> On 8 Dec 2016, at 01:27, Claes Redestad <claes.redes...@oracle.com> wrote: > > > > On 2016-12-08 10:01, Ulf Zibis wrote: >> Am 08.12.2016 um 09:28 schrieb Peter Levart: >>> >>> On 12/07/2016 11:28 PM, Roger Riggs wrote: >>>> AbstractStringBuilder: >>>> I agree with Claes' comment suggesting that IAE for negative lengths is >>>> a pain >>>> and defining it to append 0 would be natural in many use cases. >>> >>> OTOH, inserting a simple Math.max(n, 0) instead of n where n could get >>> negative would achieve the same without complicating the expression too >>> much. Java standard APIs have a tradition of being explicit rather than >>> having implicit hidden logic which surely shortens many usecases, but makes >>> them harder to read and understand for casual readers not intimately >>> familiar with such API. The logic to treat negative lengths as 0 is >>> implicit and not universally correct. >> +1 >> If we would treat negative values as 0, we loose a chance, where programmers >> could become aware about possible errors in the logic of their program. > > Right, the two of you have convinced me that some exceptional explicitness is > the better choice in this (and likely most) cases. >
+1 i think this is the correct approach. Paul.
