I've done lots of tables in UniData with VM levels of data. We only have a couple of files that have SVM level data - a co-worker of mine set at least one up that was working with Crystal reports.
For the multi-values it simply keeps the information that it needs to access it as part of the primary key. I would think that it simply does the same with subvalues. You just appear to have a lot more tables and some of them have some pretty funky primary keys. hth -- Colin Alfke Calgary, Alberta Canada "Just because something isn't broken doesn't mean that you can't fix it" Stu Pickles >-----Original Message----- >From: Geoffrey Mitchell [mailto:[EMAIL PROTECTED] >Sent: Friday, April 16, 2004 12:06 PM >To: U2 Users Discussion List >Subject: Re: UV to Text Conversion Standard? > > >To my knowledge, no. Not in Universe anyway. I *think* (from >the docs, >I've never played with it) that UniData has a way to map sub-values to >ODBC/JDBC, but UniVerse does not. Multivalues work fine, but >it doesn't >go any lower than that. > >Best answer: don't use sub-values. They're evil, anyway. > >On Fri, 2004-04-16 at 12:48, [EMAIL PROTECTED] wrote: > >> Thank you, but what is the exact method to do this? >> I have never seen a "view of SVM level data" that actually works. >> Using ODBC or any other tool >> And by "works" I mean that it understands the relationship >of the SVM data to the VM data and the relationship of that to >the AM data and properly processes table-in-a-table >configurations for editing, etc. >> >> It's one thing to use BY-EXP to understand VM level tables, >but can you really use some tool to understand SVM embedded >data at that second table level? >> Will -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users