Dear Antti,
sorry to give a very short answer only: the decoder looks into data only
structurally - i.e. it can decode when the decodingHypothesis matches
the base type. The subtype limitations need to be handled by TE in
course of the matching.
With best regards,
Ina.
Antti Hyrkkanen wrote:
> Hello
>
> Relating an email of the last winter, is it legal that as the return
> value of TCI-CD operation
>
> Value decode(in TriMessageType message,
> in Type decodingHypothesis),
>
> the decoder returns a value, that is not compatible to decodingHypothesis?
>
> It is said in the standard, that it should return null if it cannot
> decode a compatible value. Now, how does the decoder know that the return
> value is incompatible with the decodingHypothesis, if the type is a subtype,
> whose values have been limited by range or a value list?
>
> For example:
>
> type integer MyIntegerRange (1, 2, 3, 10 .. 20, 99, 100);
> type integer MyStrings ("Alice", "Bob");
>
> There is no operation with which the decoder could ask what are the legal
> values of for a certain type, nor there is no operation with which the
> the decoder could ask whether a certain value is compatible with the
> type of decodingHypothesis.
>
> Should all the the set* operations return an error value, when the
> codec tries to assign an illegal value for the type?
>
> Or should there be a separate validate() operation, with which
> the codec could ask from the TE, whether the constructed value
> is compatible with the decodingHypothesis?
>
>
> The only operations for EnumeratedValue type are
>
> TString getEnum() and
> void setEnum(in TString enumValue),
>
> which get and set the value by enumeration identifier. There is no method
> with which a decoder would get a list of the legal identifiers.
>
> At the core language level, the user can specify integer values associated
> with the string indentifiers. There are no methods for querying the
> associated integer value or for setting the EnumeratedValue by the
> integer value.
>
> "6.3.4 Enumerated type and values
>
> NOTE 1: The integer value also may be used by the system to
> encode/decode enumerated values. This, however is outside of the scope
> of the present document (with the exception that TTCN-3 allows the
> association of encoding attributes to TTCN-3 items). "
>
>
> Br
> Antti Hyrkkanen
>
>
>>
Original Message
>>From: active_ttcn3 : mts stf133 ttcn version 3 - active members only
>>[
This email address is being protected from spambots. You need JavaScript enabled to view it.]On Behalf Of ext Stephan Tobies
>>Sent: Thursday, December 11, 2003 13:06 PM
>>To:
This email address is being protected from spambots. You need JavaScript enabled to view it.
>>Subject: Re: Decoding enumerated types
>
>
> <clip>
>
>>Now what happens when you receive a value that does not match your
>>specific chosen encoding [which is probably what you had in mind with
>>you question], that is simply a 'decoding error', which means that your
>>decoder should report that the value could not be decoded as the
>>expected type. This is not necessarily a test case error, but will
>>rather cause the receive statement to fail to match. If no default are
>>active and no guard timer exists, that means that you get a deadlock of
>>the receive statement, which _is_ a test case error.
>>
>>BR
>>
>>Stephan Tobies
--
Ina Schieferdecker url:
www.fokus.fraunhofer.de/tip
Fraunhofer FOKUS email:
This email address is being protected from spambots. You need JavaScript enabled to view it.
Kaiserin-Augusta-Allee 31 tel: ++49-30-3463-7241
D-10589 Berlin fax: ++49-30-3463-8241
