win32netapi module -version 1.0beta1 (on Windows NT4/5)

[Asang Dani]

Mark Hammond <MHammond at skippinet.com.au> wrote in message
news:7icndp$sbn$1 at m2.c2.telstra-mm.net.au...
> >  It really looks quite similar. I looked at your *win32net* module and
> >found that some of the interfaces are not present in your module, but
> >are part of *win32netapi* (my module).
>
> Some of these are missing; however a few are oversights that will take all
2
> minutes to fix.

I agree.

>
> Also note that some of the NetShare functions are labelled in the SDK as
> obsolete, and the same functionality can be obtained from win32wnet.
> win32wnet is designed to work on all Windows, win32net is designed to be
NT
> only.

I think in long-run using win32wnet APIs is a better way to write the
networking applications
for win32 platform, since these APIs are suppored on all windows platforms.
(Windows NT,
Windows CE and Windows 9x). I had planned to include these in these in next
release
of win32netapi. If you are planning to fully support these in the future
releases of *win32all*
I will not work on them.

regards,
asang..

> >Please correct me if I am wrong about these differences. About the
> >info. level param, I think that the info. level numbers in most of
> >these APIs are really crazy in base APIs. (101, 102, 1024, 512 ...). So
> >I return dictionaries with keys for maximum possible info level
> >(constranied by the privileges of user executing the API), in most of
> >the cases. It's upto the user to choose which key/values he is
> >interested in. I think it is better to free the programmer from the
> >burden of remembering which fielss are present in which info. level
> >and refer constantly to MSDN to find this info.
>
> Except that the maximum info level is:
> a) Not always appropriate when creating entries.  Eg, using the lowest
info
> level when creating a user is far easier as only 2 or 3 entries are
> required.  Using the high user-info structures means I need to fill in far
> more.
>
> b) Doesnt always contain all information.  SERVER_INFO and USER_INFO seem
to
> have different info levels with unque pieces of data - eg, there is some
> data that can only be obtained by using a specific info level, and other
> data that can only be obtained using a different info level.
>
> However, I do take your point regarding freeing the user from always
needing
> to know this stuff.  The solution for me is probably to provide a default
> value for these arguments.