We are looking to upload a large amount of data to our mobile backend on Parse. Our data consists of two classes - Store and Product. A store can have multiple products, while each product belongs to only one store. We want to streamline this process by bulk uploading stores and products using either a CSV or JSON file.
After conducting some research, we have identified two possible approaches:
Use the JSON importer to import objects where we manually define the ObjectId for both classes based on our own IDs.
Add additional columns for unique productId and storeId in each class. Start by uploading stores first, then upload products by referencing the randomly generated objectId for each store using the storeId as a pointer in the product class.
Which option would be more efficient? Are there any potential drawbacks to generating our own objectIds manually?