Not everyone gets around to doing updates/upgrades immediately!
I concur by supporting this more disciplined (gradual) approach (mentionned below by Charles Campbell) of blocking out the code using #ifdef, then allowing the resultant "handicapped" code into the "wild" for at least 6 months.
After 6 months, any newer release is likely to have been installed and tested by those who have a critical vested interest in ensuring continuity of their preferred tools, especially if they know their OS (a.k.a. NextStep) is being gradually abandoned by upstream tool providers like Vim/GVim or others.
If the OS providers themselves consider the tool critical enough to their own internal processes, or as part of the User-expected toolset deployed with their OS, I am sure that they will make their voices be heard and speak up!
Charles, On Mo, 25 Aug 2025, Charles Campbell wrote:I think it would be a Good Idea to have a procedure for this sort of thing, and to publish it in the Vim Manual someplace. *1 announce that something is being considered for deprecation (comments accepted) *2 deprecate by surrounding the relevant code with #ifdef blocks and await any screams of protest. With this approach you'll get notified if anyone is using the feature/support and they can reverse it by putting a #define SOMETHING in vim.h. *3 release vim with the #ifdef SOMETHING blocks *4 next release remove the SOMETHING blocks At any point until the last one reversal of the change is easy. *2 is likely to get protestations amongst those who pay attention to the vim/vimdev groups. *3 will get protestations from those who are using the deprecated feature. Finally, *4 will remove the feature/support.Thanks, that makes a lot of sense. Yes I need to document this, it's been on my list for a while already. Thanks, Christian
No comments:
Post a Comment