Question 227515 in Mongodb Question posted in
The official documentation can be found here.

Mongodb – Mongo DB / Python – Search DB for string but limit results to 1 of each item based on specified field

I am trying to search my MongoDB of products. The dataset has multiple of each product to record price over time. I would like to search for a phrase then limit the results to 1 of each UPC. My current code works well but will return multiple of the same UPC.

Current Code, will return multiple of the same UPC:

response = self.DB.find({'$text': {'$search': f'/{search}/'}}, {'Response': 0, '_id': 0}).sort("timestamp", -1)

Example Data Set:

{
  "_id": {
    "$oid": "64cf05707844ef1a25ee57fa"
  },
  "upc": "032622013625",
  "name": "Luigi Bormioli Michelangelo Beverage 20oz Set of 4",
  "salePrice": 29.99,
  "timestamp": "2023-08-05 22:29:04 EDT-0400",

  }
}
{
  "_id": {
    "$oid": "64cf057c7844ef1a25ee57fd"
  },
  "upc": "048894970887",
  "name": "Basic Window Fan - Holmes",
  "salePrice": 54.99,
  "available": false,
  "timestamp": "2023-08-05 22:29:16 EDT-0400",
 
    }
  }
}
{
  "_id": {
    "$oid": "64cf05707844ef1a25ee57fa"
  },
  "upc": "032622013625",
  "name": "Luigi Bormioli Michelangelo Beverage 20oz Set of 4",
  "salePrice": 29.97,
  "timestamp": "2023-08-04 13:25:09 EDT-0400",

  }
}

Not sure if I should be using distinct, or find?

Tags:

2

Answers


  1. 0

    I filtered the array of results by using a dictionary storing the upc id, and append into a list of documents if upc id is not existing.

    import pymongo

myclient = pymongo.MongoClient("mongodb://mongoadmin:ansible@localhost:27017/")
mydb = myclient["mydatabase"]
mycol = mydb["product"]
mycol.drop()
 
data=[
 {
  "_id": {
    "oid": "64cf05707844ef1a25ee57fa"
  },
  "upc": "032622013625",
  "name": "Luigi Bormioli Michelangelo Beverage 20oz Set of 4",
  "salePrice": 29.99,
  "timestamp": "2023-08-05 22:29:04 EDT-0400"

 },
 {
  "_id": {
    "oid": "64cf057c7844ef1a25ee57fd"
  },
  "upc": "048894970887",
  "name": "Basic Window Fan - Holmes",
  "salePrice": 54.99,
  "available": False,
  "timestamp": "2023-08-05 22:29:16 EDT-0400"
 
},
{
  "_id": {
    "oid": "64cf05707844ef1a25ee57fb"
  },
  "upc": "032622013625",
  "name": "Luigi Bormioli Michelangelo Beverage 20oz Set of 4",
  "salePrice": 29.97,
  "timestamp": "2023-08-04 13:25:09 EDT-0400"

}
]

for d in data:   
    x = mycol.insert_one(d)
    
resp=mycol.create_index(
    [
         ("upc", "text")
    ]
)
print(resp)   
    
search="032622013625"
response = mycol.find( { "$text": { "$search": f"/{search}/"}}, {'Response': 0, '_id': 0}).sort("timestamp", -1)

list_upc_already_seen=[]
list_documents=[]

for doc in response:
    upc=doc.get("upc")
    if upc not in list_upc_already_seen:
        list_documents.append(doc)
        list_upc_already_seen.append(upc)

print(list_documents)

    [{‘upc’: ‘032622013625’, ‘name’: ‘Luigi Bormioli Michelangelo Beverage 20oz Set of 4’, ‘salePrice’: 29.99, ‘timestamp’: ‘2023-08-05 22:29:04 EDT-0400’}]

    Login or Signup to reply.
  2. 0

    You could use "$top" with "$group" in an aggregation pipeline to get your result. If you only want certain fields returned, you could use a "$project" stage.

    response = self.DB.aggregate([
  {
    "$match": {'$text': {'$search': f'/{search}/'}}
  },
  {
    "$group": {
      "_id": "$upc",
      "mostRecent": {
        "$top": {
          "sortBy": {
            "timestamp": -1
          },
          "output": "$$ROOT"
        }
      }
    }
  },
  {
    "$replaceWith": "$mostRecent"
  }
])
    Login or Signup to reply.
Please signup or login to give your own answer.

Back To Top
Search