genrowanie pdf: usnięcie bloku try..catch; xml na pewno jest poprawny, ponieważ zosta...

[wolnelektury.git] / apps / api / handlers.py

diff --git a/apps/api/handlers.py b/apps/api/handlers.py
index cca504f..685dfaf 100644 (file)
--- a/apps/api/handlers.py
+++ b/apps/api/handlers.py
@@ -2,42 +2,302 @@
 # This file is part of Wolnelektury, licensed under GNU Affero GPLv3 or later.
 # Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
 
 # This file is part of Wolnelektury, licensed under GNU Affero GPLv3 or later.
 # Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
 

-from datetime import datetime
-from django.shortcuts import get_object_or_404
-from django.contrib.auth.decorators import login_required, user_passes_test
-from django.utils import simplejson as json
+from datetime import datetime, timedelta
+
+from django.conf import settings
+from django.contrib.sites.models import Site
+from django.core.urlresolvers import reverse

 from piston.handler import BaseHandler
 from piston.handler import BaseHandler

-from piston.utils import rc, validate
+from piston.utils import rc

 
 

-from api.models import Deleted

 from api.helpers import timestamp
 from api.helpers import timestamp

-from catalogue.models import Book, Tag
-from catalogue.forms import BookImportForm
-from catalogue.views import tagged_object_list 
-"""
-class TagHandler(BaseHandler):
-   allowed_methods = ('GET',)
-   model = Tag   
-
-   def read(self, request, tags=''):
-      if tags == '':
-        return Tag.objects.all()
-      else:
-        return tagged_object_list(request, tags, api=True)
-
-class BookHandler(BaseHandler):
+from api.models import Deleted
+from catalogue.models import Book, Tag, BookMedia, Fragment
+
+
+API_BASE = WL_BASE = MEDIA_BASE = 'http://' + Site.objects.get_current().domain
+
+
+category_singular = {
+    'authors': 'author',
+    'kinds': 'kind',
+    'genres': 'genre',
+    'epochs': 'epoch',
+    'themes': 'theme',
+    'books': 'book',
+}
+category_plural={}
+for k, v in category_singular.items():
+    category_plural[v] = k
+
+
+def read_tags(tags, allowed):
+    """ Reads a path of filtering tags.
+
+    :param str tags: a path of category and slug pairs, like: authors/an-author/...
+    :returns: list of Tag objects
+    :raises: ValueError when tags can't be found
+    """
+    if not tags:
+        return []
+
+    tags = tags.strip('/').split('/')
+    real_tags = []
+    while tags:
+        category = tags.pop(0)
+        slug = tags.pop(0)
+
+        try:
+            category = category_singular[category]
+        except KeyError:
+            raise ValueError('Unknown category.')
+
+        if not category in allowed:
+            raise ValueError('Category not allowed.')
+
+        # !^%@#$^#!
+        if category == 'book':
+            slug = 'l-' + slug
+
+        try:
+            real_tags.append(Tag.objects.get(category=category, slug=slug))
+        except Tag.DoesNotExist:
+            raise ValueError('Tag not found')
+    return real_tags
+
+
+# RESTful handlers
+
+
+class BookMediaHandler(BaseHandler):
+    """ Responsible for representing media in Books. """
+
+    model = BookMedia
+    fields = ['name', 'type', 'url']
+
+    @classmethod
+    def url(cls, media):
+        """ Link to media on site. """
+
+        return MEDIA_BASE + media.file.url
+
+
+class BookDetailHandler(BaseHandler):
+    """ Main handler for Book objects.
+
+    Responsible for lists of Book objects
+    and fields used for representing Books.
+
+    """
+    allowed_methods = ['GET']
+    fields = ['title', 'parent',
+        'xml', 'html', 'pdf', 'epub', 'txt',
+        'media', 'url'] + category_singular.keys()
+
+    def read(self, request, slug):
+        """ Returns details of a book, identified by a slug. """
+
+        try:
+            return Book.objects.get(slug=slug)
+        except Book.DoesNotExist:
+            return rc.NOT_FOUND
+
+
+class BooksHandler(BaseHandler):
+    """ Main handler for Book objects.
+
+    Responsible for lists of Book objects
+    and fields used for representing Books.
+
+    """
+    allowed_methods = ('GET',)

     model = Book
     model = Book

-    #fields = ('slug', 'title')
+    fields = ['href', 'title']
+
+    categories = set(['author', 'epoch', 'kind', 'genre'])
+
+    @classmethod
+    def href(cls, book):
+        """ Returns an URI for a Book in the API. """
+        return API_BASE + reverse("api_book", args=[book.slug])
+
+    @classmethod
+    def url(cls, book):
+        """ Returns Book's URL on the site. """
+
+        return WL_BASE + book.get_absolute_url()
+
+    def read(self, request, tags, top_level=False):
+        """ Lists all books with given tags.
+
+        :param tags: filtering tags; should be a path of categories
+             and slugs, i.e.: authors/an-author/epoch/an-epoch/
+        :param top_level: if True and a book is included in the results,
+             it's children are aren't. By default all books matching the tags
+             are returned.
+        """
+        tags = read_tags(tags, allowed=self.categories)
+        if tags:
+            if top_level:
+                books = Book.tagged_top_level(tags)
+                return books if books else rc.NOT_FOUND
+            else:
+                books = Book.tagged.with_all(tags)
+        else:
+            books = Book.objects.all()
+
+        if books.exists():
+            return books
+        else:
+            return rc.NOT_FOUND
+
+    @classmethod
+    def media(self, book):
+        """ Returns all media for a book. """
+
+        return book.media.all()
+
+
+# add categorized tags fields for Book
+def _tags_getter(category):
+    @classmethod
+    def get_tags(cls, book):
+        return book.tags.filter(category=category)
+    return get_tags
+for plural, singular in category_singular.items():
+    setattr(BooksHandler, plural, _tags_getter(singular))
+
+# add fields for files in Book
+def _file_getter(format):
+    field = "%s_file" % format
+    @classmethod
+    def get_file(cls, book):
+        f = getattr(book, field)
+        if f:
+            return MEDIA_BASE + f.url
+        else:
+            return ''
+    return get_file
+for format in ('xml', 'txt', 'html', 'epub', 'pdf'):
+    setattr(BooksHandler, format, _file_getter(format))
+
+
+class TagDetailHandler(BaseHandler):
+    """ Responsible for details of a single Tag object. """
+
+    fields = ['name', 'sort_key', 'description']
+
+    def read(self, request, category, slug):
+        """ Returns details of a tag, identified by category and slug. """
+
+        try:
+            category_sng = category_singular[category]
+        except KeyError, e:
+            return rc.NOT_FOUND
+
+        try:
+            return Tag.objects.get(category=category_sng, slug=slug)
+        except Tag.DoesNotExist:
+            return rc.NOT_FOUND
+
+
+class TagsHandler(BaseHandler):
+    """ Main handler for Tag objects.
+
+    Responsible for lists of Tag objects
+    and fields used for representing Tags.
+
+    """
+    allowed_methods = ('GET',)
+    model = Tag
+    fields = ['name', 'href']
+
+    def read(self, request, category):
+        """ Lists all tags in the category (eg. all themes). """
+
+        try:
+            category_sng = category_singular[category]
+        except KeyError, e:
+            return rc.NOT_FOUND
+
+        tags = Tag.objects.filter(category=category_sng)
+        tags = [t for t in tags if t.get_count() > 0]
+        if tags:
+            return tags
+        else:
+            return rc.NOT_FOUND
+
+
+    @classmethod
+    def href(cls, tag):
+        """ Returns URI in the API for the tag. """
+
+        return API_BASE + reverse("api_tag", args=[category_plural[tag.category], tag.slug])
+
+
+class FragmentDetailHandler(BaseHandler):
+    fields = ['book', 'anchor', 'text', 'url', 'themes']
+
+    def read(self, request, slug, anchor):
+        """ Returns details of a fragment, identified by book slug and anchor. """
+
+        try:
+            return Fragment.objects.get(book__slug=slug, anchor=anchor)
+        except Fragment.DoesNotExist:
+            return rc.NOT_FOUND

 
 

-    def read(self, request, slug=None):
-        if slug:
-            return get_object_or_404(Book, slug=slug)
+
+class FragmentsHandler(BaseHandler):
+    """ Main handler for Fragments.
+
+    Responsible for lists of Fragment objects
+    and fields used for representing Fragments.
+
+    """
+    model = Fragment
+    fields = ['book', 'anchor', 'href']
+    allowed_methods = ('GET',)
+
+    categories = set(['author', 'epoch', 'kind', 'genre', 'book', 'theme'])
+
+    def read(self, request, tags):
+        """ Lists all fragments with given book, tags, themes.
+
+        :param tags: should be a path of categories and slugs, i.e.:
+             books/book-slug/authors/an-author/themes/a-theme/
+
+        """
+        tags = read_tags(tags, allowed=self.categories)
+        fragments = Fragment.tagged.with_all(tags).select_related('book')
+        if fragments.exists():
+            return fragments

         else:
         else:

-            return Book.objects.all()
-"""
+            return rc.NOT_FOUND
+
+    @classmethod
+    def href(cls, fragment):
+        """ Returns URI in the API for the fragment. """
+
+        return API_BASE + reverse("api_fragment", args=[fragment.book.slug, fragment.anchor])
+
+    @classmethod
+    def url(cls, fragment):
+        """ Returns URL on the site for the fragment. """
+
+        return WL_BASE + fragment.get_absolute_url()
+
+    @classmethod
+    def themes(cls, fragment):
+        """ Returns a list of theme tags for the fragment. """
+
+        return fragment.tags.filter(category='theme')
+
+

 
 
 
 

-class WLHandler(BaseHandler):
+# Changes handlers
+
+class CatalogueHandler(BaseHandler):

 
     @staticmethod
     def fields(request, name):
 
     @staticmethod
     def fields(request, name):


@@ -45,164 +305,239 @@ class WLHandler(BaseHandler):
         return fields_str.split(',') if fields_str is not None else None
 
     @staticmethod
         return fields_str.split(',') if fields_str is not None else None
 
     @staticmethod

-    def book_dict(book, fields=None, extra_fields=None):
+    def until(t=None):
+        """ Returns time suitable for use as upper time boundary for check.
+
+            Used to avoid issues with time between setting the change stamp
+            and actually saving the model in database.
+            Cuts the microsecond part to avoid issues with DBs where time has
+            more precision.
+
+            :param datetime t: manually sets the upper boundary
+
+        """
+        # set to five minutes ago, to avoid concurrency issues
+        if t is None:
+            t = datetime.now() - timedelta(seconds=settings.API_WAIT)
+        # set to whole second in case DB supports something smaller
+        return t.replace(microsecond=0)
+
+    @staticmethod
+    def book_dict(book, fields=None):
+        all_fields = ('url', 'title', 'description',
+                      'gazeta_link', 'wiki_link',
+                      'xml', 'epub', 'txt', 'pdf', 'html',
+                      'mp3', 'ogg', 'daisy',
+                      'parent', 'parent_number',
+                      'tags',
+                      'license', 'license_description', 'source_name',
+                      'technical_editors', 'editors',
+                      'author', 'sort_key',
+                     )
+        if fields:
+            fields = (f for f in fields if f in all_fields)
+        else:
+            fields = all_fields
+
+        extra_info = book.get_extra_info_value()
+

         obj = {}
         obj = {}

-        for field in ('slug', 'title', 'description',
-                      'extra_info', 'gazeta_link', 'wiki_link'):
-            if getattr(book, field):
-                obj[field] = getattr(book, field)
-        for field in ('created_at', 'changed_at'):
-            obj[field] = timestamp(getattr(book, field))
-        for field in ('xml', 'epub', 'txt', 'pdf', 'html'):
-            f = getattr(book, field+'_file') 
-            if f:
-                obj[field] = f.url
-        for media in book.medias.all():
-            obj.setdefault(media.type, []).append(media.file.url)
-        if book.parent:
-            obj['parent'] = book.parent.id
-            obj['parent_number'] = book.parent_number
-        if fields is not None:
-            for key in obj.keys():
-                if key not in fields:
-                    del obj[key]
-
-        # if there's still extra_info, we can parse it
-        if 'extra_info' in obj:
-            extra = json.loads(obj['extra_info'])
-            if extra_fields is not None:
-                for key in extra.keys():
-                    if key not in extra_fields:
-                        del extra[key]
-            obj['extra_info'] = extra
+        for field in fields:
+
+            if field in ('xml', 'epub', 'txt', 'pdf', 'html'):
+                f = getattr(book, field+'_file')
+                if f:
+                    obj[field] = {
+                        'url': f.url,
+                        'size': f.size,
+                    }
+
+            elif field in ('mp3', 'ogg', 'daisy'):
+                media = []
+                for m in book.media.filter(type=field):
+                    media.append({
+                        'url': m.file.url,
+                        'size': m.file.size,
+                    })
+                if media:
+                    obj[field] = media
+
+            elif field == 'url':
+                obj[field] = book.get_absolute_url()
+
+            elif field == 'tags':
+                obj[field] = [t.id for t in book.tags.exclude(category__in=('book', 'set'))]
+
+            elif field == 'author':
+                obj[field] = ", ".join(t.name for t in book.tags.filter(category='author'))
+
+            elif field == 'parent':
+                obj[field] = book.parent_id
+
+            elif field in ('license', 'license_description', 'source_name',
+                      'technical_editors', 'editors'):
+                f = extra_info.get(field)
+                if f:
+                    obj[field] = f
+
+            else:
+                f = getattr(book, field)
+                if f:
+                    obj[field] = f

 
         obj['id'] = book.id
         return obj
 
     @classmethod
 
         obj['id'] = book.id
         return obj
 
     @classmethod

-    def book_changes(cls, since=0, request=None):
-        since = datetime.fromtimestamp(float(since))
-        book_fields = cls.fields(request, 'book_fields')
-        extra_fields = cls.fields(request, 'extra_fields')
+    def book_changes(cls, request=None, since=0, until=None, fields=None):
+        since = datetime.fromtimestamp(int(since))
+        until = cls.until(until)
+
+        changes = {
+            'time_checked': timestamp(until)
+        }
+
+        if not fields:
+            fields = cls.fields(request, 'book_fields')

 
         added = []
 
         added = []

-        changed = []
+        updated = []

         deleted = []
 
         last_change = since
         deleted = []
 
         last_change = since

-        for book in Book.objects.filter(changed_at__gt=since):
-            if book.changed_at > last_change:
-                last_change = book.changed_at
-            book_d = cls.book_dict(book, book_fields, extra_fields)
-            if book.created_at > since:
-                added.append(book_d)
-            else:
-                changed.append(book_d)
-
-        for book in Deleted.objects.filter(type='Book', deleted_at__gt=since, created_at__lte=since):
-            if book.deleted_at > last_change:
-                last_change = book.deleted_at
+        for book in Book.objects.filter(changed_at__gte=since,
+                    changed_at__lt=until):
+            book_d = cls.book_dict(book, fields)
+            updated.append(book_d)
+        if updated:
+            changes['updated'] = updated
+
+        for book in Deleted.objects.filter(content_type=Book, 
+                    deleted_at__gte=since,
+                    deleted_at__lt=until,
+                    created_at__lt=since):

             deleted.append(book.id)
             deleted.append(book.id)

-        return {'added': added, 'changed': changed, 'deleted': deleted, 'last_change': timestamp(last_change)}
+        if deleted:
+            changes['deleted'] = deleted
+
+        return changes

 
     @staticmethod
     def tag_dict(tag, fields=None):
 
     @staticmethod
     def tag_dict(tag, fields=None):

+        all_fields = ('name', 'category', 'sort_key', 'description',
+                      'gazeta_link', 'wiki_link',
+                      'url', 'books',
+                     )
+
+        if fields:
+            fields = (f for f in fields if f in all_fields)
+        else:
+            fields = all_fields
+

         obj = {}
         obj = {}

-        for field in ('name', 'slug', 'sort_key', 'category', 'description', 'main_page', #'created_at', 'changed_at',
-                      'gazeta_link', 'wiki_link'):
-            if getattr(tag, field):
-                obj[field] = getattr(tag, field)
-        if fields is not None:
-            for key in obj.keys():
-                if key not in fields:
-                    del obj[key]
+        for field in fields:
+
+            if field == 'url':
+                obj[field] = tag.get_absolute_url()
+
+            elif field == 'books':
+                obj[field] = [b.id for b in Book.tagged_top_level([tag])]
+
+            elif field == 'sort_key':
+                obj[field] = tag.sort_key
+
+            else:
+                f = getattr(tag, field)
+                if f:
+                    obj[field] = f
+

         obj['id'] = tag.id
         return obj
 
     @classmethod
         obj['id'] = tag.id
         return obj
 
     @classmethod

-    def tag_changes(cls, since=0, request=None):
-        since = datetime.fromtimestamp(float(since))
-        tag_fields = cls.fields(request, 'tag_fields')
+    def tag_changes(cls, request=None, since=0, until=None, fields=None, categories=None):
+        since = datetime.fromtimestamp(int(since))
+        until = cls.until(until)

 
 

-        added = []
-        changed = []
-        deleted = []
+        changes = {
+            'time_checked': timestamp(until)
+        }

 
 

-        last_change = since
-        for tag in Tag.objects.filter(changed_at__gt=since):
-            if tag.changed_at > last_change:
-                last_change = tag.changed_at
-            tag_d = cls.tag_dict(tag, tag_fields)
-            if tag.created_at > since:
-                added.append(tag_d)
-            else:
-                changed.append(tag_d)
+        if not fields:
+            fields = cls.fields(request, 'tag_fields')
+        if not categories:
+            categories = cls.fields(request, 'tag_categories')

 
 

-        for tag in Deleted.objects.filter(type='Tag', deleted_at__gt=since, created_at__lte=since):
-            if tag.deleted_at > last_change:
-                last_change = tag.deleted_at
-            deleted.append(tag.id)
-        return {'added': added, 'changed': changed, 'deleted': deleted, 'last_change': timestamp(last_change)}
+        all_categories = ('author', 'epoch', 'kind', 'genre')
+        if categories:
+            categories = (c for c in categories if c in all_categories)
+        else:
+            categories = all_categories

 
 

+        updated = []
+        deleted = []

 
 

-class BookChangesHandler(WLHandler):
-    allowed_methods = ('GET',)
+        for tag in Tag.objects.filter(category__in=categories, 
+                    changed_at__gte=since,
+                    changed_at__lt=until):
+            # only serve non-empty tags
+            if tag.get_count():
+                tag_d = cls.tag_dict(tag, fields)
+                updated.append(tag_d)
+            elif tag.created_at < since:
+                deleted.append(tag.id)
+        if updated:
+            changes['updated'] = updated
+
+        for tag in Deleted.objects.filter(category__in=categories,
+                content_type=Tag, 
+                    deleted_at__gte=since,
+                    deleted_at__lt=until,
+                    created_at__lt=since):
+            deleted.append(tag.id)
+        if deleted:
+            changes['deleted'] = deleted

 
 

-    def read(self, request, since):
-        return self.book_changes(since, request)
+        return changes

 
 

+    @classmethod
+    def changes(cls, request=None, since=0, until=None, book_fields=None,
+                tag_fields=None, tag_categories=None):
+        until = cls.until(until)

 
 

-class TagChangesHandler(WLHandler):
-    allowed_methods = ('GET',)
+        changes = {
+            'time_checked': timestamp(until)
+        }

 
 

-    def read(self, request, since):
-        return self.tag_changes(since, request)
+        changes_by_type = {
+            'books': cls.book_changes(request, since, until, book_fields),
+            'tags': cls.tag_changes(request, since, until, tag_fields, tag_categories),
+        }
+
+        for model in changes_by_type:
+            for field in changes_by_type[model]:
+                if field == 'time_checked':
+                    continue
+                changes.setdefault(field, {})[model] = changes_by_type[model][field]
+        return changes

 
 
 
 

-class ChangesHandler(WLHandler):
+class BookChangesHandler(CatalogueHandler):

     allowed_methods = ('GET',)
 
     def read(self, request, since):
     allowed_methods = ('GET',)
 
     def read(self, request, since):

-        changes = {
-            'books': self.book_changes(since, request),
-            'tags': self.tag_changes(since, request),
-        }
+        return self.book_changes(request, since)

 
 

-        last_change = 0
-        changes_rev = {}
-        for model in changes:
-            for field in changes[model]:
-                if field == 'last_change':
-                    if changes[model][field] > last_change:
-                        last_change = changes[model][field]
-                else:
-                    changes_rev.setdefault(field, {})[model] = changes[model][field]
-        changes_rev['last_change'] = last_change
-        return changes_rev

 
 

+class TagChangesHandler(CatalogueHandler):
+    allowed_methods = ('GET',)

 
 

+    def read(self, request, since):
+        return self.tag_changes(request, since)

 
 

-# old
-"""
-staff_required = user_passes_test(lambda user: user.is_staff)

 
 

-class BookHandler(BaseHandler):
-    model = Book
-    fields = ('slug', 'title')
+class ChangesHandler(CatalogueHandler):
+    allowed_methods = ('GET',)

 
 

-    @staff_required
-    def read(self, request, slug=None):
-        if slug:
-            return get_object_or_404(Book, slug=slug)
-        else:
-            return Book.objects.all()
-
-    @staff_required
-    def create(self, request):
-        form = BookImportForm(request.POST, request.FILES)
-        if form.is_valid():
-            form.save()
-            return rc.CREATED
-        else:
-            return rc.BAD_REQUEST
-"""
+    def read(self, request, since):
+        return self.changes(request, since)