如何使用 pymongo 插入另一个嵌套文档？

Question

Ahoy,

I have a document that looks like this:

    {"_id": "123abc456def",
    "name": "John Smith",
    "address": [
        {"street": "First St.", "date": "yesterday", "last_updated": "two days ago"}
    ],
    "age": 123}

I try to add another street document using $push, it errors out with:

pymongo.errors.WriteError: The field 'address' must be an array but is of type object in document {_id: ObjectId('6049e88657e43d8801197c72')}

Code I'm using:

    mydb3 = myclient["catalogue"]
    mycolALL = mydb3["locations"]
    query = {"charID": 0}
    newvalue = {"$push": {"address": {"street": "test123", "date": "test123", "last_updated": "now123"}}}
    mycolALL.update_one(query, newvalue)

Not making an address book or anything, just edited it so it makes a bit more sense to anyone without context.

My desired output would be that the document would look like this:

    {"_id": "123abc456def",
    "name": "John Smith",
    "address": [
        {"street": "First St.", "date": "yesterday", "last_updated": "two days ago"},
        {"street": "test123", "date": "test123", "last_updated": "now123"}
    ],
    "age": 123}

Normally I can google my way to an answer that makes the coin drop and JACKPOT. but this time I'm outta luck.

$set = it just changes the existing document, effectively replacing it. Which is not what I want.

$addToSet = for arrays only, error message: "pymongo.errors.WriteError: Cannot apply $addToSet to non-array field. Field named 'address' has non-array type object"

Anyone that can help?

Answer 1

Just a guess but are you sure you're looking at the right data / database.

Based on the data you posted your update_one() won't update that record because it doesn't match your filter {"charID": 0}