| Title: | DECmcc user notes file. Does not replace IPMT. |
| Notice: | Use IPMT for problems. Newsletter location in note 6187 |
| Moderator: | TAEC::BEROUD |
| Created: | Mon Aug 21 1989 |
| Last Modified: | Wed Jun 04 1997 |
| Last Successful Update: | Fri Jun 06 1997 |
| Number of topics: | 6497 |
| Total number of notes: | 27359 |
I have a question about Constructor data types.
In SRM their definitions are given as:
MSL Syntax
<set-construct>::= "SET OF" <base-type>
or
<sequence-construct>::= "SEQUENCE OF" <base-type>
and so on. The question is: what is the <base-type>? May it be a constructor
data type or not?
For example, may I have the following:
TYPE NodeInfo = 3100 RECORD
phase = 11 : Integer16;
type = 12 : Integer16;
phase4 addr = 13 : Phase4Address;
phase4 name = 14 : Phase4Name;
full name = 15 : FullName;
cluster alias = 16 : Latin1String;
END RECORD;
TYPE PathHop = 3101 RECORD
node = 21 : NodeInfo;
circuit = 22 : SimpleName;
END RECORD;
TYPE Path = 3200 SEQUENCE OF PathHop;
and will the PM be able to handle this?
Thank you for your answers,
Gene
| T.R | Title | User | Personal Name | Date | Lines |
|---|---|---|---|---|---|
| 1890.1 | VERNA::V_GILBERT | Fri Dec 06 1991 15:18 | 7 | ||
Gene, In your example, you have a sequence of record. The record has another record as one of its fields. The Iconic Map should be able to handle this. Verna | |||||
| 1890.2 | FCL? | SWORD1::ES | Eugene Shvartsman | Fri Dec 06 1991 16:19 | 3 |
What about FCL?
Gene
| |||||
| 1890.3 | I believe FCL is ok | TOOK::MATTHEWS | Mon Dec 09 1991 09:34 | 6 | |
It has been our experience that the FCL is more accepting than
the IMPM. When doing the Circuit AM, we found that the FCL
handled some of our esoteric stuff pretty well but that the
IMPM needed some additions.
wally
| |||||
| 1890.4 | complex datatypes in the FCL | TOOK::CALLANDER | MCC = My Constant Companion | Tue Jan 14 1992 10:50 | 16 |
RE -.1
What Wally means is that the FCL shouldn't have a problem with it.
Please keep in mind when defining "esoteric" types (sequences of
records of records) that a record can only be entered by entering ALL
fields in the record, and that these complex types do not always appear
at the user interface in a "user-friendly" fashio. You should check out
how the arguments look from the Iconic Map and how they are entered
from the FCL before deciding on complex definitions.
Note: to test a construct your xM doesn't have to really support the
command. Put in the MS and give the command a shot, you will simply get
a dispatch entry not found error if we parse the input successfully.
jill
| |||||