Add the ability to use RegEx tag to extract any value from the URL

[RobertKlohr]

Implementing this feature would also take care of the open requests #8 and #14.

Example URL: https://example.com/?login=saml&email=test%40example.com with the title “Sign-In”

Example:

The tag could be {regex=”expression”} with “expression” being the RegEx to apply.

Tab title format: {title} - {protocol}://{hostnameascii} - {regex=([a-zA-Z0-9._-]+%40([a-zA-Z0-9_-]+.)+[a-zA-Z0-9_-]+)}

Result: Sign In - https://example.com/ - test%40example.com

[guilryder]

Similar complexity considerations as in #14.

Same regex implementation caveats as detailed in #8 (comment).

I will consider adding the feature if there's a use case strong enough to justify the complexity and the security implications (correctly parsing URLs with regexes is difficult, if only because of %-encoding).

[RobertKlohr]

Speaking to your comment in #14 (comment) and the discussion in #8 (comment), I agree that having you code the extension to cover all possible parsing with tags would be problematic and a burden. This is why, aside from the existing tags that will work for most users, implementing a generic Regex parser would solve my use case, and I would guess any other existing and future use cases for parsing content from a URL.

Specifically, I am not looking to have the extension handle pre-website customization in the configuration. If users had the ability to parse the URL with Regex including return capture groups, then any and all customization a user could need would be available AND up to the user to implement via Regex.

Here is my case that applies to this request and #14 as well.

As both a system administrator and developer I work with multiple (dozens) of authentication accounts which I store in KeePass. Many of these accounts use the same a small set of shared SAML or OAUTH authentication services. The current situation is that we can trigger KeePass to prompt for an entry based on the host and domain in the URL, but then have to select, sometimes scrolling many entries, to then select the KeePass entry to populate to the authentication service.

If we could parse the URL to extract information being passed in the URL as part of the authentication process, in most cases this would be the username, we would be able to associate separate authentication calls to the same service to matching KeePass entries. This would allow the user to then more easily use password managers, such as KeePass, to store and use they login credentials.

Please let me know if you want clarification on the use case I am describing.

[guilryder]

If the use case is KeePass integration, KeePass supports matching window titles with custom wildcard expressions. Configure the extension to include the full URL in the title with the {args} tag. Configure KeePass with expressions like *https://example.com/?*email=test%40example.com*. I think this is both simpler and safer than regexes.

[RobertKlohr]

KeePass was one example, and what you described is how I have been using it for years, but not the only use case I had in mind.

There are other robotic process automation tools I use, like UiPath, that rely on the window and tab titles to trigger automated actions. Most, but not all, of these can be configured to work with the full URL in the title, but the downside of putting the full URL in the tab title is then having tab titles that can be several hundred characters long and needing to edit every bookmark you create.

By addressing the issue closer to the source of the data, I was hoping to simplify the full set of tools I use that rely on this information. I think there this value to the feature and I believe it would used. I also wish I had the JavaScript coding skills and knowledge on programming extensions to help with the development, but alas, my coding skills are in other scripting languages and C.