian import ParseError, ValidationError
13
14 from documents.models import Book
15
16
17 WL_API = 'http://www.wolnelektury.pl/api/books/'
18
19
20 class Command(BaseCommand):
21     help = 'Imports XML files from WL.'
22
23     def add_arguments(self, parser):
24         parser.add_argument('-q', '--quiet', action='store_false', dest='verbose', default=True,
25             help='Less output')
26
27     def handle(self, *args, **options):
28
29         self.style = color_style()
30
31         verbose = options.get('verbose')
32
33         # Start transaction management.
34         transaction.enter_transaction_management()
35
36         if verbose:
37             print('Reading currently managed files (skipping hidden ones).')
38         slugs = defaultdict(list)
39         for b in Book.objects.exclude(slug__startswith='.').all():
40             if verbose:
41                 print(b.slug)
42             text = b.materialize().encode('utf-8')
43             try:
44                 info = BookInfo.from_bytes(text)
45             except (ParseError, ValidationError):
46                 pass
47             else:
48                 slugs[info.slug].append(b)
49
50         book_count = 0
51         commit_args = {
52             "author_name": 'Platforma',
53             "description": 'Automatycznie zaimportowane z Wolnych Lektur',
54             "publishable": True,
55         }
56
57         if verbose:
58             print('Opening books list')
59         for book in json.load(urlopen(WL_API)):
60             book_detail = json.load(urlopen(book['href']))
61             xml_text = urlopen(book_detail['xml']).read()
62             info = BookInfo.from_bytes(xml_text)
63             previous_books = slugs.get(info.slug)
64             if previous_books:
65                 if len(previous_books) > 1:
66                     print(self.style.ERROR("There is more than one book "
67                         "with slug %s:") % info.slug)
68                 previous_book = previous_books[0]
69                 comm = previous_book.slug
70             else:
71                 previous_book = None
72                 comm = '*'
73             print(book_count, info.slug , '-->', comm)
74             Book.import_xml_text(xml_text, title=info.title[:255],
75                 slug=info.slug[:128], previous_book=previous_book,
76                 commit_args=commit_args)
77             book_count += 1
78
79         # Print results
80         print()
81         print("Results:")
82         print("Imported %d books from WL:" % (
83                 book_count, ))
84         print()
85
86
87         transaction.commit()
88         transaction.leave_transaction_management()
89