implementation of suffix

[Robert Virding]

I would say a bug, but in which clause is debatable.

Robert

Ulf Wiger (AL/EAB) wrote:

>The implementation of lists:suffix/2 is as follows:
>
>
>suffix(Suffix, Suffix) -> true;
>suffix(Suffix, [_|Tail]) ->
>    suffix(Suffix, Tail);
>suffix(_, []) -> false.
>
>
>The programming style in lists.erl seems to be to be very conservative
>about using type guards. One objection to this could be that it makes 
>it more difficult for Dialyzer to derive good type information from one of 
>the most frequently used modules, but in the case of suffix/2, I would
>say that it also violates the Principle of Least Astonishment:
>
>(n@REDACTED)13> lists:suffix(a,[1,2,3|a]).
>true
>(n@REDACTED)14> lists:suffix(b,[1,2,3|a]).
>** exited: {function_clause,[{lists,suffix,[b,a]},
>                             {erl_eval,do_apply,5},
>                             {shell,exprs,6},
>                             {shell,eval_loop,3}]} **
>
>=ERROR REPORT==== 27-May-2005::15:36:23 ===
>Error in process <0.95.0> on node 'n@REDACTED' with exit value: {function_clause,[{lists,suffix,[b,a]},{erl_eval,do_apply,5},{shell,exprs,6},{shell,eval_loop,3}]}
>
>
>Bug or feature?
>
>/Uffe
>
>  
>