On Tue, 19 Apr 2005, Dominique Devienne <[EMAIL PROTECTED]> wrote: >> From: Stefan Bodewig [mailto:[EMAIL PROTECTED] >> >> as well as org/apache/tools/ant/types/antlib.xml with >> <antlib> >> <typedef resource="o/a/t/a/types/defaults.properties"/> >> <typedef resource="o/a/t/a/types/selectors/antlib.xml"/> >> <typedef resource="o/a/t/a/taskdefs/conditions/antlib.xml"/> >> </antlib> > > But won't that create some kind of warning because of the > and/or/etc... name clashes from conditions/selectors?
Yes. Maybe we should have: oata/taskdefs/conditions/unique.xml with <antlib> <typedef name="istrue" .../> <typedef name="isfalse" .../> </antlib> oata/taskdefs/conditions/antlib.xml with <antlib> <typedef resource="oata/taskdefs/conditions/unique.xml"/> <!-- these clash with selectors --> <typedef name="and" .../> <typedef name="or" .../> <typedef name="not" .../> <!-- these are duplicates from taskdefs --> <typedef name="available" .../> <typedef name="uptodate" .../> </antlib> same for selectors and then org/apache/tools/ant/types/antlib.xml with <antlib> <typedef resource="o/a/t/a/types/defaults.properties"/> <typedef resource="o/a/t/a/types/selectors/unique.xml"/> <typedef resource="o/a/t/a/taskdefs/conditions/unique.xml"/> </antlib> We could decide which <and> would get defined by default and new naming conflicts (in the default antlib) would get resolved by "the older one wins". Stefan --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
