From 5b01b088d308d34dfb119a4cee8e72d85967165f Mon Sep 17 00:00:00 2001
From: James Turk <james.p.turk@gmail.com>
Date: Wed, 15 Feb 2012 17:29:50 -0500
Subject: [PATCH] delete obsolete ExternalStoreTask

---
 design.txt      |  6 ------
 oyster/tasks.py | 33 ---------------------------------
 2 files changed, 39 deletions(-)

diff --git a/design.txt b/design.txt
index 807c027..46b6d07 100644
--- a/design.txt
+++ b/design.txt
@@ -4,12 +4,6 @@ and keeps a local copy up to date depending on user-specified criteria.
 Data Model
 ==========
 
-Oyster keeps its data in a MongoDB instance and makes use of GridFS to store the
-raw document data.
-
-In addition to the standard gridfs collections (fs.chunks, fs.files) oyster
-uses the following collections:
-
 tracked - metadata for tracked resources
     _id         : internal id
     _random     : a random integer used for sorting
diff --git a/oyster/tasks.py b/oyster/tasks.py
index e1c8d07..f659aad 100644
--- a/oyster/tasks.py
+++ b/oyster/tasks.py
@@ -36,36 +36,3 @@ class UpdateTaskScheduler(PeriodicTask):
         for doc in next_set:
             UpdateTask.delay(doc['_id'])
             kernel.db.status.update({}, {'$inc': {'update_queue': 1}})
-
-
-class ExternalStoreTask(Task):
-    """ base class for tasks that push documents to an external store
-
-        when overiding be sure to define
-            external_store
-                short string describing external store (eg. 's3')
-            upload_document(self, doc_id, filedata, metadata)
-                function that uploads the document and returns a unique ID
-    """
-
-    # results go straight to database
-    ignore_result = True
-    # used as a base class
-    abstract = True
-
-    def run(self, doc_id, extract_text=lambda x: x):
-        # get the document
-        doc = kernel.db.tracked.find_one({'_id': ObjectId(doc_id)})
-        filedata = kernel.get_version(doc['url']).read()
-        text = extract_text(filedata, doc['metadata'])
-
-        # put the document into the data store
-        result = self.upload_document(doc_id, text, doc['metadata'])
-
-        doc[self.external_store + '_id'] = result
-        kernel.db.tracked.save(doc, safe=True)
-
-
-    def upload_document(self, doc_id, filedata, metadata):
-        """ abstract method, override on implementations """
-        pass