Hi Alec -

Nice to hear from you and I'll try to answer your questions as best I can.

I was not aware that there was a "default" retrieve spec, simply because I never had any interest in one.  I think it is highly unlikely that I could have applied one even "inadvertently."  BUT, I too, thought of that and was hoping that the other records were still there and I just wasn't accessing them.  So, I looked at the file size and realized there was no way the records could be there.  Also, I sent the database immediately to Ray, so I'm confident that he would have discovered any "stupid mistake" I might have made if, indeed I made one.  Also, since I DO use a default SORT spec, I know and understand the "process" and think that I would have recognized it, as it relates to the default retrieve spec.

I don't remember what, precisely, I was doing in table view, but as memory serves me I was simply checking some information and may have changed a record or two.  In every case, I manually updated the "verified" field to reflect the current date (using control/f5 to enter the date).

Before ending the session, I switched (using the toggle table view option on the menu) back to form view.  I can't swear to it, but I believe that I also retrieved all the records (using a "universal" search with nothing in it) to be sure that all my records were there (in view of my earlier experience with lost records).

I then saved the database to my "backup" copy (which in this case was obviously a big mistake since I then had two databases with only 52 records.  I then exited out of Sesame.

I have NEVER used Save Record and Close Form or Shift-F10 in Sesame.

I do not recall whether I re-sorted the records in table view. I may have, but it's unlikely.

I only use Sesame on stand-alone computers, so a network was not involved.

I don't know how to tell "which exact version of Sesame 2.0.0 were you using (file creation date + time please)."  The "about Sesame" tab only tells me that it is version 2.0.0.  It is the copy of the program that was distributed at the conference.

I am going to make an extra "test" copy of the database to see whether I can replicate everything I did before this occurred.  I'm hoping I can and that I might then be able to provide information that will provide a "clue" as to what is happening.

I want to add, by the way, that there have been several other times that I have used table view without any problem.

Thanks for asking and let's hope I can "force" it to happen again, with my test file.

Don Newlin wrote:
Quote:

Whoa, nice price, I could use a couple of those.  What store is it?

Today Spencer kindly cooperated with me in demonstrating to me online just how he works with his application.

Yes, Spencer has discovered a bug. There's a certain sequence of steps which results in data loss when you back up your application with a result set open in a form at the time after performing several other actions. You need to perform a search, obtain a result set, go into table view, back to form view then run another search without closing the form, then make a backup (using either File / Save As, or Application Utilities / Backup Application).

Clearly this is something that is going to be fixed. In the meantime, my advice is either ..

* If you want to back up your application, close all forms and use Backup Application from the Application Utilities menu,
.. or ..
* Close the application and use Windows Explorer or My Computer to back up the application. You need to copy both the .DB and the .DAT file.

We'll get this fixed and a new release uploaded just as soon as we have this bug fixed and the requisite testing completed.
This affects Sesame 2.0.0 and 2.0.1.  Sesame 1 (all versions) is not affected.

Thank you Spencer for finding this and bringing it to our attention. Thanks also for working with me today to discover the steps to replicate it. Sorry for the trouble it has caused you.

Thanks to Spencer and to Alec for their persistence and updates.   

I am sure that Spencer can now relax and know he has not been imagining things.   

I have sent email to SANE members referencing this posting.
I think this problem is critical enough for Lantica to send a notice to all version 2 users, advising them of the potential problem and the clear workaround instructions that Alec has provided.