Hi Geoff,

I agree that if that is what is meant, it will be useful.

I'm still unclear as to WHAT the change is, and I guess you are too.

I assume that you've noticed that MSKB #Q293638 has been revised to "fixed" by VFP8.
The fix is really a prevention (trap) which certainly is better than actually losing data. An error is emitted on the END TRANSACTION command if there is an inconsistency between the header record count and the calculated vs actual file size. And this is totally separate from the SET TABLEVALIDATE feature.

cheers

>Hi Jim,
>
>This just might be part of an answer to the problem that can arise when you need to deal with alternate data sets. The path to a table (for example), is stored in the DE of a form. While that path/table exists, you cannot access another table of the same name in another data set. Only way is to either clear the stored path, or rename the original directory so that it cannot be found - then it will find it on the path on which you want it to be found.
>If it is, it will be a significant benefit.
>
>Geoff
>
>>The thread title is a subject under the topic "New Classes, Commands, and Functions" in the VFP8 Help.
>>
>>Why does the subject talk about searching for executables when:
>>1) the description refers only to "for a specific procedure or user-defined function (UDF)";
>>2 the SYS(2450) to which it refers says "for a specific procedure or user-defined function (UDF)"?
>>
>>Another thing that is confusing me is that the description for SYS(2450) states that the parameter value of 0 is default and says for it "Search along path and default locations before searching in the application."!
>>This is directly opposite to what I have known it to be in the past and opposite to what is written for search order descriptions in the Help for either the PROCEDURE Command or the FUNCTION Command.
>>
>>So the question is: What IS the change that is being communicated by this entry in the "New Classes, Commands and Functions"?