[Laszlo-dev] Swf9 and method argument signatures

Henry Minsky hminsky at laszlosystems.com
Wed Sep 10 15:04:47 PDT 2008

As I understand it we  don't have a lot of freedom in as3 as to how the
method arg signature looks.
If there are required args in the superclass method, we can't make any of
the subclass method's args optional.

If the superclass has optional args in its arglist, we have to declare them
as optional in the subclass method. .They can have different default values,
but that is the only difference alllowed.

So I think the two features we could provide  would be

1)  to massage args to be optional with default values, if the base class

<method name="foo" args="a,b=259">

and the user  declares

<method name="foo" args="a,b">

Then the compiler will declare it as "a,b=259"

2) to add extra optional args as needed, so if the base class is

<method name="foo" args="a,b=259">

and the user  declares

<method name="foo" args="a">

Then the compiler will declare it as "a,b=259"

This makes me uneasy, as there will be this arg lurking around that the user
has no
idea exists in their method.

We can get our hands on the info in the tag compiler to do this, but we
don't currently have the information
as to the arg signatures from the javascript LFC libraries.

Don has his automatic schema generator almost working, so that would make
a lot more feasible to generate this info for the base LFC schema file.

On Wed, Sep 10, 2008 at 5:01 PM, Max Carlson <max at openlaszlo.org> wrote:

> Hi,
> I wanted to get a discussion going about resolving the issue with the AS3
> requirement that superclass method argument signatures match the superclass
> implementation.   This can lead to code changes that aren't backward
> compatible with I've written 4.1.x.  I wrote about the issue here:
> http://wiki.openlaszlo.org/Runtime_Differences#Argument_signatures_on_overridden_methods_must_match_the_superclass_definition
> and filed http://jira.openlaszlo.org/jira/browse/LPP-6950 to make the
> error message more helpful.  What do you think is the best way to deal with
> this issue?
> --
> Regards,
> Max Carlson
> OpenLaszlo.org

Henry Minsky
Software Architect
hminsky at laszlosystems.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.openlaszlo.org/pipermail/laszlo-dev/attachments/20080910/443ea944/attachment.html

More information about the Laszlo-dev mailing list