refs #126. Makes data dir configurable and sets default to /sfm-data.…

… Adds documentation.

docs/install.rst

@@ -57,6 +57,17 @@ Configuration
   If setting ``DEBUG`` to false, the ``ALLOWED_HOSTS`` environment variable must be provided with a
   comma-separated list of hosts.  See the `Django documentation <https://docs.djangoproject.com/en/1.8/ref/settings/#allowed-hosts>`_
   for ``ALLOWED_HOSTS``.
+* The `data volume strategy <https://docs.docker.com/engine/userguide/dockervolumes/#creating-and-mounting-a-data-volume-container>`_
+  is used to manage the volumes that store SFM's data. By default, normal Docker volumes are used; to use
+  a host volume instead, add the host directory to the ``volumes`` field.  This will allow you to access the
+  data outside of Docker.  For example::
+
+    sfmdata:
+        image: ubuntu:14.04
+        command: /bin/true
+        volumes:
+             - /myhost/data:/sfm-data
+
 
 Installation
 ------------

sfm/sfm/settings/common.py

@@ -135,5 +135,8 @@
 RABBITMQ_USER = env.get('SFM_RABBITMQ_USER')
 RABBITMQ_PASSWORD = env.get('SFM_RABBITMQ_PASSWORD')
 
-# crispy forms bootstrap version 
+# crispy forms bootstrap version
 CRISPY_TEMPLATE_PACK = 'bootstrap3'
+
+# Directory where SFM data (e.g., harvested WARCs) is stored.
+SFM_DATA_DIR = env.get("SFM_DATA_DIR", "/sfm-data")

sfm/sfm/settings/test_settings.py

@@ -6,4 +6,6 @@
         'ENGINE': 'django.db.backends.sqlite3',
         'NAME': 'testdb'
     }
-}
+}
+
+SFM_DATA_DIR="/test-data"

sfm/ui/jobs.py

@@ -4,6 +4,7 @@
 from django.core.exceptions import ObjectDoesNotExist
 import logging
 import datetime
+from django.conf import settings
 
 log = logging.getLogger(__name__)
 
@@ -29,8 +30,7 @@ def seedset_harvest(seedset_id):
     # Collection
     collection = seedset.collection
     message["collection"]["id"] = "collection:{}".format(collection.id)
-    # TODO: Different approach for path
-    message["collection"]["path"] = "/tmp/collection/{}".format(collection.id)
+    message["collection"]["path"] = "{}/collection/{}".format(settings.SFM_DATA_DIR, collection.id)
 
     # Credential
     credentials = seedset.credential

sfm/ui/test_jobs.py

@@ -34,7 +34,7 @@ def test_seedset_harvest(self, mock_rabbit_worker_class):
         self.assertEqual("send_message", name)
         message = args[0]
         self.assertEqual("collection:{}".format(self.collection.id), message["collection"]["id"])
-        self.assertEqual("/tmp/collection/{}".format(self.collection.id), message["collection"]["path"])
+        self.assertEqual("/test-data/collection/{}".format(self.collection.id), message["collection"]["path"])
         self.assertDictEqual(self.harvest_options, message["options"])
         self.assertDictEqual({"token": "test_token1"}, message["seeds"][0])
         self.assertDictEqual({"uid": "test_uid2"}, message["seeds"][1])