Hi.

I just converted a Focus file to an XFocus. format.

This file contained several cross-reference files which are in Focus format which I did not convert. All seems to be working fine.

I read in help the following:

"JOIN commands among suffix FOC and suffix XFOCUS data sources are supported. Master File cross-references are supported to other suffix XFOCUS data sources only."

Not exactly sure what the second part is. Can someone explain. And the big question is: am I going to have problems with XFocus file and Focus cross-reference file. Should I convert the crosss-reference file to XFocus also?

This message has been edited. Last edited by: <Kathryn Henning>,

My hunch is, any join structure that you need to construct can be build as a Business View (referencing one-segment synonyms each covering a single FOCUS or XFOCUS file). -- I would expect that internally a request against such a BV will be handled as if the structure were coded with JOIN.

I just ran the Focus file with XFocus cross-reference files and it worked fine also.

Can someone please explain what is nment by:

"Master File cross-references are supported to other suffix XFOCUS data sources only"

Does it mean That Xfocus cross-references work with XFcous files only? I just proved it wrong. Very confusing.

Unsupported really means that under some circumstances this is going to fail.

Within the master file declaration there are 2 types of joins, static and dynamic.

The static join, using segtype of KU and KM, requires a REBUILD to be changed. This is because the structure holds the link information to the cross-referenced file in the parent segment of the host file; these are equivalent to the parent-child pointers within a multi-segment single FOCUS/XFOCUS data structure. This is accomplished on the first use and used thereafter for faster cross-referenced data retrieval.

The dynamic join, using segtype of DKU and DKM, can be altered at any time and does not hold link information, but this is retrieved for each record in every report request.

So why would this be unsupported for joins between FOCUS and XFOCUS? Well the pointer(link) space for FOCUS data files and for XFOCUS is different as the XFOCUS has to allow for a larger number of data instances, which is why XFOCUS can support larger data volumes.

So the question is, why would you want to do something that has been documented as unsupported? If you have space issues within the FOCUS data file there are other options available other than XFOCUS (which requires a licence). If you have a licence then you can REBUILD all sources to XFOCUS and live in peace, if there are no space issues, stick with FOCUS.

dz: Master File cross-references are coded in the Master, making the JOIN command unnecessary. I suspect that you did not understand that there were two different ways to implement. If you intend to connect the two types of files via JOIN, it appears that this method is supported. Does this address your concerns about what is and is not supported?

Here is my situation,

I have a very large master file that I converted to XFOCUS.

This master comtain 12 statically joined cross reference files (SEGNAME=SEGNAME1,SEGTYPE=KU,PARENT=PARENT1,CRFILE=FILE1,CRKEY=CRKEY1)

Without converting the cross-reference files to XFOCUS, I'm was able to access all the data from the master and cross-reference correctly.

Now, since I have many master files master files I have to convert dozens if cross reference files, which I'm trying to avoid.

I'm trying to undertsnd why do I need to convert when everything is working fine. Please help.

dz,
If everything is working, leave it as it is.
However, if you have to convert all your file from FOCUS to XFOCUS, you surely can write a procedure that does it for you, using SYSCOLUM.
Good luck!