Well to your point that this behavior of corrupted SCT files is new -- I had experienced this problem about a year ago or maybe a little more. It was happening to forms after I had saved them. I was using PEME at that time but not sure if it is/was the problem then. Since then I have reinstalled by system by re-installing WinXP and then software. So I am not sure if that fixed it (a fresh install of VFP) or if it is somehow related to PEME.

It might be related to the timer - it does affect the VFP debugger and possibly could affect other functions. Now that I only have PEME active for a change to the property settings and then close I have not seen this problem. Not sure about other users...




>I think you believe that PEM Editor is doing more than it actually is. If you are not actually clicking on it, or using its IDE features, the only thing that is happening is the timer loop which runs every quarter second. It is checking to see if the currently selected object has changed (if so, it updates).
>
>The timer does not cause anything to be written anywhere. And, it has been there since PEM Editor came out more than two years ago. If this were the cause, then I would certainly expect that with the number of users, somebody else would have experienced this problem (if so, they are always very quick to notify me of it). And that I, as probably the heaviest of its users, would not have just started seeing it within the last month or so.
>
>So that narrows the search to the newer features of PEM Editor ... which is the first place you look when things go wrong, correct? IF PEM Editor is the cause, then the problem should be there.
>
>That doesn't help a lot, of course, since PEM Editor doesn't do anything when you save a form (or class). So, my suspicion is that there is some latent problem that only appears when I try to save (which is why I think that turning off PEM Editor is not likely to be the solution).
>
>It is beginning to look like this might be something like a C5 error that is latent, and doesn't show up until some time after the actual code that caused the problem.
>
>However, based on the other responses so far, I would have to say that PEM Editor is indeed the most likely suspect. I had been pretty sure that the error occurred even when I had used none of the features of PEM Editor -- but, quite frankly, they are so automatic to me now that I am probably not likely to remember that I used them. Time for some logging, then, to see if I can identify which features I used while working on the file before it got damaged.
>
>Ah, this will be a challenge.
>
>
>
>
>>It would seem that something is trying to write to the memo at the same time as VFP -- most likely during the save operation. I would expect that VFP does not commit changes to the SCX/SCT files until the user selects save; otherwise, it would potentially leave changes in limbo in the event of a VFP crash. So I would expect the programmers to have protected the SCX/SCT files by only updating at the save request (applying changes to a copy during edits).
>>
>>We also know that VFP caches a lot of activity -- which may include PEME writes back to the form being edited. These may be cached and the save process of VFP may handle the PEME actions at times in an incorrect sequence due to unknown influences.
>>
>>The amount of activity in PEME may also contribute -- you are doing a lot of things now. I would suggest turning off any activity in PEME that is continuous in nature or at a high request rate. I regularly use PEME but do not keep it active all the time. I use the standard property form for editing property values and PEME for managing the create/rename and access/assign settings and then close PEME. So I have not experienced any problems that you are having (not to say that PEME is the cause -- only a suspect at this point). If you do turn down PEME and the problems seem to go away or at a much less frequency, then PEME would be a strong suspect -- try incrementing the functionality to see if this problem reappears...
>>
>>
>>>Yes, PEME may be part of the problem, of course, as much as I hate to admit it.
>>>
>>>If so, it would be a miserable thing to diagnose, since the problem happens so infrequently, and even when PEME was not used to modify anything on the form. It's hard to imagine, under those circumstances,, what it might be doing to cause the problem.
>>>
>>>And yet, no other possible solution has been suggested either.
>>>
>>>>Try working without using PEME -- possibly something in PEME is interferring with the normal saving that VFP would be doing. At least try stopping PEME before saving.