>This happens pretty often when the user CTRL-ALT-DELs
Of course. Chances are that many DBFs will be corrupted also.

What I meant though, was that once you zip both the DBFs and CDXs and the next time you refresh just the DBFs in the existing zip file. Well, if you did that, on decompression you'll have older valid CDXs with newer DBFs.
Sorin

>>Strange behaviour. I wonder how the decompression could have corrupted the CDX in such a manner that it would still be a valid CDX and yet miss records. That sounds more like the table was using an older CDX that did not have the latest updates. Is it possible that your zip routine did not refresh the zip file properly ??
>>Just a thought.
>>Sorin
>>
>>>The problem I had: cdx was corrupted, but VFP didn't detect that.
>>>SEEK was working ! But guess what: it didn't find the record...
>>>Alex.
>
>This happens pretty often when the user CTRL-ALT-DELs (is this a new verb in English? :)) or equivalent while in a FoxPro program. But I never saw it for (de)compression.
>
>Vlad