librarian update
[redakcja.git] / apps / catalogue / models / book.py
index ca38998..4c35e68 100755 (executable)
@@ -7,20 +7,16 @@ from django.contrib.sites.models import Site
 from django.db import models, transaction
 from django.template.loader import render_to_string
 from django.utils.translation import ugettext_lazy as _
-from django.conf import settings
 from slughifi import slughifi
 
-
 import apiclient
 from catalogue.helpers import cached_in_field, GalleryMerger
-from catalogue.models import BookPublishRecord, ChunkPublishRecord
+from catalogue.models import BookPublishRecord, ChunkPublishRecord, Project
 from catalogue.signals import post_publish
 from catalogue.tasks import refresh_instance, book_content_updated
 from catalogue.xml_tools import compile_text, split_xml
 from cover.models import Image
-import os
-import shutil
-import re
+
 
 class Book(models.Model):
     """ A document edited on the wiki """
@@ -28,21 +24,25 @@ class Book(models.Model):
     title = models.CharField(_('title'), max_length=255, db_index=True)
     slug = models.SlugField(_('slug'), max_length=128, unique=True, db_index=True)
     public = models.BooleanField(_('public'), default=True, db_index=True)
-    gallery = models.CharField(_('scan gallery name'), max_length=255, blank=True)
+    gallery = models.CharField(u'materiały', max_length=255, blank=True)
+    project = models.ForeignKey(Project, null=True, blank=True)
 
-    #wl_slug = models.CharField(_('title'), max_length=255, null=True, db_index=True, editable=False)
-    parent = models.ForeignKey('self', null=True, blank=True, verbose_name=_('parent'), related_name="children", editable=False)
+    # wl_slug = models.CharField(_('title'), max_length=255, null=True, db_index=True, editable=False)
+    parent = models.ForeignKey(
+        'self', null=True, blank=True, verbose_name=_('parent'), related_name="children", editable=False)
     parent_number = models.IntegerField(_('parent number'), null=True, blank=True, db_index=True, editable=False)
 
+    for_cybernauts = models.BooleanField(_('for Cybernauts'), default=False)
+
     # Cache
     _short_html = models.TextField(null=True, blank=True, editable=False)
     _single = models.NullBooleanField(editable=False, db_index=True)
     _new_publishable = models.NullBooleanField(editable=False)
     _published = models.NullBooleanField(editable=False)
     _on_track = models.IntegerField(null=True, blank=True, db_index=True, editable=False)
-    dc_cover_image = models.ForeignKey(Image, blank=True, null=True, db_index=True, on_delete=models.SET_NULL)
-    dc_slug = models.CharField(max_length=128, null=True, blank=True,
-            editable=False, db_index=True)
+    dc_cover_image = models.ForeignKey(
+        Image, blank=True, null=True, db_index=True, on_delete=models.SET_NULL, editable=False)
+    dc_slug = models.CharField(max_length=128, null=True, blank=True, editable=False, db_index=True)
 
     class NoTextError(BaseException):
         pass
@@ -50,9 +50,8 @@ class Book(models.Model):
     class Meta:
         app_label = 'catalogue'
         ordering = ['title', 'slug']
-        verbose_name = _('book')
-        verbose_name_plural = _('books')
-
+        verbose_name = u'moduł'
+        verbose_name_plural = u'moduły'
 
     # Representing
     # ============
@@ -77,7 +76,7 @@ class Book(models.Model):
 
     @models.permalink
     def get_absolute_url(self):
-        return ("catalogue_book", [self.slug])
+        return "catalogue_book", [self.slug]
 
     def correct_about(self):
         return "http://%s%s" % (
@@ -93,8 +92,8 @@ class Book(models.Model):
 
     @classmethod
     @transaction.commit_on_success
-    def create(cls, creator, text, *args, **kwargs):
-        b = cls.objects.create(*args, **kwargs)
+    def create(cls, creator, text, **kwargs):
+        b = cls.objects.create(**kwargs)
         b.chunk_set.all().update(creator=creator)
         b[0].commit(text, author=creator)
         return b
@@ -105,8 +104,7 @@ class Book(models.Model):
 
     @classmethod
     @transaction.commit_on_success
-    def import_xml_text(cls, text=u'', previous_book=None,
-                commit_args=None, **kwargs):
+    def import_xml_text(cls, text=u'', previous_book=None, commit_args=None, **kwargs):
         """Imports a book from XML, splitting it into chunks as necessary."""
         texts = split_xml(text)
         if previous_book:
@@ -120,7 +118,6 @@ class Book(models.Model):
         for i in range(book_len - len(texts)):
             texts.append((u'pusta część %d' % (i + 1), u''))
 
-        i = 0
         for i, (title, text) in enumerate(texts):
             if not title:
                 title = u'część %d' % (i + 1)
@@ -140,7 +137,7 @@ class Book(models.Model):
         return instance
 
     def make_chunk_slug(self, proposed):
-        """ 
+        """
             Finds a chunk slug not yet used in the book.
         """
         slugs = set(c.slug for c in self)
@@ -203,13 +200,13 @@ class Book(models.Model):
 
         # and move the gallery starts
         if gm.was_merged:
-                for chunk in self[len(self) - len_other:]:
-                        chunk.gallery_start += gm.dest_size - gm.num_deleted
-                        chunk.save()
+            for chunk in self[len(self) - len_other:]:
+                old_start = chunk.gallery_start or 1
+                chunk.gallery_start = old_start + gm.dest_size - gm.num_deleted
+                chunk.save()
 
         other.delete()
 
-
     @transaction.commit_on_success
     def prepend_history(self, other):
         """Prepend history from all the other book's chunks to own."""
@@ -228,10 +225,8 @@ class Book(models.Model):
 
     def split(self):
         """Splits all the chunks into separate books."""
-        self.title
         for chunk in self:
-            book = Book.objects.create(title=chunk.title, slug=chunk.slug,
-                    public=self.public, gallery=self.gallery)
+            book = Book.objects.create(title=chunk.title, slug=chunk.slug, public=self.public, gallery=self.gallery)
             book[0].delete()
             chunk.book = book
             chunk.number = 1
@@ -251,16 +246,18 @@ class Book(models.Model):
     def assert_publishable(self):
         assert self.chunk_set.exists(), _('No chunks in the book.')
         try:
-            changes = self.get_current_changes(publishable=True)
+            changes = self.get_current_changes()
         except self.NoTextError:
             raise AssertionError(_('Not all chunks have publishable revisions.'))
-        book_xml = self.materialize(changes=changes)
 
-        from librarian.dcparser import BookInfo
         from librarian import NoDublinCore, ParseError, ValidationError
 
         try:
-            bi = BookInfo.from_string(book_xml.encode('utf-8'), strict=True)
+            bi = self.wldocument(changes=changes, strict=True).book_info
+            if not bi.audience:
+                raise ValidationError('No audience specified')
+            if not bi.type:
+                raise ValidationError('No type specified')
         except ParseError, e:
             raise AssertionError(_('Invalid XML') + ': ' + unicode(e))
         except NoDublinCore:
@@ -271,6 +268,12 @@ class Book(models.Model):
         valid_about = self.correct_about()
         assert bi.about == valid_about, _("rdf:about is not") + " " + valid_about
 
+    def publishable_error(self):
+        try:
+            return self.assert_publishable()
+        except AssertionError, e:
+            return e
+
     def hidden(self):
         return self.slug.startswith('.')
 
@@ -300,8 +303,7 @@ class Book(models.Model):
     def get_on_track(self):
         if self.published:
             return -1
-        stages = [ch.stage.ordering if ch.stage is not None else 0
-                    for ch in self]
+        stages = [ch.stage.ordering if ch.stage is not None else 0 for ch in self]
         if not len(stages):
             return 0
         return min(stages)
@@ -317,14 +319,14 @@ class Book(models.Model):
 
     def book_info(self, publishable=True):
         try:
-            book_xml = self.materialize(publishable=publishable)
+            book_xml = self.wl1_xml(publishable=publishable)
         except self.NoTextError:
             pass
         else:
             from librarian.dcparser import BookInfo
             from librarian import NoDublinCore, ParseError, ValidationError
             try:
-                return BookInfo.from_string(book_xml.encode('utf-8'))
+                return BookInfo.from_string(book_xml)
             except (self.NoTextError, ParseError, NoDublinCore, ValidationError):
                 return None
 
@@ -340,7 +342,7 @@ class Book(models.Model):
             if info.cover_source:
                 try:
                     image = Image.objects.get(pk=int(info.cover_source.rstrip('/').rsplit('/', 1)[-1]))
-                except:
+                except Image.DoesNotExist:
                     pass
                 else:
                     if info.cover_source == image.get_full_url():
@@ -386,7 +388,7 @@ class Book(models.Model):
         return changes
 
     def materialize(self, publishable=False, changes=None):
-        """ 
+        """
             Get full text of the document compiled from chunks.
             Takes the current versions of all texts
             or versions most recently tagged for publishing,
@@ -396,25 +398,102 @@ class Book(models.Model):
             changes = self.get_current_changes(publishable)
         return compile_text(change.materialize() for change in changes)
 
-    def wldocument(self, publishable=True, changes=None, parse_dublincore=True):
+    def wldocument(self, publishable=True, changes=None, parse_dublincore=True, strict=False):
         from catalogue.ebook_utils import RedakcjaDocProvider
         from librarian.parser import WLDocument
 
         return WLDocument.from_string(
-                self.materialize(publishable=publishable, changes=changes),
+                self.wl1_xml(publishable=publishable, changes=changes),
                 provider=RedakcjaDocProvider(publishable=publishable),
-                parse_dublincore=parse_dublincore)
+                parse_dublincore=parse_dublincore,
+                strict=strict)
 
-    def publish(self, user):
+    def publish(self, user, host=None):
         """
             Publishes a book on behalf of a (local) user.
         """
+        import json
+        import os
+        from django.conf import settings
         self.assert_publishable()
-        changes = self.get_current_changes(publishable=True)
-        book_xml = self.materialize(changes=changes)
-        apiclient.api_call(user, "books/", {"book_xml": book_xml})
+        changes = self.get_current_changes()
+        data = {"lesson_xml": self.wl1_xml(changes=changes)}
+        if host:
+            gallery_url = u'%s%s%s%s/' % (host, settings.MEDIA_URL, settings.IMAGE_DIR, self.slug)
+            gallery_dir = os.path.join(settings.MEDIA_ROOT, settings.IMAGE_DIR, self.slug)
+            if os.path.isdir(gallery_dir):
+                data['gallery_url'] = gallery_url
+                data['attachments'] = json.dumps(os.listdir(gallery_dir))
+        apiclient.api_call(user, "lessons/", data)
         # record the publish
         br = BookPublishRecord.objects.create(book=self, user=user)
         for c in changes:
             ChunkPublishRecord.objects.create(book_record=br, change=c)
         post_publish.send(sender=br)
+
+    def wl1_xml(self, publishable=True, changes=None):
+        from lxml import etree
+        import re
+        from StringIO import StringIO
+        from urllib import unquote
+        import os.path
+        from django.conf import settings
+        from fnpdjango.utils.text.slughifi import slughifi
+        from librarian import ParseError, DCNS
+
+        def _register_function(f):
+            """ Register extension function with lxml """
+            ns = etree.FunctionNamespace('http://wolnelektury.pl/functions')
+            ns[f.__name__] = f
+            return f
+
+        @_register_function
+        def slugify(context, text):
+            """Remove unneeded whitespace from beginning and end"""
+            if isinstance(text, list):
+                text = ''.join(text)
+            return slughifi(text)
+
+        @_register_function
+        def rmext(context, text):
+            if isinstance(text, list):
+                text = ''.join(text)
+            text = unquote(text)
+            if '.' in text:
+                name, ext = text.rsplit('.', 1)
+                if ext.lower() in ('doc', 'docx', 'odt', 'pdf', 'jpg', 'jpeg'):
+                    text = name
+            return text
+
+        t = etree.parse(os.path.join(settings.PROJECT_ROOT, 'xslt/wl2to1.xslt'))
+        ft = self.materialize(publishable=publishable, changes=changes)
+        ft = ft.replace(' ', ' ')
+        f2 = StringIO(ft)
+        i1 = etree.parse(f2)
+
+        for sect in i1.findall('//section'):
+            if sect[0].text and sect[0].text.strip() == u'Przebieg zajęć':
+                # Prostujemy.
+                first = sect.find('section')
+                subs = first.findall('.//section')
+                for sub in subs:
+                    sect.append(sub)
+                break
+        else:
+            # print 'BRAK PRZEBIEGU'
+            dc_type = i1.findall('//dc:type', namespaces={'dc': DCNS.uri})
+            if dc_type and dc_type[0] in ('course', 'synthetic'):
+                raise ParseError('Brak przebiegu')
+
+        i1.getroot().attrib['redslug'] = self.slug
+        i1.getroot().attrib['wlslug'] = self.slug  # THIS!
+        # print '.',
+        w1t = i1.xslt(t)
+        for h in w1t.findall('//aktywnosc/opis'):
+            if len(h) == 0:
+                raise ParseError('Pusty element aktywnosc/opis')
+            # FIXME assumption that every lesson has at most 9 parts
+            if not h[0].text or not re.match(r'\d\.\s', h[0].text):
+                raise ParseError('Niepoprawny nagłówek (aktywnosc/opis): %s' % repr(h[0].text))
+            h[0].text = h[0].text[3:]
+        return etree.tostring(w1t, encoding='utf-8')