At 05:24 PM 9/25/2006 +0200, Bonzini Paolo wrote: > > I may have just overwritten my > > changes with yours. > >No problem. > > > I agree with Mike that utime would be a better name > > than touch if it only changes an already existing file. My touch created > > the file if it didn't exist, Paolo's does not. Maybe we should have both? > >My latest changes have both #touch: and #lastAccessTime:lastModifyTime: >(the latter does not create the file). Thanks. Why is it that File isn't a subclass of RealFileSystem. > >Paolo _______________________________________________ help-smalltalk mailing list [hidden email] http://lists.gnu.org/mailman/listinfo/help-smalltalk |
J Pfersich wrote:
> > Why is it that File isn't a subclass of RealFileSystem. Because it could be contained (not extracted) within an archive, or on a remote server. I must say that I don't think that the classes File and Directory have a very clear idea of what they *are*, though. To take some examples, it seems clear to me that Directory class >> #working should return a Directory and not a String, and that Directory >> contents should return a collection of Files. Unfortunately, in the latter case it becomes difficult to pick out '.' and '..' (since you usually want to ignore them). Mike _______________________________________________ help-smalltalk mailing list [hidden email] http://lists.gnu.org/mailman/listinfo/help-smalltalk |
Mike Anderson wrote:
> J Pfersich wrote: > >> Why is it that File isn't a subclass of RealFileSystem. >> File is an adaptor for VFSHandler, of which RealFileSystem is a subclass. > I must say that I don't think that the classes File and Directory have a > very clear idea of what they *are*, though. > I agree, though I'm not sure if I want to break compatibility now. If we get a 3.0 we could think about it. > latter case it becomes difficult to pick out '.' and '..' (since you > usually want to ignore them). Directory>>#contents may blow them up itself, if we are to change the interface... Paolo _______________________________________________ help-smalltalk mailing list [hidden email] http://lists.gnu.org/mailman/listinfo/help-smalltalk |
Paolo Bonzini wrote:
> Mike Anderson wrote: > >> I must say that I don't think that the classes File and Directory have a >> very clear idea of what they *are*, though. > > I agree, though I'm not sure if I want to break compatibility now. If > we get a 3.0 we could think about it. Yes (although what do you think about a loadable package in examples or unsupported?). >> latter case it becomes difficult to pick out '.' and '..' (since you >> usually want to ignore them). > > Directory>>#contents may blow them up itself, if we are to change the > interface... Good point (I think I'll apply that to my 'comfort' package right now), although I think there is a larger problem, which is that Files don't remember if they were created with a relative path. Mike _______________________________________________ help-smalltalk mailing list [hidden email] http://lists.gnu.org/mailman/listinfo/help-smalltalk |
> Yes (although what do you think about a loadable package in examples or > unsupported?). > Yes, why not. Just a file-in in unsupported would be great (with the plan to merge it later on). Actually File/Directory started in unsupported (when I was not maintainer and it was not yet called unsupported...). And alas when I moved it out of unsupported, I really forget to think about a proper design. :-( >>> latter case it becomes difficult to pick out '.' and '..' (since you >>> usually want to ignore them). >>> >> Directory>>#contents may blow them up itself, if we are to change the >> interface... >> > > Good point (I think I'll apply that to my 'comfort' package right now), > although I think there is a larger problem, which is that Files don't > remember if they were created with a relative path. > would not break backwards compatibility. Paolo _______________________________________________ help-smalltalk mailing list [hidden email] http://lists.gnu.org/mailman/listinfo/help-smalltalk |
Free forum by Nabble | Edit this page |