I am investigating different approaches to obtain select metadata (ALL objects - including custom - and ALL fields) for any org configuration, reliably.   My company works with the Net-Zero cloud and the NPSP. The metadata API does not have full coverage of all the objects and fields we require (https://developer.salesforce.com/docs/metadata-coverage/56). We will then use this information to build a .CSV.   We know this is possible, as the WORKBENCH REST API is able to retrieve the data we need, albeit on an object by object basis using REST API calls.

Approaches:

1.  Build an APEX class that uses the schema class to retrieve all objects. Loop through objects, and retrieve all fields. Convert custom object into .CSV (either directly, or using a wrapper class to convert to JSON and then to .CSV).

2. Build a node.js server to perform REST API calls and then write the data to the local file system in .CSV format.

3. Parse the .XML directly from the Salesforce org.   Any thoughts are welcome. We need this to be repeatable for any org configuration. Thankyou!