你可以做一个upsert，这个操作在MongoDB是用来保存文档到集合。 如果文档符合查询条件，则执行更新操作，否则将向集合中插入新文档。

类似如下所示

db.employees.update(
    {type:"FT"},
    {$set:{salary:200000}},
    {upsert : true}
 )

使用包含必要的点路径的属性名创建一个更新对象。(“somekey。+来自OP的例子)，然后使用它进行更新。

//the modification that's requested
updateReq = { 
   param2 : "val2_new",
   param3 : "val3_new"   
}

//build a renamed version of the update request
var update = {};
for(var field in updateReq){
    update["somekey."+field] = updateReq[field];
}

//apply the update without modifying fields not originally in the update
db.collection.update({._id:...},{$set:update},{upsert:true},function(err,result){...});

    // where clause DBObject
    DBObject query = new BasicDBObject("_id", new ObjectId(id));

    // modifications to be applied
    DBObject update = new BasicDBObject();

    // set new values
    update.put("$set", new BasicDBObject("param2","value2"));

   // update the document
    collection.update(query, update, true, false); //3rd param->upsertFlag, 4th param->updateMultiFlag

如果您有多个字段要更新

        Document doc = new Document();
        doc.put("param2","value2");
        doc.put("param3","value3");
        update.put("$set", doc);

看起来你可以设置isPartialObject，这可能完成你想要的。

你必须使用Embedded Documents (stringfy path对象)

let update = {}
Object.getOwnPropertyNames(new_info).forEach(param => {
   update['some_key.' + param] = new_info[param]
})

因此，在JavaScript中，您可以使用扩展操作符(…)进行更新

db.collection.update(  { _id:...} , { $set: { ...update  } } 

从Mongo 4.2开始，db.collection.updateMany()(或db.collection.update())可以接受一个聚合管道，它允许使用聚合操作符，如$addFields，它输出输入文档中所有现有的字段和新添加的字段:

var new_info = { param2: "val2_new", param3: "val3_new" }

// { some_key: { param1: "val1", param2: "val2", param3: "val3" } }
// { some_key: { param1: "val1", param2: "val2"                 } }
db.collection.updateMany({}, [{ $addFields: { some_key: new_info } }])
// { some_key: { param1: "val1", param2: "val2_new", param3: "val3_new" } }
// { some_key: { param1: "val1", param2: "val2_new", param3: "val3_new" } }

The first part {} is the match query, filtering which documents to update (in this case all documents). The second part [{ $addFields: { some_key: new_info } }] is the update aggregation pipeline: Note the squared brackets signifying the use of an aggregation pipeline. Since this is an aggregation pipeline, we can use $addFields. $addFields performs exactly what you need: updating the object so that the new object will overlay / merge with the existing one: In this case, { param2: "val2_new", param3: "val3_new" } will be merged into the existing some_key by keeping param1 untouched and either add or replace both param2 and param3.