[
Date Prev][
Date Next][
Thread Prev][
Thread Next][
Date Index][
Thread Index]
[
List Home]
Re: [aspectj-dev] triage on most recent bug reports - update
|
More options {with optional hints}
No interface constructor-execution join point
- use call(I+.new(..)) for implementing classes
Ambiguous binding of type A
- use one args(..) per matched join point
Wes
Andrew Clement wrote:
I've been working on the three bugs Jim highlighted, I agree they should
go
into 1.2 - the fixes are quite localized.
But I would like some input from people on the new error
messages I am introducing:
https://bugs.eclipse.org/bugs/show_bug.cgi?id=60936
This is a new xlint warning to tell people we no longer make interface
constructor join points available for selection with pointcuts:
no constructor join points for interface type XXX
[Xlint:noInterfaceCtorJoinpoint]
comments? My only concern is that I don't want the user to read the
message
and think there is a way they could mangle their pointcut or their
interface definition to 'get at' a ctor jp - so another option is:
interfaces have no exposed constructor join points. Interface type XXX
https://bugs.eclipse.org/bugs/show_bug.cgi?id=61538
fixed.
https://bugs.eclipse.org/bugs/show_bug.cgi?id=61568
This is a new error message when a user attempts to bind a second variable
to a formal parameter. For example:
before(A a,B b): call(* fred(..)) && (args(a,b) || args(b,a))
Currently we blow up at runtime with a ClassCastException if you try this
kind of thing. Although we could support this in the longer term, for
now we'll output an error message at compile time. Currently I have this:
C:\Eclipse\212\eclipse\aspectj_ws2\tests\bugs\DisjunctVarBinding.java:18
error
Cannot bind more than once to a formal. Failed to bind on type:B
(args(b,a) || args(a,b)) {
^^^^^^^
C:\Eclipse\212\eclipse\aspectj_ws2\tests\bugs\DisjunctVarBinding.java:18
error
Cannot bind more than once to a formal. Failed to bind on type:A
(args(b,a) || args(a,b)) {
^^^^^^^
Including more information than that in the message is hard without it
becoming a much larger fix.
thanks,
Andy.
"Jim Hugunin" <lists@xxxxxxxxxxx>
Sent by: aspectj-dev-admin@xxxxxxxxxxx
10/05/2004 20:27
Please respond to
aspectj-dev
To
<aspectj-dev@xxxxxxxxxxx>
cc
Subject
[aspectj-dev] triage on most recent bug reports
Looking towards 1.2, I've made a quick pass through the large group of
recent bug reports starting with #60936. Several of those bugs could be
resolved as either doc issues or invalid. Several more I believe aren't
important enough to fix until after 1.2. However, there are three
remaining
bugs that I think should be considered for the 1.2 release. See the
individual bug reports for my thoughts and to add any additional
information
or evaluation.
https://bugs.eclipse.org/bugs/show_bug.cgi?id=60936
https://bugs.eclipse.org/bugs/show_bug.cgi?id=61538
https://bugs.eclipse.org/bugs/show_bug.cgi?id=61568
-Jim
_______________________________________________
aspectj-dev mailing list
aspectj-dev@xxxxxxxxxxx
http://dev.eclipse.org/mailman/listinfo/aspectj-dev