[Bug 68596] Remaining overhead in javax.el.CompositeELResolver.convertToType

2024-02-12 Thread bugzilla
https://bz.apache.org/bugzilla/show_bug.cgi?id=68596 --- Comment #9 from John Engebretson --- Responding to these sections of comment #6: > The setPropertyResolved(boolean) method is not overwritten in subclasses. > This is a common situation that occurs everywhere, thus I wonder why this > sp

[Bug 68596] Remaining overhead in javax.el.CompositeELResolver.convertToType

2024-02-11 Thread bugzilla
https://bz.apache.org/bugzilla/show_bug.cgi?id=68596 --- Comment #8 from Mark Thomas --- I've been looking at this some more. The relevant methods on ELResolver are: getValue() getType() setValue() isReadOnly() convertToType() invoke() At the start of each of those methods, ELContext.isProperty

[Bug 68596] Remaining overhead in javax.el.CompositeELResolver.convertToType

2024-02-10 Thread bugzilla
https://bz.apache.org/bugzilla/show_bug.cgi?id=68596 --- Comment #7 from Mark Thomas --- Thanks for the additional review Konstantin. ELResolver.convertToType() is public API. It is defined by the specification and we can't change that. Your comment about setting setPropertyResolved(false) bein

[Bug 68596] Remaining overhead in javax.el.CompositeELResolver.convertToType

2024-02-10 Thread bugzilla
https://bz.apache.org/bugzilla/show_bug.cgi?id=68596 --- Comment #6 from Konstantin Kolinko --- (In reply to John Engebretson from comment #0) > context.setPropertyResolved(false); > ... > > context can be of type ELContextImpl, ELContextWrapper, EvaluationContext, > or StandardELContext

[Bug 68596] Remaining overhead in javax.el.CompositeELResolver.convertToType

2024-02-09 Thread bugzilla
https://bz.apache.org/bugzilla/show_bug.cgi?id=68596 --- Comment #5 from John Engebretson --- Yep. Thanks! -- You are receiving this mail because: You are the assignee for the bug. - To unsubscribe, e-mail: dev-unsubscr...@tom

[Bug 68596] Remaining overhead in javax.el.CompositeELResolver.convertToType

2024-02-09 Thread bugzilla
https://bz.apache.org/bugzilla/show_bug.cgi?id=68596 Mark Thomas changed: What|Removed |Added Severity|normal |enhancement --- Comment #4 from Mark Tho

[Bug 68596] Remaining overhead in javax.el.CompositeELResolver.convertToType

2024-02-07 Thread bugzilla
https://bz.apache.org/bugzilla/show_bug.cgi?id=68596 --- Comment #3 from John Engebretson --- > This needs some more thought... Thanks, I'll keep chewing on it too. > Note: This may get moved to an enhancement if there isn't an obvious way to > improve this) Makes sense, thanks. -- You are

[Bug 68596] Remaining overhead in javax.el.CompositeELResolver.convertToType

2024-02-07 Thread bugzilla
https://bz.apache.org/bugzilla/show_bug.cgi?id=68596 --- Comment #2 from Mark Thomas --- #1 isn't an option unfortunately. With more complex EL expressions ELContext.isPropertyResolved() will return true at the start of the call to convertToType(). At least one test fails if this code is removed.

[Bug 68596] Remaining overhead in javax.el.CompositeELResolver.convertToType

2024-02-07 Thread bugzilla
https://bz.apache.org/bugzilla/show_bug.cgi?id=68596 --- Comment #1 from John Engebretson --- Created attachment 39573 --> https://bz.apache.org/bugzilla/attachment.cgi?id=39573&action=edit Support class for the speed test -- You are receiving this mail because: You are the assignee for the b