LinuxQuestions.org - [SOLVED] bash - merging strings (perhaps with sort

- Linux - Newbie (https://www.linuxquestions.org/questions/linux-newbie-8/)

- - bash - merging strings (perhaps with sort | uniq) (https://www.linuxquestions.org/questions/linux-newbie-8/bash-merging-strings-perhaps-with-sort-%7C-uniq-844519/)

bash - merging strings (perhaps with sort | uniq)

Hi everyone,

I've used this site for a while an unregistered user. This is my first post.

I have multiple strings (eg. say two, firstLIST=(0 1 2) and secondLIST=(2 3)) and want to create a single string composed of their unique sorted elements. For the sample strings above, I'd like to build masterLIST=(0 1 2 3).

I suppose I could write the elements of firstLIST and secondLIST to files

Code:

echo ${firstLIST[@]} > firstFILE

echo ${secondLIST[@]} > secondFILE

then use

Code:

sort firstFILE secondFILE | uniq > masterFILE

as this gives me a file populated with the elements I'm after, but I'm not sure how to read the elements back into masterLIST... and it doesn't seem "right" to create files to accomplish this. Is there a way to do this by manipulating the strings ${firstLIST[@]} and ${secondLIST[@]} directly?

The closest I've come (not close at all) is

Code:

masterLIST=${firstLIST[@]}" "${secondLIST[@]}

but masterLIST built this way has only one element

Code:

$echo ${masterLIST[@]}

0 1 2 2 3

$echo ${#masterLIST[@]}

1

and I don't have access to the individual digits to then try to figure out how to remove duplicates.

Thanks for any help,
Chris

Here's one solution (OK for the simple numbers in the example, would need tweaking if the lines contained whitespace):

Code:

#!/bin/bash



# Load arrays

firstLIST=(0 1 2)

secondLIST=(2 3)

masterLIST=( ${firstLIST[*]} ${secondLIST[*]} )



# Generate results array

while read

do

    newLIST+=( $REPLY )

done <<< $(

        for (( i=0; i < ${#masterLIST[*]}; i++ ))

        do

            echo "${masterLIST[i]}"

        done | sort | uniq

    )



# Show results

for (( i=0; i < ${#newLIST[*]}; i++ ))

do

    echo "$i: ${newLIST[i]}"

done

Another alternative:

Code:

#!/bin/bash



# Load arrays

firstLIST=(0 1 2)

secondLIST=(2 3)

tmpLIST=( ${firstLIST[*]} ${secondLIST[*]} )



for x in ${tmpLIST[*]}

do

    for (( i=0; i< ${#newLIST[*]}; i++))

    do

        (( x == ${newLIST[i]} )) && continue 2

    done



    newLIST+=($x)

done



echo "${newLIST[*]}"

And another slight change I thought of on the way home :)

Code:

#!/bin/bash



# Load arrays

firstLIST=(0 1 2)

secondLIST=(2 3)

tmpLIST=( ${firstLIST[*]} ${secondLIST[*]} )



for x in ${tmpLIST[*]}

do

    [[ ${newLIST[*]} =~ " $x " ]] || newLIST+=($x)

done



echo "${newLIST[*]}"

solutions for sorting unique elements of bash arrays

grail and catkin,

Thanks for your solutions! I'd cobbled together the following inferior solution of writing out to files, working with the files, then reading into an array,

Code:

count1=${#firstLIST[@]}

count2=${#secondLIST[@]}



for (( i=0; i<$count1; i++ )); do

  echo ${firstLIST[$i]} >> tmp_file

done

for (( i=0; i<$count2; i++ )); do

  echo ${secondLIST[$i]} >> tmp_file

done



sort -u tmp_file > tmp_file

masterLIST=( `cat tmp_file` )

but much prefer how you have worked directly with the arrays.

Thanks again,
Chris