On Wed, Dec 1, 2021 at 7:13 PM Eric Sunshine <sunshine@xxxxxxxxxxxxxx> wrote:
> If we instead hoist ownership of `worktrees` up to execute_commands()
> -- which calls execute_commands_atomic() or
> execute_commands_non_atomic() -- then we can get by with retrieving
> the worktrees just once, and all those noise changes in update() can
> be dropped since it will no longer be responsible for allocating or
> freeing `worktrees`. For instance:
>
> static void execute_commands(...)
> {
> struct worktree **worktrees;
> ...
> worktrees = get_worktrees();
> if (use_atomic)
> execute_commands_atomic(commands, si, worktrees);
> else
> execute_commands_non_atomic(commands, si, worktrees);
> free_worktrees(worktrees);
> ...
> }
>
> and then execute_commands_atomic() and execute_commands_non_atomic()
> can pass `worktrees` along to update().
By the way, I wouldn't typically recommend this sort of change for
public API since it probably exposes too much implementation detail to
callers, but as these are all private (static) worker functions --
with very few callers, moreover -- I don't find it terribly worrying.
