Hi Roger,
> Are you taking into account that for many reads the data is not copied
> into the local buffer.
> See the comments in BufferedInputStream.read1: 277:280?
sure, I'm aware of this optimization. The buffer however is always allocated as
BufferedInputStream instantiated,
waisting by defa
Hi Sergey,
Are you taking into account that for many reads the data is not copied
into the local buffer.
See the comments in BufferedInputStream.read1: 277:280?
How much is the slowdown, when BufferedInputStreams are chained?
Thanks, Roger
On 4/15/21 7:08 AM, Pavel Rappo wrote:
On 15 Apr 20
> On 15 Apr 2021, at 08:33, Сергей Цыпанов wrote:
>
> Hello,
>
> buffering with j.i.BufferedInputStream is a handy way to improve performance
> of IO operations.
> However in many cases buffering is redundant. Consider this code snippet from
> Spring Framework:
>
> static ClassReader getClas
Hello,
buffering with j.i.BufferedInputStream is a handy way to improve performance of
IO operations.
However in many cases buffering is redundant. Consider this code snippet from
Spring Framework:
static ClassReader getClassReader(Resource rsc) throws Exception {
try (var is = new BufferedInp
