Bug in readln interface ?

monarch_dodra monarchdodra at gmail.com
Sun Jun 30 23:36:46 PDT 2013


On Sunday, 30 June 2013 at 23:10:07 UTC, Steven Schveighoffer 
wrote:
> On Sun, 30 Jun 2013 19:08:35 -0400, Steven Schveighoffer 
> <schveiguy at yahoo.com> wrote:
>
>> On Sun, 30 Jun 2013 15:12:40 -0400, monarch_dodra 
>> <monarchdodra at gmail.com> wrote:
>>
>>>
>>> So my question is: Do we *want* to keep this? Should we 
>>> deprecate it? I think we should deprecated it.
>>>
>>> Thoughts?
>>
>> Fully agree.  The fact that the condition explicitly disallows 
>> enums should clue us in that it was not intended to simply 
>> play nice with types, it really wants a mutable buffer.
>>
>> If you want an immutable result, use the version that gives 
>> you the result as a return value.
>
> BTW, this should simply be changed, not deprectated IMO.  It is 
> an "accepts invalid" bug.  The function description 
> specifically says it reuses the buffer and extends as 
> necessary.  Since it can't possibly reuse, that means this is 
> truly a bug.
>
> -Steve

Alright, thanks. Officially Filed:
http://d.puremagic.com/issues/show_bug.cgi?id=10517

And under correction.


More information about the Digitalmars-d mailing list