Hi,
forget about this patch. I prepare a different one.
Robert Schuster wrote:
> Hi,
> I am currently working to get an application running which depends on
> implementation details of the RI.
>
> This changes Container.removeAll() to make it work in a special case
> (see added note in patch).
>
> ChangeLog:
>
> 2006-05-30 Robert Schuster <[EMAIL PROTECTED]>
>
> * java/awt/Container.java:
> (removeAll): Implemented different removal mechanism, added note.
>
> cya
> Robert
>
>
> ------------------------------------------------------------------------
>
> Index: java/awt/Container.java
> ===================================================================
> RCS file: /cvsroot/classpath/classpath/java/awt/Container.java,v
> retrieving revision 1.91
> diff -u -r1.91 Container.java
> --- java/awt/Container.java 5 Apr 2006 10:31:26 -0000 1.91
> +++ java/awt/Container.java 30 May 2006 17:40:36 -0000
> @@ -457,8 +457,15 @@
> {
> synchronized (getTreeLock ())
> {
> - while (ncomponents > 0)
> + // In order to allow the same bad tricks to be used as in RI
> + // this code has to stay exactly that way: In a real-life app
> + // a Container subclass implemented its own vector for
> + // subcomponents, supplied additional addXYZ() methods
> + // and overrode remove(int) (but not removeAll).
> + for ( int i=0; i<ncomponents; i++)
> remove(0);
> +
> + ncomponents = 0;
> }
> }
>
signature.asc
Description: OpenPGP digital signature
