std.gregorian contribution

[Steven Schveighoffer]

On Mon, 17 May 2010 14:00:41 -0400, Simen kjaeraas  
<simen.kjaras at gmail.com> wrote:

> Andrei Alexandrescu <SeeWebsiteForEmail at erdani.org> wrote:
>
>> I have two unrelated suggestions about unjoin.
>>
>> First, you may want to follow the model set by splitter() instead of  
>> split() when defining unjoin(). This is because split() allocates  
>> memory whereas splitter splits lazily so it doesn't need to. If you do  
>> want split(), just call array(splitter()).
>>
>> Second, there is an ambiguity between splitting using a string as  
>> separator and splitting using a set of characters as separator. This  
>> could be solved by simply using different names:
>>
>> string str = ...;
>> foreach (splitByOneOf(str, "; ")) { ... }
>> foreach (splitter(str, "; ")) { ... }
>>
>> First look splits by one of the two, whereas the second splits by the  
>> exact string "; ".
>>
>> An idea I am toying with is to factor things out into the data types.  
>> After all, if I'm splitting by "one of" an element in a set of  
>> elements, that should be reflected in the set's type. For example:
>>
>> foreach (splitter(str, either(';', ' ')) { ... }
>> foreach (splitter(str, "; ")) { ... }
>>
>> or, using a more general notion of a set:
>>
>> foreach (splitter(str, set(';', ' ')) { ... }
>
> D could use a set type, and this is a very nice way to specify these
> different parameters.
>
> votes = -~votes;

Comparing splitByOneOf(str, "; ")) to splitter(str, set(';', ' ')), I see  
one major difference here  -- "; " is a literal, set(';', ' ') is not.

I would expect that 'set' as a generic set type would implement it's guts  
as some sort of tree/hash, which means a lot of overhead for a simple  
argument.  With the literal version, the notation is in the function, not  
the type.  While it seems rather neat, the overhead should be considered.

A compromise:

foreach(x; splitter(str, either("; ")))

Which can be implemented without heap activity.

-Steve