Aggregate query in mongo works, does not in Pymongo

问题

I encountered a problem. I try to query this document to obtain the sum the amount and group by the LOC identifier that is outside the "COL" array.

{
"_id" : ObjectId("57506d74c469888f0d631be6"),
"LOC" : "User001",
"COL" : [ 
    {
        "date" : "25/03/2016",
        "number" : "Folio009",
        "amount" : 100
    }, 
    {
        "date" : "25/04/2016",
        "number" : "Folio010",
        "amount" : 100
    }

] }

This command works in mongo but I cannot make it work in Python with the Pymongo package:

Mongo query (working)

db.perfiles.aggregate({"$unwind": "$COL"},
{ "$group": { _id: "$LOC", "sum" : {"$sum" : "$COL.amount" }}})

Pymongo (not working)

from pymongo import MongoClient

client = MongoClient()

db = client['temporal']

docs = db.perfiles


pipeline = [{"$unwind": "$COL"},
     {"$group": {"_id": "$LOC", "count": {"$sum": "$COL.amount"}}}
          ]

list(db.docs.aggregate(pipeline))

Any suggestion to query this same query but in Pymongo? Thanks!

回答1:

I assume you have a valid connection to MongoDB in Python.
The following code snippet will return a MongoDB cursor in result.

pipeline = [
    {"$unwind": "$COL"},
    {"$group": {"_id": "$LOC", "sum": {"$sum": "$COL.amount"}}}
]

cursor = collection.aggregate(pipeline)

Now you can convert cursor to list

result = list(cursor)

and if you print result's value, you'll get exactly the same result as in your Shell query.

[{u'sum': 200.0, u'_id': u'User001'}]

Update:

I see that you are calling the aggregate function in python code as db.docs.aggregate(pipeline). You need to call it as docs.aggregate... without db. See example above.

回答2:

MongoDB Enterprise > db.test.aggregate([{$match:{name:'prasad'}},{$group : {_id : "$name", age : {$min : "$age"}}}]);
{ "_id" : "prasad", "age" : "20" }
MongoDB Enterprise > db.test.find()
{ "_id" : ObjectId("5890543bce1477899c6f05e8"), "name" : "prasad", "age" : "22" }
{ "_id" : ObjectId("5890543fce1477899c6f05e9"), "name" : "prasad", "age" : "21" }
{ "_id" : ObjectId("58905443ce1477899c6f05ea"), "name" : "prasad", "age" : "20" }
{ "_id" : ObjectId("5890544bce1477899c6f05eb"), "name" : "durga", "age" : "20" }
{ "_id" : ObjectId("58905451ce1477899c6f05ec"), "name" : "durga", "age" : "21" }
{ "_id" : ObjectId("58905454ce1477899c6f05ed"), "name" : "durga", "age" : "22" }
MongoDB Enterprise >    


############code


import pymongo
from pymongo import MongoClient
client=MongoClient("localhost:27017")
db=client.prasad      #####prasad is dbname, test is collection name
nameVar='prasad'
aggregation_string=[{"$match":{"name":nameVar}},{"$group" : {"_id" : "$name", "age" : {"$min" : "$age"}}}]
x=db.test.aggregate(aggregation_string)
print x
for r in x:
        min_age=r.items()[0]
        print(min_age[1])      #######output:      20

回答3:

you are in a right track but add one more statement it will be fine.

    from pymongo import MongoClient

    client = MongoClient()

    db = client['temporal']

    docs = db.perfiles


    pipeline = [{"$unwind": "$COL"},
         {"$group": {"_id": "$LOC", "count": {"$sum": "$COL.amount"}}}
          ]

    result = list(db.docs.aggregate(pipeline))

    for i in result:

        sum += i['sum']

    print(sum)

来源：https://stackoverflow.com/questions/38133529/aggregate-query-in-mongo-works-does-not-in-pymongo

标签

python

mongodb

pymongo