Long transactions

If I'm using a dbaccess script to load data from a file is there a way I can tell the load to commit the transaction after every 'x' rows processed. I'm trying to load a file with about 180000 records. I know I can use dbload, but with the automated tools I have dbaccess works better. My script file is: database db1; load from loadfile.dat insert into table1;commit; Thanks. M

"mpw" <em-2@cox.net> wrote > If I'm using a dbaccess script to load data from a file is there a way I can > tell the load to commit the transaction after every 'x' rows processed. I'm > trying to load a file with about 180000 records. I know I can use dbload, > but with the automated tools I have dbaccess works better. > > My script file is: > > database db1; > load from loadfile.dat insert into table1;> commit; use Jonathan Leffler's sqlreload. sqlreload -d database -t table -i unload_file -n commit_after_n_rows. The default for -n is 1024. I find it to be the best as compared to dbaccess or dbload. Great for automated tools.

Something like #! /bin/ksh split loadfile.data for F in `ls x*` do dbaccess db1 <<-EOF load from $F insert into eric; EOF done mpw wrote: > > If I'm using a dbaccess script to load data from a file is there a way I can > tell the load to commit the transaction after every 'x' rows processed. I'm > trying to load a file with about 180000 records. I know I can use dbload, > but with the automated tools I have dbaccess works better. > > My script file is: > > database db1; > load from loadfile.dat insert into table1;> commit; > > Thanks. M -- Paul Watson # Oninit Ltd # Growing old is mandatory Tel: +44 1436 672201 # Growing up is optional Fax: +44 1436 678693 # Mob: +44 7818 003457 # www.oninit.com #

You could drop the indexes, alter the table to "raw", do the load, and then alter it back to "standard". "mpw" <em-2@cox.net> wrote in message news:lmXMa.564558$vU3.107079@news1.central.cox.net... > If I'm using a dbaccess script to load data from a file is there a way I can > tell the load to commit the transaction after every 'x' rows processed. I'm > trying to load a file with about 180000 records. I know I can use dbload, > but with the automated tools I have dbaccess works better. > > My script file is: > > database db1; > load from loadfile.dat insert into table1;> commit; > > Thanks. M > >

On Thu, 03 Jul 2003 14:47:45 GMT, "mpw" <em-2@cox.net> wrote: If you are running on Unix/Linux you can use split to divide your input file to smaller pices. Here is from 'man split' "The split command reads file and writes it in as many n-line pieces as necessary (default 1000), onto a set of output files. The name of the first output file is name with aa appended, and so on lexicographically. If no output name is given, x is default." So you can write script: split loadfile.dat ABC small_files=`ls ABC*` # make SQL script file echo " " > load.sql for b in $small_files; do echo "load from $b insert into table1;" >> load.sql done # run dbaccess dbaccess db1 load.sql # rm ABC* load.sql Nebojsa >If I'm using a dbaccess script to load data from a file is there a way I can >tell the load to commit the transaction after every 'x' rows processed. I'm >trying to load a file with about 180000 records. I know I can use dbload, >but with the automated tools I have dbaccess works better. > >My script file is: > >database db1; >load from loadfile.dat insert into table1;>commit; > >Thanks. M > ------------------------------------ Remove spam block (DELETE_) to reply