|
|
 Up |
|
|
  |
Author: Jules ColdingJules Colding
Date: Dec 26, 2008 01:55
On 26/12/2008, at 10.03, Johnny Willemsen wrote:
> Hi,
>
>> No problem with the test as such, but how can it possibly work? I
>> mean, the interface name as produced by my patch must be compared
>> manually with the name as produced by a native system tool on all
>> supported platforms.
>>
>> Writing a general test which can do that automatically seems
>> impossible in the absence of a universal tool whose output can be
>> grep'ed for the name.
>>
>> I can test the patch on my own system when svn builds for me, but an
>> automatic test for all platforms?
>
> You could print it with ACE_DEBUG, that way at least it can be checked
> easily.
OK, no problem.
|
| Show full article (0.69Kb) |
|
| |
2 Comments |
|
|
Author: Jules ColdingJules Colding
Date: Dec 26, 2008 01:52
Hi Johnny,
On 26/12/2008, at 10.05, Johnny Willemsen wrote:
> Is the name really something for Addr, what about only having it at
> INET_Addr.
Thought about that, but having addr_name_ initialized once in the
(single) base class constructor seemed cleaner than initializing it in
the handful of constructors in INET_Addr. Alos, a name seemed like a
vey general thing that may be useful for many kind of addresses.
Thoughts?
jules
|
| Show full article (1.20Kb) |
|
| |
no comments
|
|
|
|
|
|
Author: Johnny WillemsenJohnny Willemsen
Date: Dec 26, 2008 01:03
Hi,
> No problem with the test as such, but how can it possibly work? I
> mean, the interface name as produced by my patch must be compared
> manually with the name as produced by a native system tool on all
> supported platforms.
>
> Writing a general test which can do that automatically seems
> impossible in the absence of a universal tool whose output can be
> grep'ed for the name.
>
> I can test the patch on my own system when svn builds for me, but an
> automatic test for all platforms?
You could print it with ACE_DEBUG, that way at least it can be checked
easily. I agree you can't check the name automatically.
Johnny
|
| |
|
no comments
|
|
|
Author: Jules ColdingJules Colding
Date: Dec 26, 2008 00:42
Hi Johnny,
On 26/12/2008, at 08.35, Johnny Willemsen wrote:
> We will need a regression test for this and you should test this
> before we
> could integrate such a change. Please extend or create a new test
> and test
> it first. Then put everything in bugzilla.
No problem with the test as such, but how can it possibly work? I
mean, the interface name as produced by my patch must be compared
manually with the name as produced by a native system tool on all
supported platforms.
Writing a general test which can do that automatically seems
impossible in the absence of a universal tool whose output can be
grep'ed for the name.
I can test the patch on my own system when svn builds for me, but an
automatic test for all platforms?
Best,
jules
|
| Show full article (1.53Kb) |
|
no comments
|
|
|
Author: Johnny WillemsenJohnny Willemsen
Date: Dec 25, 2008 23:35
Hi,
We will need a regression test for this and you should test this before we
could integrate such a change. Please extend or create a new test and test
it first. Then put everything in bugzilla.
Regards,
Johnny Willemsen
Remedy IT
Postbus 101 - 2650 AC Berkel en Rodenrijs
Melkrijder 11 - 3861 SG Nijkerk
The Netherlands
www.theaceorb.nl / www.remedy.nl
*** Integrated compile and test statistics see
http://scoreboard.theaceorb.nl **
*** Commercial service and support for ACE/TAO/CIAO ***
*** Get your free TAO Programmers Guide copy from
http://www.theaceorb.nl ***
|
| Show full article (1.35Kb) |
|
no comments
|
|
|
Author: Jules ColdingJules Colding
Date: Dec 25, 2008 18:25
Hi Doug,
On 26/12/2008, at 00.29, Douglas C. Schmidt wrote:
>> Yes. The get_ip_interfaces() method was exactly my way into
>> ACE_INET_Addr. It do seem to be in need of an extension. I'll se if I
>> can work out a patch...
>
> That would be super!
What about the attached patch? MPC is not currently able to generate
Makefiles for me so I'm not able to test it yet.
Best,
jules
|
| |
|
no comments
|
|
|
Author: Douglas C. SchmidtDouglas C. Schmidt
Date: Dec 25, 2008 15:29
Hi Jules,
> Yes. The get_ip_interfaces() method was exactly my way into
> ACE_INET_Addr. It do seem to be in need of an extension. I'll se if I
> can work out a patch...
That would be super!
> Thanks, and merry Christmas btw :-)
Thanks Jules - and a very merry christmas to you and all other ACE+TAO users!!
Doug
|
| |
|
no comments
|
|
|
Author: Jules ColdingJules Colding
Date: Dec 25, 2008 10:42
On 25/12/2008, at 16.14, Douglas C. Schmidt wrote:
> Hi Jules,
>
>> Is there any way of getting the interface name (e.g eth0) using ACE?
>
> I don't think there's a way of doing this (yet), though you might
> check out the ACE::get_ip_interfaces() method (and the associated test
> at ACE_ROOT/tests/Enum_Interfaces_Test.cpp) to see if you can figure
> out how to use/extend this capability to get what you need.
Yes. The get_ip_interfaces() method was exactly my way into
ACE_INET_Addr. It do seem to be in need of an extension. I'll se if I
can work out a patch...
Thanks, and merry Christmas btw :-)
jules
|
| |
|
no comments
|
|
|
|
|
|
Author: Douglas C. SchmidtDouglas C. Schmidt
Date: Dec 25, 2008 10:25
Hi Arkady,
Thanks very much for your email. Please make sure to send all
questions related to TAO or ACE to the ACE mailing list or ACE+TAO
newsgroup, rather than to me directly since I travel frequently and
often don't have ready access to email.
> I do not try to binding to so-called reserved ports , the question is
> ,what should I do to listen more then 1 port using single
> ACE_TP_Reactor ?
"Just do it" - you simply need to declare two instances of ACE_Acceptor
that listen to two different ports. The fact that you're using
ACE_TP_Reactor is irrelevant.
Thanks,
Doug
|
| |
|
no comments
|
|
|
|
|
