Started on a JSON serializer rather than my custom JSON hackery.

bigbuild/management/commands/cachepages.py

@@ -3,29 +3,18 @@
 import io
 import os
 import six
-import json
-from datetime import datetime
+import bigbuild
 from bigbuild.models import PageList
-from bigbuild import get_archive_directory
 from django.core.management.base import BaseCommand
-
-
-def serializer(obj):
-    """
-    JSON serializer for objects not serializable by default json code
-    """
-    if isinstance(obj, datetime):
-        serial = obj.isoformat()
-        return serial
-    raise TypeError("Type not serializable")
+from bigbuild.serializers import BigBuildJSONSerializer
 
 
 class Command(BaseCommand):
     help = "Cache page metadata to increase the application speed"
 
     def handle(self, *args, **options):
         # Set the cache path for archived pages
-        archive_cache_path = os.path.join(get_archive_directory(), '.cache')
+        archive_cache_path = os.path.join(bigbuild.get_archive_directory(), '.cache')
 
         # Delete it if it already exists
         if os.path.exists(archive_cache_path):
@@ -34,11 +23,10 @@ def handle(self, *args, **options):
         # Pull the live PageList from the YAML files
         page_list = PageList()
 
+        # Create a JSON serializer
+        serializer = BigBuildJSONSerializer()
+
         # Save the archived pages out to a new cache
         with io.open(archive_cache_path, 'w', encoding='utf8') as f:
-            data = json.dumps(
-                dict(archived_pages=[p.to_json() for p in page_list.archived_pages]),
-                default=serializer,
-                ensure_ascii=False
-            )
+            data = serializer.serialize(page_list)
             f.write(six.text_type(data))

bigbuild/models/archivedpages.py

@@ -11,9 +11,6 @@ class ArchivedPage(BasePage):
     """
     An archived custom page.
     """
-    class Meta:
-        abstract = True
-
     def __str__(self):
         return self.slug
 

bigbuild/models/pagelists.py

@@ -2,7 +2,6 @@
 # -*- coding: utf-8 -*-
 import os
 import six
-import json
 import logging
 from django.conf import settings
 from collections import Sequence
@@ -11,7 +10,7 @@
     MissingRecommendedMetadataWarning
 )
 from bigbuild.models import Page, ArchivedPage
-from dateutil.parser import parse as dateparse
+from bigbuild.serializers import BigBuildJSONDeserializer
 from bigbuild import get_page_directory, get_archive_directory
 logger = logging.getLogger(__name__)
 
@@ -133,12 +132,10 @@ def get_archived_pages(self):
             logger.debug("Loading cached archived page list")
             with open(self.archived_cache_path, 'r') as f:
                 page_list = []
-                dict_list = json.load(f)['archived_pages']
-                for d in dict_list:
-                    d['pub_date'] = dateparse(d['pub_date'])
-                    page = ArchivedPage(**d)
-                    if page.should_build():
-                        page_list.append(page)
+                dobj_list = BigBuildJSONDeserializer(f.read())
+                for dobj in dobj_list:
+                    if dobj.object.should_build():
+                        page_list.append(dobj.object)
         # Otherwise get them from the YAML
         else:
             logger.debug("Retrieving YAML archived page list")

bigbuild/models/pages.py

@@ -22,9 +22,6 @@ class Page(BasePage):
     """
     A custom page published via static.latimes.com
     """
-    class Meta:
-        abstract = True
-
     def __str__(self):
         return self.slug
 

bigbuild/serializers.py

@@ -0,0 +1,32 @@
+import six
+import sys
+import json
+from django.core.serializers.json import Serializer
+from django.core.serializers.base import DeserializationError
+from django.core.serializers.python import Deserializer as PythonDeserializer
+
+
+class BigBuildJSONSerializer(Serializer):
+    """
+    A custom JSON serializer for bigbuild models.
+    """
+    pass
+
+
+def BigBuildJSONDeserializer(stream_or_string, **options):
+    """
+    Deserialize a stream or string of JSON data.
+    """
+    if not isinstance(stream_or_string, (bytes, six.string_types)):
+        stream_or_string = stream_or_string.read()
+    if isinstance(stream_or_string, bytes):
+        stream_or_string = stream_or_string.decode('utf-8')
+    try:
+        objects = json.loads(stream_or_string)
+        for obj in PythonDeserializer(objects, **options):
+            yield obj
+    except GeneratorExit:
+        raise
+    except Exception as e:
+        # Map to deserializer error
+        six.reraise(DeserializationError, DeserializationError(e), sys.exc_info()[2])