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

fix: remove html tags

Browse Source
This commit is contained in:
Maria Schreiber
2017-11-24 10:29:06 -05:00
parent 2e77d7f497
commit 5152a6bf5f
3 changed files with 5499 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5492
package-lock.json generated Normal file
View File

File diff suppressed because it is too large Load Diff

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"
} }
} }

9
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,
@@ -46,14 +47,14 @@ export async function getSubtitles({
const [, start] = startRegex.exec(line); const [, start] = startRegex.exec(line);
const [, dur] = durRegex.exec(line); const [, dur] = durRegex.exec(line);
const fontTag = new RegExp('<'+'font'+'[^><]*>|<.'+'font'+'[^><]*>','g');
const htmlText = line const htmlText = line
.replace(/<text.+>/, '') .replace(/<text.+>/, '')
.replace(/&amp;/gi, '&') .replace(/&amp;/gi, '&')
.replace(/<\/?[^>]+(>|$)/g, '') .replace(/<\/?[^>]+(>|$)/g, '');
.replace(fontTag, '');
const text = he.decode(htmlText); const strippedText = striptags(htmlText);
const text = he.decode(strippedText);
return { start, dur, text }; return { start, dur, text };
}); });