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