Get rid of pointer_default_top()

Jelmer Vernooij jelmer at
Mon Feb 4 11:00:00 GMT 2008

Hi Metze,

On Sun, Feb 03, 2008 at 02:18:45AM +0100, Stefan (metze) Metzmacher wrote:
> could you take a look at this branches?
I'll see if I can review these tonight or tomorrow afternoon.

Getting rid of pointer_default_top() would be really really nice.

> Get rid of pointer_default_top():

> gd: can you do the same in v3-2-test?

> Just remove pointer_default_top(unique) and keep adding explicit
> [unique] properties untill the pidl output is 100% the same as before.

> But always rebuild all idl files with 'make idl', because otherwise you
> sometimes get different results...

> jelmer: we should really correctly support 'import' and load the types
> of the imported idl. It took me 8 hours to figure out why pidl output
> has changed after a touch librpc/idl/foo.idl;make idl, compared
> to make idl_full. The reason was that make idl_full loads the types of
> all files, so the types are implicit known as the typelist is global.
> 'can_contain_deferred() behaives different for make idl and make
> idl_full, e.g. for GUID...
I did some initial work on that some time ago, generating our own
equivalent of Typelib files. Perhaps I can finish that code later this
week. It's almost done and shouldn't be that hard to implement in


-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 307 bytes
Desc: Digital signature
Url :

More information about the samba-technical mailing list