Added capability to import PDF, ODT and TXT files to importbook management command.
[wolnelektury.git] / apps / catalogue / models.py
1 # -*- coding: utf-8 -*-
2 from django.db import models
3 from django.db.models import permalink, Q
4 from django.utils.translation import ugettext_lazy as _
5 from django.contrib.auth.models import User
6 from django.core.files import File
7 from django.template.loader import render_to_string
8 from django.utils.safestring import mark_safe
9 from django.core.urlresolvers import reverse
10
11 from newtagging.models import TagBase
12 from newtagging import managers
13
14 from librarian import html, dcparser
15
16
17 TAG_CATEGORIES = (
18     ('author', _('author')),
19     ('epoch', _('epoch')),
20     ('kind', _('kind')),
21     ('genre', _('genre')),
22     ('theme', _('theme')),
23     ('set', _('set')),
24 )
25
26
27 class TagSubcategoryManager(models.Manager):
28     def __init__(self, subcategory):
29         super(TagSubcategoryManager, self).__init__()
30         self.subcategory = subcategory
31         
32     def get_query_set(self):
33         return super(TagSubcategoryManager, self).get_query_set().filter(category=self.subcategory)
34
35
36 class Tag(TagBase):
37     name = models.CharField(_('name'), max_length=50, db_index=True)
38     slug = models.SlugField(_('slug'), max_length=120, unique=True, db_index=True)
39     sort_key = models.SlugField(_('sort key'), max_length=120, db_index=True)
40     category = models.CharField(_('category'), max_length=50, blank=False, null=False, 
41         db_index=True, choices=TAG_CATEGORIES)
42     description = models.TextField(_('description'), blank=True)
43     main_page = models.BooleanField(_('main page'), default=False, db_index=True, help_text=_('Show tag on main page'))
44         
45     user = models.ForeignKey(User, blank=True, null=True)
46     
47     def has_description(self):
48         return len(self.description) > 0
49     has_description.short_description = _('description')
50     has_description.boolean = True
51
52     @permalink
53     def get_absolute_url(self):
54         return ('catalogue.views.tagged_object_list', [self.slug])
55     
56     class Meta:
57         ordering = ('sort_key',)
58         verbose_name = _('tag')
59         verbose_name_plural = _('tags')
60     
61     def __unicode__(self):
62         return self.name
63
64     @staticmethod
65     def get_tag_list(tags):
66         if isinstance(tags, basestring):
67             tag_slugs = tags.split('/')
68             return [Tag.objects.get(slug=slug) for slug in tag_slugs]
69         else:
70             return TagBase.get_tag_list(tags)
71
72
73 def book_upload_path(ext):
74     def get_dynamic_path(book, filename):
75         return 'lektura/%s.%s' % (book.slug, ext)
76     return get_dynamic_path
77
78
79 class Book(models.Model):
80     title = models.CharField(_('title'), max_length=120)
81     slug = models.SlugField(_('slug'), max_length=120, unique=True, db_index=True)
82     description = models.TextField(_('description'), blank=True)
83     created_at = models.DateTimeField(_('creation date'), auto_now=True)
84     _short_html = models.TextField(_('short HTML'), editable=False)
85     parent_number = models.IntegerField(_('parent number'), default=0)
86     
87     # Formats
88     xml_file = models.FileField(_('XML file'), upload_to=book_upload_path('xml'), blank=True)
89     html_file = models.FileField(_('HTML file'), upload_to=book_upload_path('html'), blank=True)
90     pdf_file = models.FileField(_('PDF file'), upload_to=book_upload_path('pdf'), blank=True)
91     odt_file = models.FileField(_('ODT file'), upload_to=book_upload_path('odt'), blank=True)
92     txt_file = models.FileField(_('TXT file'), upload_to=book_upload_path('txt'), blank=True)
93     
94     parent = models.ForeignKey('self', blank=True, null=True, related_name='children')
95     
96     objects = models.Manager()
97     tagged = managers.ModelTaggedItemManager(Tag)
98     tags = managers.TagDescriptor(Tag)
99     
100     def short_html(self):
101         if len(self._short_html):
102             return mark_safe(self._short_html)
103         else:
104             tags = self.tags.filter(~Q(category__in=('set', 'theme')))
105             tags = [u'<a href="%s">%s</a>' % (tag.get_absolute_url(), tag.name) for tag in tags]
106
107             formats = []
108             if self.html_file:
109                 formats.append(u'<a href="%s">Czytaj online</a>' % reverse('book_text', kwargs={'slug': self.slug}))
110             if self.pdf_file:
111                 formats.append(u'<a href="%s">Plik PDF</a>' % self.pdf_file.url)
112             if self.odt_file:
113                 formats.append(u'<a href="%s">Plik ODT</a>' % self.odt_file.url)
114             
115             self._short_html = unicode(render_to_string('catalogue/book_short.html',
116                 {'book': self, 'tags': tags, 'formats': formats}))
117             self.save()
118             return mark_safe(self._short_html)
119     
120     def has_description(self):
121         return len(self.description) > 0
122     has_description.short_description = _('description')
123     has_description.boolean = True
124     
125     def has_pdf_file(self):
126         return bool(self.pdf_file)
127     has_pdf_file.short_description = 'PDF'
128     has_pdf_file.boolean = True
129     
130     def has_odt_file(self):
131         return bool(self.odt_file)
132     has_odt_file.short_description = 'ODT'
133     has_odt_file.boolean = True
134     
135     def has_html_file(self):
136         return bool(self.html_file)
137     has_html_file.short_description = 'HTML'
138     has_html_file.boolean = True
139
140     class AlreadyExists(Exception):
141         pass
142     
143     @staticmethod
144     def from_xml_file(xml_file, overwrite=False):
145         from tempfile import NamedTemporaryFile
146         from slughifi import slughifi
147         from markupstring import MarkupString
148         
149         # Read book metadata
150         book_info = dcparser.parse(xml_file)
151         book_base, book_slug = book_info.url.rsplit('/', 1)
152         book, created = Book.objects.get_or_create(slug=book_slug)
153         if not created and not overwrite:
154             raise Book.AlreadyExists('Book %s already exists' % book_slug)
155         
156         book.title = book_info.title
157         book.save()
158         
159         book_tags = []
160         for category in ('kind', 'genre', 'author', 'epoch'):    
161             tag_name = getattr(book_info, category)
162             tag_sort_key = tag_name
163             if category == 'author':
164                 tag_sort_key = tag_name.last_name
165                 tag_name = ' '.join(tag_name.first_names) + ' ' + tag_name.last_name
166             tag, created = Tag.objects.get_or_create(slug=slughifi(tag_name))
167             if created:
168                 tag.name = tag_name
169                 tag.sort_key = slughifi(tag_sort_key)
170                 tag.category = category
171                 tag.save()
172             book_tags.append(tag)
173         book.tags = book_tags
174         
175         if hasattr(book_info, 'parts'):
176             for n, part_url in enumerate(book_info.parts):
177                 base, slug = part_url.rsplit('/', 1)
178                 child_book = Book.objects.get(slug=slug)
179                 child_book.parent = book
180                 child_book.parent_number = n
181                 child_book.save()
182         
183         # Save XML and HTML files
184         book.xml_file.save('%s.xml' % book.slug, File(file(xml_file)), save=False)
185         
186         html_file = NamedTemporaryFile()
187         if html.transform(book.xml_file.path, html_file):
188             book.html_file.save('%s.html' % book.slug, File(html_file), save=False)
189             
190             # Extract fragments
191             closed_fragments, open_fragments = html.extract_fragments(book.html_file.path)
192             book_themes = []
193             for fragment in closed_fragments.values():
194                 text = fragment.to_string()
195                 short_text = ''
196                 if (len(MarkupString(text)) > 240):
197                     short_text = unicode(MarkupString(text)[:160])
198                 new_fragment = Fragment(text=text, short_text=short_text, anchor=fragment.id, book=book)
199                 
200                 try:
201                     theme_names = [s.strip() for s in fragment.themes.split(',')]
202                 except AttributeError:
203                     continue
204                 themes = []
205                 for theme_name in theme_names:
206                     tag, created = Tag.objects.get_or_create(slug=slughifi(theme_name))
207                     if created:
208                         tag.name = theme_name
209                         tag.sort_key = slughifi(theme_name)
210                         tag.category = 'theme'
211                         tag.save()
212                     themes.append(tag)
213                 new_fragment.save()
214                 new_fragment.tags = list(book.tags) + themes
215                 book_themes += themes
216             
217             book_themes = set(book_themes)
218             book.tags = list(book.tags) + list(book_themes)
219         
220         book.save()
221         return book
222     
223     @permalink
224     def get_absolute_url(self):
225         return ('catalogue.views.book_detail', [self.slug])
226         
227     class Meta:
228         ordering = ('title',)
229         verbose_name = _('book')
230         verbose_name_plural = _('books')
231
232     def __unicode__(self):
233         return self.title
234
235
236 class Fragment(models.Model):
237     text = models.TextField()
238     short_text = models.TextField(editable=False)
239     _short_html = models.TextField(editable=False)
240     anchor = models.CharField(max_length=120)
241     book = models.ForeignKey(Book, related_name='fragments')
242
243     objects = models.Manager()
244     tagged = managers.ModelTaggedItemManager(Tag)
245     tags = managers.TagDescriptor(Tag)
246     
247     def short_html(self):
248         if len(self._short_html):
249             return mark_safe(self._short_html)
250         else:
251             book_authors = [u'<a href="%s">%s</a>' % (tag.get_absolute_url(), tag.name) 
252                 for tag in self.book.tags if tag.category == 'author']
253             
254             self._short_html = unicode(render_to_string('catalogue/fragment_short.html',
255                 {'fragment': self, 'book': self.book, 'book_authors': book_authors}))
256             self.save()
257             return mark_safe(self._short_html)
258     
259     def get_absolute_url(self):
260         return '%s#m%s' % (reverse('book_text', kwargs={'slug': self.book.slug}), self.anchor)
261     
262     class Meta:
263         ordering = ('book', 'anchor',)
264         verbose_name = _('fragment')
265         verbose_name_plural = _('fragments')
266