Unfortunately, I'm not going to do this anytime soon; the tradeoff is clearly not worth it within the domain of just my company and our apps.
Nor am I likely to persuade you that it's worthwhile; nor you to persuade me. The arguments over static analysis tools are religious and, even after thirty-five years, people are still arguing the pros and cons of lint, in the C world. So, really no more to say on this thread. But, I do hope that some toolmeister inside Google sees this thread and picks up the gauntlet someday. David On Jul 13, 8:45 pm, TreKing <treking...@gmail.com> wrote: > On Tue, Jul 13, 2010 at 9:00 AM, deg <d...@degel.com> wrote: > > Actually, I can imagine two very common cases: > > > 1) Code or even functionality is changed in an application, possibly > > in some minor way. The developer does not realize that he has removed > > the last SDK call using a permission. The permission lingers on in the > > manifest. It doesn't do much harm, but add minor bloat and maybe > > scares aways some users. > > I wouldn't call this common since if you're removing permissions you're > likely removing features / functionality which relied on said permission, > which is not very likely. > > > 2) A developer adds a new feature, and does not realize that a > > permission is needed. Best case, he catches this immediately when > > testing, but still wastes time and annoyance on an extra compile/ > > deploy/test cycle. > > You answered your own question. If you're adding a new feature, I would hope > you're testing it. Odds are you don't just do one compile / deploy test > cycle for any feature you add, so one more to catch a permission you forgot > to add, like any other programming error you catch while debugging, is > negligible. You'd run into a "permission required" exception almost > immediately while testing and you fix it right then and there. > > > It seems that it would not be too hard to annotate the SDK and and a static > > permissions check to the build cycle. > > I really don't think the time it would take to make this happen is worth it > versus how much time it would actually save. You seem to think it is, so you > are welcome to grab the source code and modify it to make this happen. Good > luck. > > --------------------------------------------------------------------------- > ---------------------- > TreKing - Chicago transit tracking app for Android-powered > deviceshttp://sites.google.com/site/rezmobileapps/treking -- You received this message because you are subscribed to the Google Groups "Android Developers" group. To post to this group, send email to android-developers@googlegroups.com To unsubscribe from this group, send email to android-developers+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/android-developers?hl=en
