Thank you for your post.
If you are unable to find a range of dates in a database file with FileMaker Pro 8, then the file is damaged. The Recover command is not a cure-all, but it will try to repair a file. The best solution is to go to a previous backup where the find worked properly, and import the data from the damaged file.
FileMaker Pro 10 is more strict reading files, so it is very likely the damaged file cannot be salvaged in FileMaker Pro 10.
If you do not have a backup file, open the file in FileMaker Pro 8 and export the data to a Merge text file (comma-separated file that includes the field names). You can then import this information into a new database file and it will create the field names for you. You will have to recreate the layouts again, but at least you have the data.
Let me know if you need clarification for any of the above steps.
Thank you very much for your reply.
As you recommended, I exported to *.mer file and imorted the file. About 20,000 among 31,000 records were imported. When exporting to *.mer file, there was no UNICODE options, whereas csv exporting has a UNICODE option. So, all the exported Korean characters became illegible. When importing from *.mer file, there was a Unicode option, but it is of no use because the exported Korean characters were already illegible characters.
I tried adding a new field named RecordSelected, which is a calcuated field. Calcualtion is Case (OPUdate >= "2009-01-01" and OPUdate <= "2009-03-01", 1, 0). And then, I find the records which has RecordSelected = 1. It seems to work withut error message. Temporarily, I can select out records for analysis, but eventually I do have to recover file further. Deleting all the records and trying to make a clone copy does not work.
When exporting to *.mer file, there was no UNICODE options, whereas csv exporting has a UNICODE option. So, all the exported Korean characters became illegible.
Try exporting as XML, using FMPXMLRESULT grammar and no stylesheet.