Hello, I thought that using fileset with the includesfile attribute would still trigger a source directory scan, which is what I want to avoid to begin with. Isn't that correct?
However, this <augment> task does sound very interesting to add <file name="..."/> elements to the filelist (avoiding the "space in filename" issue) or to add additional <name name="..."/> selectors in a <restrict> resource collection in order to simulate include and exclude patterns. Would it also work to add elements in a <firstmatchmapper> as suggested by Stefan ? Thanks, Patrick On Fri, Jun 25, 2010 at 4:57 PM, Matt Benson <gudnabr...@gmail.com> wrote: > > On Jun 25, 2010, at 9:48 AM, Stefan Bodewig wrote: > >> On 2010-06-25, Patrick Martin wrote: >> >>> I tried that but then, copy (verbose="true") tells me something like: >>> [copy] //host/share/dir1/dir2/one file.txt omitted as >>> //host/share/dir1/dir2/one file.txt is up to date. >> >>> So it means copy actually tried to copy the files over themselves ? It >>> did not take its "todir" attribute into account ? >> >> Yes, actually this is the surprising behavior if you sources are >> absolute paths. I forgot about that myself, sorry. See my other post >> about using mappers to turn the paths into relative paths again. >> > > Note that if your sources are relative paths to begin with, you have other > options. Most simply and backward-compatibly if the list is in a > line-delimited file you can supply that file to <fileset>'s includesfile > attribute. If for the sake of argument your list was in, say, a property or > read from a url, and you are simply too stubborn or resource-constrained to > write the includesfile to the filesystem, you could use a combination of the > <tokens> resourcecollection, antcontrib <for>, and the brand-spanking-new > <augment> task to loop over the lines (or other recognizable tokens) of your > list and add each as a separate include to your fileset. Ain't that slick? > ;P > > -Matt > >> Stefan >> >> --------------------------------------------------------------------- >> To unsubscribe, e-mail: user-unsubscr...@ant.apache.org >> For additional commands, e-mail: user-h...@ant.apache.org >> > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: user-unsubscr...@ant.apache.org > For additional commands, e-mail: user-h...@ant.apache.org > > --------------------------------------------------------------------- To unsubscribe, e-mail: user-unsubscr...@ant.apache.org For additional commands, e-mail: user-h...@ant.apache.org
