Hi Chaim,
>Can I understund clearlythat never VFP will not go to other line because index problem?
Can I guarantee this? No, there's always the chance that while attempting to read the index, VFP updates memory adresses that happen to contain the next operation pointer in VFP. But I've never seen such a case. When you have index corruption some where (in memory, on disk, etc.), VFP typically finds wrong records, trashes data during updates , raises errors or crashes.
This can cause all kind of side effects such triggers being executed for the wrong records, relations pointing to wrong records, etc. In consequence VFP would execute code that it won't if the index were correct. You might also run into an error handler that immediately returns. In this case, VFP executes some lines and skips others (that is, all that raise an error). In other words, there are many possibilities to make it look like VFP is randomly executing code or jumping to certain lines.
However, I've never seen that VFP really suddenly moves the instruction pointer to a different line because of index corruption.
--
Christof