Re: "Load from" SQL Statement ...

A very usefull command in this case is dbload. I do not know if you are using transaction logging, but this shouldn't make much of a difference. The dbload command tells you how many rows have been loaded into the table (depending on the number of records you commit at a time) and you can also skip x number of lines in a load file the second time you start the load - which you will use when a load has crashed dbload -d <database> -c <loadscript> -n 1000 -k -l <errorlog> -n 1000 - commit every 1000 records to the database -k - locks the table exclusively If you look at the syntax, you will find another option to skip x number of lines in the loadfile when you need to. I hope this helps Rick Nath <rnath@wrbc.com> wrote in article <01bd7f85$70eb5210$2fdda8c0@bis005408>... > I was wondering if anyone has stumbled upon a way in which to continue > processing a 'load' file even after it has encountered an error (such as > duplicate key) within the file? We have process written that takes a > delimited file and attempts to load. If there are no errors everything is > fine. However if there is an error such as a duplicate key, the loading of > the entire file stops and we are forced to go through the entire file to > find the bad record, remove it, and restart the process. Other options > such as SPs and 4GL are welcome. > > Rick Nath >