Bash loops

verse123 · 10-22-2011, 12:20 AM

Hello,

I have a file with 4 columns resembling this:

Col1. Col2. Col3. Col4. Quality
0 1200 3. 500. 40
1201 3000. 1220. 2030. 35
3001 9050. 3400. 5000. 3

If you notice col3 and 4 are contained within col1 and 2 (reading them by row).

I need to create a loop for this. First i need it to check if the quality is above a certain threshold (for example 30). If that is true, then I need the loop to determine if col3 is less than col2 but greater than col 1 and that col4 is less than or equal to col 2. How can I do this?

I would strongly appreciate your help.

crts · 10-22-2011, 12:44 AM

Quote:

Originally Posted by verse123

Hello,

I have a file with 4 columns resembling this:

Col1. Col2. Col3. Col4. Quality
0 1200 3. 500. 40
1201 3000. 1220. 2030. 35
3001 9050. 3400. 5000. 3

If you notice col3 and 4 are contained within col1 and 2 (reading them by row).

I need to create a loop for this. First i need it to check if the quality is above a certain threshold (for example 30). If that is true, then I need the loop to determine if col3 is less than col2 but greater than col 1 and that col4 is less than or equal to col 2. How can I do this?

I would strongly appreciate your help.

Hi,

how about

Code:

while read line;do
 set -- $line
 if [[ $(echo "$5 > 30 && $3>$1 && $3<$2 && $4<=$2" | bc) = 1 ]];then
  echo $line
  # or whatever you need to do
 fi
done < file

[EDIT]
However, I just saw that you have floating point numbers. In this case an awk solution might be more appropriate:

Code:

awk '{if ($5 > 30 && $3>$1 && $3<$2 && $4<=$2){print}}' file

verse123 · 10-22-2011, 03:19 PM

Hi crts,

Your second recommendation worked - thanks a ton!

If I wanted to take this a step further and replace the values in col2 with the values in col3 if they pass the awk criteria would I do something like this:

awk '{if ($5 > 30 && $3>$1 && $3<$2 && $4<=$2){print}}' file | sed '{if ($2=$4) s/$2/$3/} > output

PTrenholme · 10-22-2011, 04:25 PM

Quote:

Originally Posted by verse123

Hi crts,

Your second recommendation worked - thanks a ton!

If I wanted to take this a step further and replace the values in col2 with the values in col3 if they pass the awk criteria would I do something like this:

awk '{if ($5 > 30 && $3>$1 && $3<$2 && $4<=$2){print}}' file | sed '{if ($2=$4) s/$2/$3/} > output

Do it within AWK: awk '{if ($5 > 30 && $3>$1 && $3<$2 && $4<=$2){if ($2==$4) {$2=$3};print > output}}' file