2012/6/18 Tim Golden <m...@timgolden.me.uk>: > On 18/06/2012 00:58, Radek Holý wrote: >> >> My question is probably poorly formulated. >> In fact -- as I discovered -- some WMI objects reflect their values in >> the Windows Registry keys (for example there is mapping >> “root\cimv2:Win32_OSRecoveryConfiguration.AutoReboot” in >> >> “HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\CrashControl\AutoReboot”). >> Is there some *common* attribute of WMI/COM/OLE objects giving the key >> path? > > > I don't believe so. The thing about WMI is that it provides a > uniform and accessible layer on top of quite a lot of lower-level > APIs. In some cases, that's a layer over registry entries; in other > cases it's a layer over an actual API call. It may even be the only > publicly-accessible means of achieving some result. > > As it happens, for *methods* there is sometimes a clue as to the > underlying API. I have exposed this as (for want of a better name) the > provenance attribute of a method class. So if you do this: > > <code> > import wmi > > print wmi.WMI().Win32_Process.Create.provenance > > </code> > > You'll see something like this: > > Win32API|Process and Thread Functions|CreateProcess > > I'm not aware of any such thing for attributes, although it wouldn't > surprise me utterly if there were.
I know that the dependence WMI <-> WinReg is not frequent. Saying “common attribute” I thought common to this (reflecting) type of objects. So it seems that it was an exception that I managed to print this thing. Or is it possible that the attribute belongs to underlaying COM or OLE object? (In the pywin32 implementation!) I wonder which object it was... :-/ What a pity that interactive shell does not remember the history of commands after rebooting the computer… :-D -- Radek Holý Czech republic _______________________________________________ python-win32 mailing list python-win32@python.org http://mail.python.org/mailman/listinfo/python-win32