Re: [sca-j] NEW ISSUE: @Callback injection could be NULL

[David Booz <booz@us.ibm.com>]

So let's focus on the re-injection case for setter methods. If the Domain is changed such that a wire disappears (and the runtime supports re-injection), then wouldn't you expect a NULL to be re-injected? The answer has to be yes, so why then wouldn't you do the initial injection with NULL. I don't see why we should be inconsistent in this case either.

Dave Booz
STSM, BPM and SCA Architecture
Co-Chair OASIS SCA-Policy TC and SCA-J TC
"Distributed objects first, then world hunger"
Poughkeepsie, NY (845)-435-6093 or 8-295-6093
e-mail:booz@us.ibm.com

Jim Marino ---02/16/2009 11:37:42 AM---@Init and @Destroy are very important in both cases as they allow for initialization code to be ex

From:	Jim Marino <jim.marino@gmail.com>
To:	David Booz/Poughkeepsie/IBM@IBMUS
Cc:	sca-j@lists.oasis-open.org
Date:	02/16/2009 11:37 AM
Subject:	Re: [sca-j] NEW ISSUE: @Callback injection could be NULL

---

@Init and @Destroy are very important in both cases as they allow for initialization code to be executed before a service operation is executed and a component to receive a container callback to perform cleanup respectively. 

If @Init were removed, it would force implementation code to handles these cases by inelegantly inserting an initialization method call in the code block of every service operation. Note the specs should not recommend constructors as a way to perform general initialization as the latter may be required to invoke another object instance.

@Destroy is important for stateless components for the same reason as @Init: it avoids requiring implementations to insert a method call at the end of every service operation code block. In composite scope, @Destroy is important as there is no other way to receive a container callback when an instance is shut down.

On the issue of setter injection, it's useful for the following reasons:

1. It allows for reinjection wheres CDI does not
2. It makes unit testing easier with respect to when protected fields are used for injection as testing code does not need to use reflection or a subclass to set mock component dependencies. 
3. It facilitates porting component implementations to other POJO frameworks 
4. Some ideologues prefer setter injection over protected fields as it preserves encapsulation. They would argue it is not really much typing as most IDEs can auto-generate setters from a field through one click (I'm not one of the ideologues but IntelliJ, the IDE I use, will do this).

Jim      


On Feb 16, 2009, at 4:32 AM, David Booz wrote:

I have been wondering why we need @init and @destroy now that we only have composite and stateless scoped components.

That aside, what's the purpose of setter injection if not to allow for some init time code? It's extra work to the developer for no perceived value if we take your view.

Dave Booz
STSM, BPM and SCA Architecture
Co-Chair OASIS SCA-Policy TC and SCA-J TC
"Distributed objects first, then world hunger"
Poughkeepsie, NY (845)-435-6093 or 8-295-6093
e-mail:booz@us.ibm.com

<graycol.gif>Jim Marino ---02/13/2009 06:46:47 PM---On Feb 13, 2009, at 3:34 PM, Simon Nash wrote: > David Booz wrote:


<ecblank.gif> From:	<ecblank.gif> Jim Marino <jim.marino@gmail.com>
<ecblank.gif> To:	<ecblank.gif> Simon Nash <oasis@cjnash.com>
<ecblank.gif> Cc:	<ecblank.gif> sca-j@lists.oasis-open.org
<ecblank.gif> Date:	<ecblank.gif> 02/13/2009 06:46 PM
<ecblank.gif> Subject:	<ecblank.gif> Re: [sca-j] NEW ISSUE: @Callback injection could be NULL

---





On Feb 13, 2009, at 3:34 PM, Simon Nash wrote:

> David Booz wrote:
>> I'm going to try to state this again without using the word  
>> injection because it seems to have too many connotations.....and  
>> the email thread is wandering all over the place.
>> The point I'm worried about is a setter method that is annotated  
>> with @callback (or even a 0..1 @reference as pointed out earlier).
>> I'm concerned that we provide a consistent model to component  
>> developers, and therefore the runtime MUST NOT skip invocation of  
>> the setter method. It may be very important for the logic in the  
>> setter method to understand that there is no reference target or  
>> callback. It's the 'skipping of the setter' that is the problem.
> I don't think we need to drive all these setters with null arguments
> just in case they have some important initialization to do.  My reason
> for saying this is that we also have the @Init method, which is called
> after all the setters have (or have not) run.
>
> So the consistent model is that all setters with something to set
> would run, and then the @Init method would run.  This combination
> allows the implementation code to handle all possible cases.
> For example, the setters could each set a flag to say "OK, done my
> thing", and the @Init method could check these flags to find out
> if further processing is needed because some setter did not run.
>
FWIW, I prefer the approach of using the @Init method. A strong  
argument can be made that setters should not be performing  
initialization logic - a setter should just set a value and @Init  
should be used for initialization Sticking to this rule has the nice  
effect of keeping setters simple and clean (i.e. no need for null  
checks) and consolidating initialization in one place where it  
arguably should be. Also, from an implementation perspective, less  
invocations to set items increases performance. Of course the  
exception is CDI in the case where a reference is optional, but that  
is probably a very specific use-case.


> I was looking to see what Spring does.  AFAICT it does not drive
> dependency injection setters when the dependency is missing (though
> it does pass NULL for missing dependencies when calling constructors).
> See this article:
> http://blog.springsource.com/2007/07/11/setter-injection-versus-constructor-injection-and-the-use-of-required/
> The article also mentions the @Init-style approach that I described
> above as one way to check for missing dependencies.
>
>  Simon
>
>> The spec text you point out does not clearly say what to do. The  
>> text is clearly written assuming everything is wired up (I'm guilty  
>> of thinking that way myself).
>> Dave Booz
>> STSM, BPM and SCA Architecture
>> Co-Chair OASIS SCA-Policy TC and SCA-J TC
>> "Distributed objects first, then world hunger"
>> Poughkeepsie, NY (845)-435-6093 or 8-295-6093
>> e-mail:booz@us.ibm.com
>> Inactive hide details for Simon Nash ---02/13/2009 09:41:55 AM---
>> David Booz wrote: > I'm glad we're doing this in email, and noSimon  
>> Nash ---02/13/2009 09:41:55 AM---David Booz wrote: > I'm glad we're  
>> doing this in email, and not on the call. We need to do
>> From:
>> Simon Nash <oasis@cjnash.com>
>> To:
>> sca-j@lists.oasis-open.org
>> Date:
>> 02/13/2009 09:41 AM
>> Subject:
>> Re: [sca-j] NEW ISSUE: @Callback injection could be NULL
>> ------------------------------------------------------------------------
>> David Booz wrote:
>> > I'm glad we're doing this in email, and not on the call. We need  
>> to do
>> > more of this.
>> >
>> > We need to avoid uninitialized fields and so am asserting that NULL
>> > would in fact be injected, ensuring that a test for NULL in the  
>> app was
>> > always going to give the correct answer. This makes the two cases
>> > parallel (which they should be). It doesn't do the app any good  
>> to leave
>> > uninitialized fields laying around.
>> >
>> > Good catch on the "should".
>> >
>> > I knew about the 76 overlap in the example and chose to ignore it  
>> given
>> > that it hasn't been accepted/applied yet. In any case, I am fine  
>> if we
>> > omit the example updates from the proposal.
>> >
>> > This then changes the proposal to:
>> >
>> > In section 6.2.4, at the end of the first paragraph (line 395  
>> [1] ) add
>> > the following:
>> >
>> > A field or setter method annotated with @callback is injected or  
>> set to
>> > NULL if the implementation is invoked through a non-bidirectional
>> > service or when invoked as a callback. It is RECOMMENDED that the
>> > component implementation checks for null before using a callback
>> > reference in this situation (See also the getCallbackReference()  
>> method
>> > in section 8.2 "RequestContext").
>> >
>> Apart from the question of whether or not NULL gets injected, I don't
>> think we should make the component implementation logic into an  
>> RFC2119
>> conformance target.  I'd like to proposed alternative wording that
>> addresses these two points:
>> A field or setter method annotated with @Callback will not be  
>> injected
>> if the implementation is invoked through a non-bidirectional  
>> service or
>> when invoked as a callback. To allow for this, the component  
>> implementation
>> can check for null before using a callback reference in this  
>> situation
>> (See also the getCallbackReference() method in section 8.2  
>> "RequestContext").
>>  Simon
>> >
>> > Dave Booz
>> > STSM, BPM and SCA Architecture
>> > Co-Chair OASIS SCA-Policy TC and SCA-J TC
>> > "Distributed objects first, then world hunger"
>> > Poughkeepsie, NY (845)-435-6093 or 8-295-6093
>> > e-mail:booz@us.ibm.com
>> >
>> > Inactive hide details for Simon Nash ---02/10/2009 11:18:56 AM---
>> David
>> > Booz wrote: > TARGET: Java CAA, section 6.2.4 "AccessingSimon Nash
>> > ---02/10/2009 11:18:56 AM---David Booz wrote: > TARGET: Java CAA,
>> > section 6.2.4 "Accessing Callbacks" [1]
>> >
>> >
>> > From:
>> > Simon Nash <oasis@cjnash.com>
>> >
>> > To:
>> > sca-j@lists.oasis-open.org
>> >
>> > Date:
>> > 02/10/2009 11:18 AM
>> >
>> > Subject:
>> > Re: [sca-j] NEW ISSUE: @Callback injection could be NULL
>> >
>> >  
>> ------------------------------------------------------------------------
>> >
>> >
>> >
>> > David Booz wrote:
>> >  > TARGET: Java CAA, section 6.2.4 "Accessing Callbacks" [1]
>> >  >
>> >  > DESCRIPTION: There are situations where the field or setter  
>> with an
>> >  > @callback annotation (when used to obtain a reference to a  
>> callaback
>> >  > instance) can be injected with NULL. As noted in section 8.2
>> >  > RequestContext, the getCallback() method might return NULL in  
>> certain
>> >  > situations, and those situations might also occur in  
>> combination with
>> >  > the use of @callback. For example, suppose a component  
>> implements two
>> >  > services, one has a bidirectional interface and the other does  
>> not. The
>> >  > component impl could use @callback to get a callback instance.  
>> When
>> >  > invoked over the bidirectional service, the @callback will non-
>> null, but
>> >  > will be null when the non-bidirectional service is invoked.  
>> AFAIK, the
>> >  > specs don't prohibit the example (nor do I think they should).
>> >  >
>> > This case isn't quite the same as RequestContext.getCallback().
>> > In that case, the API can return NULL.  In this case, the callback
>> > would not be injected, and so the field would be uninitialized.
>> >
>> >  > PROPOSAL:
>> >  > There is no pdf for [1] so bear with me:
>> >  > In section 6.2.4, at the end of the first paragraph (line 395)  
>> add the
>> >  > following:
>> >  >
>> >  > A field or setter method annotated with @callback might be  
>> injected or
>> >  > set to NULL if the implementation is invoked through a non-
>> bidirectional
>> >  > service or when invoked as a callback.
>> >  >
>> > As I said above, it's not correct to say "...might be injected or
>> > set to NULL...".  It's really "...might not be injected...".
>> >
>> >  >                                        The component  
>> implementation
>> >  > should always check for null before using a callback reference  
>> (See also
>> >  >
>> > You've got a non-RFC2119 "should" here.  Even if we found another
>> > word, I think it's going too far to say that this should *always*
>> > be done.  IMO it should *only* be done if the component logic
>> > allows this case to arise, which I believe will be a minority of
>> > all the cases where callbacks are used.
>> >
>> >  > the getCallbackReference() method in section 8.2  
>> "RequestContext").
>> >  >
>> >  > In section 6.2.4, the first example (line 400) replace the  
>> example with
>> >  > the following:
>> >  >
>> >  > @Callback
>> >  > protected ServiceReference<MyCallback> callback;
>> >  >
>> >  >       *public void *someMethod() {
>> >  >
>> >  >       *if* (callback) {
>> >  >       MyCallback myCallback = callback.getCallback(); …
>> >  >
>> >  >       myCallback.receiveResult(theResult);
>> >  >       }
>> >  >       }
>> >  >
>> > These changes overlap with the proposed changes for JAVA-76,
>> > and they carry forward one of the problems that is the subject
>> > of JAVA-76 (incorrectly using getCallback() where getService()
>> > should be used).  Also, bearing in mind my comment above, I'm
>> > not convinced that this example needs to show the "if".  I think
>> > it's enough to describe the possibility in added text.
>> >
>> >   Simon
>> >  >
>> >  >
>> >  > [1]
>> >  >
>> > http://www.oasis-open.org/committees/download.php/31121/sca-javacaa-1.1-spec-cd02-rev2.doc
>> >  >
>> >  > Dave Booz
>> >  > STSM, BPM and SCA Architecture
>> >  > Co-Chair OASIS SCA-Policy TC and SCA-J TC
>> >  > "Distributed objects first, then world hunger"
>> >  > Poughkeepsie, NY (845)-435-6093 or 8-295-6093
>> >  > e-mail:booz@us.ibm.com
>> >
>> >
>> >
>> >  
>> ---------------------------------------------------------------------
>> > To unsubscribe from this mail list, you must leave the OASIS TC  
>> that
>> > generates this mail.  Follow this link to all your TCs in OASIS at:
>> > https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php 
>>   >
>> >
>> >
>> ---------------------------------------------------------------------
>> To unsubscribe from this mail list, you must leave the OASIS TC that
>> generates this mail.  Follow this link to all your TCs in OASIS at:
>> https://www.oasis-open.org/apps/org/workgroup/portal/ 
>> my_workgroups.php
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe from this mail list, you must leave the OASIS TC that
> generates this mail.  Follow this link to all your TCs in OASIS at:
> https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php


---------------------------------------------------------------------
To unsubscribe from this mail list, you must leave the OASIS TC that
generates this mail.  Follow this link to all your TCs in OASIS at:
https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php