Eclipse Community Forums: Eclipse Titan » Using encode function of TTCN

Help

Home

Home » Eclipse Projects » Eclipse Titan » Using encode function of TTCN_Buffer

Show: Today's Messages :: Show Polls :: Message Navigator

Using encode function of TTCN_Buffer [message #1712068]

Wed, 21 October 2015 08:17

Johannes Bro

Messages: 56
Registered: June 2015

Member

Hi all,

i'm implementing a test port for a communication standard for metering. I defined my datatypes like this :

	type record WmbusFrameA
	{
	  WmbusDLLheader	dllheader,
	  WmbusCRC			crc,
	  WmbusAPLheader 	aplheader,
	  octetstring       payload 
	}

For sending messages of these complex datatype, i want to convert it to raw format and have just a buffer with the data. I saw in the api guide the functions of the TTCN_Buffer and also a example witch do what i want (see page 42/43 of apiguide.pdf).

My outgoing_send functions for this type looks like this:

void LcmTP__Port::outgoing_send(const LibWmbus__TypesAndValues::WmbusFrameA& send_par)
{

	TTCN_Buffer buf;
	send_par.encode(LibWmbus__TypesAndValues::WmbusFrameA_descr_, buf, TTCN_EncDec::CT_RAW);
....

The Problem is:

error: 'const class LibWmbus__TypesAndValues::WmbusFrameA' has no member named 'encode

Why has my type no encode function ?

best regards

Report message to a moderator

Re: Using encode function of TTCN_Buffer [message #1712071 is a reply to message #1712068]

Wed, 21 October 2015 08:23

Johannes Bro

Messages: 56
Registered: June 2015

Member

One remark: It is possible to disable generation of encoding/coding functionalities with a compiler flag. But I have already checked, it isn't set.

Report message to a moderator

Re: Using encode function of TTCN_Buffer [message #1712073 is a reply to message #1712068]

Wed, 21 October 2015 08:32

Elemer Lelik

Messages: 1120
Registered: January 2015

Senior Member

HI Johannes , my guess is that you are missing two things:

-you need a with { encode "RAW" }; instruction at the end of the module:

module something {

:
:

}
with { encode "RAW" };

-for every type you want RAW encoded, you need a with { variant "....." };
instruction, which details how the type should be encoded/decoded
if you have no specific instructions (meaning that the defaults will kick in) still you need and empty with { variant "" }

This will indicate to the compiler that it has to generate the codec functions; else, to decrease the size of the generated code, it will not do so.

I hope this helps

Elemer

Report message to a moderator

Re: Using encode function of TTCN_Buffer [message #1712080 is a reply to message #1712073]

Wed, 21 October 2015 08:50

Johannes Bro

Messages: 56
Registered: June 2015

Member

Yes it helped, thanks ! One more question. This "with { variant ""} instructions are special for titan and not ttcn standard, aren't it ?

Report message to a moderator

Re: Using encode function of TTCN_Buffer [message #1712084 is a reply to message #1712080]

Wed, 21 October 2015 08:58

Elemer Lelik

Messages: 1120
Registered: January 2015

Senior Member

Hi Johannes,

that's right , variants are not standard, but implementation specific.
The standard leaves codec implementation details to tool manufacturers.

Regards

Elemer

Report message to a moderator

Previous Topic:	The binary RAW encoder of Titan : part III
Next Topic:	The binary RAW encoder of Titan : part IV

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

]

Current Time: Thu Apr 25 10:46:32 GMT 2024

.:: Contact :: Home ::.

Breadcrumbs

Sign up to our Newsletter