using command line tools to extract and replace texts for translations - sed

Question

For an application, I have a language file in the way
first_identifier = English words
second_identifier = more English words
and need to translate it to further languages. In a first step I'm required to extract the right side of those texts resulting in a file like ...
English words
more English words
... How can I archive that? Using grep maybe?
Next I'd use a translation tool and receive something like
German words
more German words
that need to be inserted in the first file again (replace English words with Germans) now. I was thinking about using sed maybe, but I don't know how to use it for this purpose. Or, do you have other recommendations?

Answer 1

To do it as you describe would be:
$ cat tst.sh
#!/usr/bin/env bash
tmp=$(mktemp) || exit 1
trap 'rm -f "$tmp"; exit' 0
sed 's/[^ =]* = //' "${#:--}" > "$tmp" &&
tr 'a-z' 'A-Z' < "$tmp" |
awk '
BEGIN { OFS = " = " }
NR == FNR {
ger[NR] = $0
next
}
{
sub(/ = .*/,"")
print $0, ger[FNR]
}
' - "$tmp"
$ ./tst.sh file
English words = ENGLISH WORDS
more English words = MORE ENGLISH WORDS
but you don't need a temp file for that:
$ cat tst.sh
#!/usr/bin/env bash
sed 's/[^ =]* = //' "$#" |
tr 'a-z' 'A-Z' |
awk '
BEGIN { OFS = " = " }
NR == FNR {
ger[NR] = $0
next
}
{
sub(/ = .*/,"")
print $0, ger[FNR]
}
' - "$#"
$ ./tst.sh file
first_identifier = ENGLISH WORDS
second_identifier = MORE ENGLISH WORDS
and I think this might be what you really want anyway so your translation tool can translate 1 line at a time instead of the whole input at once which might produce different results:
$ cat tst.sh
#!/usr/bin/env bash
while IFS= read -r line; do
id="${line%% = *}"
eng="${line#* = }"
ger="$(tr 'a-z' 'A-Z' <<<"$eng")"
printf '%s = %s\n' "$id" "$ger"
done < "${#:--}"
$ ./tst.sh file
first_identifier = ENGLISH WORDS
second_identifier = MORE ENGLISH WORDS
Just replace tr 'a-z' 'A-Z' < "$tmp" or tr 'a-z' 'A-Z' <<<"$eng" with the call to whatever translation tool you have in mind.