[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference azur::mcc

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

361.0. "Home brewed AM: FCL ok, Iconic map NOK" by CCIIS1::ROGGEBAND (_ �hili��e _) Thu Sep 27 1990 13:07

    Hello,
    
    I hav encountered an interesting problem today with an home-brewed
    AM. 
    
    This AM worked fine if EFT 1.0 + the IFT DECwindows interface, I
    have an Icon which i could place on the map, and using the Icon,
    I could read / set attributes.
    
    Along comes the latest IT1.1.0 version. I can still place my Icon on
    the map (in other words, I suppose the "SHOW ALL ID" works once) but
    when I try to read or set attributes from the Iconic MAP, I get a
    window with "Management Window Message : ILV Mode inconsistent with
    attributes". The interesting thing is that I can set / show whatever
    I like with the FCL PM. 
    
    Any ideas on why the two PM's might expect different ILV modes ?
    
    Cordialement,
    
    Philippe.

T.R	Title	User	Personal Name	Date	Lines
361.1	REQUEST: try a relink	GOSTE::CALLANDER		`Thu Sep 27 1990 17:58`	16
	Well the mode error comes from using the wrong mode on an ILV PUT. The FCL and Iconic Map share an awful lot of the decode logic so I am surprised that one works and the other doesn't. But...there is a possibility. When did you last relink you AM, and with what version? With the last version of the 1.0 kit, and the latest IFT release we have finally stablized the kernel_shr vector tables. Your problem may simply require you to relink with the new kernel share that you just installed, and then to reinstall your AM. Why don't you start by giving that a try if you haven't yet, if you have, then I have to go back to the drawing board or see if someone else has an idea.
361.2		CCIIS1::ROGGEBAND	_ �hili��e _	`Mon Oct 01 1990 06:54`	7
	Jill, I have relinked the AM, re-enrolled it and still have the same problem. I'll try to run some tests with the debugger later today or tomorrow morning to try and clarify the context. Philippe.
361.3	TRY: ilv dump and entry/exit points	GOSTE::CALLANDER		`Mon Oct 01 1990 12:02`	6
	Do you have a debug logical of your own in your module? If you do it would be interesting to see an ILV dump at your entry and exit points. jill
361.4	Problem is in IN_Q	CCIIS1::ROGGEBAND	_ �hili��e _	`Mon Oct 01 1990 13:18`	21
	Jill, I've narrowed it down somewhat by using the debugger both with the FCL and the Iconic map. The problem seems to be in the validation of the In_Qualifier parameter, not the IN-P or out-P as I expected. I have not used any qualifiers, but when my AM is called from the FCL PM, In_Q contains a null_pointer. When it is called from the Iconic Map, the In_Q contains a value, and the subsequent validation (ie MCC_GAB_AM_INIT_READ_ILVCTX) fails. I'll try to delve deeper into the content of the IN_Q. Meanwhile, if no qualifiers were specified, why would the Inconic Map put something in there ? Amicalement, Philippe.
361.5	In Domain always sent	TOOK::HAO		`Mon Oct 01 1990 13:32`	6
	When doing operations from the Management Window, the Iconic Map PM always sends down the IN DOMAIN qualifier. This is simply the domain whose map you're working from. Christine
361.6	INFO - to test it from FCL...	GOSTE::CALLANDER		`Mon Oct 01 1990 14:03`	9
	to see if you have the same problem in the FCL simply append the following to your command. , IN DOMAIN <a_valid_domain_name> ** actually it will parse and send down ANY domain name, it doesn't do any validation on the name.
361.7	Thanks, it was the IN DOMAIN qualifier	CCIIS1::ROGGEBAND	_ �hili��e _	`Tue Oct 02 1990 07:52`	12
	Christine, Jill, The problem was indeed the IN DOMAIN qualifier, the initial IFT did not use this qualifier. I have added a check for MCC_K_QUAL_IN_DOMAIN in my validation module, that solved the problem and I can now see money-machines on my map again ! Cordialement, Philippe.