[Newcompiler] Properties for AST

[Marcus Denker]

On 09.04.2007, at 18:25, Andrew Tween wrote:


>
> I actually think that people will like it, so long as ...
>     1. they get some extra functionality
>     2. they can stiil view and edit their methods as source.
>     3. they can still open the .changes file in a text editor and  
> search for
> stuff.
>
>>
>> Same problem, same solution. Smalltalk opted to model Classes as
>> object, with
>> no source level representation at all. I think we should do the same
>> for methods.
>
> This is almost the case anyway. CompiledMethods contain enough info  
> to browse
> senders etc.
> And you could ditch the source completely and just run with the  
> CompiledMethods.
> The problem comes when a compiledMethod cannot be decompiled into  
> its original
> source.

CompiledMethods suck.

We used CompiledMethods to be the main representation to encode sub  
method structure
for ByteSurgeon: http://www.iam.unibe.ch/~scg/Research/ByteSurgeon/

This was just using Bytecode for structural reflection (and on top of  
it, for a behavioral reflection
framework), not trying to replace text. There are a number of  
problems with this approach, e.g.
there need to be a way to destinguish inlined code from the code the  
programmer wrote,
e.g. to be able to provide a debugger that makes sense.

	Marcus
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 3947 bytes
Desc: not available
Url : http://lists.squeakfoundation.org/pipermail/newcompiler/attachments/20070409/055209b4/smime-0001.bin