basic mongo aggregation script for tim

jlivni_mongo_tim.py

@@ -0,0 +1,58 @@
+import pymongo
+import json
+
+# setup mongo client and db
+client = pymongo.MongoClient()
+db = client['homework']
+
+# load json data
+data = json.loads(open('example_tim_data.json').read())
+
+tim_col = db.tim
+# load data into mongo if we don't already have it
+if not tim_col.estimated_document_count():
+    tim_col.insert_many(data)
+print('tim has %d records' % tim_col.estimated_document_count())
+# ensure team col is empty
+team_col = db.team_col
+team_col.drop()
+
+
+# Define mongodb aggregation pipeline
+pipeline = [
+    {
+        "$group": {
+            "_id": "$team_num",
+            "average_balls_scored": { "$avg": "$num_balls" },
+            "least_balls_scored": { "$min": "$num_balls" },
+            "most_balls_scored": { "$max": "$num_balls" },
+            "num_matches_played": { "$sum": 1 },
+            "num_climbed_matches": {
+                "$sum": {
+                    "$cond": [ { "$eq": ["$climbed", True] }, 1, 0 ]
+                }
+            }
+        }
+    },
+    {
+        "$project": {
+            "_id": 1, 
+            "average_balls_scored": 1,
+            "least_balls_scored": 1,
+            "most_balls_scored": 1,
+            "num_matches_played": 1,
+            "climbed_fraction": {
+                "$divide": ["$num_climbed_matches", "$num_matches_played"]
+            }
+        }
+    },
+    {
+        # Add the $out stage as the final step, writing to "team_col" as string
+        "$out": 'team_col'
+    }
+]
+
+tim_col.aggregate(pipeline)
+team_col = db.team_col
+for td in team_col.find():
+    print(td.pop('_id'), td)