1 files changed, 51 insertions, 0 deletions
diff --git a/collections/activities.js b/collections/activities.js
new file mode 100644
index 00000000..1e24cf7c
--- /dev/null
+++ b/collections/activities.js
@@ -0,0 +1,51 @@
+// Activities don't need a schema because they are always set from the a trusted
+// environment - the server - and there is no risk that a user change the logic
+// we use with this collection. Moreover using a schema for this collection
+// would be difficult (different activities have different fields) and wouldn't
+// bring any direct advantage.
+//
+// XXX The activities API is not so nice and need some functionalities. For
+// instance if a user archive a card, and un-archive it a few seconds later we
+// should remove both activities assuming it was an error the user decided to
+// revert.
+Activities = new Mongo.Collection('activities');
+
+Activities.helpers({
+  board: function() {
+    return Boards.findOne(this.boardId);
+  },
+  user: function() {
+    return Users.findOne(this.userId);
+  },
+  member: function() {
+    return Users.findOne(this.memberId);
+  },
+  list: function() {
+    return Lists.findOne(this.listId);
+  },
+  oldList: function() {
+    return Lists.findOne(this.oldListId);
+  },
+  card: function() {
+    return Cards.findOne(this.cardId);
+  },
+  comment: function() {
+    return CardComments.findOne(this.commentId);
+  },
+  attachment: function() {
+    return Attachments.findOne(this.attachmentId);
+  }
+});
+
+Activities.before.insert(function(userId, doc) {
+  doc.createdAt = new Date();
+});
+
+// For efficiency create an index on the date of creation.
+if (Meteor.isServer) {
+  Meteor.startup(function() {
+    Activities._collection._ensureIndex({
+      createdAt: -1
+    });
+  });
+}