On Sat, Dec 16, 2006 at 11:15:43AM -0200, Otavio Salvador wrote: > But then this isn't a deprecation but a semantic change and shouldn't > be tag as a deprecation. You are right here.
> > For example, 1.8 will return a zero or a one in the file system > > probe function. 2.0 will presumably return an integer between 0 and > > 100. When using a macro, we can have flexible semantics for legacy > > apps and new semantics for upgraded apps. > > Why do you wish to change that? This shall not be the discussion in this mail, but: We need to change it to support partition guessing. You can't return a probability with only two values. > Maybe would be better just break the compilation compatibility and > write a documentation explaning what has been change and how to port > the old application to the new code. Yeah, that sounds reasonable. But how will you detect that the user hasn't adapted? Leslie -- gpg --keyserver pgp.mit.edu --recv-keys DD4EBF83 http://nic-nac-project.de/~skypher/
pgp4li6NPz92f.pgp
Description: PGP signature
_______________________________________________ parted-devel mailing list [email protected] http://lists.alioth.debian.org/mailman/listinfo/parted-devel
