>One of our customers is having some time problems with a procedure of closing the period of a particular module. We just identified the step where the function takes the longest which i am posting here
>
>

>
>*:This is a temp cursor with xx,xxx amount of records, since I do not have access to the real data
>*:I just can't really now how many records there are, but most likely there are thousands
>select CurIctfer
>go top in CurIctfer
>
>*:This is basically a wrapper routine for the SQLPrepare() VFP function
>if !PrepareSql(pnHandle, "insert into mySQlTable (cuid,cbatcno,cjeid, cacctid,cdescript,creference," +;
>   "cyear, cpdno, cstatus, csource, dtrs, ntrsamt, ctrstype, ctrsno, cobjno) values (?CurIctfer.cuid, " +;
>   "?CurIctfer.cbatcno, " +;
>   "?CurIctfer.cjeid, ?CurIctfer.cacctid, ?CurIctfer.cdescript, ?CurIctfer.creference," + ;
>   "?CurIctfer.cyear, ?CurIctfer.cpdno, ?CurIctfer.cstatus, ?CurIctfer.csource," +;
>   " ?CurIctfer.cTrs, ?CurIctfer.ntrsamt, ?CurIctfer.cTrsType, ?CurIctfer.cTrsNo, ?CurIctfer.cObjNo)")
>    return .f.
>endif
>
>*:and here is where the inserts are made, goes record by record and inserts each one to the back end
>*:database
>scan
>    if !SetSqlData(, pnHandle)
>        return .f.
>    endif
>ENDSCAN
>
>=UnprepareSql(pnHandle)
>
>

>
>The customer claims this piece of code takes over 2 hours. so I was wondering if We change the Scan..Insert..Endscan and replace it with a append from kinda like this
>
>

>*:I would create a Remote view, NoDataOnLoad, base on the mySQlTable, Table Buffering
>Select mySQlTable
>APPEND FROM CurIctfer
>TABLEUPDATE()
>

>
>do you guys think this would be faster, or slower?
>
>maybe not with a append from, maybe the "CurIctfer" could be a remote view, and then just perfom the Tableupdate(). would this be helpfull?
>
>Once again, i do not have the actual data to do the testing so this is basically Hypothetical
>
>regards

Why not make this cursor updatable and send all updates directly?