fix conversion wl2 to wl1

[redakcja.git] / apps / catalogue / models / book.py

diff --git a/apps/catalogue/models/book.py b/apps/catalogue/models/book.py
index f4e025e..b0862f0 100755 (executable)
--- a/apps/catalogue/models/book.py
+++ b/apps/catalogue/models/book.py
@@ -4,19 +4,18 @@
 # Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
 #
 from django.contrib.sites.models import Site
 # Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
 #
 from django.contrib.sites.models import Site

-from django.db import models
+from django.db import models, transaction

 from django.template.loader import render_to_string
 from django.utils.translation import ugettext_lazy as _
 from slughifi import slughifi
 from django.template.loader import render_to_string
 from django.utils.translation import ugettext_lazy as _
 from slughifi import slughifi

-from librarian import NoDublinCore, ParseError, ValidationError
-from librarian.dcparser import BookInfo

 
 import apiclient
 
 import apiclient

-from catalogue.helpers import cached_in_field
-from catalogue.models import BookPublishRecord, ChunkPublishRecord
+from catalogue.helpers import cached_in_field, GalleryMerger
+from catalogue.models import BookPublishRecord, ChunkPublishRecord, Project

 from catalogue.signals import post_publish
 from catalogue.signals import post_publish

-from catalogue.tasks import refresh_instance
+from catalogue.tasks import refresh_instance, book_content_updated

 from catalogue.xml_tools import compile_text, split_xml
 from catalogue.xml_tools import compile_text, split_xml

+from cover.models import Image

 
 
 class Book(models.Model):
 
 
 class Book(models.Model):


@@ -24,30 +23,35 @@ 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)
 
     title = models.CharField(_('title'), max_length=255, db_index=True)
     slug = models.SlugField(_('slug'), max_length=128, unique=True, db_index=True)

-    gallery = models.CharField(_('scan gallery name'), max_length=255, blank=True)
+    public = models.BooleanField(_('public'), default=True, db_index=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")
-    parent_number = models.IntegerField(_('parent number'), null=True, blank=True, db_index=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)
+    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)
 
     # 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)

-
-    # Managers
-    objects = models.Manager()
+    _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, editable=False)
+    dc_slug = models.CharField(max_length=128, null=True, blank=True, editable=False, db_index=True)

 
     class NoTextError(BaseException):
         pass
 
     class Meta:
         app_label = 'catalogue'
 
     class NoTextError(BaseException):
         pass
 
     class Meta:
         app_label = 'catalogue'

-        ordering = ['parent_number', 'title']
-        verbose_name = _('book')
-        verbose_name_plural = _('books')
-
+        ordering = ['title', 'slug']
+        verbose_name = u'moduł'
+        verbose_name_plural = u'moduły'

 
     # Representing
     # ============
 
     # Representing
     # ============


@@ -72,15 +76,24 @@ class Book(models.Model):
 
     @models.permalink
     def get_absolute_url(self):
 
     @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" % (
+            Site.objects.get_current().domain,
+            self.get_absolute_url()
+        )

 
     # Creating & manipulating
     # =======================
 
 
     # Creating & manipulating
     # =======================
 

+    def accessible(self, request):
+        return self.public or request.user.is_authenticated()
+

     @classmethod
     @classmethod

-    def create(cls, creator, text, *args, **kwargs):
-        b = cls.objects.create(*args, **kwargs)
+    @transaction.commit_on_success
+    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
         b.chunk_set.all().update(creator=creator)
         b[0].commit(text, author=creator)
         return b


@@ -90,8 +103,8 @@ class Book(models.Model):
         return self.chunk_set.reverse()[0].split(*args, **kwargs)
 
     @classmethod
         return self.chunk_set.reverse()[0].split(*args, **kwargs)
 
     @classmethod

-    def import_xml_text(cls, text=u'', previous_book=None,
-                commit_args=None, **kwargs):
+    @transaction.commit_on_success
+    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:
         """Imports a book from XML, splitting it into chunks as necessary."""
         texts = split_xml(text)
         if previous_book:


@@ -105,7 +118,6 @@ class Book(models.Model):
         for i in range(book_len - len(texts)):
             texts.append((u'pusta część %d' % (i + 1), u''))
 
         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)
         for i, (title, text) in enumerate(texts):
             if not title:
                 title = u'część %d' % (i + 1)


@@ -130,14 +142,17 @@ class Book(models.Model):
         """
         slugs = set(c.slug for c in self)
         i = 1
         """
         slugs = set(c.slug for c in self)
         i = 1

-        new_slug = proposed
+        new_slug = proposed[:50]

         while new_slug in slugs:
         while new_slug in slugs:

-            new_slug = "%s_%d" % (proposed, i)
+            new_slug = "%s_%d" % (proposed[:45], i)

             i += 1
         return new_slug
 
             i += 1
         return new_slug
 

+    @transaction.commit_on_success

     def append(self, other, slugs=None, titles=None):
         """Add all chunks of another book to self."""
     def append(self, other, slugs=None, titles=None):
         """Add all chunks of another book to self."""

+        assert self != other
+

         number = self[len(self) - 1].number + 1
         len_other = len(other)
         single = len_other == 1
         number = self[len(self) - 1].number + 1
         len_other = len(other)
         single = len_other == 1


@@ -166,12 +181,11 @@ class Book(models.Model):
                     # just use the guessed title and original book slug
                     chunk.title = other_title_part
                     if other.slug.startswith(self.slug):
                     # just use the guessed title and original book slug
                     chunk.title = other_title_part
                     if other.slug.startswith(self.slug):

-                        chunk_slug = other.slug[len(self.slug):].lstrip('-_')
+                        chunk.slug = other.slug[len(self.slug):].lstrip('-_')

                     else:
                     else:

-                        chunk_slug = other.slug
-                    chunk.slug = self.make_chunk_slug(chunk_slug)
+                        chunk.slug = other.slug

                 else:
                 else:

-                    chunk.title = "%s, %s" % (other_title_part, chunk.title)
+                    chunk.title = ("%s, %s" % (other_title_part, chunk.title))[:255]

             else:
                 chunk.slug = slugs[i]
                 chunk.title = titles[i]
             else:
                 chunk.slug = slugs[i]
                 chunk.title = titles[i]


@@ -179,8 +193,46 @@ class Book(models.Model):
             chunk.slug = self.make_chunk_slug(chunk.slug)
             chunk.save()
             number += 1
             chunk.slug = self.make_chunk_slug(chunk.slug)
             chunk.save()
             number += 1

+        assert not other.chunk_set.exists()
+
+        gm = GalleryMerger(self.gallery, other.gallery)
+        self.gallery = gm.merge()
+
+        # and move the gallery starts
+        if gm.was_merged:
+                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()
 
         other.delete()
 

+    @transaction.commit_on_success
+    def prepend_history(self, other):
+        """Prepend history from all the other book's chunks to own."""
+        assert self != other
+
+        for i in range(len(self), len(other)):
+            title = u"pusta część %d" % i
+            chunk = self.add(slughifi(title), title)
+            chunk.commit('')
+
+        for i in range(len(other)):
+            self[i].prepend_history(other[0])
+
+        assert not other.chunk_set.exists()
+        other.delete()
+
+    def split(self):
+        """Splits all the chunks into separate books."""
+        for chunk in self:
+            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
+            chunk.save()
+        assert not self.chunk_set.exists()
+        self.delete()

 
     # State & cache
     # =============
 
     # State & cache
     # =============


@@ -194,23 +246,34 @@ class Book(models.Model):
     def assert_publishable(self):
         assert self.chunk_set.exists(), _('No chunks in the book.')
         try:
     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.'))
         except self.NoTextError:
             raise AssertionError(_('Not all chunks have publishable revisions.'))

-        book_xml = self.materialize(changes=changes)
+
+        from librarian import NoDublinCore, ParseError, ValidationError

 
         try:
 
         try:

-            bi = BookInfo.from_string(book_xml.encode('utf-8'))
+            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:
         except ParseError, e:

-            raise AssertionError(_('Invalid XML') + ': ' + str(e))
+            raise AssertionError(_('Invalid XML') + ': ' + unicode(e))

         except NoDublinCore:
             raise AssertionError(_('No Dublin Core found.'))
         except ValidationError, e:
         except NoDublinCore:
             raise AssertionError(_('No Dublin Core found.'))
         except ValidationError, e:

-            raise AssertionError(_('Invalid Dublin Core') + ': ' + str(e))
+            raise AssertionError(_('Invalid Dublin Core') + ': ' + unicode(e))

 
 

-        valid_about = "http://%s%s" % (Site.objects.get_current().domain, self.get_absolute_url())
+        valid_about = self.correct_about()

         assert bi.about == valid_about, _("rdf:about is not") + " " + valid_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('.')
 
     def hidden(self):
         return self.slug.startswith('.')
 


@@ -237,6 +300,15 @@ class Book(models.Model):
         return self.publish_log.exists()
     published = cached_in_field('_published')(is_published)
 
         return self.publish_log.exists()
     published = cached_in_field('_published')(is_published)
 

+    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]
+        if not len(stages):
+            return 0
+        return min(stages)
+    on_track = cached_in_field('_on_track')(get_on_track)
+

     def is_single(self):
         return len(self) == 1
     single = cached_in_field('_single')(is_single)
     def is_single(self):
         return len(self) == 1
     single = cached_in_field('_single')(is_single)


@@ -245,11 +317,47 @@ class Book(models.Model):
     def short_html(self):
         return render_to_string('catalogue/book_list/book.html', {'book': self})
 
     def short_html(self):
         return render_to_string('catalogue/book_list/book.html', {'book': self})
 

+    def book_info(self, publishable=True):
+        try:
+            book_xml = self.materialize(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'))
+            except (self.NoTextError, ParseError, NoDublinCore, ValidationError):
+                return None
+
+    def refresh_dc_cache(self):
+        update = {
+            'dc_slug': None,
+            'dc_cover_image': None,
+        }
+
+        info = self.book_info()
+        if info is not None:
+            update['dc_slug'] = info.url.slug
+            if info.cover_source:
+                try:
+                    image = Image.objects.get(pk=int(info.cover_source.rstrip('/').rsplit('/', 1)[-1]))
+                except Image.DoesNotExist:
+                    pass
+                else:
+                    if info.cover_source == image.get_full_url():
+                        update['dc_cover_image'] = image
+        Book.objects.filter(pk=self.pk).update(**update)
+

     def touch(self):
     def touch(self):

+        # this should only really be done when text or publishable status changes
+        book_content_updated.delay(self)
+

         update = {
             "_new_publishable": self.is_new_publishable(),
             "_published": self.is_published(),
             "_single": self.is_single(),
         update = {
             "_new_publishable": self.is_new_publishable(),
             "_published": self.is_published(),
             "_single": self.is_single(),

+            "_on_track": self.get_on_track(),

             "_short_html": None,
         }
         Book.objects.filter(pk=self.pk).update(**update)
             "_short_html": None,
         }
         Book.objects.filter(pk=self.pk).update(**update)


@@ -290,16 +398,88 @@ class Book(models.Model):
             changes = self.get_current_changes(publishable)
         return compile_text(change.materialize() for change in changes)
 
             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, strict=False):
+        from catalogue.ebook_utils import RedakcjaDocProvider
+        from librarian.parser import WLDocument
+
+        return WLDocument.from_string(
+                self.wl1_xml(publishable=publishable, changes=changes),
+                provider=RedakcjaDocProvider(publishable=publishable),
+                parse_dublincore=parse_dublincore,
+                strict=strict)
+

     def publish(self, user):
         """
             Publishes a book on behalf of a (local) user.
         """
         self.assert_publishable()
     def publish(self, user):
         """
             Publishes a book on behalf of a (local) user.
         """
         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()
+        book_xml = self.wl1_xml(changes=changes)
+        apiclient.api_call(user, "lessons/", {"lesson_xml": book_xml})

         # 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)
         # 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
+
+        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'
+            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 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')