eb0e7b95a9d0d171a3898e100b4e315ee182998c
[wolnelektury.git] / apps / catalogue / views.py
1 # -*- coding: utf-8 -*-
2 # This file is part of Wolnelektury, licensed under GNU Affero GPLv3 or later.
3 # Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
4 #
5 import re
6 import itertools
7 from datetime import datetime
8
9 from django.conf import settings
10 from django.template import RequestContext
11 from django.shortcuts import render_to_response, get_object_or_404
12 from django.http import HttpResponse, HttpResponseRedirect, Http404, HttpResponsePermanentRedirect
13 from django.core.urlresolvers import reverse
14 from django.db.models import Count, Sum, Q
15 from django.contrib.auth.decorators import login_required, user_passes_test
16 from django.utils.datastructures import SortedDict
17 from django.views.decorators.http import require_POST
18 from django.contrib import auth
19 from django.contrib.auth.forms import UserCreationForm, AuthenticationForm
20 from django.utils.http import urlquote_plus
21 from django.views.decorators import cache
22 from django.utils import translation
23 from django.utils.translation import ugettext as _
24 from django.views.generic.list_detail import object_list
25
26 from ajaxable.utils import LazyEncoder, JSONResponse, AjaxableFormView
27
28 from catalogue import models
29 from catalogue import forms
30 from catalogue.utils import (split_tags, AttachmentHttpResponse,
31     async_build_pdf, MultiQuerySet)
32 from catalogue.tasks import touch_tag
33 from pdcounter import models as pdcounter_models
34 from pdcounter import views as pdcounter_views
35 from suggest.forms import PublishingSuggestForm
36 from picture.models import Picture
37
38 from os import path
39
40 staff_required = user_passes_test(lambda user: user.is_staff)
41
42
43 def catalogue(request):
44     tags = models.Tag.objects.exclude(
45         category__in=('set', 'book')).exclude(book_count=0)
46     tags = list(tags)
47     for tag in tags:
48         tag.count = tag.book_count
49     categories = split_tags(tags)
50     fragment_tags = categories.get('theme', [])
51
52     return render_to_response('catalogue/catalogue.html', locals(),
53         context_instance=RequestContext(request))
54
55
56 def book_list(request, filter=None, template_name='catalogue/book_list.html',
57         context=None):
58     """ generates a listing of all books, optionally filtered with a test function """
59
60     books_by_author, orphans, books_by_parent = models.Book.book_list(filter)
61     books_nav = SortedDict()
62     for tag in books_by_author:
63         if books_by_author[tag]:
64             books_nav.setdefault(tag.sort_key[0], []).append(tag)
65
66     return render_to_response(template_name, locals(),
67         context_instance=RequestContext(request))
68
69
70 def audiobook_list(request):
71     return book_list(request, Q(media__type='mp3') | Q(media__type='ogg'),
72                      template_name='catalogue/audiobook_list.html')
73
74
75 def daisy_list(request):
76     return book_list(request, Q(media__type='daisy'),
77                      template_name='catalogue/daisy_list.html')
78
79
80 def collection(request, slug):
81     coll = get_object_or_404(models.Collection, slug=slug)
82     slugs = coll.book_slugs.split()
83     # allow URIs
84     slugs = [slug.rstrip('/').rsplit('/', 1)[-1] if '/' in slug else slug
85                 for slug in slugs]
86     return book_list(request, Q(slug__in=slugs),
87                      template_name='catalogue/collection.html',
88                      context={'collection': coll})
89
90
91 def differentiate_tags(request, tags, ambiguous_slugs):
92     beginning = '/'.join(tag.url_chunk for tag in tags)
93     unparsed = '/'.join(ambiguous_slugs[1:])
94     options = []
95     for tag in models.Tag.objects.exclude(category='book').filter(slug=ambiguous_slugs[0]):
96         options.append({
97             'url_args': '/'.join((beginning, tag.url_chunk, unparsed)).strip('/'),
98             'tags': [tag]
99         })
100     return render_to_response('catalogue/differentiate_tags.html',
101                 {'tags': tags, 'options': options, 'unparsed': ambiguous_slugs[1:]},
102                 context_instance=RequestContext(request))
103
104
105 def tagged_object_list(request, tags=''):
106     #    import pdb; pdb.set_trace()
107     try:
108         tags = models.Tag.get_tag_list(tags)
109     except models.Tag.DoesNotExist:
110         chunks = tags.split('/')
111         if len(chunks) == 2 and chunks[0] == 'autor':
112             return pdcounter_views.author_detail(request, chunks[1])
113         else:
114             raise Http404
115     except models.Tag.MultipleObjectsReturned, e:
116         return differentiate_tags(request, e.tags, e.ambiguous_slugs)
117     except models.Tag.UrlDeprecationWarning, e:
118         return HttpResponsePermanentRedirect(reverse('tagged_object_list', args=['/'.join(tag.url_chunk for tag in e.tags)]))
119
120     try:
121         if len(tags) > settings.MAX_TAG_LIST:
122             raise Http404
123     except AttributeError:
124         pass
125
126     if len([tag for tag in tags if tag.category == 'book']):
127         raise Http404
128
129     theme_is_set = [tag for tag in tags if tag.category == 'theme']
130     shelf_is_set = [tag for tag in tags if tag.category == 'set']
131     only_shelf = shelf_is_set and len(tags) == 1
132     only_my_shelf = only_shelf and request.user.is_authenticated() and request.user == tags[0].user
133
134     objects = only_author = None
135     categories = {}
136
137     if theme_is_set:
138         shelf_tags = [tag for tag in tags if tag.category == 'set']
139         fragment_tags = [tag for tag in tags if tag.category != 'set']
140         fragments = models.Fragment.tagged.with_all(fragment_tags)
141
142         if shelf_tags:
143             books = models.Book.tagged.with_all(shelf_tags).order_by()
144             l_tags = models.Tag.objects.filter(category='book', slug__in=[book.book_tag_slug() for book in books])
145             fragments = models.Fragment.tagged.with_any(l_tags, fragments)
146
147         # newtagging goes crazy if we just try:
148         #related_tags = models.Tag.objects.usage_for_queryset(fragments, counts=True,
149         #                    extra={'where': ["catalogue_tag.category != 'book'"]})
150         fragment_keys = [fragment.pk for fragment in fragments]
151         if fragment_keys:
152             related_tags = models.Fragment.tags.usage(counts=True,
153                                 filters={'pk__in': fragment_keys},
154                                 extra={'where': ["catalogue_tag.category != 'book'"]})
155             related_tags = (tag for tag in related_tags if tag not in fragment_tags)
156             categories = split_tags(related_tags)
157
158             objects = fragments
159     else:
160         if shelf_is_set:
161             objects = models.Book.tagged.with_all(tags)
162         else:
163             objects = models.Book.tagged_top_level(tags)
164
165         # get related tags from `tag_counter` and `theme_counter`
166         related_counts = {}
167         tags_pks = [tag.pk for tag in tags]
168         for book in objects:
169             for tag_pk, value in itertools.chain(book.tag_counter.iteritems(), book.theme_counter.iteritems()):
170                 if tag_pk in tags_pks:
171                     continue
172                 related_counts[tag_pk] = related_counts.get(tag_pk, 0) + value
173         related_tags = models.Tag.objects.filter(pk__in=related_counts.keys())
174         related_tags = [tag for tag in related_tags if tag not in tags]
175         for tag in related_tags:
176             tag.count = related_counts[tag.pk]
177
178         categories = split_tags(related_tags)
179         del related_tags
180
181     if not objects:
182         only_author = len(tags) == 1 and tags[0].category == 'author'
183         objects = models.Book.objects.none()
184
185     # Add pictures
186     objects = MultiQuerySet(Picture.tagged.with_all(tags), objects)
187
188     return render_to_response('catalogue/tagged_object_list.html',
189         {
190             'object_list': objects,
191             'categories': categories,
192             'only_shelf': only_shelf,
193             'only_author': only_author,
194             'only_my_shelf': only_my_shelf,
195             'formats_form': forms.DownloadFormatsForm(),
196             'tags': tags,
197         },
198         context_instance=RequestContext(request))
199
200
201 def book_fragments(request, slug, theme_slug):
202     book = get_object_or_404(models.Book, slug=slug)
203
204     book_tag = book.book_tag()
205     theme = get_object_or_404(models.Tag, slug=theme_slug, category='theme')
206     fragments = models.Fragment.tagged.with_all([book_tag, theme])
207
208     return render_to_response('catalogue/book_fragments.html', locals(),
209         context_instance=RequestContext(request))
210
211
212 def book_detail(request, slug):
213     try:
214         book = models.Book.objects.get(slug=slug)
215     except models.Book.DoesNotExist:
216         return pdcounter_views.book_stub_detail(request, kwargs['slug'])
217
218     book_tag = book.book_tag()
219     tags = list(book.tags.filter(~Q(category='set')))
220     categories = split_tags(tags)
221     book_children = book.children.all().order_by('parent_number', 'sort_key')
222
223     _book = book
224     parents = []
225     while _book.parent:
226         parents.append(_book.parent)
227         _book = _book.parent
228     parents = reversed(parents)
229
230     theme_counter = book.theme_counter
231     book_themes = models.Tag.objects.filter(pk__in=theme_counter.keys())
232     for tag in book_themes:
233         tag.count = theme_counter[tag.pk]
234
235     extra_info = book.get_extra_info_value()
236     hide_about = extra_info.get('about', '').startswith('http://wiki.wolnepodreczniki.pl')
237
238     custom_pdf_form = forms.CustomPDFForm()
239     return render_to_response('catalogue/book_detail.html', locals(),
240         context_instance=RequestContext(request))
241
242
243 def player(request, slug):
244     book = get_object_or_404(models.Book, slug=slug)
245     if not book.has_media('mp3'):
246         raise Http404
247
248     ogg_files = {}
249     for m in book.media.filter(type='ogg').order_by():
250         ogg_files[m.name] = m
251
252     audiobooks = []
253     have_oggs = True
254     projects = set()
255     for mp3 in book.media.filter(type='mp3'):
256         # ogg files are always from the same project
257         meta = mp3.get_extra_info_value()
258         project = meta.get('project')
259         if not project:
260             # temporary fallback
261             project = u'CzytamySłuchając'
262
263         projects.add((project, meta.get('funded_by', '')))
264
265         media = {'mp3': mp3}
266
267         ogg = ogg_files.get(mp3.name)
268         if ogg:
269             media['ogg'] = ogg
270         else:
271             have_oggs = False
272         audiobooks.append(media)
273     print audiobooks
274
275     projects = sorted(projects)
276
277     return render_to_response('catalogue/player.html', locals(),
278         context_instance=RequestContext(request))
279
280
281 def book_text(request, slug):
282     book = get_object_or_404(models.Book, slug=slug)
283
284     if not book.has_html_file():
285         raise Http404
286     book_themes = {}
287     for fragment in book.fragments.all():
288         for theme in fragment.tags.filter(category='theme'):
289             book_themes.setdefault(theme, []).append(fragment)
290
291     book_themes = book_themes.items()
292     book_themes.sort(key=lambda s: s[0].sort_key)
293     return render_to_response('catalogue/book_text.html', locals(),
294         context_instance=RequestContext(request))
295
296
297 # ==========
298 # = Search =
299 # ==========
300
301 def _no_diacritics_regexp(query):
302     """ returns a regexp for searching for a query without diacritics
303
304     should be locale-aware """
305     names = {
306         u'a':u'aąĄ', u'c':u'cćĆ', u'e':u'eęĘ', u'l': u'lłŁ', u'n':u'nńŃ', u'o':u'oóÓ', u's':u'sśŚ', u'z':u'zźżŹŻ',
307         u'ą':u'ąĄ', u'ć':u'ćĆ', u'ę':u'ęĘ', u'ł': u'łŁ', u'ń':u'ńŃ', u'ó':u'óÓ', u'ś':u'śŚ', u'ź':u'źŹ', u'ż':u'żŻ'
308         }
309     def repl(m):
310         l = m.group()
311         return u"(%s)" % '|'.join(names[l])
312     return re.sub(u'[%s]' % (u''.join(names.keys())), repl, query)
313
314 def unicode_re_escape(query):
315     """ Unicode-friendly version of re.escape """
316     return re.sub('(?u)(\W)', r'\\\1', query)
317
318 def _word_starts_with(name, prefix):
319     """returns a Q object getting models having `name` contain a word
320     starting with `prefix`
321
322     We define word characters as alphanumeric and underscore, like in JS.
323
324     Works for MySQL, PostgreSQL, Oracle.
325     For SQLite, _sqlite* version is substituted for this.
326     """
327     kwargs = {}
328
329     prefix = _no_diacritics_regexp(unicode_re_escape(prefix))
330     # can't use [[:<:]] (word start),
331     # but we want both `xy` and `(xy` to catch `(xyz)`
332     kwargs['%s__iregex' % name] = u"(^|[^[:alnum:]_])%s" % prefix
333
334     return Q(**kwargs)
335
336
337 def _word_starts_with_regexp(prefix):
338     prefix = _no_diacritics_regexp(unicode_re_escape(prefix))
339     return ur"(^|(?<=[^\wąćęłńóśźżĄĆĘŁŃÓŚŹŻ]))%s" % prefix
340
341
342 def _sqlite_word_starts_with(name, prefix):
343     """ version of _word_starts_with for SQLite
344
345     SQLite in Django uses Python re module
346     """
347     kwargs = {}
348     kwargs['%s__iregex' % name] = _word_starts_with_regexp(prefix)
349     return Q(**kwargs)
350
351
352 if hasattr(settings, 'DATABASES'):
353     if settings.DATABASES['default']['ENGINE'] == 'django.db.backends.sqlite3':
354         _word_starts_with = _sqlite_word_starts_with
355 elif settings.DATABASE_ENGINE == 'sqlite3':
356     _word_starts_with = _sqlite_word_starts_with
357
358
359 class App():
360     def __init__(self, name, view):
361         self.name = name
362         self._view = view
363         self.lower = name.lower()
364         self.category = 'application'
365     def view(self):
366         return reverse(*self._view)
367
368 _apps = (
369     App(u'Leśmianator', (u'lesmianator', )),
370     )
371
372
373 def _tags_starting_with(prefix, user=None):
374     prefix = prefix.lower()
375     # PD counter
376     book_stubs = pdcounter_models.BookStub.objects.filter(_word_starts_with('title', prefix))
377     authors = pdcounter_models.Author.objects.filter(_word_starts_with('name', prefix))
378
379     books = models.Book.objects.filter(_word_starts_with('title', prefix))
380     tags = models.Tag.objects.filter(_word_starts_with('name', prefix))
381     if user and user.is_authenticated():
382         tags = tags.filter(~Q(category='book') & (~Q(category='set') | Q(user=user)))
383     else:
384         tags = tags.filter(~Q(category='book') & ~Q(category='set'))
385
386     prefix_regexp = re.compile(_word_starts_with_regexp(prefix))
387     return list(books) + list(tags) + [app for app in _apps if prefix_regexp.search(app.lower)] + list(book_stubs) + list(authors)
388
389
390 def _get_result_link(match, tag_list):
391     if isinstance(match, models.Tag):
392         return reverse('catalogue.views.tagged_object_list',
393             kwargs={'tags': '/'.join(tag.url_chunk for tag in tag_list + [match])}
394         )
395     elif isinstance(match, App):
396         return match.view()
397     else:
398         return match.get_absolute_url()
399
400
401 def _get_result_type(match):
402     if isinstance(match, models.Book) or isinstance(match, pdcounter_models.BookStub):
403         type = 'book'
404     else:
405         type = match.category
406     return type
407
408
409 def books_starting_with(prefix):
410     prefix = prefix.lower()
411     return models.Book.objects.filter(_word_starts_with('title', prefix))
412
413
414 def find_best_matches(query, user=None):
415     """ Finds a models.Book, Tag, models.BookStub or Author best matching a query.
416
417     Returns a with:
418       - zero elements when nothing is found,
419       - one element when a best result is found,
420       - more then one element on multiple exact matches
421
422     Raises a ValueError on too short a query.
423     """
424
425     query = query.lower()
426     if len(query) < 2:
427         raise ValueError("query must have at least two characters")
428
429     result = tuple(_tags_starting_with(query, user))
430     # remove pdcounter stuff
431     book_titles = set(match.pretty_title().lower() for match in result
432                       if isinstance(match, models.Book))
433     authors = set(match.name.lower() for match in result
434                   if isinstance(match, models.Tag) and match.category=='author')
435     result = tuple(res for res in result if not (
436                  (isinstance(res, pdcounter_models.BookStub) and res.pretty_title().lower() in book_titles)
437                  or (isinstance(res, pdcounter_models.Author) and res.name.lower() in authors)
438              ))
439
440     exact_matches = tuple(res for res in result if res.name.lower() == query)
441     if exact_matches:
442         return exact_matches
443     else:
444         return tuple(result)[:1]
445
446
447 def search(request):
448     tags = request.GET.get('tags', '')
449     prefix = request.GET.get('q', '')
450
451     try:
452         tag_list = models.Tag.get_tag_list(tags)
453     except:
454         tag_list = []
455
456     try:
457         result = find_best_matches(prefix, request.user)
458     except ValueError:
459         return render_to_response('catalogue/search_too_short.html', {'tags':tag_list, 'prefix':prefix},
460             context_instance=RequestContext(request))
461
462     if len(result) == 1:
463         return HttpResponseRedirect(_get_result_link(result[0], tag_list))
464     elif len(result) > 1:
465         return render_to_response('catalogue/search_multiple_hits.html',
466             {'tags':tag_list, 'prefix':prefix, 'results':((x, _get_result_link(x, tag_list), _get_result_type(x)) for x in result)},
467             context_instance=RequestContext(request))
468     else:
469         form = PublishingSuggestForm(initial={"books": prefix + ", "})
470         return render_to_response('catalogue/search_no_hits.html',
471             {'tags':tag_list, 'prefix':prefix, "pubsuggest_form": form},
472             context_instance=RequestContext(request))
473
474
475 def tags_starting_with(request):
476     prefix = request.GET.get('q', '')
477     # Prefix must have at least 2 characters
478     if len(prefix) < 2:
479         return HttpResponse('')
480     tags_list = []
481     result = ""
482     for tag in _tags_starting_with(prefix, request.user):
483         if not tag.name in tags_list:
484             result += "\n" + tag.name
485             tags_list.append(tag.name)
486     return HttpResponse(result)
487
488 def json_tags_starting_with(request, callback=None):
489     # Callback for JSONP
490     prefix = request.GET.get('q', '')
491     callback = request.GET.get('callback', '')
492     # Prefix must have at least 2 characters
493     if len(prefix) < 2:
494         return HttpResponse('')
495     tags_list = []
496     for tag in _tags_starting_with(prefix, request.user):
497         if not tag.name in tags_list:
498             tags_list.append(tag.name)
499     if request.GET.get('mozhint', ''):
500         result = [prefix, tags_list]
501     else:
502         result = {"matches": tags_list}
503     return JSONResponse(result, callback)
504
505
506 # =========
507 # = Admin =
508 # =========
509 @login_required
510 @staff_required
511 def import_book(request):
512     """docstring for import_book"""
513     book_import_form = forms.BookImportForm(request.POST, request.FILES)
514     if book_import_form.is_valid():
515         try:
516             book_import_form.save()
517         except:
518             import sys
519             import pprint
520             import traceback
521             info = sys.exc_info()
522             exception = pprint.pformat(info[1])
523             tb = '\n'.join(traceback.format_tb(info[2]))
524             return HttpResponse(_("An error occurred: %(exception)s\n\n%(tb)s") % {'exception':exception, 'tb':tb}, mimetype='text/plain')
525         return HttpResponse(_("Book imported successfully"))
526     else:
527         return HttpResponse(_("Error importing file: %r") % book_import_form.errors)
528
529
530 # info views for API
531
532 def book_info(request, id, lang='pl'):
533     book = get_object_or_404(models.Book, id=id)
534     # set language by hand
535     translation.activate(lang)
536     return render_to_response('catalogue/book_info.html', locals(),
537         context_instance=RequestContext(request))
538
539
540 def tag_info(request, id):
541     tag = get_object_or_404(models.Tag, id=id)
542     return HttpResponse(tag.description)
543
544
545 def download_zip(request, format, slug=None):
546     url = None
547     if format in models.Book.ebook_formats:
548         url = models.Book.zip_format(format)
549     elif format in ('mp3', 'ogg') and slug is not None:
550         book = get_object_or_404(models.Book, slug=slug)
551         url = book.zip_audiobooks(format)
552     else:
553         raise Http404('No format specified for zip package')
554     return HttpResponseRedirect(urlquote_plus(settings.MEDIA_URL + url, safe='/?='))
555
556
557 def download_custom_pdf(request, slug, method='GET'):
558     book = get_object_or_404(models.Book, slug=slug)
559
560     if request.method == method:
561         form = forms.CustomPDFForm(method == 'GET' and request.GET or request.POST)
562         if form.is_valid():
563             cust = form.customizations
564             pdf_file = models.get_customized_pdf_path(book, cust)
565
566             if not path.exists(pdf_file):
567                 result = async_build_pdf.delay(book.id, cust, pdf_file)
568                 result.wait()
569             return AttachmentHttpResponse(file_name=("%s.pdf" % book.slug), file_path=pdf_file, mimetype="application/pdf")
570         else:
571             raise Http404(_('Incorrect customization options for PDF'))
572     else:
573         raise Http404(_('Bad method'))
574
575
576 class CustomPDFFormView(AjaxableFormView):
577     form_class = forms.CustomPDFForm
578     title = _('Download custom PDF')
579     submit = _('Download')
580
581     def __call__(self, request):
582         from copy import copy
583         if request.method == 'POST':
584             request.GET = copy(request.GET)
585             request.GET['next'] = "%s?%s" % (reverse('catalogue.views.download_custom_pdf', args=[request.GET['slug']]),
586                                              request.POST.urlencode())
587         return super(CustomPDFFormView, self).__call__(request)
588         
589
590     def success(self, *args):
591         pass