feat: add public_hash_list table definition

[max-ostapenko]

Hash list saved monthly into a public BQ table.

Query source: https://github.com/tomayac/public-hash-list/blob/main/queries/http-archive.sql

[max-ostapenko]

@tomayac how about public BQ table instead of Spreadsheet?

You can download it like this:

bq query --format=csv \
  "SELECT * FROM \`httparchive.performance.public_hash_list\`" > data/http-archive-hashes.csv

[tomayac]

Thanks a lot! I just checked, and it looks like bq requires a Google Cloud project and being authenticated to gcloud. It's all doable, but reduces the reproducibility of the data. Would it be much more work for you to provide the CSV directly on GitHub, so it could be downloaded anonymously? Or else, the Google Spreadsheet export would also be a viable option. If either of those would be viable options, I'd prefer this over bq. What do you think?

[max-ostapenko]

@tomayac agree that CSV in a public repo is much more accessible.
But committing a table data to Github is a very specific scenario and takes more effort.

My idea is to replace the Spreadsheet with this table, in order to simplify the CSV-download-script into a one-liner.

There will be a publicly available CSV in a repo, and anybody interested can look into a table and raw crawl for details.

[tunetheweb]

Why not just save to our CDN like the JSON files we create?

[tomayac]

Why not just save to our CDN like the JSON files we create?

This sounds like a great idea, as long as the URL is predictable.