Source code

This project relies on a data loader that reads all the source files and outputs a single summary file, minimized to contain only a subset of the source information:

import hljs from "npm:highlight.js";

data/presse.parquet.sh

const pre = display(document.createElement("pre"));
FileAttachment("data/presse.parquet.sh")
  .text()
  .then(
    (text) => (pre.innerHTML = hljs.highlight(text, { language: "bash" }).value)
  );

This is the file that the other pages load with DuckDB, as a FileAttachment:

import { DuckDBClient } from "npm:@observablehq/duckdb";
const db = DuckDBClient.of({ presse: FileAttachment("data/presse.parquet") });