Hi that would be great. Maybe, just maybe, either start a discussion whether we should assume lexical-binding as a default for the user configuration (which was my understanding of a desirable future for Emacs) and, of course, make sure this assumption holds. As stated somewhere recently, *scratch* assumes lexical binding and many users test out their own code on scratch before adding it to their configuration. In a list, options to discuss would be: 1.- should init.el be assumed to use lexical binding? 2.- should files loaded by it be assumed to use lexical binding? 3.- should custom.el and early-init.el be assumed to use lexical binding? 4.- should we assume lexical binding by default and allow it to be disabled with a mode line. My .02 cents, /PA On Sun, 5 May 2024 at 19:32, Jim Porter wrote: > On 5/5/2024 1:43 AM, Mattias EngdegÄrd wrote: > > Angelo and Pedro, thank you for telling us! The warning hasn't been > fine-tuned yet, but your reports may help us do that. > > Maybe warn only once about files under user-emacs-directory? Or only > once for all files loaded before Emacs has fully-completed startup? > > As a bonus, we could offer a command that will add the lexical-binding > cookie to all the files Emacs has noticed it's missing from. (Here, we > could ask the user whether they want to keep lexical-binding off in > those files for compatibility, or if they're prefer to upgrade to use > lexical-binding, with the small risk of some breakage.) > > Getting spammed with warnings is surely annoying, but a single warning > with a "fix it" button seems like it would be a lot better. > > And, as mentioned elsewhere in another thread, we could probably do with > making it more obvious that you can suppress warnings. I was in a > previous discussion about just that, and *still* forgot about it until > someone reminded me. > -- Fragen sind nicht da, um beantwortet zu werden, Fragen sind da um gestellt zu werden Georg Kreisler Headaches with a Juju log: unit-basic-16: 09:17:36 WARNING juju.worker.uniter.operation we should run a leader-deposed hook here, but we can't yet