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

Merge pull request #1 from algolia/remove-font-tag

Browse Source 
fix: remove font tags
This commit is contained in:
Maxime Janton
2017-11-27 10:43:04 +01:00
committed by GitHub
parent e3deddf3c2 6925762349
commit 5d78891abe
2 changed files with 6 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
package.json
View File

@@ -40,6 +40,7 @@
"dependencies": { "dependencies": {
"axios": "^0.17.1", "axios": "^0.17.1",
"he": "^1.1.1", "he": "^1.1.1",
"lodash": "^4.17.4" "lodash": "^4.17.4",
"striptags": "^3.1.0"
} }
} }

5
src/index.js
View File

@@ -3,6 +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';
export async function getSubtitles({ export async function getSubtitles({
videoID, videoID,
@@ -51,7 +52,9 @@ export async function getSubtitles({
.replace(/&/gi, '&') .replace(/&/gi, '&')
.replace(/<\/?[^>]+(>|$)/g, ''); .replace(/<\/?[^>]+(>|$)/g, '');
const text = he.decode(htmlText); const strippedText = striptags(htmlText);
const text = he.decode(strippedText);
return { start, dur, text }; return { start, dur, text };
}); });