I have a list of 1000 potential jurors. I need to be able to select, say 100, random jurors and put them into a separate database. Eventually, the unused jurors will return to the 1000 potential jurors. I have tried using RANDOM but the Random field must be able to be recalculated at will and also I can't get it to create unique integers.

(Geaorge, I sorry. Ididn't explain myself very well.)

You can use the following formula to create random numbers between 1 and 1000 in a calculation:

Round(Random*1000, 0)

The next thing to to would be to use this number to select a potential juror (by ID number I'm assuming) to be added to your new database. Make the ID unique so that you can compare the data (such as a concantenation between the random number and a social security number or other unique data - "136-892-34-2902" etc.

You could then use another sub-routine to compare the potential juror with those already added to your database.

Perhaps the best way would be to set up a relationship between the two files (using the selection date or something like that) so that you can use a global field from the first database in a sub-script nested in the second database. You could use a global field like PotentialJuror in the first database and a Loop sub-script to compare the potential juror to all those selected and add it if it was unique. You could also use copy-paste without having to set up a relationship.

Thanks for the help. I think I can get it now. The concatenation idea really helped. I modified the Round(Random*1000, 0)formula to make sure there wasn't a 0. Now its just a matter of finishing the last 2 paragraphs.

Again, Thanks.

quote:

---

Originally posted by TyGuy:

You can use the following formula to create random numbers between 1 and 1000 in a calculation:

Round(Random*1000, 0)

The next thing to to would be to use this number to select a potential juror (by ID number I'm assuming) to be added to your new database. Make the ID unique so that you can compare the data (such as a concantenation between the random number and a social security number or other unique data - "136-892-34-2902" etc.

You could then use another sub-routine to compare the potential juror with those already added to your database.

Perhaps the best way would be to set up a relationship between the two files (using the selection date or something like that) so that you can use a global field from the first database in a sub-script nested in the second database. You could use a global field like PotentialJuror in the first database and a Loop sub-script to compare the potential juror to all those selected and add it if it was unique. You could also use copy-paste without having to set up a relationship.

---

Here's another approach.

1.)Add a number field to the database. Call it "Selected".

2.)Add a global number field. Call it gRandom

Add a summary field. Call it Total Selected. Sum(Selected)

3.)Write the following script:

Loop

Set field ["gRandom", "Round(Random * 1000 +1, 0)"]

Go to Record/Request/Page "gRandom"

Set Field ["Selected", 1]

End Loop If(Total Selected = 100)

End Loop

Run this script. It will find 100 random records, mark each one with a "1". The summary field will keep count and cause the script to exit after 100 entries (no duplicates this way).

Next, just export these records to a text file or import from your other database.

It worked like a charm. Thanks for the help and the education. I have used Alpha4 and the old DB3. FMP is just another learning curve.

Again, thanks for the help.

quote:

---

Originally posted by Nearo:

Here's another approach.

1.)Add a number field to the database. Call it "Selected".

2.)Add a global number field. Call it gRandom

Add a summary field. Call it Total Selected. Sum(Selected)

3.)Write the following script:

Loop

Set field ["gRandom", "Round(Random * 1000 +1, 0)"]

Go to Record/Request/Page "gRandom"

Set Field ["Selected", 1]

End Loop If(Total Selected = 100)

End Loop

Run this script. It will find 100 random records, mark each one with a "1". The summary field will keep count and cause the script to exit after 100 entries (no duplicates this way).

Next, just export these records to a text file or import from your other database.

---