You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@beehive.apache.org by "Kenneth Tam (JIRA)" <be...@incubator.apache.org> on 2005/03/01 20:59:49 UTC
[jira] Created: (BEEHIVE-386) TwoPhaseAnnotationProcessor processing algorithm is suboptimal
TwoPhaseAnnotationProcessor processing algorithm is suboptimal
--------------------------------------------------------------
Key: BEEHIVE-386
URL: http://issues.apache.org/jira/browse/BEEHIVE-386
Project: Beehive
Type: Task
Components: Controls
Versions: V1Alpha
Reporter: Kenneth Tam
Fix For: V1
TPAP currently uses the following loop:
for ( all annotations that a processor is registered to handle )
get collection of decls annotated with a particular annotation
check that decl
This means that a decl could be checked multiple times, if it has multiple interesting annotations on it. Furthermore, each check pass currently has no context being passed wrt what annotation the TPAP considers interesting (the "check that decl" call above doesn't pass the annotation in question). We should add that to the signature.
This labelled as task-level work because it could cause some significant ripple effects on downstream processors.
A workaround for this that we've been using has been to have annotation processors register for a minimal set of annotations, and internally do explicit checks for other annotations. This makes it difficult to be comprehensive about robustly covering error cases for annotation usage (since situations where we didn't explicitly expect to see an annotation will probably just be ignored), and it also means apt reports a ton of annotations as not having any processors registered for it. See http://issues.apache.org/jira/browse/BEEHIVE-385.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
http://issues.apache.org/jira/secure/Administrators.jspa
-
If you want more information on JIRA, or have a bug to report see:
http://www.atlassian.com/software/jira
[jira] Updated: (BEEHIVE-386) TwoPhaseAnnotationProcessor processing algorithm is suboptimal
Posted by "Kenneth Tam (JIRA)" <be...@incubator.apache.org>.
[ http://issues.apache.org/jira/browse/BEEHIVE-386?page=history ]
Kenneth Tam updated BEEHIVE-386:
--------------------------------
Assign To: Kenneth Tam
Version: V1Alpha
Component: Controls
Fix Version: V1
> TwoPhaseAnnotationProcessor processing algorithm is suboptimal
> --------------------------------------------------------------
>
> Key: BEEHIVE-386
> URL: http://issues.apache.org/jira/browse/BEEHIVE-386
> Project: Beehive
> Type: Task
> Components: Controls
> Versions: V1Alpha
> Reporter: Kenneth Tam
> Assignee: Kenneth Tam
> Fix For: V1
>
> TPAP currently uses the following loop:
> for ( all annotations that a processor is registered to handle )
> get collection of decls annotated with a particular annotation
> check that decl
> This means that a decl could be checked multiple times, if it has multiple interesting annotations on it. Furthermore, each check pass currently has no context being passed wrt what annotation the TPAP considers interesting (the "check that decl" call above doesn't pass the annotation in question). We should add that to the signature.
> This labelled as task-level work because it could cause some significant ripple effects on downstream processors.
> A workaround for this that we've been using has been to have annotation processors register for a minimal set of annotations, and internally do explicit checks for other annotations. This makes it difficult to be comprehensive about robustly covering error cases for annotation usage (since situations where we didn't explicitly expect to see an annotation will probably just be ignored), and it also means apt reports a ton of annotations as not having any processors registered for it. See http://issues.apache.org/jira/browse/BEEHIVE-385.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
http://issues.apache.org/jira/secure/Administrators.jspa
-
If you want more information on JIRA, or have a bug to report see:
http://www.atlassian.com/software/jira
[jira] Updated: (BEEHIVE-386) TwoPhaseAnnotationProcessor processing algorithm is suboptimal
Posted by "Kenneth Tam (JIRA)" <be...@incubator.apache.org>.
[ http://issues.apache.org/jira/browse/BEEHIVE-386?page=all ]
Kenneth Tam updated BEEHIVE-386:
--------------------------------
Fix Version: TBD
(was: V1)
I don't see a compelling reason to embark on this rearchitecture of the AP framework at this time when stability is a high priority.
> TwoPhaseAnnotationProcessor processing algorithm is suboptimal
> --------------------------------------------------------------
>
> Key: BEEHIVE-386
> URL: http://issues.apache.org/jira/browse/BEEHIVE-386
> Project: Beehive
> Type: Task
> Components: Controls
> Versions: V1Alpha
> Reporter: Kenneth Tam
> Assignee: Kenneth Tam
> Fix For: TBD
>
> TPAP currently uses the following loop:
> for ( all annotations that a processor is registered to handle )
> get collection of decls annotated with a particular annotation
> check that decl
> This means that a decl could be checked multiple times, if it has multiple interesting annotations on it. Furthermore, each check pass currently has no context being passed wrt what annotation the TPAP considers interesting (the "check that decl" call above doesn't pass the annotation in question). We should add that to the signature.
> This labelled as task-level work because it could cause some significant ripple effects on downstream processors.
> A workaround for this that we've been using has been to have annotation processors register for a minimal set of annotations, and internally do explicit checks for other annotations. This makes it difficult to be comprehensive about robustly covering error cases for annotation usage (since situations where we didn't explicitly expect to see an annotation will probably just be ignored), and it also means apt reports a ton of annotations as not having any processors registered for it. See http://issues.apache.org/jira/browse/BEEHIVE-385.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
http://issues.apache.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see:
http://www.atlassian.com/software/jira