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