cb64a156ba6e41204eab1fb3a8fe6df45df286fd
[redakcja.git] / src / depot / publishers / woblink.py
1 from datetime import date
2 import io
3 import json
4 import re
5 from time import sleep
6 from django.conf import settings
7 from django.utils.html import escape, format_html
8 from django.utils.safestring import mark_safe
9 from librarian.builders.html import SnippetHtmlBuilder
10 from librarian.functions import lang_code_3to2
11 from catalogue.models import Audience, Author, Thema
12 from .. import models
13 from .base import BasePublisher
14 from .woblink_constants import WOBLINK_CATEGORIES
15
16
17 class WoblinkError(ValueError):
18     pass
19
20 class NoPrice(WoblinkError):
21     def as_html(self):
22         return format_html(
23             'Brak <a href="/admin/depot/shop/{price}">określonej ceny</a>.',
24             price=self.args[0].id
25         )
26
27 class NoIsbn(WoblinkError):
28     def as_html(self):
29         return 'Brak ISBN.'
30
31 class AuthorLiteralForeign(WoblinkError):
32     def as_html(self):
33         return format_html(
34             'Nie obsługiwane: autor „{author}” w języku {lang}.',
35             author=str(self.args[0]),
36             lang=self.args[0].lang,
37         )
38
39 class AuthorNotInCatalogue(WoblinkError):
40     def as_html(self):
41         return format_html(
42             'Brak autora „{author}” w katalogu.',
43             author=str(self.args[0])
44         )
45
46 class AuthorNoWoblink(WoblinkError):
47     def as_html(self):
48         return format_html(
49             'Autor <a href="/admin/catalogue/author/{author_id}/">{author}</a> bez identyfikatora Woblink.',
50             author_id=self.args[0].id,
51             author=self.args[0].name
52         )
53
54 class NoThema(WoblinkError):
55     def as_html(self):
56         return format_html('Brak Thema.')
57
58 class UnknownThema(WoblinkError):
59     def as_html(self):
60         return format_html(
61             'Nieznana Thema {code}.',
62             code=self.args[0]
63         )
64
65
66 class ThemaUnknownWoblink(WoblinkError):
67     def as_html(self):
68         return format_html(
69             'Thema <a href="/admin/catalogue/thema/{id}/">{code}</a> przypisana do nieznanej kategorii Woblink.',
70             id=self.args[0].id,
71             code=self.args[0].code,
72         )
73
74 class NoWoblinkCategory(WoblinkError):
75     def as_html(self):
76         return 'Brak kategorii Woblink.'
77
78 class WoblinkWarning(Warning):
79     pass
80
81 class NoMainThemaWarning(WoblinkWarning):
82     def as_html(self):
83         return format_html(
84             'Brak głównej kategorii Thema.'
85         )
86
87 class ThemaNoWoblink(WoblinkWarning):
88     def as_html(self):
89         return format_html(
90             'Thema <a href="/admin/catalogue/thema/{id}/">{code}</a> nie przypisana do kategorii Woblink.',
91             id=self.args[0].id,
92             code=self.args[0].code,
93         )
94
95 class AuthorLiteralForeignWarning(WoblinkWarning):
96     def as_html(self):
97         return format_html(
98             'Nie obsługiwane: autor „{author}” w języku {lang}.',
99             author=str(self.args[0]),
100             lang=self.args[0].lang,
101         )
102
103 class AuthorNotInCatalogueWarning(WoblinkWarning):
104     def as_html(self):
105         return format_html(
106             'Brak autora „{author}” w katalogu.',
107             author=str(self.args[0])
108         )
109
110 class AuthorNoWoblinkWarning(WoblinkWarning):
111     def as_html(self):
112         return format_html(
113             'Autor <a href="/admin/catalogue/author/{author_id}/">{author}</a> bez identyfikatora Woblink.',
114             author_id=self.args[0].id,
115             author=self.args[0].name
116         )
117
118
119
120
121 class Woblink(BasePublisher):
122     BASE_URL = 'https://publisher.woblink.com/'
123     ADD_URL = BASE_URL + 'catalog/add'
124     STEP1_URL = BASE_URL + 'catalog/edit/%s'
125     STEP2_URL = BASE_URL + 'catalog/edit/%s/2'
126     STEP3_URL = BASE_URL + 'catalog/edit/%s/3'
127     UPLOAD_URL = BASE_URL + 'file/upload-%s'
128     JOB_STATUS_URL = BASE_URL + 'task/status'
129     GENERATE_DEMO_URL = BASE_URL + 'task/run/generate-%s-demo/%s/%d'
130     CHECK_DEMO_URL = BASE_URL + 'task/run/check-%s-demo/%s'
131
132     SEARCH_CATALOGUE_URL = BASE_URL + '{category}/autocomplete/{term}'
133
134     ROLE_AUTHOR = 1
135     ROLE_TRANSLATOR = 4
136
137     def login(self):
138         response = self.session.get('https://publisher.woblink.com/login')
139         token = re.search(
140             r'name="_csrf_token" value="([^"]+)"',
141             response.text
142         ).group(1)
143         data = {
144             '_csrf_token': token,
145             '_username': self.username,
146             '_password': self.password,
147         }
148         response = self.session.post(
149             'https://publisher.woblink.com/login_check',
150             data=data,
151         )
152
153     def search_catalogue(self, category, term):
154         return self.session.get(
155             self.SEARCH_CATALOGUE_URL.format(category=category, term=term)
156         ).json()
157
158     def search_author_catalogue(self, term):
159         return [
160             {
161                 'id': item['autId'],
162                 'text': item['autFullname']
163             }
164             for item in self.search_catalogue('author', term)
165         ]
166     def search_series_catalogue(self, term):
167         return [
168             {
169                 'id': item['id'],
170                 'text': item['name']
171             }
172             for item in self.search_catalogue('series', term)
173         ]
174         
175     def get_isbn(self, meta, errors=None):
176         if not meta.isbn_epub:
177             if errors is not None:
178                 errors.append(NoIsbn())
179         return meta.isbn_epub
180
181     def get_authors_data(self, meta, errors=None):
182         authors = []
183         for role, items, obligatory in [
184                 (self.ROLE_AUTHOR, meta.authors, True),
185                 (self.ROLE_TRANSLATOR, meta.translators, False)
186         ]:
187             for person_literal in items:
188                 if person_literal.lang != 'pl':
189                     if errors is not None:
190                         if obligatory:
191                              errors.append(AuthorLiteralForeign(person_literal))
192                         else:
193                             errors.append(AuthorLiteralForeignWarning(person_literal))
194                     continue
195                 aobj = Author.get_by_literal(str(person_literal))
196                 if aobj is None:
197                     if errors is not None:
198                         if obligatory:
199                              errors.append(AuthorNotInCatalogue(person_literal))
200                         else:
201                             errors.append(AuthorNotInCatalogueWarning(person_literal))
202                     continue
203                 if not aobj.woblink:
204                     if errors is not None:
205                         if obligatory:
206                              errors.append(AuthorNoWoblink(aobj))
207                         else:
208                             errors.append(AuthorNoWoblinkWarning(aobj))
209                     continue
210                 authors.append((role, aobj.woblink))
211         return authors
212
213     def get_genres(self, meta, errors=None):
214         thema_codes = []
215         if meta.thema_main:
216             thema_codes.append(meta.thema_main)
217         else:
218             if errors is not None:
219                 errors.append(NoMainThemaWarning())
220         thema_codes.extend(meta.thema)
221         if not thema_codes:
222             if errors is not None:
223                 errors.append(NoThema())
224         category_ids = []
225         for code in thema_codes:
226             try:
227                 thema = Thema.objects.get(code=code)
228             except Thema.DoesNotExist:
229                 if errors is not None:
230                     errors.append(UnknownThema(code))
231             else:
232                 if thema.woblink_category is None:
233                     if errors is not None:
234                         errors.append(ThemaNoWoblink(thema))
235                 elif thema.woblink_category not in WOBLINK_CATEGORIES:
236                     if errors is not None:
237                         errors.append(ThemaUnknownWoblink(thema))
238                 elif thema.woblink_category not in category_ids:
239                     category_ids.append(thema.woblink_category)
240         if not category_ids:
241             if errors is not None:
242                 errors.append(NoWoblinkCategory())
243         return category_ids
244
245     def get_series(self, meta, errors=None):
246         return list(Audience.objects.filter(code__in=meta.audiences).exclude(
247             woblink=None).values_list('woblink', flat=True))
248
249     def get_abstract(self, wldoc, errors=None, description_add=None):
250         description = self.get_description(wldoc, description_add)
251         parts = description.split('\n', 1)
252         if len(parts) == 1 or len(parts[0]) > 240:
253             # No newline found here.
254             # Try to find last sentence end..
255             parts = re.split(r' \.', description[240::-1], 1)
256             if len(parts) == 2:
257                 p1 = parts[1][::-1] + '.'
258                 p2 = description[len(p1) + 1:]
259             else:
260                 # No sentence end found.
261                 # Just find a space.
262                 p1 = description[:240].rsplit(' ', 1)[0]
263                 p2 = description[len(p1) + 1:]
264                 p1 += '…'
265                 p2 = '…' + p2
266             parts = [p1, p2]
267
268         m = re.search(r'<[^>]+$', parts[0])
269         if m is not None:
270             parts[0] = parts[:-len(m.group(0))]
271             parts[1] = m.group(0) + parts[1]
272
273         opened = []
274         for tag in re.findall(r'<[^>]+[^/>]>', parts[0]):
275             if tag[1] == '/':
276                 opened.pop()
277             else:
278                 opened.append(tag)
279         for tag in reversed(opened):
280             parts[0] += '</' + tag[1:-1].split()[0] + '>'
281             parts[1] = tag + parts[1]
282         return {
283             'header': parts[0],
284             'rest': parts[1],
285         }
286
287     def get_lang2code(self, meta, errors=None):
288         return lang_code_3to2(meta.language)
289
290     def get_price(self, shop, wldoc, errors=None):
291         stats = wldoc.get_statistics()['total']
292         words = stats['words_with_fn']
293         pages = stats['chars_with_fn'] / 1800
294         price = shop.get_price(words, pages)
295         if price is None:
296             if errors:
297                 errors.append(NoPrice(shop))
298             return 0
299
300         return price
301
302     def can_publish(self, shop, book):
303         wldoc = book.wldocument(librarian2=True)
304         d = {
305             'warnings': [],
306             'errors': [],
307         }
308         errors = []
309         book_data = self.get_book_data(shop, wldoc, errors)
310         for error in errors:
311             if not isinstance(error, Warning):
312                 errlist = d['errors']
313             else:
314                 errlist = d['warnings']
315             errlist.append(error.as_html())
316
317         if book_data.get('genres'):
318             d['comment'] = format_html(
319                 'W kategoriach: {cat} ({price} zł)',
320                 cat=', '.join(self.describe_category(g) for g in book_data['genres']),
321                 price=book_data['price']
322             )
323
324         return d
325
326     def describe_category(self, category):
327         t = []
328         while category:
329             c = WOBLINK_CATEGORIES[category]
330             t.append(c['name'])
331             category = c.get('parent')
332         return ' / '.join(reversed(t))
333
334     def create_book(self, isbn):
335         isbn = ''.join(c for c in isbn if c.isdigit())
336         assert len(isbn) == 13
337         response = self.session.post(
338             self.ADD_URL,
339             data={
340                 'AddPublication[pubType]': 'ebook',
341                 'AddPublication[pubHasIsbn]': '1',
342                 'AddPublication[pubIsbn]': isbn,
343                  ##AddPubation[save]
344             }
345         )
346         m = re.search(r'/(\d+)$', response.url)
347         if m is not None:
348             return m.group(1)
349
350     def send_book(self, shop, book, changes=None):
351         wldoc = book.wldocument(librarian2=True, changes=changes, publishable=False) # TODO pub
352         meta = wldoc.meta
353
354         book_data = self.get_book_data(shop, wldoc)
355
356         if not book.woblink_id:
357             #book.woblink_id = 2959868
358             woblink_id = self.create_book(book_data['isbn'])
359             assert woblink_id
360             book.woblink_id = woblink_id
361             book.save(update_fields=['woblink_id'])
362
363         self.edit_step1(book.woblink_id, book_data)
364         self.edit_step2(book.woblink_id, book_data)
365         self.edit_step3(book.woblink_id, book_data)
366         self.send_cover(book.woblink_id, wldoc)
367         texts = shop.get_texts()
368         self.send_epub(
369             book.woblink_id, wldoc, book.gallery_path(),
370             fundraising=texts
371         )
372         self.send_mobi(
373             book.woblink_id, wldoc, book.gallery_path(),
374             fundraising=texts
375         )
376
377     def get_book_data(self, shop, wldoc, errors=None):
378         return {
379             "title": wldoc.meta.title,
380             "isbn": self.get_isbn(wldoc.meta, errors=errors),
381             "authors": self.get_authors_data(wldoc.meta, errors=errors),
382             "abstract": self.get_abstract(
383                 wldoc, errors=errors, description_add=shop.description_add
384             ),
385             "lang2code": self.get_lang2code(wldoc.meta, errors=errors),
386             "genres": self.get_genres(wldoc.meta, errors=errors),
387             "price": self.get_price(shop, wldoc, errors=errors),
388             "series": self.get_series(wldoc.meta, errors=errors),
389         }
390
391     def with_form_name(self, data, name):
392         return {
393             f"{name}[{k}]": v
394             for (k, v) in data.items()
395         }
396
397     def edit_step1(self, woblink_id, book_data):
398         data = book_data
399
400         authors_data = [
401             {
402                 "AhpPubId": woblink_id,
403                 "AhpAutId": author_id,
404                 "AhpType": author_type,
405             }
406             for (author_type, author_id) in data['authors']
407         ]
408
409         series_data = [
410             {
411                 'PublicationId': woblink_id,
412                 'SeriesId': series_id,
413             }
414             for series_id in data['series']
415         ]
416
417         d = {
418             'pubTitle': book_data['title'],
419             'npwAuthorHasPublications': json.dumps(authors_data),
420             'pubShortNote': data['abstract']['header'],
421             'pubNote': data['abstract']['rest'],
422             'pubCulture': data['lang2code'],
423             'npwPublicationHasAwards': '[]',
424             'npwPublicationHasSeriess': json.dumps(series_data),
425         }
426         d = self.with_form_name(d, 'EditPublicationStep1')
427         d['roles'] = [author_type for (author_type, author_id) in data['authors']]
428         r = self.session.post(self.STEP1_URL % woblink_id, data=d)
429         return r
430
431
432     def edit_step2(self, woblink_id, book_data):
433         gd = {}
434         legacy = None
435         for i, g in enumerate(book_data['genres']):
436             gdata = WOBLINK_CATEGORIES[g]
437             if legacy is None:
438                 legacy = gdata.get('legacy')
439             if p := gdata.get('parent'):
440                 gd.setdefault(p, {'isMain': False})
441                 gd[p].setdefault('children', [])
442                 gd[p]['children'].append(str(g))
443                 gd[p].setdefault('mainChild', str(g))
444                 if legacy is None:
445                     legacy = WOBLINK_CATEGORIES[p].get('legacy')
446             else:
447                 gd.setdefault(g, {})
448                 gd[g]['isMain'] = True
449         gd = [
450             {
451                 "pubId": woblink_id,
452                 "category": str(k),
453                 **v
454             }
455             for k, v in gd.items()
456         ]
457
458         data = {
459             'npwPublicationHasNewGenres': json.dumps(gd),
460             'genre': legacy or '',
461         }
462         data = self.with_form_name(data, 'AddPublicationStep2')
463         return self.session.post(self.STEP2_URL % woblink_id, data=data)
464
465     def edit_step3(self, woblink_id, book_data):
466         d = {
467             'pubBasePrice': book_data['price'],
468             'pubPremiereDate': date.today().isoformat(),
469             'pubIsLicenseIndefinite': '1',
470             'pubFileFormat': 'epub+mobi',
471             'pubIsAcs': '0',
472             'pubPublisherIndex': '',
473         }
474         d = self.with_form_name(d, 'EditPublicationStep3')
475         return self.session.post(self.STEP3_URL % woblink_id, data=d)
476
477     def wait_for_job(self, job_id):
478         while True:
479             response = self.session.post(
480                 self.JOB_STATUS_URL,
481                 data={'ids[]': job_id}
482             )
483             data = response.json()[job_id]
484             if data['ready']:
485                 assert data['successful']
486                 return
487             sleep(2)
488
489     def upload_file(self, woblink_id, filename, content, form_name, field_name, mime_type):
490         data = {
491             'pubId': woblink_id,
492         }
493         files = {
494             field_name: (filename, content, mime_type)
495         }
496         response = self.session.post(
497             self.UPLOAD_URL % field_name,
498             data=self.with_form_name(data, form_name),
499             files=self.with_form_name(files, form_name),
500         )
501         resp_data = response.json()
502         assert resp_data['success'] is True
503         if 'jobId' in resp_data:
504             self.wait_for_job(resp_data['jobId'])
505
506     def generate_demo(self, woblink_id, file_format, check=True):
507         percent = 10
508         while True:
509             job_id = self.session.get(
510                 self.GENERATE_DEMO_URL % (file_format, woblink_id, percent),
511             ).json()['jobId']
512             try:
513                 self.wait_for_job(job_id)
514             except AssertionError:
515                 if percent < 50:
516                     percent += 10
517                 else:
518                     raise
519             else:
520                 break
521
522         if check:
523             self.wait_for_job(
524                 self.session.get(
525                     self.CHECK_DEMO_URL % (file_format, woblink_id)
526                 ).json()['jobId']
527             )
528
529     def send_epub(self, woblink_id, doc, gallery_path, fundraising=None):
530         from librarian.builders import EpubBuilder
531         content = EpubBuilder(
532             base_url='file://' + gallery_path + '/',
533             fundraising=fundraising or [],
534         ).build(doc).get_file()
535         self.upload_file(
536             woblink_id,
537             doc.meta.url.slug + '.epub',
538             content,
539             'UploadEpub',
540             'epub',
541             'application/epub+zip'
542         )
543         self.generate_demo(woblink_id, 'epub')
544
545     def send_mobi(self, woblink_id, doc, gallery_path, fundraising=None):
546         from librarian.builders import MobiBuilder
547         content = MobiBuilder(
548             base_url='file://' + gallery_path + '/',
549             fundraising=fundraising or [],
550         ).build(doc).get_file()
551         self.upload_file(
552             woblink_id,
553             doc.meta.url.slug + '.mobi',
554             content,
555             'UploadMobi',
556             'mobi',
557             'application/x-mobipocket-ebook'
558         )
559         self.generate_demo(woblink_id, 'mobi', check=False)
560
561     def send_cover(self, woblink_id, doc):
562         from librarian.cover import make_cover
563         # TODO Labe
564         # A5 @ 300ppi.
565         cover = make_cover(doc.meta, cover_class='m-label', width=1748, height=2480)
566         content = io.BytesIO()
567         cover.final_image().save(content, cover.format)
568         content.seek(0)
569         self.upload_file(
570             woblink_id,
571             doc.meta.url.slug + '.jpeg',
572             content,
573             'UploadCover',
574             'cover',
575             cover.mime_type()
576         )