From ddacfc0b8a3412fc47e615fd91b30430d037e5bb Mon Sep 17 00:00:00 2001 From: Dan Stillman Date: Tue, 19 Nov 2024 05:03:09 -0500 Subject: [PATCH] Strip invalid quotes around `charset` in feed responses E.g., https://bmcmedethics.biomedcentral.com/articles/most-recent/rss.xml --- resource/feeds/SAXXMLReader.js | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/resource/feeds/SAXXMLReader.js b/resource/feeds/SAXXMLReader.js index eaba91509a..89c6b41d0d 100644 --- a/resource/feeds/SAXXMLReader.js +++ b/resource/feeds/SAXXMLReader.js @@ -67,7 +67,11 @@ class SAXXMLReader { let charset = response.headers.get("Content-Type") ?.match(/charset=([^;]+)/) ?.[1]; - if (!charset) { + if (charset) { + // Strip surrounding quotes, which are invalid + charset = charset.replace(/^["']|["']$/g, ''); + } + else { charset = 'utf-8'; } this._data = new TextDecoder(charset).decode(buf);