1.3.0 Candidate

[Grant Ingersoll <gs...@apache.org>]

Proposed 1.3.0 candidate available at http://people.apache.org/~gsingers/solr/1.3.0/

Note, this is not the official release until we have a vote on it.

Have a look, and if it seems right, I suggest we call a vote tomorrow.

-Grant

Re: 1.3.0 Candidate

[Shalin Shekhar Mangar <sh...@gmail.com>]

On Wed, Sep 10, 2008 at 11:10 PM, Grant Ingersoll <gs...@apache.org>wrote:

> So, how we doing on the Woodstox stuff?  Have people had a chance to try it
> out on trunk?  If so, can we port to the 1.3 branch?


I ran a few indexing jobs on my end. Things look good.

+1 for porting to branch.

-- 
Regards,
Shalin Shekhar Mangar.

Re: 1.3.0 Candidate

[Grant Ingersoll <gs...@apache.org>]

So, how we doing on the Woodstox stuff?  Have people had a chance to  
try it out on trunk?  If so, can we port to the 1.3 branch?

On Sep 9, 2008, at 9:19 AM, Grant Ingersoll wrote:

>
> On Sep 7, 2008, at 8:59 PM, Chris Hostetter wrote:
>>
>> (Curiousity: Grant, is the hudson config for doing releases not  
>> working
>> out?  it seemed like a really sweet way to automate everything and  
>> i'm
>> just wondering if you found something that made it not  
>> worthwhile ... it
>> seems to be gone now)
>>
>
> FWIW, I added a "prepare-release" target to the build, that I think  
> does most of the steps automatically.  Now, don't get me wrong, one  
> should still check out the docs, etc.  It requires svn, forrest and  
> gpg to be on the path.  If anyone is so inclined, they could put  
> <available> checks in.  Would probably be useful to have other  
> validation steps, too.
>
>
>

Re: 1.3.0 Candidate

[Grant Ingersoll <gs...@apache.org>]

On Sep 7, 2008, at 8:59 PM, Chris Hostetter wrote:
>
> (Curiousity: Grant, is the hudson config for doing releases not  
> working
> out?  it seemed like a really sweet way to automate everything and i'm
> just wondering if you found something that made it not  
> worthwhile ... it
> seems to be gone now)
>

FWIW, I added a "prepare-release" target to the build, that I think  
does most of the steps automatically.  Now, don't get me wrong, one  
should still check out the docs, etc.  It requires svn, forrest and  
gpg to be on the path.  If anyone is so inclined, they could put  
<available> checks in.  Would probably be useful to have other  
validation steps, too.

Re: 1.3.0 Candidate

[Grant Ingersoll <gs...@apache.org>]

I took this down, looks like I need to generate another for Koji's  
changes.

On Sep 7, 2008, at 10:28 PM, Grant Ingersoll wrote:

>
> On Sep 7, 2008, at 8:59 PM, Chris Hostetter wrote:
>
>>
>> : Proposed 1.3.0 candidate available at
>> : http://people.apache.org/~gsingers/solr/1.3.0/
>>
>> Minor thing: the manifest indicates that this is a mixed revision  
>> build
>> (692862:692877). i'm guessing grant made some local changes,  
>> commited, and
>> then didn't "svn update" at the root level before doing the build.
>>
>> The diffs between those revisions are fairly insignificant, so i  
>> doubt it
>> matters, but it does mean that the build isn't completley  
>> reproducable.
>> ... and it also means we can't apply a tag and be 100% certain we are
>> tagging what was actually built.
>
> OK, good point.  I thought I had done svn up first, but...  I will  
> update the wiki page and publish another.
>
>>
>>
>> (Curiousity: Grant, is the hudson config for doing releases not  
>> working
>> out?  it seemed like a really sweet way to automate everything and  
>> i'm
>> just wondering if you found something that made it not  
>> worthwhile ... it
>> seems to be gone now)
>
> It would be really cool, but I haven't spent the time to figure out  
> how to securely sign the artifacts, so that is the only hold up.

Re: 1.3.0 Candidate

[Grant Ingersoll <gs...@apache.org>]

On Sep 7, 2008, at 8:59 PM, Chris Hostetter wrote:

>
> : Proposed 1.3.0 candidate available at
> : http://people.apache.org/~gsingers/solr/1.3.0/
>
> Minor thing: the manifest indicates that this is a mixed revision  
> build
> (692862:692877). i'm guessing grant made some local changes,  
> commited, and
> then didn't "svn update" at the root level before doing the build.
>
> The diffs between those revisions are fairly insignificant, so i  
> doubt it
> matters, but it does mean that the build isn't completley  
> reproducable.
> ... and it also means we can't apply a tag and be 100% certain we are
> tagging what was actually built.

OK, good point.  I thought I had done svn up first, but...  I will  
update the wiki page and publish another.

>
>
> (Curiousity: Grant, is the hudson config for doing releases not  
> working
> out?  it seemed like a really sweet way to automate everything and i'm
> just wondering if you found something that made it not  
> worthwhile ... it
> seems to be gone now)

It would be really cool, but I haven't spent the time to figure out  
how to securely sign the artifacts, so that is the only hold up.

Re: 1.3.0 Candidate

[Chris Hostetter <ho...@fucit.org>]

: Proposed 1.3.0 candidate available at
: http://people.apache.org/~gsingers/solr/1.3.0/

Minor thing: the manifest indicates that this is a mixed revision build 
(692862:692877). i'm guessing grant made some local changes, commited, and 
then didn't "svn update" at the root level before doing the build.

The diffs between those revisions are fairly insignificant, so i doubt it 
matters, but it does mean that the build isn't completley reproducable. 
... and it also means we can't apply a tag and be 100% certain we are 
tagging what was actually built.

(Curiousity: Grant, is the hudson config for doing releases not working 
out?  it seemed like a really sweet way to automate everything and i'm 
just wondering if you found something that made it not worthwhile ... it 
seems to be gone now)



-Hoss