Rejected patches needing review.

[Michael Karcher]

Am Mittwoch, den 17.09.2008, 01:17 +0200 schrieb Peter Dons Tychsen:
> > Peter, are you sure that windows will handle REG_SZ *only* in the
> > DependOnService case and fail if you have a REG_SZ in DependOnGroup?
> No. But i did not want to change the behavior for anything else than the
> scenario that i had been investigating and testing, to avoid
> regressions.

I can definitely understand your motivation. The problem is that this
patch seems to add unneeded complexity to wine which might cause you
problems getting it committed. Experimenting with native windows showed
strange results. If I replace the DependOnService entry in my
RemoteAccess service entry (it has a REG_MULTI_SZ hat contains just one
entry) in the registry by an REG_SZ, services.msc still shows the
dependency in the properties dialog. If I replace the DependOnGroup
(also a REG_MULTI_SZ with one entry) by a REG_SZ entry, it gets ignored
completely by services.msc in the properties dialog.

So, it might be correct to handle the two cases different, as I see
different behaviour on native. But this kind of thing should be tested
(at least manually as I did) and documented (perhaps in your commit
message) that the different loading behaviour for DependOnService and
DependOnGroup is verified. Just for reference: WinXP SP3 in KVM here.

Regards,
  Michael Karcher