Hi Jason, Jason McIntyre wrote on Sun, Aug 15, 2021 at 11:30:05PM +0100:
> can't we take a stance that where options override each other, > the last one wins? Yes, that is possible. Cases exist where one option overrides another and order does not matter - for example, "lock -n -t -1" is the same as just "lock -n", even though -t comes after -n. But if options override *each other*, that means the last one wins unless otherwise stated. A utility can also be designed such that some options override each other and the first one wins, but usually, i wouldn't consider that ideal UI design in a command line utility. In configuration files, on the other hand, both "first one wins" and "last one wins" exist. For that reason, nothing much seems wrong with being explicit that the last one wins, where that is the case. > and then not document this fact every time. or at least document > exceptions only? > > ...and continue to document where options are mutually exclusive? > > also the text "overrides earlier" is unclear. In which sense do you consider -f Frobnicate. Overrides earlier -u. -l [...] -n [...] -u Unfrobnicate. Overrides earlier -f. unclear? Options exclusively appear on the command line, so i don't see how "earlier" or "later" could possibly refer to anything else than the position on the command line. Note that this seems only loosely related to martijn@'s diff, which i think is undesirable for other, stronger reasons. Yours, Ingo
