topic Re: Select rows #x to #y in a table in Discussions

Select rows #x to #y in a table

markschahl — Thu, 20 Jun 2013 14:24:49 GMT

A colleague that is a new JMP user came to me with the following question:

"How to select a given range of rows in a data table?"

Example: I want to select rows 10000-35000.

We searched through the help, manuals, and I searched jmp.com and this forum. Could not find the answer. With a large data table, you can't easily select the first row and then scroll to the last desired row.

Thanks!

Afterthought to the developers: if this can't be done, how about adding row number to the Data Filter dialog or Select By Row Number to Row Selection?

Re: Select rows #x to #y in a table

peterw — Tue, 18 Oct 2016 20:13:12 GMT

You can add another column, and add a formula with the function "row()" - then you can use the Data Filter in the Rows menu item to select the rows you want. See screenshot below

Re: Select rows #x to #y in a table

pmroz — Thu, 20 Jun 2013 15:23:57 GMT

You can also use a JSL script to do this:

dt = open("$sample_data\Big Class.jmp");

select_matrix =index(20,35);

dt << select rows(select_matrix);

Change dt to point to your table, and change the rows mentioned in the call to index to the rows you desire. E.g. for your situation you would say:

dt = open("your table here");

select_matrix =index(10000,35000);

dt << select rows(select_matrix);

Re: Select rows #x to #y in a table

mtl — Thu, 20 Jun 2013 15:33:33 GMT

One possible method, at least in the BASE SAS world, would be like below. Don't know if it directly applies to your case in JMP:

242 data a;

243 do i=1 to 10;

244 row=i;

245 output;

246 end;

247 run;

NOTE: The data set WORK.A has 10 observations and 2 variables.

NOTE: DATA statement used (Total process time):

real time 0.00 seconds

cpu time 0.00 seconds

248

249 proc sql noprint;

250 create table b as

251 select *, monotonic() as rownum

252 from a

253 where monotonic()<=7 and monotonic()>=4;

NOTE: Table WORK.B created, with 4 rows and 3 columns.

254 quit;

NOTE: PROCEDURE SQL used (Total process time):

real time 0.00 seconds

cpu time 0.00 seconds

255

256 data _null_;

257 set b;

258 put row=;

259 run;

row=4

row=5

row=6

row=7

NOTE: There were 4 observations read from the data set WORK.B.

NOTE: DATA statement used (Total process time):

real time 0.00 seconds

cpu time 0.00 seconds

Re: Select rows #x to #y in a table

paramaribo — Thu, 20 Jun 2013 17:44:13 GMT

At least in Base SAS you can do this below. Just not sure in JPM.

Here is an example code:

Data subset_of_cars;

row_no +1; /* Just for demo/checking purpose*/

set sashelp.cars;

if _n_ >= 100 and _n_ <= 200;

run;