IF YOU WOULD LIKE TO GET AN ACCOUNT, please write an
email to Administrator. User accounts are meant only to access repo
and report issues and/or generate pull requests.
This is a purpose-specific Git hosting for
BaseALT
projects. Thank you for your understanding!
Только зарегистрированные пользователи имеют доступ к сервису!
Для получения аккаунта, обратитесь к администратору.
This reduces the total size of the samba binaries from 119 Mb to 73 Mb.
Next step will be to have the build system obtain some of this information
by itself, so that we don't have to write ~10 lines per interface manually.
rather then a large table in librpc/gen_ndr/tables.c. This will allow us
to only link in only the required gen_ndr files (speeds up linking quite a
bit, makes binaries smaller).
Each gen_ndr_* file now has a init function that calls the init functions
of the interfaces it contains. I did it this way to keep pidl's code simple,
though it might hurt startup time a bit. I'd be happy to change it if
people like one function better.
- Work on server side and local COM support (should work, just no
example classes yet)
- Use vtables so that local and remote calls can be used transparently
- Generate 'proxies and stubs' rather then heavily modified code in client.pm and server.pm. proxies (client side code) are generated in proxy.pm, stubs (server side dispatchers) are generated in stubs.pm
- Support registering classes and interfaces
- DCOM interfaces no longer have to be in the same IDL file as their
base interface, which will allow us to split up dcom.idl
The torture test DCOM-SIMPLE now successfully does an
IStream_Read and a IStream_Write call.
This test can now be run successfully against the "Simple DCOM" Visual
Studio example.
(You have to quote out line 337 in pidl. pidl complains if the variable
that contains the array size follows the array. I still need to fix this
properly)
Next goals:
- Clean up code
- Server side support
- Support custom marshalling
- Support DCOM interfaces in files other then dcom.idl
- OXID tables work now. IOXIDResolver is used if there is used for getting a STRINGBINDING if none is known yet
- Add custom dissectors for STRINGARRAY and DUALSTRINGARRAY. If there's a way to get rid of these later on (by supporting them thru pidl somehow), I'd be happy to use that instead of doing it manually.
I can now get to the point where we have created an object and are connected to
it. The only thing left to do is being able to set the Object UUID properly..
Both subsystems and modules can now have init functions, which can be
specified in .mk files (INIT_FUNCTION = ...)
The build system will define :
- SUBSYSTEM_init_static_modules that calls the init functions of all statically compiled modules. Failing to load will generate an error which is not fatal
- BINARY_init_subsystems that calls the init functions (if defined) for the subsystems the binary depends on
This removes the hack with the "static bool Initialised = " and the
"lazy_init" functions
- Support for sending over the object UUID in DCERPC calls
- Simple torture test for the DCOM "Simple" object
- Generate extra argument for "object" interfaces in pidl
- Some stubs for common DCOM functions
describes a COM class. A coclass is the implementation of one or more
interfaces. It has a UUID referred to as it's CLSID (Class ID).
Also adding an example coclass called "CoffeeMachine". You can give
it a string (or a cup, whatever you like ;-) and it will fill it with
"COFFEE" (kind of the like the echo pipe is for regular RPC). CoffeeMachine's
Windows implementation already works, a torture test for Samba will follow
soon.
the DCOM calls are wrappers around several local calls, so you get things like:
WERROR foobar ( [in] int num_ifaces,
[in,size_is(num_ifaces)] IID *ifaces,
[out,size_is(num_ifaces)] WERROR *results);
The thing that finally convinced me that minimal includes was worth
pursuing for rpc was a compiler (tcc) that failed to build Samba due
to reaching internal limits of the size of include files. Also the
fact that includes.h.gch was 16MB, which really seems excessive. This
patch brings it back to 12M, which is still too large, but
better. Note that this patch speeds up compile times for both the pch
and non-pch case.
This change also includes the addition iof a "depends()" option in our
IDL files, allowing you to specify that one IDL file depends on
another. This capability was needed for the auto-includes generation.
uuid(1ff70682-0a51-30e8-076d-740be8cee98b) so we now accept
uuid("1ff70682-0a51-30e8-076d-740be8cee98b") in pidl, and
automagically add quotes only if needed
specifying a endpoint is now also 'endpoint' instead of 'endpoints'. The
default endpoint (if none is specified) is still "ncacn_np:[\\pipe\\ifacename]",
where ifacename is the name of the interface.
Examples:
[
uuid(60a15ec5-4de8-11d7-a637-005056a20182),
endpoint("ncacn_np:[\\pipe\\rpcecho]", "ncacn_ip_tcp:")
]
interface rpcecho
{
void dummy();
}
dcerpc_binding is now converted to ep_description in the server, but I hope to
completely eliminate ep_description later on.
The eventual goal of all these changes is to make it easier to add
transports as I'm going to add support for
ncalrpc (local RPC over named pipes) and ncacn_unix_stream (Unix sockets).
from inside a swig %exception block and into the argout typemap. This
will allow us to wrap functions that don't require exception handling, and
also get rid of some ugly code in dcerpc.i
the [gensize] property to a struct or union will make pidl generate a
ndr_size_*() function.
(not all nasty bits of NDR are completely covered yet by the
ndr_size*() functions, support for those will be added when necessary)
I also have a local patch (not applied now) that simplifies the pidl output
and eliminates the number of functions required. It would, however, make
pidl more complex.
