IcedTea build instructions

Jiri Vanek jvanek at
Thu Oct 6 02:47:47 PDT 2011

On 10/06/2011 11:21 AM, Andrew Haley wrote:
> On 10/06/2011 08:53 AM, Jiri Vanek wrote:
>> On 10/05/2011 05:13 PM, Andrew Haley wrote:
>>> On 10/05/2011 04:08 PM, Andrew Haley wrote:
>>>>>  sez:
>>>>>      hg clone -r icedtea6-1.10
>>>>>   Surely this is wrong -- you really want
>>>>>      hg clone -r icedtea6-1.10.3 ssh://
>>> I meant
>>>>>   Don't you?
>>>>>   Or maybe even
>>>>>      hg clone
>> I believe the last one is the most correct. The first is also correct.
> How is the first one correct?  It's an old release on the
> icedtea6-1.10 branch.
>> But hg clone -r icedtea6-1.10.3 is not correct
>> for sure. As far as I know there is no branching on
>> minor-minor-vrsions - and have no sense.
> I'm not sure what you mean by this.  icedtea6-1.10.3 is the most
> recent release tag.

Yes - it is tagged, but not branched.

>> X.Y.Z - Y is always branched, but Z just signify that in this point
>> tarbals (for rpms...) were released.
> icedtea6-1.10   is 2487:9f4eced56544
> icedtea6-1.10.3 is 2531:7ba3ed55131e
> Why are we telling people to check out an obsolete release?  It
> doesn't make any sense.

IMHO - checkout from hg clone will checkout nearest future release (in this case 4). Although you can checkout this earlier  "Z" release, it should be forever equals with  stored tarball stored somewhere (eg Z=2 <=> icedtea6-1.10.2.tar.gz...). This "tagged ones" are not dedicated for pushing (probably also not even for checking out) as there are no fixes for exact x.y.z version. All fixes go to whole branch and are delivered as next "Z" release. In this case fixes for 10.1,2,3 are going to head of 1.10 and will be delivered in 10.4.

Regards J.

> Andrew.

More information about the distro-pkg-dev mailing list