>>>>>>Did you not see message #1173916 ?
>>>>>>
>>>>>This message doesn't explain why I'm getting wrong results using mask for searching.
>>>>
>>>>Naomi
>>>>
>>>>There is no need to explain why. Those functions simply do not work as you wish.
>>>>
>>>
>>>BTW, There IS a reason to explain why. It's either problem with our Server and FAT system, or a bug in ADIR/SYS(2000)/Filer functionality. It would be nice if someone can run similar tests for directories with huge number of files (say, Internet temp directory).
>>
>>The code I posted proves that what you see is the behavior of adir/sys(2000). If the mask is "*E*.JPG" it will find files where E is anywhere. It does not limit it to .JPG.
>>
>>It is not a problem with the FAT since I got the same behavior as you with only 39 files. It looks like a bug to me or my expecation (and yours) is not what adir/sys(2000) is doing.
>
>I wish someone from MS would clarify about masks, which we can use, and how these functions are working with such masks. Because right now I'm quite confused and not sure, what would be a good solution.
>
>AFAIK we should seriously consider an idea of having a table with the files and a Directory monitor. I know I can write Directory monitor with VFP9, but I don't have this version available here.

I'm not trying to be offensive, but you're not being very scientific in your testing.

If you can prove that a single file does not work as you expect with ADIR or SYS(2000) then you know you cannot use those functions without modification no matter if there are 10 files or 100,000. That being the case, you must lower your speed expectations.

C:\*Y*.* should not return "config.sys" right? Well it does.

BTW, if all of the above is true, then Randall's tests were not irrelevant.