nsf

Clone
Gustaf Neumann <neumann@wu-wien.ac.at>
committed
7d197d8
on 07 Sep 11
nsf.c: * fixed recration of object alias with a alias to a different cmd
2-1-0-rc + 65 more
TODO (+2 -0)
3925 3925   and the variable resolvers. The variable resolver was not able to
3926 3926   resolve variables, while the command resolver was still working
3927 3927   correctly. 
3928 3928 - Extended regression test
3929 3929
3930 3930 nsf.c:
3931 3931 - added object parameter option "slotinitialize"
3932 3932 - renamed object parameter option "invokesetter" -> "slotassign"
3933 3933 - call slot.assign instead of setter of object
3934 3934 - removed restriction on nosetter/invokesetter: nosetter can
3935 3935   be used in connection with slotassign
3936 3936 - added regression test for slot.initialize
3937 3937
3938 3938 nsf.c:
3939 3939 - pass property name to slot "initialize" method to conform
3940 3940   with the interface to "assign", "get" ... (all these receive
3941 3941   the property name as well)
3942 3942 - allow slot "initialize" method to be protected
3943 3943   (handled similarly to "init")
3944 3944
  3945 - extended regression tests for yield
  3946 - implemented "next" for ruby-like enumerators (each needs still more work)
3945 3947
3946 3948 ========================================================================
3947 3949 TODO:
3948 3950
3949 3951 - document new setable object properties perobjectdispatch and keepcallerself
3950 3952 - pertaining perobjectdispatch and keepcallerself in serializer
3951 3953
3952 3954 - behavior on keepcallerself on ordinary dispatches with implicit/explicit
3953 3955   receiver (currently the flag is ignored, the code just commented out)
3954 3956
3955 3957 - Aliases and forwards are not handled by NsfNSCopyCmdsCmd; object
3956 3958   cloning/copying remains incomplete; also, there might be object and
3957 3959   method properties not handled as well (as the "returns" method
3958 3960   property before the recent fix)
3959 3961
3960 3962 - Check the following todos such as "todo: don't hard-code registering
3961 3963   command name "method" / NSF_METHOD"; parts of the introspection
3962 3964   machinery in NSF rely on NX/script-level specifics ... how should we
3963 3965   handle this? For the system methods (create, ...), we can resolve
3964 3966   the situation; for the NX or XOTcl method delegates (method) of