Cutting stuff out.
[wolnelektury.git] / src / catalogue / models / book.py
index 6b9fd72..3eb6023 100644 (file)
@@ -15,46 +15,24 @@ import django.dispatch
 from django.contrib.contenttypes.fields import GenericRelation
 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
+from django.utils.translation import gettext_lazy as _, get_language
 from fnpdjango.storage import BofhFileSystemStorage
 from lxml import html
 from librarian.cover import WLCover
 from librarian.html import transform_abstrakt
+from librarian.builders import builders
 from newtagging import managers
 from catalogue import constants
-from catalogue.fields import EbookField
+from catalogue import fields
 from catalogue.models import Tag, Fragment, BookMedia
 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, cached_render, clear_cached_renders
 
 bofh_storage = BofhFileSystemStorage()
 
 
-@deconstructible
-class UploadToPath(object):
-    def __init__(self, path):
-        self.path = path
-
-    def __call__(self, instance, filename):
-        return self.path % instance.slug
-
-
-_cover_upload_to = UploadToPath('book/cover/%s.jpg')
-_cover_clean_upload_to = UploadToPath('book/cover_clean/%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):
-    return UploadToPath(upload_path)
-
-
 class Book(models.Model):
     """Represents a book imported from WL-XML."""
     title = models.CharField(_('title'), max_length=32767)
@@ -82,44 +60,24 @@ class Book(models.Model):
     findable = models.BooleanField(_('findable'), default=True, db_index=True)
 
     # files generated during publication
-    cover = EbookField(
-        'cover', _('cover'),
-        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)
+    xml_file = fields.XmlField(storage=bofh_storage, with_etag=False)
+    html_file = fields.HtmlField(storage=bofh_storage)
+    fb2_file = fields.Fb2Field(storage=bofh_storage)
+    txt_file = fields.TxtField(storage=bofh_storage)
+    epub_file = fields.EpubField(storage=bofh_storage)
+    mobi_file = fields.MobiField(storage=bofh_storage)
+    pdf_file = fields.PdfField(storage=bofh_storage)
+
+    cover = fields.CoverField(_('cover'), storage=bofh_storage)
     # Cleaner version of cover for thumbs
-    cover_clean = EbookField(
-        'cover_clean', _('clean cover'),
-        null=True, blank=True,
-        upload_to=_cover_clean_upload_to,
-        max_length=255
-    )
-    cover_clean_etag = models.CharField(max_length=255, editable=False, default='', db_index=True)
-    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)
+    cover_clean = fields.CoverCleanField(_('clean cover'))
+    cover_thumb = fields.CoverThumbField(_('cover thumbnail'))
+    cover_api_thumb = fields.CoverApiThumbField(
+        _('cover thumbnail for mobile app'))
+    simple_cover = fields.SimpleCoverField(_('cover for mobile app'))
+    cover_ebookpoint = fields.CoverEbookpointField(
+        _('cover for Ebookpoint'))
+
     ebook_formats = constants.EBOOK_FORMATS
     formats = ebook_formats + ['html', 'xml']
 
@@ -139,6 +97,8 @@ class Book(models.Model):
 
     SORT_KEY_SEP = '$'
 
+    is_book = True
+
     class AlreadyExists(Exception):
         pass
 
@@ -293,7 +253,7 @@ class Book(models.Model):
 
         if self.parent.html_file:
             return self.parent
-        
+
         return self.parent.get_prev_text()
 
     def get_next_text(self, inside=True):
@@ -327,7 +287,7 @@ class Book(models.Model):
 
     def get_children(self):
         return self.children.all().order_by('parent_number')
-    
+
     @property
     def name(self):
         return self.title
@@ -401,6 +361,9 @@ class Book(models.Model):
     def get_daisy(self):
         return self.get_media("daisy")
 
+    def get_audio_epub(self):
+        return self.get_media("audio.epub")
+
     def media_url(self, format_):
         media = self.get_media(format_)
         if media:
@@ -452,10 +415,32 @@ class Book(models.Model):
     has_daisy_file.short_description = 'DAISY'
     has_daisy_file.boolean = True
 
+    def has_sync_file(self):
+        return self.has_media("sync")
+
+    def get_sync(self):
+        with self.get_media('sync').first().file.open('r') as f:
+            sync = f.read().split('\n')
+        offset = float(sync[0])
+        items = []
+        for line in sync[1:]:
+            if not line:
+                continue
+            start, end, elid = line.split()
+            items.append([elid, float(start) + offset])
+        return json.dumps(items)
+    
+    def has_audio_epub_file(self):
+        return self.has_media("audio.epub")
+
     @property
     def media_daisy(self):
         return self.get_media('daisy')
-    
+
+    @property
+    def media_audio_epub(self):
+        return self.get_media('audio.epub')
+
     def get_audiobooks(self):
         ogg_files = {}
         for m in self.media.filter(type='ogg').order_by().iterator():
@@ -524,9 +509,10 @@ class Book(models.Model):
                 format_)
 
         field_name = "%s_file" % format_
+        field = getattr(Book, field_name)
         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_])
+        return create_zip(paths, field.ZIP)
 
     def zip_audiobooks(self, format_):
         bm = BookMedia.objects.filter(book=self, type=format_)
@@ -540,24 +526,15 @@ class Book(models.Model):
                 licenses.add(license)
         readme = render_to_string('catalogue/audiobook_zip_readme.txt', {
             'licenses': licenses,
+            'meta': self.wldocument2().meta,
         })
         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):
+    def search_index(self, index=None):
         if not self.findable:
             return
-        if index is None:
-            from search.index import Index
-            index = Index()
-        try:
-            index.index_book(self, book_info)
-            if index_tags:
-                index.index_tags()
-            if commit:
-                index.index.commit()
-        except Exception as e:
-            index.index.rollback()
-            raise e
+        from search.index import Index
+        Index.index_book(self)
 
     # will make problems in conjunction with paid previews
     def download_pictures(self, remote_gallery_url):
@@ -598,7 +575,7 @@ class Book(models.Model):
                 a.attrib['href'] = html_link + a.attrib['href']
             self.toc = html.tostring(toc, encoding='unicode')
             # div#toc
-            
+
     @classmethod
     def from_xml_file(cls, xml_file, **kwargs):
         from django.core.files import File
@@ -617,7 +594,9 @@ 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, findable=True):
+                           remote_gallery_url=None, days=0, findable=True):
+        from catalogue import tasks
+
         if dont_build is None:
             dont_build = set()
         dont_build = set.union(set(dont_build), set(app_settings.DONT_BUILD))
@@ -724,7 +703,7 @@ class Book(models.Model):
                 getattr(book, '%s_file' % format_).build_delay()
 
         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)
+            tasks.index_book.delay(book.id)
 
         for child in notify_cover_changed:
             child.parent_cover_changed()
@@ -735,46 +714,43 @@ class Book(models.Model):
         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()
-        if master is None:
-            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()
-    
+        Entity = apps.get_model('references', 'Entity')
+        doc = self.wldocument2()
+        doc._compat_assign_section_ids()
+        doc._compat_assign_ordered_ids()
+        refs = {}
+        for ref_elem in doc.references():
+            uri = ref_elem.attrib.get('href', '')
+            if not uri:
+                continue
+            if uri in refs:
+                ref = refs[uri]
+            else:
+                entity, entity_created = Entity.objects.get_or_create(uri=uri)
+                if entity_created:
+                    entity.populate()
+                    entity.save()
+                ref, ref_created = entity.reference_set.get_or_create(book=self)
+                refs[uri] = ref
+                if not ref_created:
+                    ref.occurence_set.all().delete()
+            sec = ref_elem.get_link()
+            m = re.match(r'sec(\d+)', sec)
+            assert m is not None
+            sec = int(m.group(1))
+            snippet = ref_elem.get_snippet()
+            b = builders['html-snippet']()
+            for s in snippet:
+                s.html_build(b)
+            html = b.output().get_bytes().decode('utf-8')
+
+            ref.occurence_set.create(
+                section=sec,
+                html=html
+            )
+        self.reference_set.exclude(entity__uri__in=refs).delete()
+
     @property
     def references(self):
         return self.reference_set.all().select_related('entity')
@@ -820,7 +796,7 @@ class Book(models.Model):
             yield self.parent
         else:
             return []
-                    
+
     def clear_cache(self):
         clear_cached_renders(self.mini_box)
         clear_cached_renders(self.mini_box_nolink)
@@ -991,7 +967,7 @@ class Book(models.Model):
             return fragments[0]
         else:
             return None
-        
+
     def fragment_data(self):
         fragment = self.choose_fragment()
         if fragment:
@@ -1044,28 +1020,6 @@ class Book(models.Model):
             'no_link': True,
         }
 
-def add_file_fields():
-    for format_ in Book.formats:
-        field_name = "%s_file" % format_
-        # This weird globals() assignment makes Django migrations comfortable.
-        _upload_to = _ebook_upload_to('book/%s/%%s.%s' % (format_, format_))
-        _upload_to.__name__ = '_%s_upload_to' % format_
-        globals()[_upload_to.__name__] = _upload_to
-
-        EbookField(
-            format_, _("%s file" % format_.upper()),
-            upload_to=_upload_to,
-            storage=bofh_storage,
-            max_length=255,
-            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, models.CASCADE, related_name='popularity')