shanefully_done/youtube-captions-scraper
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Revert to regex

Browse Source
This commit is contained in:
Maria Schreiber
2017-11-30 22:29:45 -05:00
parent dfe7c2a6fa
commit 4594650aab
1 changed files with 6 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
src/index.js
View File

@@ -3,7 +3,7 @@
import he from 'he'; import he from 'he';
import axios from 'axios'; import axios from 'axios';
import { find } from 'lodash'; import { find } from 'lodash';
import striptags from 'striptags'; // import striptags from 'striptags';
export async function getSubtitles({ export async function getSubtitles({
videoID, videoID,
@@ -44,17 +44,19 @@ export async function getSubtitles({
const startRegex = /start="([\d.]+)"/; const startRegex = /start="([\d.]+)"/;
const durRegex = /dur="([\d.]+)"/; const durRegex = /dur="([\d.]+)"/;
const fontTag = new RegExp('<'+'font'+'[^><]*>|<.'+'font'+'[^><]*>','g');
const [, start] = startRegex.exec(line); const [, start] = startRegex.exec(line);
const [, dur] = durRegex.exec(line); const [, dur] = durRegex.exec(line);
const htmlText = line const htmlText = line
.replace(/<text.+>/, '') .replace(/<text.+>/, '')
.replace(/&amp;/gi, '&') .replace(/&amp;/gi, '&')
.replace(/<\/?[^>]+(>|$)/g, ''); .replace(/<\/?[^>]+(>|$)/g, '')
.replace(fontTag, '');
const strippedText = striptags(htmlText); // const strippedText = striptags(htmlText);
const text = he.decode(strippedText); const text = he.decode(htmlText);
return { start, dur, text }; return { start, dur, text };
}); });