OMG publishes V1.0 version of the IDL to C++11 Language Mapping

[Johnny Willemsen]

Hi all,

The OMG has published the V1.0 version of the new IDL to C++11 Language
Mapping at http://www.omg.org/spec/CPP11/. This new language mapping
greatly simplifies the development of CORBA, DDS, and CCM applications
compared to the already existing IDL to C++ Language mapping.

At the upcoming OMG Component Information Day (see
http://www.omg.org/com) I will present a session which will highlight
the simplifications and new features of this new language mapping.

Best regards,

Johnny Willemsen
Remedy IT
http://www.theaceorb.nl

[Wil Evers]

Post by Johnny Willemsen
The OMG has published the V1.0 version of the new IDL to C++11 Language
Mapping at http://www.omg.org/spec/CPP11/. This new language mapping
greatly simplifies the development of CORBA, DDS, and CCM applications
compared to the already existing IDL to C++ Language mapping.

This is great news; the old mapping predated C++98 and was full of
wrinkles. Congratulations, Johnny!

- Wil

[Christof Meerwald]

Post by Johnny Willemsen
Hi all,
The OMG has published the V1.0 version of the new IDL to C++11 Language
Mapping at http://www.omg.org/spec/CPP11/. This new language mapping
greatly simplifies the development of CORBA, DDS, and CCM applications
compared to the already existing IDL to C++ Language mapping.

There appear to be a few embarassing typos and inaccuracies in that
document, e.g.

2.1 refers to ISO/IEC 14822:2011 (which should be 14882:2011)

6.30 instead refers to ISO/IEC 14882:011 (again should be 14882:2011)

the table in 6.30 contains typos (alinas, alineof, constrexpr), names
that aren't keywords (int32_t, ...) and misses some keywords
(noexcept, char16_t, char32_t).


Christof

[Johnny Willemsen]

Hi,

Post by Christof Meerwald

Post by Johnny Willemsen
The OMG has published the V1.0 version of the new IDL to C++11 Language
Mapping at http://www.omg.org/spec/CPP11/. This new language mapping
greatly simplifies the development of CORBA, DDS, and CCM applications
compared to the already existing IDL to C++ Language mapping.

There appear to be a few embarassing typos and inaccuracies in that
document, e.g.
2.1 refers to ISO/IEC 14822:2011 (which should be 14882:2011)
6.30 instead refers to ISO/IEC 14882:011 (again should be 14882:2011)
the table in 6.30 contains typos (alinas, alineof, constrexpr), names
that aren't keywords (int32_t, ...) and misses some keywords
(noexcept, char16_t, char32_t).

Thanks for pointing these out, the ISO number, it seems that slipped
through all review, section 6.30 will be renamed to something like
"Protected C++11 names. I have just reported these comments to the OMG
so that we can address them in the current RTF which is scheduling to
deliver a V1.1 this June.

Thanks for your comments,

Best regads,

Johnny

Post by Christof Meerwald
Christof

[Bruce Visscher]

Post by Johnny Willemsen
Hi,

Post by Christof Meerwald

Post by Johnny Willemsen
The OMG has published the V1.0 version of the new IDL to C++11 Language
Mapping at http://www.omg.org/spec/CPP11/. This new language mapping
greatly simplifies the development of CORBA, DDS, and CCM applications
compared to the already existing IDL to C++ Language mapping.

There appear to be a few embarassing typos and inaccuracies in that
document, e.g.

[...]

Post by Christof Meerwald
the table in 6.30 contains typos (alinas, alineof, constrexpr), names
that aren't keywords (int32_t, ...) and misses some keywords
(noexcept, char16_t, char32_t).

Thanks for pointing these out, the ISO number, it seems that slipped
through all review, section 6.30 will be renamed to something like
"Protected C++11 names. [...]

I am curious why the integer type names from #include <cstdint> were included in this list in the first place. In the draft standard these names are shown to be in the std namespace so they should not need special handling. I.e., shouldn't the list of identifiers that need escaping just be the union of Table 4 (keywords) and Table 5 (Alternative representations) in the draft standard?

[Johnny Willemsen]

Hi,

Post by Bruce Visscher

Post by Johnny Willemsen
Thanks for pointing these out, the ISO number, it seems that slipped
through all review, section 6.30 will be renamed to something like
"Protected C++11 names. [...]

I am curious why the integer type names from #include <cstdint> were included in this list in the first place. In the draft standard these names are shown to be in the std namespace so they should not need special handling. I.e., shouldn't the list of identifiers that need escaping just be the union of Table 4 (keywords) and Table 5 (Alternative representations) in the draft standard?

Any usage of one of the cstdint integer types in IDL has to get a prefix
in order to not cause compile problems, for example a typedef to int8_t
in IDL can't be a typedef to int8_t in C++, that causes a conflict with
the cstdint type.

We could have referred to the table 4 and 5 of the C++11 spec for the
keywords, maybe something we can do in the next revision of the C++11 spec.

If you want to try IDL2C++11 yourself, see http://swsupport.remedy.nl
for TAOX11, we have free evaluation licenses.

Best regards,

Johnny Willemsen
Remedy IT