Problems ... again ... [message #40150] |
Sun, 18 July 2004 16:03 |
rats
Messages: 11 Registered: June 2004
|
Junior Member |
|
|
Hi,
I am having some problems with a huge file that I have to manipulate
...
This file has more than 5 million rows by 4 columns ...
One of this columns is the ATTRIBUTE, and the other ones are: X, Y and
Z (all values in meters).
First I have separated this file in 4 new ones ... each one for one
attribute.
But the files are still huge.
I have to make some statistics with the points in these files, but the
USER have to specify different SPACINGS ... so the statistic will
exist to make some MEAN, MEDIAN, etc... with the points inside this
spacing.
What I have done till now is:
For example: Attribute 3
1) I have created a NEW INDEX COLUMN in this file ... Beginning with
value 1 and the last value will be the last line number.
2) Created 2 ARRAYS . One with the INDEX column, X column and Z
column.
The other one with the INDEX column, Y column and Z column.
I have sorted this two arrays by the X and Y columns, so that this two
columns begin with the lowerst value. I did that to find the same
points inside the spacing (using some intersection with the two arrays
by the INDEX COLUMN)
3) I am "moving" the "grid" by the spacing value that the user choose
... so if he chooses 2 meters for the spacing: I first FIX the X axis
by two meters and then I "walk" thru the Y Axis (2 by 2 meters) and
then I make the statistic for the "intersection points" inside the
grid(2x2m) till it reaches the end. After that I move another 2 meters
in the X axis and do the same thing for the Y axis ... I do that till
the end of the X and Y axis.
I know that this method is too slow ... it works ... but it takes a
looong time to terminate.
Any ideas ?
Thank you very much ...
Rafael
|
|
|