ETSI's Bug Tracker - Part 01: TTCN-3 Core Language |
| View Issue Details |
|
| ID | Project | Category | View Status | Date Submitted | Last Update |
| 0007790 | Part 01: TTCN-3 Core Language | Clarification | public | 07-09-2018 12:44 | 06-01-2019 09:23 |
|
| Reporter | Tomas Urban | |
| Assigned To | Gyorgy Rethy | |
| Priority | normal | Severity | minor | Reproducibility | have not tried |
| Status | closed | Resolution | fixed | |
| Platform | | OS | | OS Version | |
| Product Version | | |
| Target Version | | Fixed in Version | 4.11.1 (published 2019-05) | |
| Clause Reference(s) | 16.1.4 |
| Source (company - Author) | STF 548 |
|
| Summary | 0007790: Using fuzzy templates in reception statements |
| Description | The current specification doesn't restrict the use of fuzzy templates in reception statements. There are two rules related to this issue:
16.1.4.k doesn't allow using fuzzy variables in functions called from special places
11.2.j sets a restriction on fuzzy template variables content when these variables are used in deterministic context.
When discussing the details with Jacob, the following problem emerged:
1. There's absolutely no rule for using fuzzy templates that are neither variables nor parameters (static global, local or component-specific templates) in special places listed in 16.1.4
2. There seem to be different understanding of what the expression "shall not be used" mean in 16.1.4 - does it mean that the listed statements cannot be present at all in called functions (making a static check possible) or that they just cannot be executed during evaluation of the snapshot(implying runtime check for most cases)
3. It would be useful to allow adding @deterministic modifier to fuzzy template to mark templates that do recalculate their value, but the result of this calculation is always deterministic. It would be especially beneficial if we decide in favour of the static check. |
| Steps To Reproduce | |
| Additional Information | |
| Tags | No tags attached. |
| Relationships | |
| Attached Files |  CR7790.docx (193,698) 10-10-2018 16:27
http://oldforge.etsi.org/mantis/file_download.php?file_id=3796&type=bug
CR7790-v2.docx (248,796) 11-10-2018 10:10
http://oldforge.etsi.org/mantis/file_download.php?file_id=3799&type=bug |
|
| Issue History |
| Date Modified | Username | Field | Change |
| 07-09-2018 12:44 | Tomas Urban | New Issue | |
| 08-10-2018 13:15 | Jens Grabowski | Assigned To | => Jacob Wieland - Spirent |
| 08-10-2018 13:15 | Jens Grabowski | Status | new => assigned |
| 08-10-2018 13:19 | Jens Grabowski | Note Added: 0015201 | |
| 08-10-2018 13:34 | Jens Grabowski | Note Added: 0015202 | |
| 10-10-2018 16:27 | Jacob Wieland - Spirent | File Added: CR7790.docx | |
| 10-10-2018 16:29 | Jacob Wieland - Spirent | Note Added: 0015243 | |
| 10-10-2018 16:29 | Jacob Wieland - Spirent | Assigned To | Jacob Wieland - Spirent => Tomas Urban |
| 10-10-2018 16:29 | Jacob Wieland - Spirent | Status | assigned => confirmed |
| 11-10-2018 10:10 | Tomas Urban | File Added: CR7790-v2.docx | |
| 11-10-2018 10:18 | Tomas Urban | Note Added: 0015248 | |
| 11-10-2018 10:18 | Tomas Urban | Assigned To | Tomas Urban => Jacob Wieland - Spirent |
| 11-10-2018 11:17 | Jacob Wieland - Spirent | Note Added: 0015251 | |
| 11-10-2018 11:17 | Jacob Wieland - Spirent | Status | confirmed => resolved |
| 11-10-2018 11:17 | Jacob Wieland - Spirent | Resolution | open => fixed |
| 11-10-2018 11:17 | Jacob Wieland - Spirent | Assigned To | Jacob Wieland - Spirent => Jens Grabowski |
| 06-01-2019 08:34 | Gyorgy Rethy | Project | TTCN-3 Change Requests => Part 01: TTCN-3 Core Language |
| 06-01-2019 09:23 | Gyorgy Rethy | Note Added: 0015327 | |
| 06-01-2019 09:23 | Gyorgy Rethy | Status | resolved => closed |
| 06-01-2019 09:23 | Gyorgy Rethy | Assigned To | Jens Grabowski => Gyorgy Rethy |
| 06-01-2019 09:23 | Gyorgy Rethy | Fixed in Version | => 4.11.1 (published 2019-05) |
|
Notes |
|
|
|
STF discussion: "shall not be used" will be changed to "shall not be present"
(Please have a short look on other usages of "shall not be used". Alternative solution: if "shall not be used" is used consistently used in the standard, then we should define the meaning in the definitions clause.) |
|
|
|
|
|
STF discussion: Changes may also be needed in clause 15.0 (add restriction: non-deterministic fuzzy templates cannot be used in deterministic contexts). Changes may also be needed 11.2 (j). |
|
|
|
|
I have added the deterministic modifier to variables (value/template) template definitions, and parameter declarations (value/template). also updated the grammar for this.
Please review |
|
|
|
|
I made some minor corrections:
1. I removed the restriction on actual parameters passed to fuzzy and lazy formal parameters from the section 5.4.1. (formal parameters) and put it with some minor modifications to the section 5.4.2 (actual parameters).
2. Changed the word "value" to "expression" in the restriction on deterministic fuzzy and lazy variable, because IMHO it better describes the content of the RHS (value could also mean the result of evaluation).
Please check and if you are fine with the changes, the CR can be marked as resolved. |
|
|
|
|
|
|
|
|
|