On Sat, Aug 23, 2025 at 11:41 PM Martin D Kealey <mar...@kurahaupo.gen.nz> wrote: > > Ignoring -i when binding (only applying it when assigning) would seem like > a reasonable backwards-compatible extension, but can anyone think of a > circumstance where failing to leave the name unbound, and/or failing to > return a non-zero exit status, could conceivably break something useful?
-i may as well work the same as -l, -u, and -r, when combined with -n. What could the user expect to happen when -i and -n are combined? Something they actually want? Why would they want that? If your answer is to silently ignore -i and allow the declare command to succeed, the user still isn't getting what they expect to happen, assuming they expected something useful. An error message is more helpful. >From the manual: -n Give each name the nameref attribute, making it a name reference to another variable. That other variable is defined by the value of name. All references, assignments, and attribute modifications to name, except those using or changing the -n attribute itself, are performed on the variable referenced by name's value. The nameref attribute cannot be applied to array variables. I might have expected $ declare -ni reference=variable to have the same effect as $ declare -n reference=variable $ declare -i reference That being: $ declare -p reference variable declare -n reference="variable" declare -i variable Not going to argue that the behavior should be changed to work like that.
