AI/reader
1
0
Fork 0
mirror of https://git.mirrors.martin98.com/https://github.com/jina-ai/reader synced 2025-08-17 00:45:55 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix: truncate svg in jsdom

Browse Source
This commit is contained in:
yanlong.wang 2024-07-31 13:12:57 +08:00
parent 4e5aff3332
commit a08218506e
No known key found for this signature in database
GPG Key ID: C0A623C0BADF9F37
1 changed files with 2 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
backend/functions/src/services/jsdom.ts
View File

@ -39,6 +39,7 @@ export class JSDomControl extends AsyncService {
const t0 = Date.now(); const t0 = Date.now();
const jsdom = new JSDOM(snapshot.html, { url: snapshot.href, virtualConsole }); const jsdom = new JSDOM(snapshot.html, { url: snapshot.href, virtualConsole });
const allNodes: Node[] = []; const allNodes: Node[] = [];
jsdom.window.document.querySelectorAll('svg').forEach((x) => x.innerHTML = '');
if (options?.withIframe) { if (options?.withIframe) {
jsdom.window.document.querySelectorAll('iframe[src],frame[src]').forEach((x) => { jsdom.window.document.querySelectorAll('iframe[src],frame[src]').forEach((x) => {
const src = x.getAttribute('src'); const src = x.getAttribute('src');
@ -151,6 +152,7 @@ export class JSDomControl extends AsyncService {
const extendedSnapshot = { ...snapshot } as ExtendedSnapshot; const extendedSnapshot = { ...snapshot } as ExtendedSnapshot;
try { try {
const jsdom = new JSDOM(snapshot.html, { url: snapshot.href, virtualConsole }); const jsdom = new JSDOM(snapshot.html, { url: snapshot.href, virtualConsole });
jsdom.window.document.querySelectorAll('svg').forEach((x) => x.innerHTML = '');
const links = Array.from(jsdom.window.document.querySelectorAll('a[href]')) const links = Array.from(jsdom.window.document.querySelectorAll('a[href]'))
.map((x: any) => [x.getAttribute('href'), x.textContent.replace(/\s+/g, ' ').trim()]) .map((x: any) => [x.getAttribute('href'), x.textContent.replace(/\s+/g, ' ').trim()])
.map(([href, text]) => { .map(([href, text]) => {