Re: NTDLL: implement NtSignalAndWaitForSingleObject

2005-04-18 Thread Mike McCormack
Actually, I just thought about one more solution. You could open the object with using an access right of 0, meaning that anything is accepted, then check the access right manually. Still a bit of a kludge, but I think it is better than the brute force chaining. Yeah, I'm not really happy with

Re: NTDLL: implement NtSignalAndWaitForSingleObject

2005-04-18 Thread Robert Shearman
Robert Shearman wrote: Mike McCormack wrote: I think this is a little ugly. I'd prefer to see this code do a switch on obj->ops and call the relevant function with the raw object instead of a handle. I know it's ugly; I rewrote that bit three times, including once with an ops->signal() method.

Re: NTDLL: implement NtSignalAndWaitForSingleObject

2005-04-18 Thread Robert Shearman
Mike McCormack wrote: I think this is a little ugly. I'd prefer to see this code do a switch on obj->ops and call the relevant function with the raw object instead of a handle. I know it's ugly; I rewrote that bit three times, including once with an ops->signal() method. Implementing a object

Re: NTDLL: implement NtSignalAndWaitForSingleObject

2005-04-18 Thread Mike McCormack
I think this is a little ugly. I'd prefer to see this code do a switch on obj->ops and call the relevant function with the raw object instead of a handle. I know it's ugly; I rewrote that bit three times, including once with an ops->signal() method. Implementing a object operation requires tha

Re: NTDLL: implement NtSignalAndWaitForSingleObject

2005-04-18 Thread Robert Shearman
Mike McCormack wrote: ChangeLog: * implement NtSignalAndWaitForSingleObject Index: server/thread.c === RCS file: /home/wine/wine/server/thread.c,v retrieving revision 1.110 diff -u -p -r1.110 thread.c --- server/thread.c 4 Mar 200