Kamil Śliwak

Yes, this was already possible in 1.1.4. I am doing ``` ruby has_widgets do |root| root

If you mean requiring user to always provide unique ID then no, its not possible. In BWidget you only name the nested CWidget once, but there may be multiple instances...

@matheusaaguiar Just a quick hint, you can use `--output-format markdown` with `benchmark_diff.py` to get a properly formatted markdown table you can paste on github. The one you have here is...

Weird. Could happen with some changes in Hardhat but then it would affect all tests. I wonder how changes in ENS alone could have done that... One possibility is that...

It is supported :) They both accept name of the config file. Try putting the new config file name in `local config_file="hardhat.config.js"` at the top of the script. These functions...

Ah, that's because in TS the config exports a variable and getting the name of that automatically is tricky (and there seems to be no universal default; it differs wildly)....

Needs rebase on latest `develop` to have the OZ test pass. And the chainlink one looks just like a transient networking error.

The failing OpenZeppelin test is already fixed on latest `develop`. I rebased the PR and pushed to the original branch to make it pass so that we can merge this....

Here's the gist of my discussion with @ekpyron on Gitter: The reason that the stack compressor and the `g` and `fDnTOc g` parts are hard-coded is that it's just final...

We discussed it today and decided that making stack compressor a step and the whole sequence configurable is a good idea. Also, we want to keep it simple and have...