On 6/19/15 2:44 PM, Timon Gehr wrote:
On 06/19/2015 11:37 PM, Timon Gehr wrote:
On 06/19/2015 05:27 PM, Andrei Alexandrescu wrote:
On 6/19/15 8:01 AM, "Ulrich =?UTF-8?B?S8O8dHRsZXIi?=
<[email protected]>" wrote:
If opAssign is allowed, a major point of functional data structures is
lost. Client code is so much better if rebinding is off the table.
I have the same instinct but not enough rationale. What would be an
example in favor of the argument?
I think he is just arguing for immutable by default. The obvious
counter-argument is: Show me the implementation of the
'length(T)(SList!T)' function for SList!T with disabled opAssign.
Bad example, you'd just use the range, I guess. :o)
Append?
Appending to a functional singly-linked list would necessitate wholesale
duplication, whether or not the list has mutable elements. Indeed if the
reference count is 1 for _all_ elements in the list, there's no need for
copying. That takes O(n) time to evaluate, but walking through to the
end of the list is needed anyway.
But that's due to the topology of the list. You can append to a slice of
immutable objects no problem (we do that with strings all the time).
I still am confused as to where your vision aims; I'll reply to your
other message.
Andrei
