Would a partial disable based on `ExtensionState.DISABLING` which only reverts patches be permissible?

A requirement to only change monkey-patching on enable/disable would allow extension rebasing to work correctly, and prevent that class of errors.

All these restrictions look nice on paper, but in reality they can bring other problems and worse user experience. Extension authors have no control over the order in which extensions are enabled, and no control over extensions that override their code. If I don’t take care of conflicts with other extensions myself, users know exactly whose fault it is. My extension is buggy, nobody cares that DtD overrides my code. To minimize conflicts, I need to be able to override the code whenever needed. We cannot rely on support from the authors of conflicting extensions. It’s all time consuming and often frustrating.
However, for the rebase purposes, I save the original code when enabled.

Created an issue for this one in gjs.guide repo since we need to decide whether this should be added to the ego review guidelines.

1 Like

This topic was automatically closed 45 days after the last reply. New replies are no longer allowed.