Using IF statements with maths where the input is not an integer

Hi All

I've made a few scripts which using GDAL extract the value of a pixel within a given raster. The purpose is to work out the combine value of every pixel. I thought there may have been an easier way to do this but alas!

The code below extracts the pixel value at position X Y. The value is a floating point rather than an integer and thats where my problem lies. The if statement simply exports the value to a txt file. Other scripts take care of the rest ie change values of X Y and summing the resultant txt files.

#!/bin/sh
input=`gdallocationinfo -valonly 210_1515_volume.env X Y`  # The value of the pixel at coordinate X Y
if [ $input -gt 0 ] ;then
	echo $input > colrow.txt
fi

However, I get the following message back when ever my statement is true because my value isn't an integer;

extract2.sh: line 4: [: 2.15787172317505: integer expression expected

Please note the raster file mostly contains zero values, and its only the values greater than zero I'm interested in, which is why I added the expression.

Any ideas how I can alter this code so the IF statement excepts integers? Alternatively if anyone knows of a faster method of quickly summing the values of a raster file, that would be equally useful.

Many thanks in advance for any assistance you can offer.

Andy

Try string comparison:

if [ "$input" != "0" ]; then
...

Yes this works, Thank you Yazu!

Whilst re-examining what I'm trying to do I've realised that I may be over complicating things, I can turn my original raster file into a ascii txt file with one command. So now all I need to do is work out how to sum all the values within a txt file.... I am sure there must be a post on that around here somewhere!

More practice at shell scripting anyway!

---------- Post updated at 02:39 PM ---------- Previous update was at 01:50 PM ----------

Ok maybe this is harder than I thought it might be. It seems its not easy to sum up floating point values in a txt file as many programs are designed for use with integers.

A sample of my txt file;
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2.7223560810089111328 2.4221210479736328125 0 0 2.0098361968994140625 2.9117717742919921875 2.4130835533142089844 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

So I'm trying to find the total of all these values. Ive tried a few things picked up from other forum posts;

Code:

awk '{s+=$1} END {print s}' 210.txt

This produces a value of zero, so I'm guessing its not recognising floating point values.

Code:

i=0
for n in `cat 210.txt`
do
    i=`expr $i + $n`
done
echo $i

But I get a syntax error.

Still searching for other solutions... any help would be appreciated!

---------- Post updated at 04:22 PM ---------- Previous update was at 02:39 PM ----------

Ok I think I've found a solution, in case anyone else has a similar problem in the future.

Create the following script 'sumall.awk'.

#sumall.awk
{   for (i=1; i<=NF; i++) { sum+= $i }   }

END { for (i=1; i<=NF; i++ ) 
	{ print sum } 
          }

then 

awk -f sumall.awk < 210.txt > output.txt

to output the sum of each col into an output.txt file.

then

awk '{ print "a = a + " $1 "; a" }' output.txt | bc -l | tail -1 > 210_final.txt

to print the sum of output.txt into a new file '210_final.txt'

This method is much much faster than my original plan. AWK ftw!

Andy

You really over complicated things. Awk can work with floating numbers (even "old" awk, afaik).

You just need this:

awk  '{ for (i=1; i<=NF; i++)  s+= $i  } END { print s }'

In your first attempt you just add zero with the first field and this field was zero.