Re: [csswg-drafts] [css-variables] Substitution of invalid variables into other variables (#5370)

@emilio Acknowledged on all points, thank you for the corrections.

Here is minimal var component example:,css,output

The inner most --smaller-input, and any other instances nested within, will always use --bigger-input behavior now because no fallback behavior is possible with initial at computed-value-time now. (unless it's on the root or has no parents with the same var name)

The library I'm working on is of course much more complex than that and has many many layers of "internal" variables that may compute to initial in any number of ways so I could use fallback behavior, all based on a small number of end-user input options.


@andruud Please know I respect your opinion here even though I disagree with the conclusion. It wouldn't be a reasonable request for me to ask this if we can't coordinate with all 3 browsers to move forward. Perhaps an alternative path to allow the previous computed-value-time initial behavior could be added independently? I haven't given much thought to that but would you be more comfortable going in that direction?

I think everyone will agree with both of you that it's _really nice_ to have all 3 browsers on the same page (and agreeing with the current spec). 

GitHub Notification of comment by James0x57
Please view or discuss this issue at using your GitHub account

Sent via github-notify-ml as configured in

Received on Wednesday, 29 July 2020 12:31:21 UTC