Problem: Inefficient fetching of data from multiple sources. Particularly with paginated endpoints.

Inefficient by:

1. needing to manually type out pagination
2. requiring manual updating to the action when there are more pages in the future
3. doing extra fetching that isn't required (some of the data I'm fetching is static, so data that has been fetched previously doesn't need refetching, only new data)

Context:
I'm working on an interactive data visualization that is getting data from 3 different endpoints.

I figured out how to run multiple steps by following the COVID example.

And... while I do have it working right now, it's a little less than ideal, because 2 of them are fetching data from paginated endpoints, meaning I have a bunch of this happening:

# 0000-0099
- name: Fetch data 0000-0049
  uses: githubocto/flat@v3
  with:
    http_url: https://api.opensea.io/api/v1/assets?asset_contract_address=0x585a2c37858d3b03824bc683829e4dbbf58969ee&order_direction=asc&offset=0&limit=50
    downloaded_filename: flatdata/opensea_junks_00a.json
- name: Fetch data 0050-0099
  uses: githubocto/flat@v3
  with:
    http_url: https://api.opensea.io/api/v1/assets?asset_contract_address=0x585a2c37858d3b03824bc683829e4dbbf58969ee&order_direction=asc&offset=50&limit=50
    downloaded_filename: flatdata/opensea_junks_00b.json
# this endpoint only returns 50, and I need to fetch 2000... XD

The second one is more tedious because I have to manually specify the pagination using block numbers

# 12153412-12355753
- name: Fetch data etherscan logs
  uses: githubocto/flat@v3
  with:
    http_url: https://api.etherscan.io/api?module=logs&action=getLogs&fromBlock=12153412&toBlock=12355753&address=0x585A2C37858D3B03824BC683829e4DBbF58969ee&apikey=APIKEY
    downloaded_filename: flatdata/etherscan_logs_00.json
# this returns 1000, so I only have a couple of them

Then I have several post processing steps to parse, extract, filter, and merge the data for how I need it.

I understand that I am... slightly pushing the limits of what flat data was meant to do. But what I would really like to be able to do is write up a JS file, kind of like the post processing one, that readJSONs the files I already have, and then performs only the fetches it needs to.

(Then run the postprocessing normally, to merge it all back together)

Any input or advice is very welcome!

PS. I was also hoping to specify different schedules for each of the endpoints... so I tried making several yml files, but it created some issues with git commit/merging so I moved them back into a single file with one schedule.

PPS. @Wattenberger big fan of your work! I absolutely love this flat data project & I already have 7! different use-cases for it across multiple projects.