proper tail calls

Brendan Eich brendan at mozilla.org
Mon Jan 21 22:52:24 PST 2008


On Jan 21, 2008, at 8:02 PM, Maciej Stachowiak wrote:

> On Jan 21, 2008, at 12:35 PM, Brendan Eich wrote:
>
>> Conversions (implicit and hardcoded among the
>> built-in types representing and wrapping primitives) that might
>> defeat PTC may not be evident until runtime, where the result would
>> be a TypeError or possibly a new Error subtype.
>
> Isn't this case (implicit conversion) exactly what motivated the idea
> that programmers may not be able to easily tell if a call is in tail
> position?


Indeed:

"ES4 has proper tail calls, but their constraints are sometimes  
subtle, especially with regard to conversions or type checks inserted  
at the return point. It may be that the "Explicit Is Better Than  
Implicit" principle once again finds application here."

First paragraph in http://bugs.ecmascript.org/ticket/323. Again, the  
ticket is just sitting there, you don't need me transcribing it into  
this list :-/.

/be



More information about the Es4-discuss mailing list