ant-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From J Pai <jai.forums2...@gmail.com>
Subject Re: IVY-1475 - cachefileset task and its inherent limitation
Date Fri, 02 Jun 2017 10:05:41 GMT
So that does mean cachefileset has a role to play in at least some cases. In that case, I think
the approach we could take is to _not_ deprecate it and instead include this limitation as
part of the documentation *and* enhance the code of this task to explicit fail with a proper
error when it can’t determine a common base directory.

-Jaikiran
On 02-Jun-2017, at 2:55 PM, Nicolas Lalevée <nicolas.lalevee@hibnet.org> wrote:


> Le 2 juin 2017 à 08:22, Jan Matèrne (jhm) <apache@materne.de> a écrit :
> 
> From a naive user point of view, it doesnt matter (to me) if I use
> ivy:chachefileset or ivy:resources.
> I want to specify the dependency and have a 'thing' which contains all required jars,
so I could
> use external tasks/antlibs.
> 
> Ant itself moved from fileset to resource collection some years ago and Ivy could follow.
> But I am not sure that we could use RCs _everywhere_.
> In the few exceptions you have to use ivy:cachefileset - maybe multiple …

One limitation of resource collections is that they doesn’t have necessarily a basedir,
contrary to a fileset :)
For instance a basedir it is quite useful for the copy task, so a set of files be be copied
with their relative paths to the basedir, rather than with their absolute paths.

Nicolas

> Jan
> 
>> -----Ursprüngliche Nachricht-----
>> Von: J Pai [mailto:jai.forums2013@gmail.com]
>> Gesendet: Freitag, 2. Juni 2017 05:29
>> An: Ant Developers List
>> Betreff: IVY-1475 - cachefileset task and its inherent limitation
>> 
>> One of the Ivy users has pointed out to an issue in cachefileset
>> task[1] of Ivy here https://issues.apache.org/jira/browse/IVY-1475.
>> 
>> To summarize, the cachefileset task is expected to create a Ant Fileset
>> of the resolved artifacts in the cache(s). Ant Fileset requires a
>> (single) basedir to work on and the Ivy cachefileset has a piece of
>> logic which tries to determine a common base directory for the resolved
>> artifacts. It’s very much possible that there won’t be a common base
>> directory for artifacts if the caches have been configured to be on
>> multiple different filesystem roots, as noted in that JIRA. So
>> essentially, to me, it looks like this cachefileset has an inherent
>> deficiency which can’t really be fixed.
>> 
>> The user in that JIRA notes that we can deprecate this task (and also
>> add a note about this limitation) in favour of “resources” task [2]
>> which provides a similar functionality but is much more flexible and
>> doesn’t suffer this limitation.
>> 
>> Any thoughts on how we should go about this JIRA and the cachefileset
>> task?
>> 
>> [1] https://ant.apache.org/ivy/history/latest-
>> milestone/use/cachefileset.html
>> [2] https://ant.apache.org/ivy/history/latest-
>> milestone/use/resources.html
>> 
>> -Jaikiran
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: dev-unsubscribe@ant.apache.org For additional
>> commands, e-mail: dev-help@ant.apache.org
> 
> 
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@ant.apache.org
> For additional commands, e-mail: dev-help@ant.apache.org
> 


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@ant.apache.org
For additional commands, e-mail: dev-help@ant.apache.org



---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@ant.apache.org
For additional commands, e-mail: dev-help@ant.apache.org


Mime
View raw message