Sender: |
|
Subject: |
|
From: |
|
Date: |
Tue, 25 Jan 2000 21:27:01 +0100 |
In-Reply-To: |
<v03110700b4b38b004405@[195.100.226.140]> |
Reply-To: |
|
Parts/Attachments: |
|
|
Hans,
> -- Perhaps this input can be used for the template discussion:
>
> One idea that comes to my mind in connection with the "environments with
> hooks" is to build up two layers, one which takes care of the semantics,
> and one which takes of the syntax by which the environments are called.
not really since I was talking only about the semantic layer; what you propose
is (under a different label) the separation as it is provided by the interface
of xparse on one hand (syntax side) and the template on the other hand
(semantics)
but what i was getting at is that the semantics layer need further
clarification (and they would in your code as well the question isn't
different there)
> Here, some arguments should (when required by the environment definition)
> be passed to the commands \<name>/before, \<name>/begin, \<name>/end,
i was trying to pin down the nature of those "some arguments" this was the
question posed in the thread not whether or not separation of syntax and
semantics is helpful/sensible.
frank
|
|
|