monkeytypegame/monkeytype
1
1
Fork 0
mirror of https://github.com/monkeytypegame/monkeytype.git synced 2025-02-05 13:27:49 +08:00
Code Issues Projects Releases Packages Wiki Activity

Completely Fix Wikipedia Spacing (#2195) by Ferotiq

Browse source
This commit is contained in:
Ferotiq 2021-12-20 07:05:02 -06:00 committed by GitHub
parent 0ece272c5e
commit 8156c483a5
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
1 changed files with 7 additions and 5 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

12
src/js/test/wikipedia.js
View file

@ -30,7 +30,7 @@ export async function getSection() {
rej(randomPostReq.status);
}
const sectionURL = `https://en.wikipedia.org/w/api.php?action=query&format=json&pageids=${pageid}&prop=extracts&exintro=true&explaintext=true&origin=*`;
const sectionURL = `https://en.wikipedia.org/w/api.php?action=query&format=json&pageids=${pageid}&prop=extracts&exintro=true&origin=*`;
var sectionReq = new XMLHttpRequest();
sectionReq.onload = () => {
@ -42,14 +42,16 @@ export async function getSection() {
let words = [];
// Remove non-ascii characters, double whitespaces and finally trailing whitespaces.
sectionText = sectionText.replace(/<\/p><p>+/g, " ");
sectionText = $("<div/>").html(sectionText).text();
sectionText = sectionText.replace(/[\u{0080}-\u{10FFFF}]/gu, "");
sectionText = sectionText.replace(/\s+/g, " ");
sectionText = sectionText.trim();
// Add spaces
sectionText = sectionText.replace(/[a-zA-Z0-9]{3,}\.[a-zA-Z]/g, (x) =>
x.replace(/\./, ". ")
);
// // Add spaces
// sectionText = sectionText.replace(/[a-zA-Z0-9]{3,}\.[a-zA-Z]/g, (x) =>
// x.replace(/\./, ". ")
// );
sectionText.split(" ").forEach((word) => {
words.push(word);