mass_shooting_feed/index.mjs
2023-10-16 17:50:47 +02:00

129 lines
3.5 KiB
JavaScript
Raw Permalink Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

import fs from "fs";
import xml from "xml";
import crypto from "crypto";
class Shooting {
date;
location;
state;
victims_dead;
victims_injured;
victims_total;
description;
source_urls;
/**
* @param {string} input
*/
constructor(input) {
const lines = input.trim().split("\n");
this.date = lines[0].replace(/(^\|+)|(\{\{[\w\s]*\|)|(\}\})|((-|)[0-9]+)/g, "");
this.location = lines[1].replace(/(^\|+)|(\[\[(.*\|)?)|(\]\].*)/g, "");
this.state = lines[2].replace(/(^\|+)|(\[\[(.*\|)?)|(\]\])/g, "");
this.victims_dead = Number(lines[3].replace(/^\|+/, "")) || 0;
this.victims_injured = Number(lines[4].replace(/^\|+/, "")) || 0;
this.victims_total = Number(lines[5].replace(/(^\|+)|(''')/g, "")) || 0;
this.description = lines[6]
.replace(/^\|+/, "")
.replace(/(\[\[(.*\|)?)|(\]\])/g, "")
.replace(/\<ref\>.*\<\/ref\>/g, "")
.replace(/\{\{.*\}\}/g, "")
.replace(/\}\}.*/g, "")
.trim();
this.source_urls = /\|url=^[\s|]*/g
.exec(lines[6])
?.map((match) => match.replace("|url=", ""));
}
}
const input = fs.readFileSync("/tmp/feed_input.txt").toString("utf8");
const text = input
.substring(
input.indexOf("== List =="),
input.indexOf("== Monthly statistics ==") - 1
)
.trim();
const segments = text
.split(/\n\|-.*\n/)
.slice(2)
.filter((segment) => segment.length);
const shootings = [];
for (const segment of segments) {
const shooting = new Shooting(segment);
shootings.push(shooting);
}
// fs.writeFileSync("./shootings.json", JSON.stringify(shootings, null, 4));
const feed = {
rss: [
{
_attr: {
version: "2.0",
"xmlns:atom": "http://www.w3.org/2005/Atom",
},
},
{
channel: [
{
"atom:link": {
_attr: {
href: "https://futacockinside.me/files/mass_shootings.rss",
rel: "self",
type: "application/rss+xml",
},
},
},
{
title: "Mass shootings in America (2023)",
},
{
link: "https://en.m.wikipedia.org/w/index.php?title=List_of_mass_shootings_in_the_United_States_in_2023",
},
{
description:
"Feed of mass shootings in America, updated automatically from Wikipedia",
},
{ language: "en-US" },
...shootings.map((shooting) => {
const guid = crypto
.createHash("sha256")
.update(
`${shooting.date} ${shooting.location} ${shooting.state} ${shooting.victims_dead}/${shooting.victims_injured}/${shooting.victims_total}`
)
.digest("hex");
return {
item: [
{
title: `${shooting.date} - ${shooting.location}, ${shooting.state}`,
},
{ pubDate: new Date(`${shooting.date} 2023`).toUTCString() },
{ guid: guid },
{ link: "https://en.m.wikipedia.org/w/index.php?title=List_of_mass_shootings_in_the_United_States_in_2023" },
{
description: `${shooting.victims_dead} dead, ${
shooting.victims_injured
} injured.\n\n${shooting.description}\n${
shooting.source_urls?.join(" ") || ""
}`.trim(),
},
],
};
}),
],
},
],
};
fs.writeFileSync(
"/tmp/feed_output.rss",
xml(feed, { indent: " " })
);