On Sep 9, 2011, at 10:41 AM, Dave Miner wrote: > On 09/09/11 13:32, Seth Goldberg wrote: >> >> On Sep 9, 2011, at 7:17 AM, Dave Miner wrote: >> >>> On 09/09/11 10:14, Seth Goldberg wrote: >>>> >>>> >>>> On Sep 9, 2011, at 7:10 AM, Dave Miner<[email protected]> wrote: >>>> >>>>> On 09/09/11 05:03, Niall Power wrote: >>>>>> Hi Seth, >>>>>> >>>>>> I've taken your suggestions on board - but kept the "bios" part in the >>>>>> file name :-) >>>>>> So now it is ".bios-eltorito-img" instead of "bios-eltorito-boot" >>>>>> >>>>>> New webrev: >>>>>> https://cr.opensolaris.org/action/browse/caiman/niall/7052879-1/webrev-7052879-1/ >>>>>> >>>>> >>>>> The question that is unanswered for me is why pybootmgmt cannot keep its >>>>> own mess hidden rather than requiring DC to clean up after it. Are there >>>>> other uses of pybootmgmt that will result in debris elsewhere in the >>>>> system? >>>> >>>> This isn't pybootmgmt hiding debris-- it's up to the caller to take the >>>> eltorito boot image and to move it somewhere appropriate, including >>>> renaming it. >>>> >>> >>> OK, let me rephrase: why doesn't pybootmgmt provide an API that does >>> exactly that? >> >> The API allows you to specify the directory, but not the filename because >> the caller doesn't know what files are going to be created (since you might >> be on any number of platforms). >> > > I guess my naive view is that the library should know what filenames it wants > to standardize on and just do it. Having DC randomly inventing some naming > scheme that could just as easily be buried from it doesn't seem like a win.
The main concern was overwriting existing files -- I could have easily chosen a hard-coded name (and if you guys feel strongly about it, I can do that pretty easily), but my original thought was that it was the consumer who would rename the file to something they wanted. --S _______________________________________________ caiman-discuss mailing list [email protected] http://mail.opensolaris.org/mailman/listinfo/caiman-discuss
