In HBASE-24966, we found that in AsyncTableRegionLocator, we accidentally
declared 3 methods
getStartKeys
getEndKeys
getStartEndKeys
to throw IOException directly.
This should be a copy paste mistake, as typically, for a method which
returns CompletableFuture, the exception should be returned th
I think we should change what they throw directly and label it
incompatible. I think this is in line with our previous expectation setting
about how we'll handle mistakes in the API.
That change would be source incompatible but would still be binary
compatible.
I think we should do it in a major
I think it could introduce compilation error when removing the throws part
of a method signature. As if there is no exception thrown but you have a
'try...catch' then there will be a compilation error...
It requires a code change sometimes but anyway, you just need to remove the
'try...catch', no
If no other objections tomorrow, will follow what Sean proposed above, to
include the patch in 3.0.0 and mark the issue as incompatible change.
Thanks.
张铎(Duo Zhang) 于2020年11月30日周一 上午11:04写道：
> I think it could introduce compilation error when removing the throws part
> of a method signature. A
On Tue, Dec 1, 2020 at 6:45 AM 张铎(Duo Zhang) wrote:
>
> If no other objections tomorrow, will follow what Sean proposed above, to
> include the patch in 3.0.0 and mark the issue as incompatible change.
Belated +1.
> 张铎(Duo Zhang) 于2020年11月30日周一 上午11:04写道：
>
> > I think it could introduce compil
