Hi Jochen!

The code is now indeed self-contained. I did not really look at the code but 
like to start with just a few small observations:

1.) the repo contains the whole eclipse project files. I'd rather remove those 
from the repos and add them (+ idea files) to svn:ignore.

2.) javax-inject.api. I'd probably replace this via our own impl:
http://repo1.maven.org/maven2/org/apache/geronimo/specs/geronimo-atinject_1.0_spec/1.0/

3.) the atinject (JSR-330) TCK. I compiled the project and it only said it 
passes 2 tests. I double checked with OpenWebBeans and over there we run (and 
pass of course) 50 tests. Maybe there is something wrong with the integration?

4.) the Scopes.
You currently have a Enum for this. I guess it would be pretty easy to switch 
this to using scope annotations which are meta-annotated with @Scope instead? 
And also implement the @Singleton scope based on that approach instead of 
rolling your own?


Just a few ideas. 


LieGrue,
strub







> On Wednesday, 17 December 2014, 13:56, Jochen Wiedmann 
> <jochen.wiedm...@gmail.com> wrote:
> > Well spotted. I had added Guice as a Maven dependency so as to
> validate certain things while implementing. It's now removed. This
> should eliminate your concerns. Also, please note that the remaining
> dependencies are all provided, with the exception of
> javax.inject-1.jar and javax.inject-tck-1.jar, which are required for
> obvious reasons. (After all, this is the implemented standard.)
> 
> 
> 
> On Wed, Dec 17, 2014 at 11:52 AM, Benedikt Ritter <brit...@apache.org> 
> wrote:
>>  2014-11-19 8:44 GMT+01:00 Mark Struberg <strub...@yahoo.de>:
>>> 
>>>  Jochen, I might have done something wrong so please help me.
>>> 
>>>  I've checked out your svn link and built it.
>>> 
>>>  Then I did a
>>> 
>>>  $> mvn clean -DincludeScope=runtime dependency:copy-dependencies
>>>  -rw-r--r--+ 1 struberg staff    4467 19. Nov 08:41 aopalliance-1.0.jar
>>>  -rw-r--r--+ 1 struberg staff 2228009 19. Nov 08:41 guava-16.0.1.jar
>>>  -rw-r--r--+ 1 struberg staff  642741 19. Nov 08:41 guice-4.0-beta5.jar
>>>  -rw-r--r--+ 1 struberg staff    2497 19. Nov 08:41 javax.inject-1.jar
>>> 
>>> 
>>>  $> du -hs target/dependencies
>>> 
>>>  show me 2.8 MB
>>> 
>>> 
>>>  Plus your own jar which is 76k.
>>>  Is there something wrong? What are you using guice and guava for?
>>>  Also there is an own ASF package for atinject [1].
>>> 
>> 
>>  Jochen, can you please comment?
>> 
>>  Benedikt
>> 
>> 
>>> 
>>>  LieGrue,
>>>  strub
>>> 
>>> 
>>> 
>>>  [1]
>>> 
> http://repo1.maven.org/maven2/org/apache/geronimo/specs/geronimo-atinject_1.0_spec/1.0/
>>> 
>>> 
>>> 
>>> 
>>>  On Wednesday, 19 November 2014, 8:34, Mark Struberg 
> <strub...@yahoo.de>
>>>  wrote:
>>>  >> Sorry, did not mean to step on somebody's toes.
>>>  >
>>>  >No worries you didn't. It's most probably our fault as our 
> (OpenWebBeans)
>>>  documentation sucks and we did not properly document all this stuff ;)
>>>  >
>>>  >If one of you guys is at ApacheCon in Budapest right now, then 
> I'd love
>>>  to give you a quick rush through our code an see if there is something 
> you
>>>  could use and also if we could improve something in OWB.
>>>  >
>>>  >LieGrue,
>>>  >strub
>>>  >
>>>  >
>>>  >
>>>  >
>>>  >----- Original Message -----
>>>  >> From: Oliver Heger <oliver.he...@oliver-heger.de>
>>>  >> To: Commons Developers List <dev@commons.apache.org>
>>>  >> Cc:
>>>  >> Sent: Sunday, 9 November 2014, 11:45
>>>  >> Subject: Re: Announce: Commons Inject
>>>  >>
>>>  >>
>>>  >>
>>>  >> On 08.11.2014 21:51, Romain Manni-Bucau wrote:
>>>  >>>  Le 8 nov. 2014 19:51, "Oliver Heger"
>>>  >> <oliver.he...@oliver-heger.de> a écrit
>>>  >>>  :
>>>  >>>>
>>>  >>>>  Hi Jochen,
>>>  >>>>
>>>  >>>>  do you intend to position this framework for specific 
> use cases? In
>>>  >>>>  which way is it different or special from other 
> implementations?
>>>  >>>>
>>>  >>>>  Just as one example: In the company I am working for, 
> we are using
>>>  CDI
>>>  >>>>  in a pretty large JSE application. Due to the huge 
> class path the
>>>  setup
>>>  >>>>  of the CDI container takes a long time and consumes a 
> lot of memory
>>>  >>>>  (tested with both Weld SE and OpenWebBeans). So a 
> fast and
>>>  lightweight
>>>  >>>>  implementation for this special purpose would be 
> interesting.
>>>  >>>>
>>>  >>>
>>>  >>>  Surely a bad example since you can achieve it with owb, 
> just
>>>  configure the
>>>  >>>  scanner service
>>>  >>
>>>  >> Sorry, did not mean to step on somebody's toes.
>>>  >>
>>>  >> Oliver
>>>  >>
>>>  >>>
>>>  >>>>  Oliver
>>>  >>>>
>>>  >>>>  Am 04.11.2014 um 15:55 schrieb Jochen Wiedmann:
>>>  >>>>>  Hi,
>>>  >>>>>
>>>  >>>>>  As some of you (hopefulyl not all) may have 
> noticed, have added a
>>>  >>>>>  project called Commons Inject to the Sandbox [1] 
> today. Commons
>>>  >> Inject
>>>  >>>>>  is a JSR 330 compliant dependency injection 
> framework. It is
>>>  >> something
>>>  >>>>>  I had in the works for quite some time, but now 
> it has reached a
>>>  >>>>>  decent state with my preliminary milestones 
> reached:
>>>  >>>>>
>>>  >>>>>  - Passes the JSR 330 TCK.
>>>  >>>>>  - Integrated lifecycle handling via 
> @PostConstruct and @PreDestroy.
>>>  >>>>>  Works with all scopes,
>>>  >>>>>     including lazy singletons. (I never got this 
> to work with Guice,
>>>  >>>>>  which has been my major
>>>  >>>>>     driver for doing this.)
>>>  >>>>>  - Integrated logger injection framework for 
> Log4J, SLF4J, and
>>>  >> Commons
>>>  >>>>>  Logging. Others can
>>>  >>>>>     easily be added.
>>>  >>>>>
>>>  >>>>>  I intend to use this for serious works from now 
> on and consequently
>>>  >>>>>  hope to create a release real soon, at which 
> point I'll ask to
>>>  >> move
>>>  >>>>>  this to proper.
>>>  >>>>>
>>>  >>>>>  Any feedback welcome.
>>>  >>>>>
>>>  >>>>>  Jochen
>>>  >>>>>
>>>  >>>>>  [1]
>>>  >> 
> https://svn.apache.org/viewvc/commons/sandbox/commons-inject/trunk/
>>>  >>>>>
>>>  >>>>>
>>>  >>>>
>>>  >>>>
>>>  >>>>  
> ---------------------------------------------------------------------
>>>  >>>>  To unsubscribe, e-mail: 
> dev-unsubscr...@commons.apache.org
>>>  >>>>  For additional commands, e-mail: 
> dev-h...@commons.apache.org
>>>  >
>>>  >>
>>>  >>>>
>>>  >>>
>>>  >>
>>>  >> 
> ---------------------------------------------------------------------
>>>  >> To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org
>>>  >> For additional commands, e-mail: dev-h...@commons.apache.org
>>>  >>
>>>  >
>>> 
>> ---------------------------------------------------------------------
>>>  >To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org
>>>  >For additional commands, e-mail: dev-h...@commons.apache.org
>>>  >
>>>  >
>>>  >
>>> 
>>>  ---------------------------------------------------------------------
>>>  To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org
>>>  For additional commands, e-mail: dev-h...@commons.apache.org
>>> 
>>> 
>> 
>>  --
>>  http://people.apache.org/~britter/
>>  http://www.systemoutprintln.de/
>>  http://twitter.com/BenediktRitter
>>  http://github.com/britter
> 
> 
> 
> -- 
> Our time is just a point along a line that runs forever with no end.
> (Al Stewart, Lord Grenville)
>

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org
For additional commands, e-mail: dev-h...@commons.apache.org

Reply via email to