You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@commons.apache.org by PeterLee <pe...@apache.org> on 2021/07/30 04:00:26 UTC
[COMPRESS] About Compress-583
Hi all,
There is a new issue Compress-583 reporting Compress 1.21 generated tars
are not byte-to-byte
same with older versions.And the generated tars are also different in
different OSes in 1.21.
This is caused by Github PR #97(Compress-404) : the UID/GID is set in Unix
in the constructor of
TarArchiveEntry - but we didn't do it before 1.21.. And please note that
this is the default behavior
of tar(1).
We didn't record this in the change log of 1.21 - and it was my bad.
IMO this is not a breaking change - there's no API change. Only the
generated tars is changed. But
I agree that this should be documented.
I want to hear about your ideas. WDYT?
cheers,
Lee
Re: [COMPRESS] About Compress-583
Posted by Gary Gregory <ga...@gmail.com>.
I updated the release notes and site.
Gary
On Fri, Jul 30, 2021 at 7:26 AM Bruno P. Kinoshita <ki...@apache.org> wrote:
>
> Agreed. Just document in the release notes.
>
>
>
>
> On Friday, 30 July 2021, 4:21:01 pm NZST, Gary Gregory <ga...@gmail.com> wrote:
>
> It seems to me we just need to update our release notes.
>
> Gary
>
> On Fri, Jul 30, 2021, 00:00 PeterLee <pe...@apache.org> wrote:
>
> > Hi all,
> >
> > There is a new issue Compress-583 reporting Compress 1.21 generated tars
> > are not byte-to-byte
> > same with older versions.And the generated tars are also different in
> > different OSes in 1.21.
> >
> > This is caused by Github PR #97(Compress-404) : the UID/GID is set in Unix
> > in the constructor of
> > TarArchiveEntry - but we didn't do it before 1.21.. And please note that
> > this is the default behavior
> > of tar(1).
> > We didn't record this in the change log of 1.21 - and it was my bad.
> >
> > IMO this is not a breaking change - there's no API change. Only the
> > generated tars is changed. But
> > I agree that this should be documented.
> >
> > I want to hear about your ideas. WDYT?
> >
> > cheers,
> > Lee
> >
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
For additional commands, e-mail: dev-help@commons.apache.org
Re: [COMPRESS] About Compress-583
Posted by "Bruno P. Kinoshita" <ki...@apache.org>.
Agreed. Just document in the release notes.
On Friday, 30 July 2021, 4:21:01 pm NZST, Gary Gregory <ga...@gmail.com> wrote:
It seems to me we just need to update our release notes.
Gary
On Fri, Jul 30, 2021, 00:00 PeterLee <pe...@apache.org> wrote:
> Hi all,
>
> There is a new issue Compress-583 reporting Compress 1.21 generated tars
> are not byte-to-byte
> same with older versions.And the generated tars are also different in
> different OSes in 1.21.
>
> This is caused by Github PR #97(Compress-404) : the UID/GID is set in Unix
> in the constructor of
> TarArchiveEntry - but we didn't do it before 1.21.. And please note that
> this is the default behavior
> of tar(1).
> We didn't record this in the change log of 1.21 - and it was my bad.
>
> IMO this is not a breaking change - there's no API change. Only the
> generated tars is changed. But
> I agree that this should be documented.
>
> I want to hear about your ideas. WDYT?
>
> cheers,
> Lee
>
Re: [COMPRESS] About Compress-583
Posted by Gary Gregory <ga...@gmail.com>.
It seems to me we just need to update our release notes.
Gary
On Fri, Jul 30, 2021, 00:00 PeterLee <pe...@apache.org> wrote:
> Hi all,
>
> There is a new issue Compress-583 reporting Compress 1.21 generated tars
> are not byte-to-byte
> same with older versions.And the generated tars are also different in
> different OSes in 1.21.
>
> This is caused by Github PR #97(Compress-404) : the UID/GID is set in Unix
> in the constructor of
> TarArchiveEntry - but we didn't do it before 1.21.. And please note that
> this is the default behavior
> of tar(1).
> We didn't record this in the change log of 1.21 - and it was my bad.
>
> IMO this is not a breaking change - there's no API change. Only the
> generated tars is changed. But
> I agree that this should be documented.
>
> I want to hear about your ideas. WDYT?
>
> cheers,
> Lee
>