Add support for markdown links. #835
Conversation
There was a problem hiding this comment.
Thanks for opening the PR @skilldeliver.
Would you be up for adding some unit tests to test/client/lib/textUtils.ts?
| // Match http or https then domain name (with optional port) then any text that ends with letter or number. | ||
| export const urlRegex = /(https?:\/\/[A-Za-z\d][A-Za-z\d-.]*(?!\.)(?::\d+)?(?:\/[^\s]*)?[\w\d/])/; | ||
| // Match http or https then domain name and capture markdown link text and URL separately | ||
| const urlRegex = /(?:\[(.*?)\]\()?https?:\/\/[A-Za-z\d][A-Za-z\d-.]*\.[A-Za-z]{2,}(?::\d+)?(?:\/[^\s)]*)?(?:\))?/; |
There was a problem hiding this comment.
I think this would match strings like [Google](https://www.google.com or https://www.google.com).
There was a problem hiding this comment.
You are right @georgegevoian - the regex needs fix;
I will take a look how can I add tests.
Btw one of the CI tests of nbrowser is failing and I can't understand why.
There was a problem hiding this comment.
DetailView does not opens cell for editing when clicked on link looks to be failing because a localhost URL isn't being treated as a link (most likely a regression in the regex).
The other failures look unrelated - we'll be landing a fix on main soon.
Thanks.
This PR introduces markdown link detection and rendering. To achieve this, the regex has been updated, and an additional property, link, has been added to the tokens. If it's a URL, the URL remains the same, with the actual text rendered. However, for markdown, we aim to have a named link where the text differs from the link.
Screenshot examples:
Here is the raw text content
Here is the rendered
Special thanks to Darius for recognizing the need for named links.