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 from collections import OrderedDict
   6 import re
   7
   8 from django.conf import settings
   9 from django.core.cache import get_cache
  10 from django.template import RequestContext
  11 from django.template.loader import render_to_string
  12 from django.shortcuts import render_to_response, get_object_or_404
  13 from django.http import HttpResponse, HttpResponseRedirect, Http404, HttpResponsePermanentRedirect, JsonResponse
  14 from django.core.urlresolvers import reverse
  15 from django.db.models import Q
  16 from django.contrib.auth.decorators import login_required, user_passes_test
  17 from django.utils.http import urlquote_plus
  18 from django.utils import translation
  19 from django.utils.translation import get_language, ugettext as _, ugettext_lazy
  20 from django.views.decorators.vary import vary_on_headers
  21
  22 from ajaxable.utils import AjaxableFormView
  23 from catalogue import models
  24 from catalogue import forms
  25 from .helpers import get_related_tags, get_fragment_related_tags, tags_usage_for_books, tags_usage_for_works, tags_usage_for_fragments
  26 from catalogue.utils import split_tags, MultiQuerySet, SortedMultiQuerySet
  27 from catalogue.templatetags.catalogue_tags import tag_list, collection_list
  28 from pdcounter import models as pdcounter_models
  29 from pdcounter import views as pdcounter_views
  30 from suggest.forms import PublishingSuggestForm
  31 from picture.models import Picture, PictureArea
  32 from picture.views import picture_list_thumb
  33
  34 staff_required = user_passes_test(lambda user: user.is_staff)
  35 permanent_cache = get_cache('permanent')
  36
  37
  38 @vary_on_headers('X-Requested-With')
  39 def catalogue(request):
  40     #cache_key = 'catalogue.catalogue/' + get_language()
  41     #output = permanent_cache.get(cache_key)
  42     output = None
  43
  44     if output is None:
  45         common_categories = ('author',)
  46         split_categories = ('epoch', 'genre', 'kind')
  47
  48         categories = split_tags(tags_usage_for_works(common_categories))
  49         book_categories = split_tags(tags_usage_for_books(split_categories))
  50         picture_categories = split_tags(
  51             models.Tag.objects.usage_for_model(Picture, counts=True).filter(
  52                 category__in=split_categories))
  53         # we want global usage for themes
  54         fragment_tags = list(tags_usage_for_fragments(('theme',)))
  55         collections = models.Collection.objects.all()
  56
  57         render_tag_list = lambda x: render_to_string(
  58             'catalogue/tag_list.html', tag_list(x))
  59
  60         def render_split(with_books, with_pictures):
  61             ctx = {}
  62             if with_books:
  63                 ctx['books'] = render_tag_list(with_books)
  64             if with_pictures:
  65                 ctx['pictures'] = render_tag_list(with_pictures)
  66             return render_to_string('catalogue/tag_list_split.html', ctx)
  67
  68         output = {}
  69         output['theme'] = render_tag_list(fragment_tags)
  70         for category in common_categories:
  71             output[category] = render_tag_list(categories.get(category, []))
  72         for category in split_categories:
  73             output[category] = render_split(
  74                 book_categories.get(category, []),
  75                 picture_categories.get(category, []))
  76
  77         output['collections'] = render_to_string(
  78             'catalogue/collection_list.html', collection_list(collections))
  79         #permanent_cache.set(cache_key, output)
  80     if request.is_ajax():
  81         return JsonResponse(output)
  82     else:
  83         return render_to_response('catalogue/catalogue.html', locals(),
  84             context_instance=RequestContext(request))
  85
  86
  87 def book_list(request, filter=None, get_filter=None,
  88         template_name='catalogue/book_list.html',
  89         nav_template_name='catalogue/snippets/book_list_nav.html',
  90         list_template_name='catalogue/snippets/book_list.html',
  91         cache_key='catalogue.book_list',
  92         context=None,
  93         ):
  94     """ generates a listing of all books, optionally filtered with a test function """
  95     cache_key = "%s/%s" % (cache_key, get_language())
  96     cached = permanent_cache.get(cache_key)
  97     if cached is not None:
  98         rendered_nav, rendered_book_list = cached
  99     else:
 100         if get_filter:
 101             filter = get_filter()
 102         books_by_author, orphans, books_by_parent = models.Book.book_list(filter)
 103         books_nav = OrderedDict()
 104         for tag in books_by_author:
 105             if books_by_author[tag]:
 106                 books_nav.setdefault(tag.sort_key[0], []).append(tag)
 107         rendered_nav = render_to_string(nav_template_name, locals())
 108         rendered_book_list = render_to_string(list_template_name, locals())
 109         permanent_cache.set(cache_key, (rendered_nav, rendered_book_list))
 110     return render_to_response(template_name, locals(),
 111         context_instance=RequestContext(request))
 112
 113
 114 def audiobook_list(request):
 115     return book_list(request, Q(media__type='mp3') | Q(media__type='ogg'),
 116                      template_name='catalogue/audiobook_list.html',
 117                      list_template_name='catalogue/snippets/audiobook_list.html',
 118                      cache_key='catalogue.audiobook_list')
 119
 120
 121 def daisy_list(request):
 122     return book_list(request, Q(media__type='daisy'),
 123                      template_name='catalogue/daisy_list.html',
 124                      cache_key='catalogue.daisy_list')
 125
 126
 127 def collection(request, slug):
 128     coll = get_object_or_404(models.Collection, slug=slug)
 129     if coll.kind == 'book':
 130         view = book_list
 131         tmpl = "catalogue/collection.html"
 132     elif coll.kind == 'picture':
 133         view = picture_list_thumb
 134         tmpl = "picture/collection.html"
 135     else:
 136         raise ValueError('How do I show this kind of collection? %s' % coll.kind)
 137     return view(request, get_filter=coll.get_query,
 138                      template_name=tmpl,
 139                      cache_key='catalogue.collection:%s' % coll.slug,
 140                      context={'collection': coll})
 141
 142
 143 def differentiate_tags(request, tags, ambiguous_slugs):
 144     beginning = '/'.join(tag.url_chunk for tag in tags)
 145     unparsed = '/'.join(ambiguous_slugs[1:])
 146     options = []
 147     for tag in models.Tag.objects.filter(slug=ambiguous_slugs[0]):
 148         options.append({
 149             'url_args': '/'.join((beginning, tag.url_chunk, unparsed)).strip('/'),
 150             'tags': [tag]
 151         })
 152     return render_to_response('catalogue/differentiate_tags.html',
 153                 {'tags': tags, 'options': options, 'unparsed': ambiguous_slugs[1:]},
 154                 context_instance=RequestContext(request))
 155
 156
 157 # TODO: Rewrite this hellish piece of code which tries to do everything
 158 def tagged_object_list(request, tags=''):
 159     # preliminary tests and conditions
 160     try:
 161         tags = models.Tag.get_tag_list(tags)
 162     except models.Tag.DoesNotExist:
 163         # Perhaps the user is asking about an author in Public Domain
 164         # counter (they are not represented in tags)
 165         chunks = tags.split('/')
 166         if len(chunks) == 2 and chunks[0] == 'autor':
 167             return pdcounter_views.author_detail(request, chunks[1])
 168         else:
 169             raise Http404
 170     except models.Tag.MultipleObjectsReturned, e:
 171         # Ask the user to disambiguate
 172         return differentiate_tags(request, e.tags, e.ambiguous_slugs)
 173     except models.Tag.UrlDeprecationWarning, e:
 174         return HttpResponsePermanentRedirect(reverse('tagged_object_list', args=['/'.join(tag.url_chunk for tag in e.tags)]))
 175
 176     try:
 177         if len(tags) > settings.MAX_TAG_LIST:
 178             raise Http404
 179     except AttributeError:
 180         pass
 181
 182     if len([tag for tag in tags if tag.category == 'book']):
 183         raise Http404
 184
 185     # beginning of digestion
 186     theme_is_set = [tag for tag in tags if tag.category == 'theme']
 187     shelf_is_set = [tag for tag in tags if tag.category == 'set']
 188     only_shelf = shelf_is_set and len(tags) == 1
 189     only_my_shelf = only_shelf and request.user.is_authenticated() and request.user == tags[0].user
 190
 191
 192     objects =  None
 193     categories = {}
 194     object_queries = []
 195
 196     if theme_is_set:
 197         shelf_tags = [tag for tag in tags if tag.category == 'set']
 198         fragment_tags = [tag for tag in tags if tag.category != 'set']
 199         fragments = models.Fragment.tagged.with_all(fragment_tags)
 200         areas = PictureArea.tagged.with_all(fragment_tags)
 201
 202         if shelf_tags:
 203             # FIXME: book tags here
 204             books = models.Book.tagged.with_all(shelf_tags).order_by()
 205             l_tags = models.Tag.objects.filter(category='book',
 206                 slug__in=[book.book_tag_slug() for book in books.iterator()])
 207             fragments = models.Fragment.tagged.with_any(l_tags, fragments)
 208
 209         related_tags = get_fragment_related_tags(tags)
 210         categories = split_tags(related_tags, categories)
 211         object_queries.insert(0, fragments)
 212
 213         area_keys = [area.pk for area in areas.iterator()]
 214         if area_keys:
 215             related_tags = PictureArea.tags.usage(counts=True,
 216                                                          filters={'pk__in': area_keys})
 217             related_tags = (tag for tag in related_tags if tag not in fragment_tags)
 218
 219             categories = split_tags(related_tags, categories)
 220
 221         # we want the Pictures to go first
 222         object_queries.insert(0, areas)
 223         objects = MultiQuerySet(*object_queries)
 224     else:
 225         if shelf_is_set:
 226             books = models.Book.tagged.with_all(tags).order_by(
 227                 'sort_key_author', 'title')
 228         else:
 229             books = models.Book.tagged_top_level(tags).order_by(
 230                 'sort_key_author', 'title')
 231
 232         pictures = Picture.tagged.with_all(tags).order_by(
 233             'sort_key_author', 'title')
 234
 235         categories = split_tags(get_related_tags(tags))
 236
 237         objects = SortedMultiQuerySet(pictures, books,
 238             order_by=('sort_key_author', 'title'))
 239
 240
 241     if not objects:
 242         objects = models.Book.objects.none()
 243
 244     return render_to_response('catalogue/tagged_object_list.html',
 245         {
 246             'object_list': objects,
 247             'categories': categories,
 248             'only_shelf': only_shelf,
 249             'only_my_shelf': only_my_shelf,
 250             'formats_form': forms.DownloadFormatsForm(),
 251             'tags': tags,
 252             'theme_is_set': theme_is_set,
 253         },
 254         context_instance=RequestContext(request))
 255
 256
 257 def book_fragments(request, slug, theme_slug):
 258     book = get_object_or_404(models.Book, slug=slug)
 259     theme = get_object_or_404(models.Tag, slug=theme_slug, category='theme')
 260     fragments = models.Fragment.tagged.with_all([theme]).filter(
 261         Q(book=book) | Q(book__ancestor=book))
 262
 263     return render_to_response('catalogue/book_fragments.html', locals(),
 264         context_instance=RequestContext(request))
 265
 266
 267 def book_detail(request, slug):
 268     try:
 269         book = models.Book.objects.get(slug=slug)
 270     except models.Book.DoesNotExist:
 271         return pdcounter_views.book_stub_detail(request, slug)
 272
 273     book_children = book.children.all().order_by('parent_number', 'sort_key')
 274     return render_to_response('catalogue/book_detail.html', locals(),
 275         context_instance=RequestContext(request))
 276
 277
 278 def player(request, slug):
 279     book = get_object_or_404(models.Book, slug=slug)
 280     if not book.has_media('mp3'):
 281         raise Http404
 282
 283     ogg_files = {}
 284     for m in book.media.filter(type='ogg').order_by().iterator():
 285         ogg_files[m.name] = m
 286
 287     audiobooks = []
 288     have_oggs = True
 289     projects = set()
 290     for mp3 in book.media.filter(type='mp3').iterator():
 291         # ogg files are always from the same project
 292         meta = mp3.extra_info
 293         project = meta.get('project')
 294         if not project:
 295             # temporary fallback
 296             project = u'CzytamySłuchając'
 297
 298         projects.add((project, meta.get('funded_by', '')))
 299
 300         media = {'mp3': mp3}
 301
 302         ogg = ogg_files.get(mp3.name)
 303         if ogg:
 304             media['ogg'] = ogg
 305         else:
 306             have_oggs = False
 307         audiobooks.append(media)
 308
 309     projects = sorted(projects)
 310
 311     extra_info = book.extra_info
 312
 313     return render_to_response('catalogue/player.html', locals(),
 314         context_instance=RequestContext(request))
 315
 316
 317 def book_text(request, slug):
 318     book = get_object_or_404(models.Book, slug=slug)
 319
 320     if not book.has_html_file():
 321         raise Http404
 322     return render_to_response('catalogue/book_text.html', locals(),
 323         context_instance=RequestContext(request))
 324
 325
 326 # ==========
 327 # = Search =
 328 # ==========
 329
 330 def _no_diacritics_regexp(query):
 331     """ returns a regexp for searching for a query without diacritics
 332
 333     should be locale-aware """
 334     names = {
 335         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',
 336         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'
 337         }
 338     def repl(m):
 339         l = m.group()
 340         return u"(%s)" % '|'.join(names[l])
 341     return re.sub(u'[%s]' % (u''.join(names.keys())), repl, query)
 342
 343 def unicode_re_escape(query):
 344     """ Unicode-friendly version of re.escape """
 345     return re.sub('(?u)(\W)', r'\\\1', query)
 346
 347 def _word_starts_with(name, prefix):
 348     """returns a Q object getting models having `name` contain a word
 349     starting with `prefix`
 350
 351     We define word characters as alphanumeric and underscore, like in JS.
 352
 353     Works for MySQL, PostgreSQL, Oracle.
 354     For SQLite, _sqlite* version is substituted for this.
 355     """
 356     kwargs = {}
 357
 358     prefix = _no_diacritics_regexp(unicode_re_escape(prefix))
 359     # can't use [[:<:]] (word start),
 360     # but we want both `xy` and `(xy` to catch `(xyz)`
 361     kwargs['%s__iregex' % name] = u"(^|[^[:alnum:]_])%s" % prefix
 362
 363     return Q(**kwargs)
 364
 365
 366 def _word_starts_with_regexp(prefix):
 367     prefix = _no_diacritics_regexp(unicode_re_escape(prefix))
 368     return ur"(^|(?<=[^\wąćęłńóśźżĄĆĘŁŃÓŚŹŻ]))%s" % prefix
 369
 370
 371 def _sqlite_word_starts_with(name, prefix):
 372     """ version of _word_starts_with for SQLite
 373
 374     SQLite in Django uses Python re module
 375     """
 376     kwargs = {}
 377     kwargs['%s__iregex' % name] = _word_starts_with_regexp(prefix)
 378     return Q(**kwargs)
 379
 380
 381 if hasattr(settings, 'DATABASES'):
 382     if settings.DATABASES['default']['ENGINE'] == 'django.db.backends.sqlite3':
 383         _word_starts_with = _sqlite_word_starts_with
 384 elif settings.DATABASE_ENGINE == 'sqlite3':
 385     _word_starts_with = _sqlite_word_starts_with
 386
 387
 388 class App():
 389     def __init__(self, name, view):
 390         self.name = name
 391         self._view = view
 392         self.lower = name.lower()
 393         self.category = 'application'
 394     def view(self):
 395         return reverse(*self._view)
 396
 397 _apps = (
 398     App(u'Leśmianator', (u'lesmianator', )),
 399     )
 400
 401
 402 def _tags_starting_with(prefix, user=None):
 403     prefix = prefix.lower()
 404     # PD counter
 405     book_stubs = pdcounter_models.BookStub.objects.filter(_word_starts_with('title', prefix))
 406     authors = pdcounter_models.Author.objects.filter(_word_starts_with('name', prefix))
 407
 408     books = models.Book.objects.filter(_word_starts_with('title', prefix))
 409     tags = models.Tag.objects.filter(_word_starts_with('name', prefix))
 410     if user and user.is_authenticated():
 411         tags = tags.filter(~Q(category='set') | Q(user=user))
 412     else:
 413         tags = tags.exclude(category='set')
 414
 415     prefix_regexp = re.compile(_word_starts_with_regexp(prefix))
 416     return list(books) + list(tags) + [app for app in _apps if prefix_regexp.search(app.lower)] + list(book_stubs) + list(authors)
 417
 418
 419 def _get_result_link(match, tag_list):
 420     if isinstance(match, models.Tag):
 421         return reverse('catalogue.views.tagged_object_list',
 422             kwargs={'tags': '/'.join(tag.url_chunk for tag in tag_list + [match])}
 423         )
 424     elif isinstance(match, App):
 425         return match.view()
 426     else:
 427         return match.get_absolute_url()
 428
 429
 430 def _get_result_type(match):
 431     if isinstance(match, models.Book) or isinstance(match, pdcounter_models.BookStub):
 432         type = 'book'
 433     else:
 434         type = match.category
 435     return type
 436
 437
 438 def books_starting_with(prefix):
 439     prefix = prefix.lower()
 440     return models.Book.objects.filter(_word_starts_with('title', prefix))
 441
 442
 443 def find_best_matches(query, user=None):
 444     """ Finds a models.Book, Tag, models.BookStub or Author best matching a query.
 445
 446     Returns a with:
 447       - zero elements when nothing is found,
 448       - one element when a best result is found,
 449       - more then one element on multiple exact matches
 450
 451     Raises a ValueError on too short a query.
 452     """
 453
 454     query = query.lower()
 455     if len(query) < 2:
 456         raise ValueError("query must have at least two characters")
 457
 458     result = tuple(_tags_starting_with(query, user))
 459     # remove pdcounter stuff
 460     book_titles = set(match.pretty_title().lower() for match in result
 461                       if isinstance(match, models.Book))
 462     authors = set(match.name.lower() for match in result
 463                   if isinstance(match, models.Tag) and match.category == 'author')
 464     result = tuple(res for res in result if not (
 465                  (isinstance(res, pdcounter_models.BookStub) and res.pretty_title().lower() in book_titles)
 466                  or (isinstance(res, pdcounter_models.Author) and res.name.lower() in authors)
 467              ))
 468
 469     exact_matches = tuple(res for res in result if res.name.lower() == query)
 470     if exact_matches:
 471         return exact_matches
 472     else:
 473         return tuple(result)[:1]
 474
 475
 476 def search(request):
 477     tags = request.GET.get('tags', '')
 478     prefix = request.GET.get('q', '')
 479
 480     try:
 481         tag_list = models.Tag.get_tag_list(tags)
 482     except:
 483         tag_list = []
 484
 485     try:
 486         result = find_best_matches(prefix, request.user)
 487     except ValueError:
 488         return render_to_response('catalogue/search_too_short.html', {'tags':tag_list, 'prefix':prefix},
 489             context_instance=RequestContext(request))
 490
 491     if len(result) == 1:
 492         return HttpResponseRedirect(_get_result_link(result[0], tag_list))
 493     elif len(result) > 1:
 494         return render_to_response('catalogue/search_multiple_hits.html',
 495             {'tags':tag_list, 'prefix':prefix, 'results':((x, _get_result_link(x, tag_list), _get_result_type(x)) for x in result)},
 496             context_instance=RequestContext(request))
 497     else:
 498         form = PublishingSuggestForm(initial={"books": prefix + ", "})
 499         return render_to_response('catalogue/search_no_hits.html',
 500             {'tags':tag_list, 'prefix':prefix, "pubsuggest_form": form},
 501             context_instance=RequestContext(request))
 502
 503
 504 def tags_starting_with(request):
 505     prefix = request.GET.get('q', '')
 506     # Prefix must have at least 2 characters
 507     if len(prefix) < 2:
 508         return HttpResponse('')
 509     tags_list = []
 510     result = ""
 511     for tag in _tags_starting_with(prefix, request.user):
 512         if not tag.name in tags_list:
 513             result += "\n" + tag.name
 514             tags_list.append(tag.name)
 515     return HttpResponse(result)
 516
 517 def json_tags_starting_with(request, callback=None):
 518     # Callback for JSONP
 519     prefix = request.GET.get('q', '')
 520     callback = request.GET.get('callback', '')
 521     # Prefix must have at least 2 characters
 522     if len(prefix) < 2:
 523         return HttpResponse('')
 524     tags_list = []
 525     for tag in _tags_starting_with(prefix, request.user):
 526         if not tag.name in tags_list:
 527             tags_list.append(tag.name)
 528     if request.GET.get('mozhint', ''):
 529         result = [prefix, tags_list]
 530     else:
 531         result = {"matches": tags_list}
 532     return JsonResponse(result, callback)
 533
 534
 535 # =========
 536 # = Admin =
 537 # =========
 538 @login_required
 539 @staff_required
 540 def import_book(request):
 541     """docstring for import_book"""
 542     book_import_form = forms.BookImportForm(request.POST, request.FILES)
 543     if book_import_form.is_valid():
 544         try:
 545             book_import_form.save()
 546         except:
 547             import sys
 548             import pprint
 549             import traceback
 550             info = sys.exc_info()
 551             exception = pprint.pformat(info[1])
 552             tb = '\n'.join(traceback.format_tb(info[2]))
 553             return HttpResponse(_("An error occurred: %(exception)s\n\n%(tb)s") % {'exception':exception, 'tb':tb}, mimetype='text/plain')
 554         return HttpResponse(_("Book imported successfully"))
 555     else:
 556         return HttpResponse(_("Error importing file: %r") % book_import_form.errors)
 557
 558
 559 # info views for API
 560
 561 def book_info(request, id, lang='pl'):
 562     book = get_object_or_404(models.Book, id=id)
 563     # set language by hand
 564     translation.activate(lang)
 565     return render_to_response('catalogue/book_info.html', locals(),
 566         context_instance=RequestContext(request))
 567
 568
 569 def tag_info(request, id):
 570     tag = get_object_or_404(models.Tag, id=id)
 571     return HttpResponse(tag.description)
 572
 573
 574 def download_zip(request, format, slug=None):
 575     url = None
 576     if format in models.Book.ebook_formats:
 577         url = models.Book.zip_format(format)
 578     elif format in ('mp3', 'ogg') and slug is not None:
 579         book = get_object_or_404(models.Book, slug=slug)
 580         url = book.zip_audiobooks(format)
 581     else:
 582         raise Http404('No format specified for zip package')
 583     return HttpResponseRedirect(urlquote_plus(settings.MEDIA_URL + url, safe='/?='))
 584
 585
 586 class CustomPDFFormView(AjaxableFormView):
 587     form_class = forms.CustomPDFForm
 588     title = ugettext_lazy('Download custom PDF')
 589     submit = ugettext_lazy('Download')
 590     honeypot = True
 591
 592     def __call__(self, *args, **kwargs):
 593         if settings.NO_CUSTOM_PDF:
 594             raise Http404('Custom PDF is disabled')
 595         return super(CustomPDFFormView, self).__call__(*args, **kwargs)
 596
 597     def form_args(self, request, obj):
 598         """Override to parse view args and give additional args to the form."""
 599         return (obj,), {}
 600
 601     def get_object(self, request, slug, *args, **kwargs):
 602         return get_object_or_404(models.Book, slug=slug)
 603
 604     def context_description(self, request, obj):
 605         return obj.pretty_title()