You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When we stream data (in my case from a chatbot) to the frontend and then parse the markdown using markdown-to-jsx if the chatbot starts outputting a code block the app would crash (only for HTML). This happens as:
Since we stream the responses from chatbot the chatbot starts the code with opening the code block and then starts writing its HTML but since the code block isn't closed markdown-to-jsx detects it as plain HTML and renders it inside a <p> tag which causes DOM errors as the <html> tag cannot be a child of the <p> tag.
I believe markdown-to-jsx should detect the starting of the code block and till the code block is not closed by '```' it should just detect the chars after the starting as the code.
The text was updated successfully, but these errors were encountered:
We're facing a similar issue where invalid markdown causes a CPU blowout. We essentially need a way to handle markdown which contains invalid html/markdown. Perhaps there's a way to strip this invalid markup or escape it?
When we stream data (in my case from a chatbot) to the frontend and then parse the markdown using
markdown-to-jsx
if the chatbot starts outputting a code block the app would crash (only for HTML). This happens as:Since we stream the responses from chatbot the chatbot starts the code with opening the code block and then starts writing its HTML but since the code block isn't closed
markdown-to-jsx
detects it as plain HTML and renders it inside a<p>
tag which causes DOM errors as the<html>
tag cannot be a child of the<p>
tag.I believe
markdown-to-jsx
should detect the starting of the code block and till the code block is not closed by '```' it should just detect the chars after the starting as the code.The text was updated successfully, but these errors were encountered: