ETSI's Bug Tracker - Part 01: TTCN-3 Core Language
View Issue Details
0004507Part 01: TTCN-3 Core LanguageClarificationpublic28-11-2008 14:0719-12-2008 13:51
Thomas Deiß 
Ina Schieferdecker 
normalmajorN/A
closedfixed 
 
v4.1.1 (published 2009-06)v4.1.1 (published 2009-06) 
part 1, 7.1.3, 6.2.3
Thomas Deiß, Nokia Siemens Networks
0004507: omit in setof constant
An example to explain equality of set values with set of values uses a set of constant that itself contains 'omit' for one of its fields.

const SetOf conSetOf1 := { 0, omit, 2 };

The usage of 'omit' for within record of and set of values is not defined at all.

It can be interpreted as deleting a value from a record of, set of, but this is nowhere explicitly defined.

It has to be clarified
a) whether omit can be used at all for record of/set of values
b) whether omit or - can be used for constants of record of and set of types.

I suggest to actually exclude case a)

The concept of a partially initialized constant (case b) is somehow strange, but it would not hurt. So one could allow to use '-' in a constant.
from discussion with Elena de Vega, MTP.
No tags attached.
doc CR4507_Omit.doc (54,272) 09-12-2008 11:42
http://oldforge.etsi.org/mantis/file_download.php?file_id=1868&type=bug
doc CR4507_Omit_01.doc (54,784) 18-12-2008 15:46
http://oldforge.etsi.org/mantis/file_download.php?file_id=1916&type=bug
Issue History
28-11-2008 14:07Thomas DeißNew Issue
28-11-2008 14:07Thomas DeißStatusnew => assigned
28-11-2008 14:07Thomas DeißAssigned To => Thomas Deiß
28-11-2008 14:07Thomas DeißClause Reference(s) => part 1, 7.1.3, 6.2.3
28-11-2008 14:07Thomas DeißSource (company - Author) => Thomas Deiß, Nokia Siemens Networks
28-11-2008 14:07Thomas DeißProjectTTCN-3 Change Requests => Part 01: TTCN-3 Core Language
28-11-2008 14:07Thomas DeißAssigned ToThomas Deiß => Ina Schieferdecker
09-12-2008 11:27Ina SchieferdeckerNote Added: 0007591
09-12-2008 11:42Ina SchieferdeckerFile Added: CR4507_Omit.doc
09-12-2008 11:43Ina SchieferdeckerNote Added: 0007592
09-12-2008 11:43Ina SchieferdeckerAssigned ToIna Schieferdecker => Thomas Deiß
09-12-2008 11:43Ina SchieferdeckerResolutionopen => fixed
09-12-2008 11:43Ina SchieferdeckerTarget Version => Edition 4.1.1 (not yet published)
09-12-2008 12:33Thomas DeißNote Added: 0007593
09-12-2008 12:33Thomas DeißAssigned ToThomas Deiß => Ina Schieferdecker
09-12-2008 13:34Ina SchieferdeckerNote Added: 0007598
09-12-2008 13:34Ina SchieferdeckerAssigned ToIna Schieferdecker => Gyorgy Rethy
18-12-2008 15:40Gyorgy RethyNote Added: 0007742
18-12-2008 15:46Gyorgy RethyFile Added: CR4507_Omit_01.doc
18-12-2008 15:46Gyorgy RethyAssigned ToGyorgy Rethy => Ina Schieferdecker
19-12-2008 13:51Ina SchieferdeckerStatusassigned => resolved
19-12-2008 13:51Ina SchieferdeckerFixed in Version => Edition 4.1.1 (not yet published)
19-12-2008 13:51Ina SchieferdeckerStatusresolved => closed

Notes
(0007591)
Ina Schieferdecker   
09-12-2008 11:27   
The STF agreed that omit should not be used record of or set of values.

However, "-" could be used which may lead to partially intialized values.
(0007592)
Ina Schieferdecker   
09-12-2008 11:43   
Please check
(0007593)
Thomas Deiß   
09-12-2008 12:33   
proposal is ok.

Gyorgy wanted to have a look whether the BNF can be changed such that omit is not a value.
(0007598)
Ina Schieferdecker   
09-12-2008 13:34   
See note by Thomas.
(0007742)
Gyorgy Rethy   
18-12-2008 15:40   
OK with one amendment, in clause 7.1.3 only the set of example should be changed but not the set and record examples. In case of sets and records omit is allowed and makes the field initialized. The dash leaves an unitialized field uninitialized, thus is disallowed in constants (all examples in $7.1.3 are using constants. See changes in CR4507_Omit_01.doc