Need to develop a Quality checking software Urgent!!!

[Farooq Ali]

Hello everybody
I need some help in devloping a qualiyt control software.
My company records are kept in a lot on excel spreads sheets.
I need to create a software which picks some records at random for checking from these files.
But how do I load these separate (lot of) files & then pick out random records from them on the whole.
Like for example there are 10000 records in say 5 files 200 each
I need a random sample of 70 from the 10000 files ( not from indivual files)
Any suggestions as to how I do it??
Your help will be greatly appriciated
Farooq

[Cindy Glass]

If you give each file input an identifier (say number them from 1 to 5, or it variable name, whatever) and each record a number, then the combination of the file number and record number uniquely identifies each record.
Load each identifier up into a SortedMap, giving them each a key which is a random number. When the random number key scrambles them up, just pull the top 70 records from the stack and deal with them.

[Farooq Ali]

sorry cindy I didn't get it
First of all how should I read from the excel files??
should it be like connecting thru jdbc & extracting the recods from all of the files??
I wan to simultaneously access somewhere like 10-15 separate excel files & treat them as one??
if I first read them thru jdbc & save the data its going to be quite cumbersome don't you think
please let me know
Farooq

[Cindy Glass]

Yes, you can use JDBC to get at the excel files, there are quite a few discussions of this in the JDBC forum. Here is a sample one: http://www.javaranch.com/ubb/Forum3/HTML/000342.html
However I wouldn't think that you need to read in ALL of the records from the excel files, just create your random selection mechanism, and only read the selected records.

[Farooq Ali]

thanks Cindy...
but still without JDBC how do I read MS EXcel files in my java program??
Farooq

[Mike Curwen]

You don't.

Unless you want to study the *.xls file format, and use a whole bunch of byte offset manipulation to find rows and columns.