Using Mongoose with a rich document? - mongodb

Question

I'm working on a prototype that will be used for reporting (read only) where the record is a very rich set of objects embedded into a single document. Essentially the document structure is this (edited for brevity):
{
"_id": ObjectId("56b3af6f84ef45c8903acc51"),
"id": "7815dd97-e895-46e5-b6c9-45184c6eae89",
"survey": {
"id": "1fb21c69-6a5c-4805-b1cf-fabef7a5d0e6",
"type": "Survey",
"data": {
"description": "Testing reporting and data ouput",
"id": "1fb21c69-6a5c-4805-b1cf-fabef7a5d0e6",
"start_date": "2016-02-04T11:12:46Z",
"questions": [
{
"sequence": 1,
"modified_at": "2016-02-04T16:11:04.505849+00:00",
"id": "2a77921b-6853-463b-80e7-5713c82c51ca",
"previous_question": null,
"created_at": "2016-02-04T16:10:56.647746+00:00",
"parent_question": "",
"next_question": "",
"validators": [
"required",
"email"
],
"question_data": {
"modified_at": "2016-02-04T16:10:37.542715+00:00",
"type": "open-ended",
"text": "Please provide your email address",
"id": "27aa00db-4a56-4a3e-bc30-226179062af0",
"reporting_name": "email address",
"created_at": "2016-02-04T16:10:37.542695+00:00"
}
},
{
"sequence": 2,
"modified_at": "2016-02-04T16:09:53.539073+00:00",
"id": "c034819d-9281-4943-801f-c53f4047d03e",
"previous_question": null,
"created_at": "2016-02-04T16:09:53.539051+00:00",
"parent_question": "",
"next_question": null,
"validators": [
"alpha-numeric"
],
"question_data": {
"modified_at": "2016-02-04T16:05:31.008363+00:00",
"type": "open-ended",
"text": "Is there anything else that we could have done to improve your experience?",
"id": "e33c7804-20cb-4473-abfa-77b3c2a3113c",
"reporting_name": "more info open-ended",
"created_at": "2016-02-01T20:19:55.036899+00:00"
}
},
{
"sequence": 1,
"modified_at": "2016-02-04T16:08:55.681461+00:00",
"id": "f91fd70e-f204-4c38-9a56-dd6ff25e4cd8",
"previous_question": "",
"created_at": "2016-02-04T16:08:55.681441+00:00",
"parent_question": "",
"next_question": null,
"validators": [
"required"
],
"question_data": {
"modified_at": "2016-02-04T16:04:56.848528+00:00",
"type": "nps",
"text": "On a scale of 0-10 how likely are you to recommend us to a friend?",
"id": "fdb6b74d-96a3-4680-af35-8b2f6aa2bbc9",
"reporting_name": "key nps",
"created_at": "2016-02-01T20:19:27.371920+00:00"
}
}
],
"name": "Reporting Survey",
"end_date": "2016-02-11T11:12:47Z",
"trigger_active": false,
"created_at": "2016-02-04T16:13:16.808108Z",
"url": "http://www.peoplemetrics.com",
"fatigue_limit": "monthly",
"modified_at": "2016-02-04T16:13:16.808132Z",
"template": {
"id": "0ea02379-c80b-4e17-b0a6-d621d49076b9",
"type": "Template"
},
"landing_page": null,
"trigger": null,
"slug": "test-reporting-survey"
}
},
"invite_code": "7801",
"end_date": null,
"created_at": "2016-02-04T19:38:31.931147Z",
"url": "http://127.0.0.1:8000/api/v0/responses/7815dd97-e895-46e5-b6c9-45184c6eae89",
"answers": {
"data": [
{
"id": "bcc3d0dd-5419-4661-9900-ccda3ac9a308",
"end_datetime": "2016-01-22T19:57:03Z",
"survey_question": {
"id": "662fcdf9-3c92-415e-b779-ac5b0fd330d3",
"type": "SurveyQuestion"
},
"response": {
"id": "7815dd97-e895-46e5-b6c9-45184c6eae89",
"type": "Response"
},
"modified_at": "2016-02-04T19:38:31.972717Z",
"value_type": "number",
"created_at": "2016-02-04T19:38:31.972687Z",
"value": "10",
"slug": "",
"start_datetime": "2016-01-21T10:10:21Z"
},
{
"id": "8696f11e-679a-43da-b6e2-aee72a70ca9b",
"end_datetime": "2016-01-28T13:45:37Z",
"survey_question": {
"id": "f118c9dd-1c03-47e0-80ef-2a36eb3b9a29",
"type": "SurveyQuestion"
},
"response": {
"id": "7815dd97-e895-46e5-b6c9-45184c6eae89",
"type": "Response"
},
"modified_at": "2016-02-04T19:38:32.001970Z",
"value_type": "boolean",
"created_at": "2016-02-04T19:38:32.001939Z",
"value": "True",
"slug": "",
"start_datetime": "2016-02-15T04:51:24Z"
}
]
},
"modified_at": "2016-02-04T19:38:31.931171Z",
"start_date": "2016-02-01T16:14:13Z",
"invite_date": "2016-02-01T13:14:08Z",
"contact": {
"id": "94833455-b9b8-4206-9bc9-a2f96c1706ca",
"type": "Contact",
"external_contactid": null,
"name": "Miss Marceline Herzog PhD"
},
"referring_source": "web"
}
given a structure in that format, I'm unsure the best path forward using Mongoose as the ORM. Again, this is read-only, so I was it would seem that creating a nested schema would work, but the mapping itself seems tedious to say the least. Is there a better/different option available for something with embedded?

Answer 1

Interesting. First, I would think if I need all the document and its embedded subdocuments fields. You said it will be read-only, so will each call needs the entire document?
If not, I recommend taking a look at the mongo drivers (node.js, .NET, Python, etc.) and using their aggregation pipelines to simplify the document if possible.
If you're using Mongoose, you will probably end up with two or three Schemas, and with schemas inside a list. Mongoose docs e.g.
var surveySchema = new Schema(
{ "type" : string,
"data" : [dataSchema],
"invite_code" : string,
"end_date" : DateTime,
"created_at" : DateTime,
"url" : string,
"answers" : { "data": [answersSchema]},
"modified_at" : DateTime,
"start_date" : DateTime,
"invite_date" : DateTime,
"contact" : [ContactSchema],
"referring_source" : string
});
Or, you can use mongoose references and build your own schema depending on what data you need to use for your report. A simple example:
var surveySchema = {
"id" : { type: Schema.Types.ObjectId }
"description" : { type: string , ref: dataSchema },
"contactSchema" : { type: string , ref: contactSchema }
}