cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Browse apps to extend the software in the new JMP Marketplace
Choose Language Hide Translation Bar
emmablue
Level III

find all rows with value>50 and subsequent 10 rows

I need to write a script to find all rows in the table with value > 50 and return 10 rows right after those rows. Could someone help me with this script please? Thank you!

1 ACCEPTED SOLUTION

Accepted Solutions
jthi
Super User

Re: find all rows with value>50 and subsequent 10 rows

Without any knowledge of the data you have here is one option you could start with (it looks for values over 120 not 50)

Names Default To Here(1);

dt = Open("$SAMPLE_DATA/Big Class.jmp");

r = As List(dt << Get Rows Where(:weight > 120));

rs = {};

max_rows = N Rows(dt);
For Each({cur_r}, r,
	Insert Into(rs, Index(cur_r + 1, Min(max_rows, cur_r + 10)));
);

show(r);
show(rs);

// r = {2, 4, 7, 18, 30, 37, 39, 40};
// rs = {[2 3 4 5 6 7 8 9 10 11], [4 5 6 7 8 9 10 11 12 13], [7 8 9 10 11 12 13 14 15 16], [18 19 20 21 22 23 24 25 26 27], [30 31 32 33 34 35 36 37 38 39], [37 38 39 40], [39 40], [40]};

-Jarmo

View solution in original post

5 REPLIES 5
jthi
Super User

Re: find all rows with value>50 and subsequent 10 rows

Without any knowledge of the data you have here is one option you could start with (it looks for values over 120 not 50)

Names Default To Here(1);

dt = Open("$SAMPLE_DATA/Big Class.jmp");

r = As List(dt << Get Rows Where(:weight > 120));

rs = {};

max_rows = N Rows(dt);
For Each({cur_r}, r,
	Insert Into(rs, Index(cur_r + 1, Min(max_rows, cur_r + 10)));
);

show(r);
show(rs);

// r = {2, 4, 7, 18, 30, 37, 39, 40};
// rs = {[2 3 4 5 6 7 8 9 10 11], [4 5 6 7 8 9 10 11 12 13], [7 8 9 10 11 12 13 14 15 16], [18 19 20 21 22 23 24 25 26 27], [30 31 32 33 34 35 36 37 38 39], [37 38 39 40], [39 40], [40]};

-Jarmo
jthi
Super User

Re: find all rows with value>50 and subsequent 10 rows

Here are three different options, they might capture incorrect rows due to off-by-one errors

// Option1
Names Default To Here(1);

dt = Open("$SAMPLE_DATA/Big Class.jmp");

// https://leetcode.com/problems/merge-intervals/solutions/3161191/solution/
merge_intervals = function({intervals}, {Default Local},
	// Assumes array is sorted by first index
	stack = {};
	For Each({interval}, intervals,
		If(N Items(stack) == 0 | stack[N Items(stack)][2] < interval[1],
			Insert Into(stack, Eval List({interval}));
		,
			stack[N Items(stack)][2] = max(stack[N Items(stack)][2], interval[2]);
		);
	);
	return(stack);
);

r_start = (dt << Get Rows Where(:weight > 120)) + 1;
r_end = r_start  + 10;

toohigh = Loc(r_start > N Rows(dt));
r_start[toohigh] = N Rows(dt);

toohigh = Loc(r_end > N Rows(dt));
r_end[toohigh] = N Rows(dt);

rs = As List(r_start || r_end);

res = merge_intervals(rs);
// Option2
Names Default To Here(1);

dt = Open("$SAMPLE_DATA/Big Class.jmp");

r_start = dt << Get Rows Where(:weight > 120);
res2 = dt << get rows where(Any(Row() > r_start & Row() <= r_start + 10));
// Option3
Names Default To Here(1);

dt = Open("$SAMPLE_DATA/Big Class.jmp");

r = As List(dt << Get Rows Where(:weight > 120));
rs = {};

max_rows = N Rows(dt);
For Each({cur_r}, r,
	Insert Into(rs, As List(Index(Min(max_rows, cur_r + 1), Min(max_rows, cur_r + 10))));
);
-Jarmo
emmablue
Level III

Re: find all rows with value>50 and subsequent 10 rows

Thank you so much! How can I make rs (all rows selected) into a new data table?

jthi
Super User

Re: find all rows with value>50 and subsequent 10 rows

Select message << Select Rows() to data table.

Names Default To Here(1);

dt = Open("$SAMPLE_DATA/Big Class.jmp");

r_start = dt << Get Rows Where(:weight > 120);
res2 = dt << get rows where(Any(Row() > r_start & Row() <= r_start + 10));

dt << Select Rows(res2);
-Jarmo
txnelson
Super User

Re: find all rows with value>50 and subsequent 10 rows

To make the selected rows into a new data table one just needs to use the Subset Platform.

Names Default To Here(1);

dt = Open("$SAMPLE_DATA/Big Class.jmp");

r_start = dt << Get Rows Where(:weight > 120);
res2 = dt << get rows where(Any(Row() > r_start & Row() <= r_start + 10));

dt << Select Rows(res2);

dtSub = dt << subset( selected rows(1), selected columns(0));
Jim