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
   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, redirect
  13 from django.http import HttpResponse, HttpResponseRedirect, Http404, HttpResponsePermanentRedirect
  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.datastructures import SortedDict
  18 from django.utils.http import urlquote_plus
  19 from django.utils import translation
  20 from django.utils.translation import ugettext as _, ugettext_lazy
  21 from django.views.decorators.vary import vary_on_headers
  22
  23 from ajaxable.utils import JSONResponse, AjaxableFormView
  24 from catalogue import models
  25 from catalogue import forms
  26 from catalogue.utils import split_tags, MultiQuerySet
  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
  32
  33 staff_required = user_passes_test(lambda user: user.is_staff)
  34 permanent_cache = get_cache('permanent')
  35
  36
  37 @vary_on_headers('X-Requested-With')
  38 def catalogue(request):
  39     cache_key='catalogue.catalogue'
  40     output = permanent_cache.get(cache_key)
  41     if output is None:
  42         tags = models.Tag.objects.exclude(
  43             category__in=('set', 'book')).exclude(book_count=0)
  44         tags = list(tags)
  45         for tag in tags:
  46             tag.count = tag.book_count
  47         categories = split_tags(tags)
  48         fragment_tags = categories.get('theme', [])
  49         collections = models.Collection.objects.all()
  50         render_tag_list = lambda x: render_to_string(
  51             'catalogue/tag_list.html', tag_list(x))
  52         output = {'theme': render_tag_list(fragment_tags)}
  53         for category, tags in categories.items():
  54             output[category] = render_tag_list(tags)
  55         output['collections'] = render_to_string(
  56             'catalogue/collection_list.html', collection_list(collections))
  57         permanent_cache.set(cache_key, output)
  58     if request.is_ajax():
  59         return JSONResponse(output)
  60     else:
  61         return render_to_response('catalogue/catalogue.html', locals(),
  62             context_instance=RequestContext(request))
  63
  64
  65 def book_list(request, filter=None, get_filter=None,
  66         template_name='catalogue/book_list.html',
  67         nav_template_name='catalogue/snippets/book_list_nav.html',
  68         list_template_name='catalogue/snippets/book_list.html',
  69         cache_key='catalogue.book_list',
  70         context=None,
  71         ):
  72     """ generates a listing of all books, optionally filtered with a test function """
  73     cached = permanent_cache.get(cache_key)
  74     if cached is not None:
  75         rendered_nav, rendered_book_list = cached
  76     else:
  77         if get_filter:
  78             filter = get_filter()
  79         books_by_author, orphans, books_by_parent = models.Book.book_list(filter)
  80         books_nav = SortedDict()
  81         for tag in books_by_author:
  82             if books_by_author[tag]:
  83                 books_nav.setdefault(tag.sort_key[0], []).append(tag)
  84         rendered_nav = render_to_string(nav_template_name, locals())
  85         rendered_book_list = render_to_string(list_template_name, locals())
  86         permanent_cache.set(cache_key, (rendered_nav, rendered_book_list))
  87     return render_to_response(template_name, locals(),
  88         context_instance=RequestContext(request))
  89
  90
  91 def audiobook_list(request):
  92     return book_list(request, Q(media__type='mp3') | Q(media__type='ogg'),
  93                      template_name='catalogue/audiobook_list.html',
  94                      list_template_name='catalogue/snippets/audiobook_list.html',
  95                      cache_key='catalogue.audiobook_list')
  96
  97
  98 def daisy_list(request):
  99     return book_list(request, Q(media__type='daisy'),
 100                      template_name='catalogue/daisy_list.html',
 101                      cache_key='catalogue.daisy_list')
 102
 103
 104 def collection(request, slug):
 105     coll = get_object_or_404(models.Collection, slug=slug)
 106     return book_list(request, get_filter=coll.get_query,
 107                      template_name='catalogue/collection.html',
 108                      cache_key='catalogue.collection:%s' % coll.slug,
 109                      context={'collection': coll})
 110
 111
 112 def differentiate_tags(request, tags, ambiguous_slugs):
 113     beginning = '/'.join(tag.url_chunk for tag in tags)
 114     unparsed = '/'.join(ambiguous_slugs[1:])
 115     options = []
 116     for tag in models.Tag.objects.exclude(category='book').filter(slug=ambiguous_slugs[0]):
 117         options.append({
 118             'url_args': '/'.join((beginning, tag.url_chunk, unparsed)).strip('/'),
 119             'tags': [tag]
 120         })
 121     return render_to_response('catalogue/differentiate_tags.html',
 122                 {'tags': tags, 'options': options, 'unparsed': ambiguous_slugs[1:]},
 123                 context_instance=RequestContext(request))
 124
 125
 126 def tagged_object_list(request, tags=''):
 127     try:
 128         tags = models.Tag.get_tag_list(tags)
 129     except models.Tag.DoesNotExist:
 130         chunks = tags.split('/')
 131         if len(chunks) == 2 and chunks[0] == 'autor':
 132             return pdcounter_views.author_detail(request, chunks[1])
 133         else:
 134             raise Http404
 135     except models.Tag.MultipleObjectsReturned, e:
 136         return differentiate_tags(request, e.tags, e.ambiguous_slugs)
 137     except models.Tag.UrlDeprecationWarning, e:
 138         return HttpResponsePermanentRedirect(reverse('tagged_object_list', args=['/'.join(tag.url_chunk for tag in e.tags)]))
 139
 140     try:
 141         if len(tags) > settings.MAX_TAG_LIST:
 142             raise Http404
 143     except AttributeError:
 144         pass
 145
 146     if len([tag for tag in tags if tag.category == 'book']):
 147         raise Http404
 148
 149     theme_is_set = [tag for tag in tags if tag.category == 'theme']
 150     shelf_is_set = [tag for tag in tags if tag.category == 'set']
 151     only_shelf = shelf_is_set and len(tags) == 1
 152     only_my_shelf = only_shelf and request.user.is_authenticated() and request.user == tags[0].user
 153
 154     objects = only_author = None
 155     categories = {}
 156
 157     if theme_is_set:
 158         shelf_tags = [tag for tag in tags if tag.category == 'set']
 159         fragment_tags = [tag for tag in tags if tag.category != 'set']
 160         fragments = models.Fragment.tagged.with_all(fragment_tags)
 161
 162         if shelf_tags:
 163             books = models.Book.tagged.with_all(shelf_tags).order_by()
 164             l_tags = models.Tag.objects.filter(category='book',
 165                 slug__in=[book.book_tag_slug() for book in books.iterator()])
 166             fragments = models.Fragment.tagged.with_any(l_tags, fragments)
 167
 168         # newtagging goes crazy if we just try:
 169         #related_tags = models.Tag.objects.usage_for_queryset(fragments, counts=True,
 170         #                    extra={'where': ["catalogue_tag.category != 'book'"]})
 171         fragment_keys = [fragment.pk for fragment in fragments.iterator()]
 172         if fragment_keys:
 173             related_tags = models.Fragment.tags.usage(counts=True,
 174                                 filters={'pk__in': fragment_keys},
 175                                 extra={'where': ["catalogue_tag.category != 'book'"]})
 176             related_tags = (tag for tag in related_tags if tag not in fragment_tags)
 177             categories = split_tags(related_tags)
 178
 179             objects = fragments
 180     else:
 181         if shelf_is_set:
 182             objects = models.Book.tagged.with_all(tags)
 183         else:
 184             objects = models.Book.tagged_top_level(tags)
 185
 186         # get related tags from `tag_counter` and `theme_counter`
 187         related_counts = {}
 188         tags_pks = [tag.pk for tag in tags]
 189         for book in objects:
 190             for tag_pk, value in itertools.chain(book.tag_counter.iteritems(), book.theme_counter.iteritems()):
 191                 if tag_pk in tags_pks:
 192                     continue
 193                 related_counts[tag_pk] = related_counts.get(tag_pk, 0) + value
 194         related_tags = models.Tag.objects.filter(pk__in=related_counts.keys())
 195         related_tags = [tag for tag in related_tags if tag not in tags]
 196         for tag in related_tags:
 197             tag.count = related_counts[tag.pk]
 198
 199         categories = split_tags(related_tags)
 200         del related_tags
 201
 202     if not objects:
 203         only_author = len(tags) == 1 and tags[0].category == 'author'
 204         objects = models.Book.objects.none()
 205
 206     # Add pictures
 207     objects = MultiQuerySet(Picture.tagged.with_all(tags), objects)
 208
 209     return render_to_response('catalogue/tagged_object_list.html',
 210         {
 211             'object_list': objects,
 212             'categories': categories,
 213             'only_shelf': only_shelf,
 214             'only_author': only_author,
 215             'only_my_shelf': only_my_shelf,
 216             'formats_form': forms.DownloadFormatsForm(),
 217             'tags': tags,
 218             'theme_is_set': theme_is_set,
 219         },
 220         context_instance=RequestContext(request))
 221
 222
 223 def book_fragments(request, slug, theme_slug):
 224     book = get_object_or_404(models.Book, slug=slug)
 225
 226     book_tag = book.book_tag()
 227     theme = get_object_or_404(models.Tag, slug=theme_slug, category='theme')
 228     fragments = models.Fragment.tagged.with_all([book_tag, theme])
 229
 230     return render_to_response('catalogue/book_fragments.html', locals(),
 231         context_instance=RequestContext(request))
 232
 233
 234 def book_detail(request, slug):
 235     try:
 236         book = models.Book.objects.get(slug=slug)
 237     except models.Book.DoesNotExist:
 238         return pdcounter_views.book_stub_detail(request, slug)
 239
 240     book_children = book.children.all().order_by('parent_number', 'sort_key')
 241     return render_to_response('catalogue/book_detail.html', locals(),
 242         context_instance=RequestContext(request))
 243
 244
 245 def player(request, slug):
 246     book = get_object_or_404(models.Book, slug=slug)
 247     if not book.has_media('mp3'):
 248         raise Http404
 249
 250     ogg_files = {}
 251     for m in book.media.filter(type='ogg').order_by().iterator():
 252         ogg_files[m.name] = m
 253
 254     audiobooks = []
 255     have_oggs = True
 256     projects = set()
 257     for mp3 in book.media.filter(type='mp3').iterator():
 258         # ogg files are always from the same project
 259         meta = mp3.extra_info
 260         project = meta.get('project')
 261         if not project:
 262             # temporary fallback
 263             project = u'CzytamySłuchając'
 264
 265         projects.add((project, meta.get('funded_by', '')))
 266
 267         media = {'mp3': mp3}
 268
 269         ogg = ogg_files.get(mp3.name)
 270         if ogg:
 271             media['ogg'] = ogg
 272         else:
 273             have_oggs = False
 274         audiobooks.append(media)
 275
 276     projects = sorted(projects)
 277
 278     extra_info = book.extra_info
 279
 280     return render_to_response('catalogue/player.html', locals(),
 281         context_instance=RequestContext(request))
 282
 283
 284 def book_text(request, slug):
 285     book = get_object_or_404(models.Book, slug=slug)
 286
 287     if not book.has_html_file():
 288         raise Http404
 289     related = book.related_info()
 290     return render_to_response('catalogue/book_text.html', locals(),
 291         context_instance=RequestContext(request))
 292 def book_text2(request, slug):
 293     book = get_object_or_404(models.Book, slug=slug)
 294
 295     if not book.has_html_file():
 296         raise Http404
 297     related = book.related_info()
 298     return render_to_response('catalogue/book_text_new.html', locals(),
 299         context_instance=RequestContext(request))
 300
 301
 302 # ==========
 303 # = Search =
 304 # ==========
 305
 306 def _no_diacritics_regexp(query):
 307     """ returns a regexp for searching for a query without diacritics
 308
 309     should be locale-aware """
 310     names = {
 311         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',
 312         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'
 313         }
 314     def repl(m):
 315         l = m.group()
 316         return u"(%s)" % '|'.join(names[l])
 317     return re.sub(u'[%s]' % (u''.join(names.keys())), repl, query)
 318
 319 def unicode_re_escape(query):
 320     """ Unicode-friendly version of re.escape """
 321     return re.sub('(?u)(\W)', r'\\\1', query)
 322
 323 def _word_starts_with(name, prefix):
 324     """returns a Q object getting models having `name` contain a word
 325     starting with `prefix`
 326
 327     We define word characters as alphanumeric and underscore, like in JS.
 328
 329     Works for MySQL, PostgreSQL, Oracle.
 330     For SQLite, _sqlite* version is substituted for this.
 331     """
 332     kwargs = {}
 333
 334     prefix = _no_diacritics_regexp(unicode_re_escape(prefix))
 335     # can't use [[:<:]] (word start),
 336     # but we want both `xy` and `(xy` to catch `(xyz)`
 337     kwargs['%s__iregex' % name] = u"(^|[^[:alnum:]_])%s" % prefix
 338
 339     return Q(**kwargs)
 340
 341
 342 def _word_starts_with_regexp(prefix):
 343     prefix = _no_diacritics_regexp(unicode_re_escape(prefix))
 344     return ur"(^|(?<=[^\wąćęłńóśźżĄĆĘŁŃÓŚŹŻ]))%s" % prefix
 345
 346
 347 def _sqlite_word_starts_with(name, prefix):
 348     """ version of _word_starts_with for SQLite
 349
 350     SQLite in Django uses Python re module
 351     """
 352     kwargs = {}
 353     kwargs['%s__iregex' % name] = _word_starts_with_regexp(prefix)
 354     return Q(**kwargs)
 355
 356
 357 if hasattr(settings, 'DATABASES'):
 358     if settings.DATABASES['default']['ENGINE'] == 'django.db.backends.sqlite3':
 359         _word_starts_with = _sqlite_word_starts_with
 360 elif settings.DATABASE_ENGINE == 'sqlite3':
 361     _word_starts_with = _sqlite_word_starts_with
 362
 363
 364 class App():
 365     def __init__(self, name, view):
 366         self.name = name
 367         self._view = view
 368         self.lower = name.lower()
 369         self.category = 'application'
 370     def view(self):
 371         return reverse(*self._view)
 372
 373 _apps = (
 374     App(u'Leśmianator', (u'lesmianator', )),
 375     )
 376
 377
 378 def _tags_starting_with(prefix, user=None):
 379     prefix = prefix.lower()
 380     # PD counter
 381     book_stubs = pdcounter_models.BookStub.objects.filter(_word_starts_with('title', prefix))
 382     authors = pdcounter_models.Author.objects.filter(_word_starts_with('name', prefix))
 383
 384     books = models.Book.objects.filter(_word_starts_with('title', prefix))
 385     tags = models.Tag.objects.filter(_word_starts_with('name', prefix))
 386     if user and user.is_authenticated():
 387         tags = tags.filter(~Q(category='book') & (~Q(category='set') | Q(user=user)))
 388     else:
 389         tags = tags.filter(~Q(category='book') & ~Q(category='set'))
 390
 391     prefix_regexp = re.compile(_word_starts_with_regexp(prefix))
 392     return list(books) + list(tags) + [app for app in _apps if prefix_regexp.search(app.lower)] + list(book_stubs) + list(authors)
 393
 394
 395 def _get_result_link(match, tag_list):
 396     if isinstance(match, models.Tag):
 397         return reverse('catalogue.views.tagged_object_list',
 398             kwargs={'tags': '/'.join(tag.url_chunk for tag in tag_list + [match])}
 399         )
 400     elif isinstance(match, App):
 401         return match.view()
 402     else:
 403         return match.get_absolute_url()
 404
 405
 406 def _get_result_type(match):
 407     if isinstance(match, models.Book) or isinstance(match, pdcounter_models.BookStub):
 408         type = 'book'
 409     else:
 410         type = match.category
 411     return type
 412
 413
 414 def books_starting_with(prefix):
 415     prefix = prefix.lower()
 416     return models.Book.objects.filter(_word_starts_with('title', prefix))
 417
 418
 419 def find_best_matches(query, user=None):
 420     """ Finds a models.Book, Tag, models.BookStub or Author best matching a query.
 421
 422     Returns a with:
 423       - zero elements when nothing is found,
 424       - one element when a best result is found,
 425       - more then one element on multiple exact matches
 426
 427     Raises a ValueError on too short a query.
 428     """
 429
 430     query = query.lower()
 431     if len(query) < 2:
 432         raise ValueError("query must have at least two characters")
 433
 434     result = tuple(_tags_starting_with(query, user))
 435     # remove pdcounter stuff
 436     book_titles = set(match.pretty_title().lower() for match in result
 437                       if isinstance(match, models.Book))
 438     authors = set(match.name.lower() for match in result
 439                   if isinstance(match, models.Tag) and match.category=='author')
 440     result = tuple(res for res in result if not (
 441                  (isinstance(res, pdcounter_models.BookStub) and res.pretty_title().lower() in book_titles)
 442                  or (isinstance(res, pdcounter_models.Author) and res.name.lower() in authors)
 443              ))
 444
 445     exact_matches = tuple(res for res in result if res.name.lower() == query)
 446     if exact_matches:
 447         return exact_matches
 448     else:
 449         return tuple(result)[:1]
 450
 451
 452 def search(request):
 453     tags = request.GET.get('tags', '')
 454     prefix = request.GET.get('q', '')
 455
 456     try:
 457         tag_list = models.Tag.get_tag_list(tags)
 458     except:
 459         tag_list = []
 460
 461     try:
 462         result = find_best_matches(prefix, request.user)
 463     except ValueError:
 464         return render_to_response('catalogue/search_too_short.html', {'tags':tag_list, 'prefix':prefix},
 465             context_instance=RequestContext(request))
 466
 467     if len(result) == 1:
 468         return HttpResponseRedirect(_get_result_link(result[0], tag_list))
 469     elif len(result) > 1:
 470         return render_to_response('catalogue/search_multiple_hits.html',
 471             {'tags':tag_list, 'prefix':prefix, 'results':((x, _get_result_link(x, tag_list), _get_result_type(x)) for x in result)},
 472             context_instance=RequestContext(request))
 473     else:
 474         form = PublishingSuggestForm(initial={"books": prefix + ", "})
 475         return render_to_response('catalogue/search_no_hits.html',
 476             {'tags':tag_list, 'prefix':prefix, "pubsuggest_form": form},
 477             context_instance=RequestContext(request))
 478
 479
 480 def tags_starting_with(request):
 481     prefix = request.GET.get('q', '')
 482     # Prefix must have at least 2 characters
 483     if len(prefix) < 2:
 484         return HttpResponse('')
 485     tags_list = []
 486     result = ""
 487     for tag in _tags_starting_with(prefix, request.user):
 488         if not tag.name in tags_list:
 489             result += "\n" + tag.name
 490             tags_list.append(tag.name)
 491     return HttpResponse(result)
 492
 493 def json_tags_starting_with(request, callback=None):
 494     # Callback for JSONP
 495     prefix = request.GET.get('q', '')
 496     callback = request.GET.get('callback', '')
 497     # Prefix must have at least 2 characters
 498     if len(prefix) < 2:
 499         return HttpResponse('')
 500     tags_list = []
 501     for tag in _tags_starting_with(prefix, request.user):
 502         if not tag.name in tags_list:
 503             tags_list.append(tag.name)
 504     if request.GET.get('mozhint', ''):
 505         result = [prefix, tags_list]
 506     else:
 507         result = {"matches": tags_list}
 508     return JSONResponse(result, callback)
 509
 510
 511 # =========
 512 # = Admin =
 513 # =========
 514 @login_required
 515 @staff_required
 516 def import_book(request):
 517     """docstring for import_book"""
 518     book_import_form = forms.BookImportForm(request.POST, request.FILES)
 519     if book_import_form.is_valid():
 520         try:
 521             book_import_form.save()
 522         except:
 523             import sys
 524             import pprint
 525             import traceback
 526             info = sys.exc_info()
 527             exception = pprint.pformat(info[1])
 528             tb = '\n'.join(traceback.format_tb(info[2]))
 529             return HttpResponse(_("An error occurred: %(exception)s\n\n%(tb)s") % {'exception':exception, 'tb':tb}, mimetype='text/plain')
 530         return HttpResponse(_("Book imported successfully"))
 531     else:
 532         return HttpResponse(_("Error importing file: %r") % book_import_form.errors)
 533
 534
 535 # info views for API
 536
 537 def book_info(request, id, lang='pl'):
 538     book = get_object_or_404(models.Book, id=id)
 539     # set language by hand
 540     translation.activate(lang)
 541     return render_to_response('catalogue/book_info.html', locals(),
 542         context_instance=RequestContext(request))
 543
 544
 545 def tag_info(request, id):
 546     tag = get_object_or_404(models.Tag, id=id)
 547     return HttpResponse(tag.description)
 548
 549
 550 def download_zip(request, format, slug=None):
 551     url = None
 552     if format in models.Book.ebook_formats:
 553         url = models.Book.zip_format(format)
 554     elif format in ('mp3', 'ogg') and slug is not None:
 555         book = get_object_or_404(models.Book, slug=slug)
 556         url = book.zip_audiobooks(format)
 557     else:
 558         raise Http404('No format specified for zip package')
 559     return HttpResponseRedirect(urlquote_plus(settings.MEDIA_URL + url, safe='/?='))
 560
 561
 562 class CustomPDFFormView(AjaxableFormView):
 563     form_class = forms.CustomPDFForm
 564     title = ugettext_lazy('Download custom PDF')
 565     submit = ugettext_lazy('Download')
 566     honeypot = True
 567
 568     def __call__(self, *args, **kwargs):
 569         if settings.NO_CUSTOM_PDF:
 570             raise Http404('Custom PDF is disabled')
 571         return super(CustomPDFFormView, self).__call__(*args, **kwargs)
 572
 573     def form_args(self, request, obj):
 574         """Override to parse view args and give additional args to the form."""
 575         return (obj,), {}
 576
 577     def get_object(self, request, slug, *args, **kwargs):
 578         return get_object_or_404(models.Book, slug=slug)
 579
 580     def context_description(self, request, obj):
 581         return obj.pretty_title()