[erlang-questions] If Condition vs. Multiple Function Clauses

Steve Davis steven.charles.davis@REDACTED
Mon Jun 17 15:06:29 CEST 2013


Hi Richard,

Yep, you changed my mind (again). I wasn't thinking about this correctly. 

Thanks very much for your replies,

Steve

On Jun 16, 2013, at 6:19 PM, "Richard A. O'Keefe" <ok@REDACTED> wrote:

> 
> On 16/06/2013, at 1:30 AM, Steve Davis wrote:
> 
>> Yep, that would be a better, more consistent way to go...
>> 
>> if  M == N -> a();
>>    M > N -> b();
>>    _ -> c()
>> end.
> 
> Let's see, this says
> 	"Should the test M == N succeed, compute a().
> 	 Otherwise, should the test M >N succeed, compute b().
> 	 Otherwise, take a new variable that has never been given
> 	 any value whatsoever; should that variable, contrary to
> 	 all sane expectation, somehow magically turn out to be
> 	 bound to a succeeding test, compute c()."
> 
> Just how is the "_" supposed to get a value here?
> 
> By the way, anyone who is really suffering from the lack of 'else'
> can just
> 	-define(else, true).
> and then
> 	if M < N -> c()
> 	 ; M > N -> b()
> 	 ; ?else -> a()
> 	end
> and be done with it, although putting M == N there instead would
> make it more obvious to one's readers, should one _have_ readers.
> 




More information about the erlang-questions mailing list