ant-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Steve Loughran <>
Subject Re: Ant dependencies task
Date Mon, 15 Aug 2005 09:32:29 GMT
Frank W. Zammetti wrote:
> I do it that way for two reasons mainly:
> (1) Many times during development I need to have certain tasks disabled, 
> and sometimes what is enabled and disabled changes, and I find it easier 
> to comment out an antcall as shown.

or you make every target conditional


and then turn on and off on the command line

> (2) It's a little more explicit in my mind as to what is happening and 
> in what order.  I prefer to not let the runtime work things out for 
> itself as much as possible, I prefer being able to tell exactly what is 
> going to happen, and it feels like I can do that more this way.

declaring dependencies is the core of ant's dependency logic.

> I haven't been using Ant all that long frankly, I was previously a batch 
> file guy :)  So I have not problem believing this isn't the "right" way 
> to do things if that's what anyone wants to tell me... I know it works 
> well for me though :)

well, you can start that way, but before long you will reach the limitations

You cannot pass information from one antcall to another; you take the 
performance hit of every target requiring the entire project being 
cloned and rerun.

Try the other way, try using dependencies. trust the runtime


To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message