OpenGov: Description of failure condition seems wrong about polkadot-wiki

[maltekliemann]

Thanks, that settles it then!

The question came up in a discussion with the Polkassembly team about why their app shows a proposal passing outside of the decision period. Played around with the tests of pallet-referendum and then verified using the Polkadot dev node.

[DrW3RK]

Thanks for reporting. I skimmed through the source code and observed that the time elapsed could be greater than the deciding period, provided the referendum is confirming. So, your claim "if a proposal starts confirming one block before the end of the decision period and stays in confirming state for the duration of the confirmation period, it is approved. " makes sense to me.

Time elapsed is calculated here https://github.com/paritytech/polkadot-sdk/blob/9bf1a5e23884921498b381728bfddaae93f83744/substrate/frame/referenda/src/lib.rs#L1153

If the referendum is confirming, then it lets elapsed time be greater than the deciding period here https://github.com/paritytech/polkadot-sdk/blob/9bf1a5e23884921498b381728bfddaae93f83744/substrate/frame/referenda/src/lib.rs#L1300 and if the voting tally reverses, the referendum is no longer confirming and gets rejected as the decision period has elapsed.

We will update the Wiki. Just curious how you experimented with this!