I would like to read a file into a script, line by line. Each line in the file is multiple values separated by a tab, I'd like to read each line into an array.
Typical bash "read file by line" example;
while read line do echo $line; done < "myfile"
For me though, myfile looks like this (tab separated values);
value1 value2 value3 value4 value5 value6
On each iteration of the loop, I'd like each line to go into an array so I can
while read line into myArray do echo myArray[0] echo myArray[1] echo myArray[2] done < "myfile"
This would print the following on the first loop iteration;
value1 value2 value3
Then on the second iteration it would print
value4 value5 value6
Is this possible? The only way I can see is to write a small function to break out the values manually, is there built in support in bash for this?
You're very close:
while IFS=$'\t' read -r -a myArray do echo "${myArray[0]}" echo "${myArray[1]}" echo "${myArray[2]}" done < myfile
(The -r
tells read
that \
isn't special in the input data; the -a myArray
tells it to split the input-line into words and store the results in myArray
; and the IFS=$'\t'
tells it to use only tabs to split words, instead of the regular Bash default of also allowing spaces to split words as well. Note that this approach will treat one or more tabs as the delimiter, so if any field is blank, later fields will be "shifted" into earlier positions in the array. Is that O.K.?)
If you really want to split every word (bash meaning) into a different array index completely changing the array in every while loop iteration, @ruakh's answer is the correct approach. But you can use the read property to split every read word into different variables column1
, column2
, column3
like in this code snippet
while IFS=$'\t' read -r column1 column2 column3 ; do printf "%b\n" "column1<${column1}>" printf "%b\n" "column2<${column2}>" printf "%b\n" "column3<${column3}>" done < "myfile"
to reach a similar result avoiding array index access and improving your code readability by using meaningful variable names (of course using columnN
is not a good idea to do so).
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With