ant-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Peter Donald <>
Subject Re: A DAG for all occasions
Date Sun, 07 Jan 2001 07:32:58 GMT
At 12:56  7/1/01 -0500, Jason Rosenberg wrote:
>Now you are beginning to see.  Templates do require their
>own isolated context, which is local.  This is because they
>are sub-routines, which have their own local values on
>the stack (speaking in procedural terms).


>You are simply running into the problem of trying to
>do templates from a pure declarative model.  The dependency
>graph idea just isn't really buying much, when we talk
>about parameterized, reusable code.

a lot of people differ with that opinion, among them some of the XML mob,
DB mob etc.

>The <template> tag really complicates the code unnecessarily,
>since now we have to be thinking about having completely
>separately declared code.  Code which can be part of a
>template, or code which cannot.

you have to think about it any-way. This just makes it explicit.

>In the end, things will likely evolve such that 95% of the script
>is being executed by templates, and so the business of building
>a DAG representation for the non-template part is superfluous.

Possibly for some cases thou not all. So what was your point ?



| "Faced with the choice between changing one's mind, |
| and proving that there is no need to do so - almost |
| everyone gets busy on the proof."                   |
|              - John Kenneth Galbraith               |

View raw message