Conference azur::mcc

    You seem to have hit the limits of what the Dictionary, DAP and ILV can
    do.  The ACCVIO is occuring in an mcc_ilv_put() routine call inside the
    DAP code which is encoding that very complex datatype definition.  I
    looked at the DAP .COM file which was produced by your .MS file, and it
    looks like the MSL Translator is generating an ILV encoding which is
    about 15 levels deep.  That may be too much for ILV to handle.  In any
    case, the code should never ACCVIO, and I will QAR it.  Meanwhile, I
    think you will just have to try and figure out a simplier structure.
    
    -Matt.
    

    After a thorough analysis of the problem, we have discovered that the
    mcc_ilv_put_cons_begin() routine only supports a nesting of 11 levels
    of ILV encoding, anything greater than that causes an ACCVIO.  The
    datatypes which you define in your MS file are stored in the Dictionary
    as ILV encodings, so you have indirectly stumbled upon this limitation.
    
    Since other datatype information is stored with each nested datatype
    declaration, it works out to about 5 nested datatypes (two ILV levels
    for each datatype declaration) within a complex datatype declaration. 
    We will be extending this at some future release, however, for now, the
    limit to the number of nested datatype declarations in the MCC MSL
    translator is 5.  MSL does not check, nor does DAP check if the limit
    is exceeded, we just ACCVIO on you when you go to load the dictionary
    with it.  We will come up with a nicer way of telling you that you have
    hit the limit for V1.2, unfortunately, V1.1 is already in SSB land.
    
    -Matt.

Thanks for the detailed reply.

I have simplified my TYPE definition by suppressing the subrange definition
for hourtyp and minutetyp ( they now are simple Unsigned8 ).

This allow my definition to be accepted by the DAP.

I have done my ILV encoding for this type and tested it with both
Iconic Map PM and FCL.

I had no problem with the Iconic Map PM, but the FCL goes into an endless
loop when trying to display the output (parsing in input is fine).

My test is a SET, the response contains the value of the attribute.

Follow my session with FCL :


TENERE_$ mana/ent
DECmcc (V1.1.0)

MCC> set gma 1 e 1 wee p = -
_MCC>  ( starttime = 8:00:00 ,-
_MCC>    endtime = 8:00:00 ,-
_MCC>    weekmask = ( daysofweek = ( monday , sunday ), -
_MCC>      intvlsofdays = (( start = ( hour = 8 , minute = 0) ,-
_MCC>                        end = ( hour = 18 , minute = 0) )) ))


*****************************************************
*     FCL PM Arguments before call_function:        *
*****************************************************


VERB code:      2
PARTITION code: 4
AES dump of ENTITY IN:

depth=2 class code= 9501 instance = 1 class code= 125 instance = 1

ILV dump of IN_P:

[  0 ] (
    [  1 ] (
        [  6 ] (
            [  1 ]             0d
            [  3 ]             00
            [  4 ] (
                [  1 ]                 60 61 69 4e b3 da c9 01 ff ff ff ff ff ff
 60 1a
                [  2 ]                 00 e8 6a 4e b3 da c9 01 ff ff ff ff ff ff
 60 1a
                [  3 ] (
                    [  1 ] (
                        [  1 ]                         01
                        [  1 ]                         40  -- @
                        )
                    [  2 ] (
                        [  1 ] (
                            [  1 ] (
                                [  1 ]                                 08
                                [  2 ]                                 00
                                )
                            [  2 ] (
                                [  1 ]                                 12
                                [  2 ]                                 00
                                )
                            )
                        )
                    )
                )
            )
        )
    )
ILV dump of IN_Q: in_q is NULL
TIME SPEC is: 0, NOW

**********************************************
FCL PM Arguments on return from call_function:

CVR returned:
%MCC-S-RESPONSE, success with response reply

ILV dump of OUT_P:

[  1 ] (
    [  1 ] (
        [  6 ] (
            [  1 ]             0d
            [  3 ]             00
            [  4 ] (
                [  1 ]                 60 61 69 4e b3 da c9 01 ff ff ff ff ff ff
 60 1a
                [  2 ]                 00 e8 6a 4e b3 da c9 01 ff ff ff ff ff ff
 60 1a
                [  3 ] (
                    [  1 ] (
                        [  1 ]                         01
                        [  1 ]                         40  -- @
                        )
                    [  2 ] (
                        [  1 ] (
                            [  1 ] (
                                [  1 ]                                 08
                                [  2 ]                                 00
                                )
                            [  2 ] (
                                [  1 ]                                 12
                                [  2 ]                                 00
                                )
                            )
                        )
                    )
                )
            )
        )
    )
AES dump of ENTITY OUT:

depth=2 class code= 9501 instance = 1 class code= 125 instance = 1


GMA_TEST 1 EVENTLOG 1
AT  4-MAR-1991 10:11:14 Characteristics


Examination of attribute shows:
                         Weekly Package = (       starttime =  4-MAR-1991
                                                              08:00:00.63,
                                                    endtime =  4-MAR-1991
                                                              08:00:00.64,
                                                   weekmask = (      daysofweek
= (


[ Display blank lines forever ... ]


%MCC-E-ABORTCTRLY, image aborted by Ctrl\Y
TENERE_$


Regards

Pierre.

See notes 759.3, (I think...or one of the other replies to 759).  The Console 
has some limitations around datatypes that MSL and DAP do not.  Jill Callander
can probably say what they are.

Ruth Kohls

Hi,

Can you do me a favor by doing the test in .3 with your terminal width set
to 132?  

Or do a 

MCC> USE INDENTATION 12  (or something similarly low)

before doing your SET.

I think FCL is fumbling while trying to maintain an indentation level that
exceeds the screens width.

thanks,

Pete

Thanks, it worked !!!

With a 132 col. terminal I got :

GMA_TEST 1 EVENTLOG 1
AT  5-MAR-1991 09:56:40 Characteristics

Examination of attribute shows:
                         Weekly Package = (       starttime =  5-MAR-1991 08:00:00.33,
                                                    endtime =  5-MAR-1991 08:00:00.34,
                                                   weekmask = (      daysofweek = ( Monday,
                                                                                    Sunday ),
                                                                   Intvlsofdays = { (           start = (            hour = 8,
                                                                                                                   minute = 0 ),
                                                                                                  end = (            hour = 18,
                                                                                                                   minute = 0 ) ) }
) )

So it is not a FCL limitation on ILV depth, but on terminal width ? ^)

I have tried with 80 col. and USE INDENTATION 20 (min. value) with no success.

Thanks

Pierre

thanks.

>So it is not a FCL limitation on ILV depth, but on terminal width ? ^)

Correct.  FCL's wrapping/indentation algorithm isn't smart enough to know what
to do when the indentation level is greater than the screen's width.  We need
a modulo function in there somewhere.

that's quite the record structure you've got there.

    
    I love it when I go away on vacation and all these new qars pop
    up....
    
    This has been entered in the MCC_INTERNAL data base as qar 561.
    
    Thanks for finding it.