Re: [Opencxx-users] future directions

[SF M. E. <el...@us...>]

> besides being a 'tree object model' they don't have anything in common.

I mean that this is enough to generalize the node interfaces.


> I don't believe that XML will help us much in this context, it's just
> another syntax to express trees.

I agree.
But I see some benefits by using XSLT with its XPath expressions to analyze and process the parse trees by DOM methods.


> Semantics have to be added on top.

I agree.
The meaning and its validation should be done by the proposed API enhancements for the type management.


> It has its value, but not here. (you may be interested into this on-spot
> and entertaining article: http://www.venge.net/graydon/markup-abuse.html)

Nice.   ;-)
I hope that a door can be left open for use cases that do not use only binary processing for C++ source files.


> In what way do you think 'internet browsers' would help us ?

Such software contains function or class libraries for the work with DOM and XSLT that can be adapted to OpenC++.
I suggest to enable "browsing" of Ptrees by those well defined APIs.
http://www.w3.org/TR/DOM-Level-3-Core/introduction.html
http://www.w3.org/TR/DOM-Level-3-Core/core.html#ID-1590626202