Thread: looking for large dataset

I was woundering where could I find a nice large dataset.  Perhaps 50
thousand records or more
--
Antoine <asolomon15@nyc.rr.com>

That's a very small dataset :)

Chris

On 3 May 2003, Antoine wrote:

> I was woundering where could I find a nice large dataset.  Perhaps 50
> thousand records or more
> --
> Antoine <asolomon15@nyc.rr.com>
>
>
> ---------------------------(end of broadcast)---------------------------
> TIP 2: you can get off all lists at once with the unregister command
>     (send "unregister YourEmailAddressHere" to majordomo@postgresql.org)
>

If you can create a flat file with some rows, it's pretty easy to
duplicate them as many times as you need to get up to 50k (which, as
previously mentioned, is relatively small)

This might not work if you need "real" data - but I started with 67k rows
of real data in my table, then copied them to a temp table,
updated the 3 key fields with previous value + max value,
and inserted back into the original table.  (Just to ensure my new rows
had new values for those 3 fields.)

On Sat, 3 May 2003, Christopher Kings-Lynne wrote:

> That's a very small dataset :)
>
> Chris
>
> On 3 May 2003, Antoine wrote:
>
> > I was woundering where could I find a nice large dataset.  Perhaps 50
> > thousand records or more
> > --
> > Antoine <asolomon15@nyc.rr.com>
> >
> >
> > ---------------------------(end of broadcast)---------------------------
> > TIP 2: you can get off all lists at once with the unregister command
> >     (send "unregister YourEmailAddressHere" to majordomo@postgresql.org)
> >
>
>
> ---------------------------(end of broadcast)---------------------------
> TIP 5: Have you checked our extensive FAQ?
>
> http://www.postgresql.org/docs/faqs/FAQ.html
>

On 3 May 2003, Antoine wrote:

> I was woundering where could I find a nice large dataset.  Perhaps 50
> thousand records or more

I've attached a PHP script called mktestdb that reads in the dictionary at
/usr/share/dict/words,  and inserts a user defined number of rows into a
user defined number of columns.

It's ugly and simple.  Just pipe the output to a text file or psql and off
you go.

usage:

mktestdb tablename [rows [cols]]

default of 1 column and 1000 rows.

It would be easy enough to rewrite this in something more portable if
someone wanted to.