Stability of object ID's in System tables?

[Guest]

I'm trying to create a method for the user to create a reporting chain for
particular clients. I create a list of reports and/or queries that the user
wants to execute if a client is selected, storing in a table called
"auto_outputs" the following:

ID (really not needed, but what the heck?)
Client ID
Output Type (Q or R...used to choose the proper VBA code lines to execute)
Object ID (from MSysObjects)
Order (order of execution)

The system seems to run OK for 1 day of development work, but I'm suspecting
that the MSysObjects.ID field doesn't stay static if I make a modification to
an object. If this is the case, what could I use to permanently refer to a
query or report in a system such as this (seeing as the names could easily be
changed)?

TIA

 

[Michel Walsh]

Hi,


Refer to it by its name? (Note that the question seems more about the
persistence than about the stability).


Vanderghast, Access MVP

 

[Guest]

I'm trying to create a method for the user to create a reporting chain for
particular clients. I create a list of reports and/or queries that the user
wants to execute if a client is selected, storing in a table called
"auto_outputs" the following:

ID (really not needed, but what the heck?)
Client ID
Output Type (Q or R...used to choose the proper VBA code lines to execute)
Object ID (from MSysObjects)
Order (order of execution)

The system seems to run OK for 1 day of development work, but I'm suspecting
that the MSysObjects.ID field doesn't stay static if I make a modification to
an object. If this is the case, what could I use to permanently refer to a
query or report in a system such as this (seeing as the names could easily be
changed)?

Ed, I wiould expect the IDs in system tables to be, if anything, _more_
volitile than the object names. I would imagine that every Compact&Repair
cycle replaces the whole lot if ID valies in the system tables (or. at least,
it _can_).

Better idea: leave the system tables alone, and institute a "we smack
knuckles with rulers" rule if/when people change query/report names without
sufficient thought/authorization.

 

[Guest]

Ed, I wiould expect the IDs in system tables to be, if anything, _more_
volitile than the object names. I would imagine that every Compact&Repair
cycle replaces the whole lot if ID valies in the system tables (or. at least,
it _can_).

ewww for my idea...I've set the db to compact on close...

Better idea: leave the system tables alone, and institute a "we smack
knuckles with rulers" rule if/when people change query/report names without
sufficient thought/authorization.

I hear that...the idea was to allow for a great deal of flexibility, as I am
part of a team of three analysts. I certainly have the best
programming/database skill set, so I suppose maintain things like this might
not be too difficult, as the people I'm programming for now are a step above
the usual user.

I did a little searching last night before going home, but haven't had much
of a chance today to really de-bug it and see what is going on. The query I
had to pull the names sems to work, but in order to have the ability for
users to organize the output list I had to change the list box from being
bound to the query I wrote to using a value list...so at this moment I
suspect that this part of the process is introducing some problems.

I'll post back what I find out (shoudl get a chance to look at it later
today)...