Split form intro.
[wolnelektury.git] / src / catalogue / models / book.py
index dbbee78..eb87f10 100644 (file)
@@ -1,35 +1,34 @@
-# -*- coding: utf-8 -*-
 # This file is part of Wolnelektury, licensed under GNU Affero GPLv3 or later.
 # Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
 #
 from collections import OrderedDict
+import json
 from datetime import date, timedelta
 from random import randint
 import os.path
 import re
-import urllib
+from urllib.request import urlretrieve
 from django.conf import settings
 from django.db import connection, models, transaction
-from django.db.models import permalink
 import django.dispatch
 from django.contrib.contenttypes.fields import GenericRelation
-from django.core.urlresolvers import reverse
+from django.template.loader import render_to_string
+from django.urls import reverse
 from django.utils.translation import ugettext_lazy as _, get_language
 from django.utils.deconstruct import deconstructible
-import jsonfield
 from fnpdjango.storage import BofhFileSystemStorage
-from ssify import flush_ssi_includes
 
+from librarian.cover import WLCover
 from librarian.html import transform_abstrakt
 from newtagging import managers
 from catalogue import constants
 from catalogue.fields import EbookField
 from catalogue.models import Tag, Fragment, BookMedia
-from catalogue.utils import create_zip, gallery_url, gallery_path, split_tags
+from catalogue.utils import create_zip, gallery_url, gallery_path, split_tags, get_random_hash
 from catalogue.models.tag import prefetched_relations
 from catalogue import app_settings
 from catalogue import tasks
-from wolnelektury.utils import makedirs
+from wolnelektury.utils import makedirs, cached_render, clear_cached_renders
 
 bofh_storage = BofhFileSystemStorage()
 
@@ -47,6 +46,7 @@ _cover_upload_to = UploadToPath('book/cover/%s.jpg')
 _cover_thumb_upload_to = UploadToPath('book/cover_thumb/%s.jpg')
 _cover_api_thumb_upload_to = UploadToPath('book/cover_api_thumb/%s.jpg')
 _simple_cover_upload_to = UploadToPath('book/cover_simple/%s.jpg')
+_cover_ebookpoint_upload_to = UploadToPath('book/cover_ebookpoint/%s.jpg')
 
 
 def _ebook_upload_to(upload_path):
@@ -58,7 +58,7 @@ class Book(models.Model):
     title = models.CharField(_('title'), max_length=32767)
     sort_key = models.CharField(_('sort key'), max_length=120, db_index=True, editable=False)
     sort_key_author = models.CharField(
-        _('sort key by author'), max_length=120, db_index=True, editable=False, default=u'')
+        _('sort key by author'), max_length=120, db_index=True, editable=False, default='')
     slug = models.SlugField(_('slug'), max_length=120, db_index=True, unique=True)
     common_slug = models.SlugField(_('slug'), max_length=120, db_index=True)
     language = models.CharField(_('language code'), max_length=3, db_index=True, default=app_settings.DEFAULT_LANGUAGE)
@@ -67,7 +67,7 @@ class Book(models.Model):
     created_at = models.DateTimeField(_('creation date'), auto_now_add=True, db_index=True)
     changed_at = models.DateTimeField(_('change date'), auto_now=True, db_index=True)
     parent_number = models.IntegerField(_('parent number'), default=0)
-    extra_info = jsonfield.JSONField(_('extra information'), default={})
+    extra_info = models.TextField(_('extra information'), default='{}')
     gazeta_link = models.CharField(blank=True, max_length=240)
     wiki_link = models.CharField(blank=True, max_length=240)
     print_on_demand = models.BooleanField(_('print on demand'), default=False)
@@ -75,6 +75,8 @@ class Book(models.Model):
     audio_length = models.CharField(_('audio length'), blank=True, max_length=8)
     preview = models.BooleanField(_('preview'), default=False)
     preview_until = models.DateField(_('preview until'), blank=True, null=True)
+    preview_key = models.CharField(max_length=32, blank=True, null=True)
+    findable = models.BooleanField(_('findable'), default=True, db_index=True)
 
     # files generated during publication
     cover = EbookField(
@@ -82,26 +84,36 @@ class Book(models.Model):
         null=True, blank=True,
         upload_to=_cover_upload_to,
         storage=bofh_storage, max_length=255)
+    cover_etag = models.CharField(max_length=255, editable=False, default='', db_index=True)
     # Cleaner version of cover for thumbs
     cover_thumb = EbookField(
         'cover_thumb', _('cover thumbnail'),
         null=True, blank=True,
         upload_to=_cover_thumb_upload_to,
         max_length=255)
+    cover_thumb_etag = models.CharField(max_length=255, editable=False, default='', db_index=True)
     cover_api_thumb = EbookField(
         'cover_api_thumb', _('cover thumbnail for mobile app'),
         null=True, blank=True,
         upload_to=_cover_api_thumb_upload_to,
         max_length=255)
+    cover_api_thumb_etag = models.CharField(max_length=255, editable=False, default='', db_index=True)
     simple_cover = EbookField(
         'simple_cover', _('cover for mobile app'),
         null=True, blank=True,
         upload_to=_simple_cover_upload_to,
         max_length=255)
+    simple_cover_etag = models.CharField(max_length=255, editable=False, default='', db_index=True)
+    cover_ebookpoint = EbookField(
+        'cover_ebookpoint', _('cover for Ebookpoint'),
+        null=True, blank=True,
+        upload_to=_cover_ebookpoint_upload_to,
+        max_length=255)
+    cover_ebookpoint_etag = models.CharField(max_length=255, editable=False, default='', db_index=True)
     ebook_formats = constants.EBOOK_FORMATS
     formats = ebook_formats + ['html', 'xml']
 
-    parent = models.ForeignKey('self', blank=True, null=True, related_name='children')
+    parent = models.ForeignKey('self', models.CASCADE, blank=True, null=True, related_name='children')
     ancestor = models.ManyToManyField('self', blank=True, editable=False, related_name='descendant', symmetrical=False)
 
     cached_author = models.CharField(blank=True, max_length=240, db_index=True)
@@ -115,7 +127,7 @@ class Book(models.Model):
     html_built = django.dispatch.Signal()
     published = django.dispatch.Signal()
 
-    short_html_url_name = 'catalogue_book_short'
+    SORT_KEY_SEP = '$'
 
     class AlreadyExists(Exception):
         pass
@@ -126,9 +138,12 @@ class Book(models.Model):
         verbose_name_plural = _('books')
         app_label = 'catalogue'
 
-    def __unicode__(self):
+    def __str__(self):
         return self.title
 
+    def get_extra_info_json(self):
+        return json.loads(self.extra_info or '{}')
+
     def get_initial(self):
         try:
             return re.search(r'\w', self.title, re.U).group(0)
@@ -138,6 +153,15 @@ class Book(models.Model):
     def authors(self):
         return self.tags.filter(category='author')
 
+    def epochs(self):
+        return self.tags.filter(category='epoch')
+
+    def genres(self):
+        return self.tags.filter(category='genre')
+
+    def kinds(self):
+        return self.tags.filter(category='kind')
+
     def tag_unicode(self, category):
         relations = prefetched_relations(self, category)
         if relations:
@@ -151,8 +175,23 @@ class Book(models.Model):
     def author_unicode(self):
         return self.cached_author
 
+    def kind_unicode(self):
+        return self.tag_unicode('kind')
+
+    def epoch_unicode(self):
+        return self.tag_unicode('epoch')
+
+    def genre_unicode(self):
+        return self.tag_unicode('genre')
+
+    def translators(self):
+        translators = self.get_extra_info_json().get('translators') or []
+        return [
+            '\xa0'.join(reversed(translator.split(', ', 1))) for translator in translators
+        ]
+
     def translator(self):
-        translators = self.extra_info.get('translators')
+        translators = self.get_extra_info_json().get('translators')
         if not translators:
             return None
         if len(translators) > 3:
@@ -160,38 +199,48 @@ class Book(models.Model):
             others = ' i inni'
         else:
             others = ''
-        return ', '.join(u'\xa0'.join(reversed(translator.split(', ', 1))) for translator in translators) + others
+        return ', '.join('\xa0'.join(reversed(translator.split(', ', 1))) for translator in translators) + others
 
     def cover_source(self):
-        return self.extra_info.get('cover_source', self.parent.cover_source() if self.parent else '')
+        return self.get_extra_info_json().get('cover_source', self.parent.cover_source() if self.parent else '')
+
+    @property
+    def isbn_pdf(self):
+        return self.get_extra_info_json().get('isbn_pdf')
+
+    @property
+    def isbn_epub(self):
+        return self.get_extra_info_json().get('isbn_epub')
+
+    @property
+    def isbn_mobi(self):
+        return self.get_extra_info_json().get('isbn_mobi')
+
 
     def save(self, force_insert=False, force_update=False, **kwargs):
         from sortify import sortify
 
         self.sort_key = sortify(self.title)[:120]
-        self.title = unicode(self.title)  # ???
+        self.title = str(self.title)  # ???
 
         try:
             author = self.authors().first().sort_key
         except AttributeError:
-            author = u''
+            author = ''
         self.sort_key_author = author
 
         self.cached_author = self.tag_unicode('author')
-        self.has_audience = 'audience' in self.extra_info
+        self.has_audience = 'audience' in self.get_extra_info_json()
+
+        if self.preview and not self.preview_key:
+            self.preview_key = get_random_hash(self.slug)[:32]
 
         ret = super(Book, self).save(force_insert, force_update, **kwargs)
 
         return ret
 
-    @permalink
     def get_absolute_url(self):
-        return 'catalogue.views.book_detail', [self.slug]
-
-    @staticmethod
-    @permalink
-    def create_url(slug):
-        return 'catalogue.views.book_detail', [slug]
+        return reverse('book_detail', args=[self.slug])
 
     def gallery_path(self):
         return gallery_path(self.slug)
@@ -199,6 +248,41 @@ class Book(models.Model):
     def gallery_url(self):
         return gallery_url(self.slug)
 
+    def get_first_text(self):
+        if self.html_file:
+            return self
+        child = self.children.all().order_by('parent_number').first()
+        if child is not None:
+            return child.get_first_text()
+
+    def get_last_text(self):
+        if self.html_file:
+            return self
+        child = self.children.all().order_by('parent_number').last()
+        if child is not None:
+            return child.get_last_text()
+
+    def get_prev_text(self):
+        if not self.parent:
+            return None
+        sibling = self.parent.children.filter(parent_number__lt=self.parent_number).order_by('-parent_number').first()
+        if sibling is not None:
+            return sibling.get_last_text()
+        return self.parent.get_prev_text()
+
+    def get_next_text(self):
+        if not self.parent:
+            return None
+        sibling = self.parent.children.filter(parent_number__gt=self.parent_number).order_by('parent_number').first()
+        if sibling is not None:
+            return sibling.get_first_text()
+        return self.parent.get_next_text()
+
+    def get_siblings(self):
+        if not self.parent:
+            return []
+        return self.parent.children.all().order_by('parent_number')
+
     @property
     def name(self):
         return self.title
@@ -220,6 +304,12 @@ class Book(models.Model):
 
     @staticmethod
     def format_audio_length(seconds):
+        """
+        >>> Book.format_audio_length(1)
+        '0:01'
+        >>> Book.format_audio_length(3661)
+        '1:01:01'
+        """
         if seconds < 60*60:
             minutes = seconds // 60
             seconds = seconds % 60
@@ -231,11 +321,9 @@ class Book(models.Model):
             return '%d:%02d:%02d' % (hours, minutes, seconds)
 
     def get_audio_length(self):
-        from mutagen.mp3 import MP3
         total = 0
         for media in self.get_mp3() or ():
-            audio = MP3(media.file.path)
-            total += audio.info.length
+            total += app_settings.GET_MP3_LENGTH(media.file.path)
         return int(total)
 
     def has_media(self, type_):
@@ -272,7 +360,7 @@ class Book(models.Model):
         media = self.get_media(format_)
         if media:
             if self.preview:
-                return reverse('embargo_link', kwargs={'slug': self.slug, 'format_': format_})
+                return reverse('embargo_link', kwargs={'key': self.preview_key, 'slug': self.slug, 'format_': format_})
             else:
                 return media.url
         else:
@@ -328,11 +416,11 @@ class Book(models.Model):
         projects = set()
         for mp3 in self.media.filter(type='mp3').iterator():
             # ogg files are always from the same project
-            meta = mp3.extra_info
+            meta = mp3.get_extra_info_json()
             project = meta.get('project')
             if not project:
                 # temporary fallback
-                project = u'CzytamySłuchając'
+                project = 'CzytamySłuchając'
 
             projects.add((project, meta.get('funded_by', '')))
 
@@ -361,25 +449,48 @@ class Book(models.Model):
             parse_dublincore=parse_dublincore,
             meta_fallbacks=meta_fallbacks)
 
+    def wldocument2(self):
+        from catalogue.import_utils import ORMDocProvider
+        from librarian.document import WLDocument
+        doc = WLDocument(
+            self.xml_file.path,
+            provider=ORMDocProvider(self)
+        )
+        doc.meta.update(self.cover_info())
+        return doc
+
+
     @staticmethod
     def zip_format(format_):
         def pretty_file_name(book):
             return "%s/%s.%s" % (
-                book.extra_info['author'],
+                book.get_extra_info_json()['author'],
                 book.slug,
                 format_)
 
         field_name = "%s_file" % format_
-        books = Book.objects.filter(parent=None).exclude(**{field_name: ""}).exclude(preview=True)
+        books = Book.objects.filter(parent=None).exclude(**{field_name: ""}).exclude(preview=True).exclude(findable=False)
         paths = [(pretty_file_name(b), getattr(b, field_name).path) for b in books.iterator()]
         return create_zip(paths, app_settings.FORMAT_ZIPS[format_])
 
     def zip_audiobooks(self, format_):
         bm = BookMedia.objects.filter(book=self, type=format_)
         paths = map(lambda bm: (None, bm.file.path), bm)
-        return create_zip(paths, "%s_%s" % (self.slug, format_))
+        licenses = set()
+        for m in bm:
+            license = constants.LICENSES.get(
+                m.get_extra_info_json().get('license'), {}
+            ).get('locative')
+            if license:
+                licenses.add(license)
+        readme = render_to_string('catalogue/audiobook_zip_readme.txt', {
+            'licenses': licenses,
+        })
+        return create_zip(paths, "%s_%s" % (self.slug, format_), {'informacje.txt': readme})
 
     def search_index(self, book_info=None, index=None, index_tags=True, commit=True):
+        if not self.findable:
+            return
         if index is None:
             from search.index import Index
             index = Index()
@@ -389,7 +500,7 @@ class Book(models.Model):
                 index.index_tags()
             if commit:
                 index.index.commit()
-        except Exception, e:
+        except Exception as e:
             index.index.rollback()
             raise e
 
@@ -407,7 +518,7 @@ class Book(models.Model):
             for ilustr in ilustr_elements:
                 ilustr_src = ilustr.get('src')
                 ilustr_path = os.path.join(gallery_path, ilustr_src)
-                urllib.urlretrieve('%s/%s' % (remote_gallery_url, ilustr_src), ilustr_path)
+                urlretrieve('%s/%s' % (remote_gallery_url, ilustr_src), ilustr_path)
 
     def load_abstract(self):
         abstract = self.wldocument(parse_dublincore=False).edoc.getroot().find('.//abstrakt')
@@ -434,7 +545,7 @@ class Book(models.Model):
 
     @classmethod
     def from_text_and_meta(cls, raw_file, book_info, overwrite=False, dont_build=None, search_index=True,
-                           search_index_tags=True, remote_gallery_url=None, days=0):
+                           search_index_tags=True, remote_gallery_url=None, days=0, findable=True):
         if dont_build is None:
             dont_build = set()
         dont_build = set.union(set(dont_build), set(app_settings.DONT_BUILD))
@@ -472,13 +583,14 @@ class Book(models.Model):
         if book.preview:
             book.xml_file.set_readable(False)
 
+        book.findable = findable
         book.language = book_info.language
         book.title = book_info.title
         if book_info.variant_of:
             book.common_slug = book_info.variant_of.slug
         else:
             book.common_slug = book.slug
-        book.extra_info = book_info.to_dict()
+        book.extra_info = json.dumps(book_info.to_dict())
         book.load_abstract()
         book.save()
 
@@ -490,6 +602,7 @@ class Book(models.Model):
                 tag.save()
 
         book.tags = set(meta_tags + book_shelves)
+        book.save()  # update sort_key_author
 
         cover_changed = old_cover != book.cover_info()
         obsolete_children = set(b for b in book.children.all()
@@ -524,6 +637,7 @@ class Book(models.Model):
             book.cover_thumb.build_delay()
             book.cover_api_thumb.build_delay()
             book.simple_cover.build_delay()
+            book.cover_ebookpoint.build_delay()
 
         # Build HTML and ebooks.
         book.html_file.build_delay()
@@ -535,17 +649,60 @@ class Book(models.Model):
             if format_ not in dont_build:
                 getattr(book, '%s_file' % format_).build_delay()
 
-        if not settings.NO_SEARCH_INDEX and search_index:
+        if not settings.NO_SEARCH_INDEX and search_index and findable:
             tasks.index_book.delay(book.id, book_info=book_info, index_tags=search_index_tags)
 
         for child in notify_cover_changed:
             child.parent_cover_changed()
 
-        book.save()  # update sort_key_author
         book.update_popularity()
+        tasks.update_references.delay(book.id)
+
         cls.published.send(sender=cls, instance=book)
         return book
 
+    def get_master(self):
+        master_tags = [
+            'opowiadanie',
+            'powiesc',
+            'dramat_wierszowany_l',
+            'dramat_wierszowany_lp',
+            'dramat_wspolczesny', 'liryka_l', 'liryka_lp',
+            'wywiad',
+        ]
+        from librarian.parser import WLDocument
+        wld = WLDocument.from_file(self.xml_file.path, parse_dublincore=False)
+        root = wld.edoc.getroot()
+        for master in root.iter():
+            if master.tag in master_tags:
+                return master
+    
+    def update_references(self):
+        from references.models import Entity, Reference
+        master = self.get_master()
+        found = set()
+        for i, sec in enumerate(master):
+            for ref in sec.findall('.//ref'):
+                href = ref.attrib.get('href', '')
+                if not href or href in found:
+                    continue
+                found.add(href)
+                entity, created = Entity.objects.get_or_create(
+                    uri=href
+                )
+                ref, created = Reference.objects.get_or_create(
+                    book=self,
+                    entity=entity
+                )
+                ref.first_section = 'sec%d' % (i + 1)
+                entity.populate()
+                entity.save()
+        Reference.objects.filter(book=self).exclude(entity__uri__in=found).delete()
+    
+    @property
+    def references(self):
+        return self.reference_set.all().select_related('entity')
+
     @classmethod
     @transaction.atomic
     def repopulate_ancestors(cls):
@@ -579,23 +736,9 @@ class Book(models.Model):
                     b.ancestor.add(parent)
                     parent = parent.parent
 
-    def flush_includes(self, languages=True):
-        if not languages:
-            return
-        if languages is True:
-            languages = [lc for (lc, _ln) in settings.LANGUAGES]
-        flush_ssi_includes([
-            template % (self.pk, lang)
-            for template in [
-                '/katalog/b/%d/mini.%s.html',
-                '/katalog/b/%d/mini_nolink.%s.html',
-                '/katalog/b/%d/short.%s.html',
-                '/katalog/b/%d/wide.%s.html',
-                '/api/include/book/%d.%s.json',
-                '/api/include/book/%d.%s.xml',
-                ]
-            for lang in languages
-            ])
+    def clear_cache(self):
+        clear_cached_renders(self.mini_box)
+        clear_cached_renders(self.mini_box_nolink)
 
     def cover_info(self, inherit=True):
         """Returns a dictionary to serve as fallback for BookInfo.
@@ -605,7 +748,7 @@ class Book(models.Model):
         need = False
         info = {}
         for field in ('cover_url', 'cover_by', 'cover_source'):
-            val = self.extra_info.get(field)
+            val = self.get_extra_info_json().get(field)
             if val:
                 info[field] = val
             else:
@@ -637,7 +780,7 @@ class Book(models.Model):
 
     def other_versions(self):
         """Find other versions (i.e. in other languages) of the book."""
-        return type(self).objects.filter(common_slug=self.common_slug).exclude(pk=self.pk)
+        return type(self).objects.filter(common_slug=self.common_slug, findable=True).exclude(pk=self.pk)
 
     def parents(self):
         books = []
@@ -659,8 +802,8 @@ class Book(models.Model):
         return ', '.join(names)
 
     def publisher(self):
-        publisher = self.extra_info['publisher']
-        if isinstance(publisher, basestring):
+        publisher = self.get_extra_info_json()['publisher']
+        if isinstance(publisher, str):
             return publisher
         elif isinstance(publisher, list):
             return ', '.join(publisher)
@@ -674,7 +817,7 @@ class Book(models.Model):
 
         """
         objects = cls.tagged.with_all(tags)
-        return objects.exclude(ancestor__in=objects)
+        return objects.filter(findable=True).exclude(ancestor__in=objects)
 
     @classmethod
     def book_list(cls, book_filter=None):
@@ -685,7 +828,7 @@ class Book(models.Model):
         """
 
         books_by_parent = {}
-        books = cls.objects.order_by('parent_number', 'sort_key').only('title', 'parent', 'slug')
+        books = cls.objects.filter(findable=True).order_by('parent_number', 'sort_key').only('title', 'parent', 'slug', 'extra_info')
         if book_filter:
             books = books.filter(book_filter).distinct()
 
@@ -715,23 +858,23 @@ class Book(models.Model):
         return books_by_author, orphans, books_by_parent
 
     _audiences_pl = {
-        "SP": (1, u"szkoła podstawowa"),
-        "SP1": (1, u"szkoła podstawowa"),
-        "SP2": (1, u"szkoła podstawowa"),
-        "SP3": (1, u"szkoła podstawowa"),
-        "P": (1, u"szkoła podstawowa"),
-        "G": (2, u"gimnazjum"),
-        "L": (3, u"liceum"),
-        "LP": (3, u"liceum"),
+        "SP": (1, "szkoła podstawowa"),
+        "SP1": (1, "szkoła podstawowa"),
+        "SP2": (1, "szkoła podstawowa"),
+        "SP3": (1, "szkoła podstawowa"),
+        "P": (1, "szkoła podstawowa"),
+        "G": (2, "gimnazjum"),
+        "L": (3, "liceum"),
+        "LP": (3, "liceum"),
     }
 
     def audiences_pl(self):
-        audiences = self.extra_info.get('audiences', [])
+        audiences = self.get_extra_info_json().get('audiences', [])
         audiences = sorted(set([self._audiences_pl.get(a, (99, a)) for a in audiences]))
         return [a[1] for a in audiences]
 
     def stage_note(self):
-        stage = self.extra_info.get('stage')
+        stage = self.get_extra_info_json().get('stage')
         if stage and stage < '0.4':
             return (_('This work needs modernisation'),
                     reverse('infopage', args=['wymagajace-uwspolczesnienia']))
@@ -784,6 +927,24 @@ class Book(models.Model):
         if likes(user, self):
             set_sets(user, self, [])
 
+    def full_sort_key(self):
+        return self.SORT_KEY_SEP.join((self.sort_key_author, self.sort_key, str(self.id)))
+
+    def cover_color(self):
+        return WLCover.epoch_colors.get(self.get_extra_info_json().get('epoch'), '#000000')
+
+    @cached_render('catalogue/book_mini_box.html')
+    def mini_box(self):
+        return {
+            'book': self
+        }
+
+    @cached_render('catalogue/book_mini_box.html')
+    def mini_box_nolink(self):
+        return {
+            'book': self,
+            'no_link': True,
+        }
 
 def add_file_fields():
     for format_ in Book.formats:
@@ -801,11 +962,13 @@ def add_file_fields():
             blank=True,
             default=''
         ).contribute_to_class(Book, field_name)
+        if format_ != 'xml':
+            models.CharField(max_length=255, editable=False, default='', db_index=True).contribute_to_class(Book, f'{field_name}_etag')
 
 
 add_file_fields()
 
 
 class BookPopularity(models.Model):
-    book = models.OneToOneField(Book, related_name='popularity')
+    book = models.OneToOneField(Book, models.CASCADE, related_name='popularity')
     count = models.IntegerField(default=0, db_index=True)