In the docs directory of the binary distribution, the sca-api and
tuscany-sca-spi subdirectories both have identical contents (i.e.,
javadocs for org.osoa.* as well as javadocs for org.apache.tuscany.*).
Presumably the intention was to separate these two sets of javadocs
into different top-level directories. However, I think it is just
as easy to combine them, i.e., to have
docs
org
osoa
...
apache
tuscany
...
rather than
docs
sca-api
org
osoa
...
tuscany-sca-spi
org
apache
tuscany
...
Also, the contents of the org.apache.tuscany.* section are incorrect
as they don't reflect the new style(s) for naming SPIs.
Simon
ant elder wrote:
Ok things are looking pretty good now. I've uploaded the latest binary and
src distributions of the current trunk code, please try them out and report
anything that looks odd: http://people.apache.org/~antelder/tuscany/latest/
Please try them out and report anything that looks odd.
There's still things to clean up and text required for readme's and
releases
notes etc, lets move to the more controlled commit mode now until the
0.90branch is cut. Say before you make any big changes, and when
committing
anything make sure it all builds including all the samples and itests, and
for non-trivial changes also build the distribution (mvn clean install
-o in
the sca/distribution directory) and check things still work.
...ant
On 5/10/07, Jean-Sebastien Delfino <[EMAIL PROTECTED]> wrote:
ant elder wrote:
> On 5/10/07, Simon Laws <[EMAIL PROTECTED]> wrote:
>>
>> What's the process for controlled commits. Patches?
>>
>> The release process takes a resonable amount of time when you include
>> the
>> time to vote on it and the possibiliy that faults will be found and we
>> have
>> to fix and vote again. So when do we think we will be in the position
to
>> roll a release candidate? Is it when the renaming is done? Are we
>> happy to
>> live with a locked down head until the release is out?
>>
>> If not I think we should go with a branch.
>>
>> Simon
>>
>
> One thing that puts me off cutting a branch right now is that everyone
> can
> then go off starting on new fun stuff on the trunk and there's less
> motivation to help polish things off to get the release out :) So while
> there's not so many active committers pushing to start working on new
> stuff
> I was hoping to avoid a branch just yet. Also I'd like to put off
> creating a
> branch until the package rename is done so things stay really close
> together.
>
> When I said a "more controlled commit mode" I didn't mean a code
freeze,
> more like a code chill - not to make non-trivial changes without
> mentioning
> them on the ML first and to always make sure a full build with all
itests
> and samples works before committing and maybe even build the distro and
> check the samples still work ok.
>
> We're really close now, so how about:
>
> - fix the last few problems now
> - make a sample 0.90 RC distro's in a couple of hours and switch to
code
> chill
> - test out the distros for a day or so and clean up the included doc
> and fix
> whatever problems
> - do the package renames tonight/tomorrow
> - and look at doing a 0.90 branch/tag over the weekend and vote on
> that as
> 0.90 final
>
> How does this sound?
>
> ...ant
>
+1, Sounds good to me!
--
Jean-Sebastien
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
--
Simon C Nash IBM Distinguished Engineer
Hursley Park, Winchester, UK [EMAIL PROTECTED]
Tel. +44-1962-815156 Fax +44-1962-818999
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
