Don't worry, I'll just make a new post everyday. All jokes aside I've looked around and I can't find a solution to this. It seems with every report I need to make there is a problem within InfoAssist that ultimately stops me from finishing a report. Anyways, on to the problem:

I'm making a document with several reports on it. I have to make several hold files using the main data source and a plethora of define fields attached to it, the problem is each time I hold a report all my define fields disappear. Anybody know why this is happening or how to prevent it?
I even have some of these define fields in other reports on the document and you can see them in that reports query but not in the data pane.

Screen Shots:

Before hold


After Hold

This message has been edited. Last edited by: Tamra,

When you create hold files, by default only fields you reference in your report will be HELD. If you want those defined fields to also show up in the hold file make sure you put them in your report.

BabakNYC,

The issue is he creates the DEFINEs for and within the original MSSQL-based data source file request he starts to work with. Down the road, he tries to create a .foc-type hold file based off that source with all the defines. As soon as he does, in the ribbon under the Data tab, he tries to switch back to that original data source request where all the DEFINEs were created in the first place, all the DEFINEs are now missing. Obviously, they aren't in the HOLD file if they weren't added to the HOLD request for such. The DEFINEs are actually missing from the Data panel when viewing the original MSSQL-based data source request columns.

Hopefully that makes more sense. Any ideas?

The DEFINEs are specific to the data source under the Query/Files. So, if you right click on FPP(rf_pallets) and pick Edit, you should see the original query with all the defined fields. If these defined fields require more permanence, they should probably be put into the master file.

BabakNYC,

True. I agree with you. The thing is, when we begin a new multi-request report, we start with a MSSQL data source and then create DEFINEs for it while in Document Mode. We then take from that added upon data source and create our first report that is output within a frame on the document. We then go to create 1 or more intermediate HOLD requests by inserting a new report and dragging fields (whether actual or DEFINEd) into the frame generated for us to populate the fields for the HOLD request. As soon as we finish a HOLD request (output = .foc) and save it, then go back to utilize additional fields from the original MSSQL data source, they are no longer present. I guess they disappear because they were originally "defined" within/for the first report developed for the document? I guess that makes sense, but that makes for a lot of repetitive DEFINE field creation for the IA user. We don't want these DEFINEs in the master file, so I guess we will just have to have our advanced IA users recreate any DEFINEs/COMPUTEs they may want to utilize in successive requests in a multi-request document. With code access you can just copy/paste DEFINEs from one request in your procedure to another if needed. I wonder if you can multiselect and copy/paste fields within IA?? Anyone tried that before??

Thanks BabakNYC for your insights!

I get it. So, if it's not practical to put the DEFINEs in the MSSQL master, maybe another way to deal with this is to put them in Reporting Objects. I haven't tried it but it might be worth a try.

Letting them use a RO to create the DEFINE fields would be nice, but then they would have the ability add their own pre/post-processor code, and we can't allow our advanced users to access code in any way at this point. It would be nice to just let them do such at some point, but we can't currently for security reasons (SQL OUT statements, etc.).

Brainstorming here but maybe you can create the hold file with a batch focexec that has all the defines in it, then let them create their content using this hold file. You could run the batch job as often as the data changes. Almost a quick ETL job that lets them skip the first HOLD file.

Yeah, that is an option, but that would require us to maintain additional processes beyond just master file descriptions. I don't think we want to go that route. We've already got a similar setup that pulls data from several MSSQL source descriptions and match/merges them into a .foc held clustered hierarchy. We are actually trying to replace that datamart because it's caused us a lot of headache.

Good thought though! Thanks!

When you look at the FOCUS code generated by this, the DEFINEs aren't cleared, so they should be there. Have you opened a case with tech support? This might already be a reported issue.

Haven't yet. Been swamped with other things, but was thinking about doing that. Cimmerian doesn't have that ability, so I will have to open a case if we go that route.

Thanks for the suggestion.

Thanks Tamra for getting a case started and tying it to the project in programming at IBI. Really appreciate your help!