2i2c Data

Public views of the datasets published by 2i2c-org/data. Each dataset is published as a GitHub Release; this site renders it so community members can read it without leaving the browser.

Datasets¶

• Cloud - monthly active users on each 2i2c cluster and hub.

How to find raw data¶

This site is built with MyST, and raw data for every plot is downloadable from the page that shows it. File names include a content hash (e.g. cloudbank-unique-[hash].csv), so the exact URL changes whenever the underlying data does. This means you can’t hard-code the link.

To resolve the current link, parse the page’s MyST AST. Every rendered page has a .json sibling at the same URL (e.g. cloud.cloudbank.json for /cloud/cloudbank/). Each static-file link has a urlSource property (the pre-hash filename) and a url property (the hashed path actually served). To find a download link, walk the AST, match on urlSource, and return url. Here’s a little example of a recursive function that does this in Python:

import json
from urllib.request import urlopen

PAGE = "https://2i2c-org.github.io/data/cloud.cloudbank.json"
WANTED = "cloudbank-unique.csv"

def find_url(node, source):
    if isinstance(node, dict):
        # If we find the source, return its URL
        if node.get("urlSource") == source:
            return node["url"]
        # Iterate through dictionary values otherwise
        for v in node.values():
            found = find_url(v, source)
            if found:
                return found
    elif isinstance(node, list):
        # Iterate through list items if it's a list
        for item in node:
            found = find_url(item, source)
            if found:
                return found

ast = json.load(urlopen(PAGE))
print(find_url(ast, WANTED))
# /build/cloudbank-unique-<hash>.csv

Prefix with the site origin (e.g. https://2i2c.org/data) to get the full download URL.