如何仅将文件中的必要行写入 bash 脚本中的数组？

Question

I wrote the following code to read a file and write it in a list using bash script:

index=0
while read line
do
  array[$index]="$line"
  index=$(($index+1))
done < ../../file.xml

However, I only need to write lines into the array if they contain the word "icon". An array element should look like this:

<icon height="36" width="36" density="ldpi" src="res/icon/android/ldpi.png"/>

Could anyone help me to fix this problem?

Answer 1

Trivially, with a condition.

case $line in *icon*) ... do stuff;;

You should probably fix the syntax to use read -r and the index variable is really unnecessary.

array=()
while read -r line
do
  case $line in *icon*) array+=("$line");; esac
done < ../../file.xml

More sensibly, do it all in one fell swoop, in Bash 4+

readarray index < <(grep 'icon'  ../../file.xml)

Probably most sensibly, if the file really is XML, use an XML parser like xmlstarlet to properly identify and extract the structure you want to examine.

readarray index < <(xmlstarlet sel -t -m //icon -c . -n ../../file.xml)

Answer 2

You can use regex:

regex="icon"
index=0
while read line
do
  if [[ $line =~ $regex ]]; then
    array[$index]="$line"
    #If you need cut address uncommend next line and comment before line
    #array[$index]="$(sed 's/.*src=\"\(.*\)\".*/\1/' <<< $line)"
    index=$(($index+1))
  fi
done < ../../file.xml