`access -e` is confusing

[jpco]

This is a tiny design question.

Currently, the -e flag in $&access is used along with the -1 flag to generate an exception if none of the given files are accessible. This is used to good effect in %pathsearch.

However, without the -1, $&access -e quietly has no effect at all. That seems a bit strange, and not very useful. Should we instead treat it as an invalid option, OR make it significant in that case by throwing an exception if the $&access operation would have a falsey result?

[memreflect]

I consider -e to be an optional modifier for -1, not a separate flag.

On the other hand, the only reason it's necessary is because -1 changes the "success" return value from 0 to a path and the "failure" return value from an error message to an empty list.
That is, there is no way to obtain the error message without a second invocation of $&access if an empty list is returned.
Combine that rationale with -1 limiting the output to a single path, and -e as a modifier for -1 has utility.

If one could return all matching paths or some custom number of paths greater than 1, rather than a single path like -1, customization of -e behavior may be desirable:

• You might only want an exception thrown if all paths provided would result in error messages without -e.
  • Should all error messages be included in the exception arguments or just the first/last error message?
• Alternatively, one might want to verify all paths provided satisfy some set of criteria—possibly all criteria, possibly one or more subsets of several "valid" subsets of criteria.
  That is, -e should throw an exception when even a single path results in a falsey value.

Those are just two possible -e behaviors, with the simpler "error if the -1 flag is omitted" being a third.