Activity for ORBit-Ada
-
Oliver Kellogg committed [573fd9] on Code
CORBA/IDLtree.pm : Synchronize with version 2.06.
-
CORBA/IDL2Ada.pm : In sub `e` remove unintended leftover from debugging.
-
CORBA/IDLtree/Comment_Directives.pm
-
CORBA/IDL2Ada.pm version 2025-04-08
-
CORBA/IDL2Ada.pm
-
CORBA/IDLtree.pm (parse_annotation_app) : fix error from commit cced735
-
CORBA/IDLtree.pm
-
CORBA/IDLtree.pm
-
CORBA/IDLtree : Synchronize with release 2.05
-
Harmonize FATFIFI text representation of arrays with Ada2022:
-
CORBA/IDL2Ada.pm version 2024-03-15:
-
orbit-ada-include/fatfifi-io.adb : Declare local Text `constant` where possible.
-
CORBA/IDLtree.pm : In sub get_numeric use Carp::cluck for warning about undefined symbol.
-
CORBA/IDL2Ada.pm followup to commit 91472e8:
-
tests/cosnaming/Makefile
-
CORBA/IDL2Ada.pm
-
idl2ada.pl
-
CORBA/IDLtree.pm
-
Daniel S Henry posted a comment on discussion Help
FYI, we are still considering your ORBIT-ADA, we did run into a few snags, but i was getting there, My Customer is currently evaluating where they want to go, i am working on other projects while we wait for the government to decide, if they want to move forward or drop this project. thnx for your help, i know i was a bit scatter brained at first, but it was starting to work. i did get several dialog boxes up, and that's how far we wanted to get at that point. danielh On 4/29/20 7:23 PM, Oliver Kellogg...
-
[...] all i had to do was change all define calls to presume a procedure and get the IOR from the variable RETURNS, moving forward.... crosss your fingers Have you been able to move forward - or have there been more problems?
-
idl2ada.pl
-
ok, i figured it out, OrbRiver treats all defines as functions, and returns the RETURNS via the function call, orbit-ada treats all defines as procedures and returns the RETURNS via an out varialbe called RETURNS, all i had to do was change all define calls to presume a procedure and get the IOR from the variable RETURNS, moving forward.... crosss your fingers
-
Hmm that's strange. I the same commit attempted a test case for the problem, see tests/tao_param_test/param_test-impl.adb : procedure test_object (Self : access Object; obj : out CORBA.Object.Ref'Class; Returns : out CORBA.Object.Ref'Class) is Nil_Ref : CORBA.Object.Ref'Class := CORBA.Object.Null_Ref; begin Debug ("Param_Test.Impl.test_object is called"); obj := Nil_Ref; Returns := Nil_Ref; end test_object; Could you send a stripped down reproducer that I can integrate into the test suite?
-
Hmm that's strange. I the same commit attempted a test case for the problem, see tests/tao_param_test/param_test-impl.adb : procedure test_object (Self : access Object; obj : out CORBA.Object.Ref'Class; Returns : out CORBA.Object.Ref'Class) is Nil_Ref : CORBA.Object.Ref'Class := CORBA.Object.Null_Ref; begin Debug ("Param_Test.Impl.test_object is called"); obj := Nil_Ref; Returns := Nil_Ref; end test_object; Could you send a stripped down reproducer that I can integrate into the test suite?
-
so first off, i don't think it can be a constant, because it's an out value 2nd, it still whines about being a package or procedure name, 3rd, i trie both a direct reference and a variable in the procedure and a private in the spec, all three same result smu_pull_down.adb:398:40: expected private type "Ref" defined at corba-object.ads:6 smu_pull_down.adb:398:40: found package or procedure name dsp_display_list.adb:1113:42: expected private type "Ref" defined at corba-object.ads:6 dsp_display_list.adb:1113:42:...
-
Oh, I see. Could you try again with https://sourceforge.net/p/orbitada/gnack/ci/855660e5a01884861ed830b9d3c698ffa5b8f8c6/ ?
-
orbit-ada-include/corba-object.ads
-
smu_pull_down.adb:393:40: expected private type "Ref" defined at corba-object.ads:6 smu_pull_down.adb:393:40: found package or procedure name dsp_display_list.adb:1107:42: expected private type "Ref" defined at corba-object.ads:6 dsp_display_list.adb:1107:42: found package or procedure name
-
Could you try out https://sourceforge.net/p/orbitada/gnack/ci/1d4a13accf1edd68e0fe5e0f1c48ef657341dec8/ ? note that i created a private NIL_REF, but the compile still doesn't like it. Hmm, that's strange. What is the compiler error message?
-
orbit-ada-include/corba-object.{ads,adb}
-
yep, that fixed it, thnx much, while i have your attention, i had one more question 1. i keep getting the error expected private type "Ref" defined at corba-object.ads:6 found package or procedure name. 2. the last variable in the Call is called RETURNS. (mycorba.PULL_DOWN.define_PULL_DOWN(self,text,enabled,item_num,font_rec,RETUNS) 3. in the OLD ORBRiver code we had a DEFAULT NULL_REF that we put here. 4. we don't need this value, but i can't get it to compile 5. what do i put in the RETURNS variable?...
-
yep, that fixed it, thnx much, while i have your attention, i had one more question 1. i keep getting the error expected private type "Ref" defined at corba-object.ads:6 found package or procedure name. 2. the last variable in the Call is called RETURNS. (mycorba.PULL_DOWN.define_PULL_DOWN(self,text,enabled,item_num,font_rec,RETUNS) 3. in the OLD ORBRiver code we had a DEFAULT NULL_REF that we put here. 4. we don't need this value, but i can't get it to compile 5. what do i put in the RETURNS variable?...
-
CORBA/IDL2Ada.pm
-
CORBA/IDLtree.pm : Sync with public version 2.03
-
CORBA/IDLtree.pm
-
CORBA/IDLtree.pm
-
CORBA/IDLtree.pm
-
CORBA/IDLtree/Comment_Directives.pm followup to IDLtree.pm v1.6 change,
-
CORBA/IDLtree.pm
-
CORBA/IDLtree.pm
-
CORBA/IDL2Ada.pm
-
CORBA/IDLtree.pm
-
CORBA/IDLtree.pm
-
CORBA/IDLtree.pm
-
idl2ada.pl
-
CORBA/IDLtree.pm
-
CORBA/IDL2Ada.pm
-
tests/fatfifi/tst.idl
-
CORBA/IDLtree.pm
-
Initial commit - corresponds to gnack-1.1c
-
Initial revision
-
Add generic function G_Seq_allocbuf.
-
CORBA::IDLtree adds a file scope which we need to avoid
-
r119771
-
CORBA::IDLtree::Elem - Information about elementary types
-
Adapt to ORBit2.
-
Utilities for processing directives embedded in IDL comments
-
Adjust libs to ORBit2.
-
Add GMGPL, sync with wavefront.
-
r116920
-
Formatted Ada Text File I/o From Idl
-
sync with internal wavefront.
-
Add functions small_alloc and small_allocbuf.
-
Finally, a proper Makefile :)
-
r120389
-
Use CORBA.C_Types.G_Seq_allocbuf.
-
Comment out functions that are not yet available in ORBit2-2.12.0.
-
start using CORB::IDLtree::Comment_Directives
-
PortableServer.POA.Activate_Object_With_Id doesn't seem to work with ORBit2,
-
Adjust to ORBit2.
-
%keywords: Add CORBA 3.0 keywords.
-
Set the directive object in CORBA::IDL2Ada.
-
First step towards ORBit2.
-
sync with internal wavefront.
-
r46955
-
Remove unused packages.
-
Use new package ORBit and new function CORBA.C_Types.G_Seq_allocbuf.
-
sub prefix: Fix condition for avoiding file scope.
-
ugh, finally make it usable
-
at last, a Makefile
-
fix placement of rules
-
Change $gen_tc_any to true by default. ORBit2 requires the typecodes
-
adapt handling of sequences to ORBit2
-
sub helper_prefix: I don't remember what this "avoid file scope" jazz was
-
sync with svn r204004
-
add info about ORBIIOPIPv4=1
-
fix GEN_C_HDRS
-
Makefile obsoletes {c,s}link
-
don't need typecodes here
-
fix typos
-
adapt handling of exceptions to ORBit2
-
lose a few words about ORBIT2_DEBUG
-
add Makefile
-
idl2ada.pl switch -T suppresses typecode generation. By default, typecodes are generated.
-
Supply -T at idl2ada. Depend on code generator files.
-
at last, a Makefile
-
update
-
finally, a Makefile
-
enable debugging the generated C code
-
generate proxy only, clean up rules
1 >
