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