Retrieve Double precision Values from a database

I have a database table that contains a two columns (X and Y) of double precision data. I wanted to retrieve these data to Matlab. I used the following code for doing this (Note that the ‘conn’ is the connection name.

curs = exec(conn, 'select x, y from Mytable');

setdbprefs('DataReturnFormat','cellarray');

 curs = fetch(curs, 10);

 AA = curs.Data;

Then I wanted to see all the X values and used the following code.

x=[AA{:,1}]'

The result is given below.

x =

1.0e+006 *

   -2.6859
   -2.6859
   -2.6859
   -2.6859
   -2.6859
   -2.6859
   -2.6859
   -2.6859
   -2.6859
   -2.6859

Please note that the result given by Matlab contains the TRUNCATED values of X.

Given below is the list of exact values for X in the database table.

-2685857.83152909 -2685873.29058222 -2685892.47755554 -2685878.77198631 -2685853.37410012 -2685864.72544327 -2685876.71371916 -2685886.58031066 -2685875.18191007 -2685856.28515618

I want to get these exact values to Matlab instead of the truncated values. Please help.

[Devinya Herath]

Given below is the exact values of X in the database table.

-2685857.83152909

-2685873.29058222

-2685892.47755554

-2685878.77198631

-2685853.37410012

-2685864.72544327

-2685876.71371916

-2685886.58031066

-2685875.18191007

-2685856.28515618

Answer 1

[Daniel Shub]

MATLAB is most likely fetching the complete data, but providing you a truncated view.

doc format

you could also check do

x(1) == -2685857.83152909
x(1) -(-2685857.83152909)

ideally the first would return 0, but there might be some small machine precision differences. Ideally, x(1) -(-2685857.83152909) should be the same magnitude as eps(x(1)).