Having trouble with bash syntax logic

[acampbe2]

0 down vote favorite
share [g+] share [fb] share [tw]


I need help reading a txt file line by line, and then looping through each element in bash script. After hitting a element I need to cut the # of the column represented by the current numerical element, and then assign it to a file. In the end I need to paste all these files together. The columns are in a separate file. The first txt file reads as such

1,6,62,6

7,7,52,24

3,145,25,69

Then for example I need to cut column 5 in a separate file, and then assign to a new file. One the line is done the process starts again.This process would be easy in python or java, but I'm just very new to bash script.

This is a sample column:

1:45

1:56

1:56

1:69

ccData_t.nnv
while read line; do
cut -f2 ccData_t.nnv >p
for i in $(line);

do

cut -fi+1 ccData_t.nnv |sed "s/^/"i":/" >ti
echo paste p t1 t2 t3 t4 |grep -v pro >svm_train_model1.tsv
done < FILE

In the end I have to have all columns appended together as such.

1:45 6:45 62:45

1:56 6:34 62:23

1:56 6:32 62:52

1:69 6:12 62:123

[disciple]

I see you have another thread:
http://www.murga-linux.com/puppy/viewtopic.php?t=78947

Or is this a howto after all?
If it isn't, it isn't immediately obvious exactly what you're asking for help with. Could you clarify?

[Flash]

It looks to me like acampbe2 copied that from another forum and pasted it in this forum. He even left some junk from the other forum at the top.

[sunburnt]

Hi acampbe2;

Code: Select all

for L in `cat File.txt`
do
( Do what you need to...)
done

paste File-1.txt File-2.txt > Final-File.txt

[Karl Godt]

Your problem is probably that column 5 is not always there :

1,6,62,6 has 4 columns ...


while read line;do
column_5=`echo "$line" |cut -f 5 -d ','`
if [ "$column_5" = '' ];then
column_5='0'
fi
done <FILE

Code: Select all

echo "
1,2,3,4
2,3,4,1,
4,3,2,1,,
5,4,3,2,1
" >test_file.lst

while read line;do [ "$line" ] || continue;column_5=`echo "$line" | cut -f 5 -d ','`; [ "$column_5" ] || column_5=0; echo "$line | $column_5"; done <test_file.lst

1,2,3,4 | 0
2,3,4,1, | 0
4,3,2,1,, | 0
5,4,3,2,1 | 1

[Moose On The Loose]

I edited the Karl Godt comment a little:

Your problem is probably that column 5 is not always there :

1,6,62,6 has 4 columns ...

Code: Select all

while read line;do
column_5=`echo "$line" |cut -f 5 -d ','`
if [ "$column_5" = '' ];then
column_5='0'
fi
done <FILE

There is a slightly easier way to deal with the issue of too few columns.

Code: Select all

column_5=`echo "${line},0,0,0,0,0,0,0" |cut -f 5 -d ','`

This way cut always has the needed column.

[Karl Godt]

I edited the Karl Godt comment a little:

Your problem is probably that column 5 is not always there :

1,6,62,6 has 4 columns ...

Code: Select all

while read line;do
column_5=`echo "$line" |cut -f 5 -d ','`
if [ "$column_5" = '' ];then
column_5='0'
fi
done <FILE

There is a slightly easier way to deal with the issue of too few columns.

Code: Select all

column_5=`echo "${line},0,0,0,0,0,0,0" |cut -f 5 -d ','`

This way cut always has the needed column.

But watch out $line has an entry. If $line would be NULL , then cut and gawk -F "," would treat ",1,2,3,4,5" differently . Field 1 would be NULL then and field 5 would get the value '4' .