The css value "unset" is not being reported for browsers that do not support it.

[narthollis]

Many browsers do not support the CSS value unset which is used for resetting a css attribute as back to its default inherit/initial value.

Currently doiuse is not reporting this when testing with browsers who do not have this value.

http://caniuse.com/#feat=css-unset-value

[ZacharyRSmith]

From what I can tell looking at the source code, no unsupported values are being checked -- only unsupported property-value pairs.

It might be worth it to have the Detector check ALL features, and whenever an unsupported value is used (such as unset or initial for IE), throw a warning.

[ZacharyRSmith]

My workaround:

I'm using stylelint to run doiuse, in addition to other linting. To warn about "initial" and "unset", I have the below rule in my .stylelintrc. The built-in rule 'declaration-property-value-blacklist' blacklists the property-value pair I give it. The pair I give it is:

property: "/^.*/" (match all regex)
value: ["initial", "unset"]

So, all properties with a value of "initial" or "unset" will throw a violation.

I set the violation severity to warning.

		"declaration-property-value-blacklist": [{
			"/^.*/": ["initial", "unset"]
		}, { "severity": "warning" }],

[ismay]

From what I can tell looking at the source code, no unsupported values are being checked -- only unsupported properties.

By the way, it isn't just properties that are checked. I do get warnings for using rem for example. No idea why unset isn't being reported.

[ZacharyRSmith]

I should have said "unsupported property-value pairs". My guess is the code checks all properties known to use rem, then complains if it sees rem. But yeah, weird that it wouldn't do the same for unset etc.

[rincedd]

Hi @anandthakker,

I fixed this in #93. I'd be happy to work on a few other issues as well. Could you provide write access to the repo, as mentioned in the contribution guideline? Otherwise please consider the PR.

Thanks,
Gerd

[rincedd]

Maybe @pkuczynski could help?

[pkuczynski]

I just merged your PR and can release the new version soon. When it comes to writing access to the repo, you have to talk with @anandthakker directly as I do not have permissions to grant rights...

[rincedd]

Cool, thanks a lot!