Adam Murdoch wrote:
Steve Appling wrote:
I have updated the copy branch.
Completed:
* Unit tests for BreadthFirstDirectoryWalker and CopyVisitor
* Fixed error path for CopyVisitor and CopySpecImpl.filter()
* You can use single files (not just directories) as sources. I did
not make
a special attribute for a single target with a rename.
* Changed the signature of the CopySpec.from methods to work like (and
delegate to) Project.file
* Removed CopySpecImpl's use of Ant
* Changed PatternMatcher to Spec<RelativePath>
This all looks good. I'll apply it to the trunk if you think it's ready.
I'm still not sure about adding Project.copy(). It feels out of place
there. Maybe a better place for it would be on File and/or FileCollection.
I'm not really sure what to do about this either. There are a lot of subclasses
of FileCollection (including Configuration which already has a copy). I think
this would be confusing here.
Do you think that it is useful to be able to access other Tasks in a similar
way? Should you be able to use tasks.copy { } or tasks.zip { }? I still don't
really like any of these ideas, but we would like to be able to use this as a
replacement for ant.copy and still see if anything was copied. I'm just not
sure where to put it.
I think this is ready to be applied to the trunk. The only additional changes I
am planning are Project.copy - but I don't know what to do there yet, and
changing the renaming so that it doesn't filter (or that is perhaps optional).
Those should be easy to add later.
I can write a little more about the copy task in the userguide if you think it
is needed.
I did not do the following:
* Reuse or move walker/pattern classes. While I fully intended these
to be
re-usable, I think it might be best to only use them for the Copy task
in 0.7.
I would prefer to use any remaining time before 0.7 to look into
making use of
Copy.didWork to set the appropriate onlyIf for the JavaPlugin tasks.
We can make
better use of these classes in 0.8.
I might find some time to move these around.
Adam
---------------------------------------------------------------------
To unsubscribe from this list, please visit:
http://xircles.codehaus.org/manage_email
--
Steve Appling
Automated Logic Research Team
---------------------------------------------------------------------
To unsubscribe from this list, please visit:
http://xircles.codehaus.org/manage_email
