Bash script awk - unix

Question

I am new to Bash scripting. I am struggling to understand this particular line of code. Please help.
old_tag = awk -v search="$new_tag" -F" " '$1==search { a[count] = $2; count++; } END { srand();print a[int(rand()*(count-1))+1] }' $tag_dir/$file
[ -z "$new_tag" ] && break

Answer 1

The code seems to be incorrect. With old_tag = awk the code tries to out the results of the awk command in the var old_tag. An assignment of a var should be done without spaces around the =, and the command should be enclosed in $(..). It might have been backtics in the original code, these are depreciated and backtics are used for formatting in SO.
Your question would have been easier to answer with an example inputfile, but try to explain assuming inputlines like
apple x1
car a
rotten apple
tree sf
apple x5
car a4
apple x3
I switched old_tag and new_tag, that seems to make more sense.
new_tag=$(awk -v search="$old_tag" -F" " '
$1==search { a[count] = $2; count++; }
END { srand(); print a[int(rand()*(count-1))+1] }
' $tag_dir/$file)
[ -z "$new_tag" ] && break
This cod tries to replace to find a new tag by searching the old tag in $tag_dir/$file. When the tag occurs more than once, take one of the lines random.
The code explained in more detail:
# assign output to variable new_tag
new_tag=$(..)
# use awk program
awk ..
# Assign the valuo of old_tag to a variable "search" that can be used in awk
-v search="$old_tag"
# Different fields seperated by spaces
-F" "
# The awk programming lines
' .. '
# Check first field of line with the variable search
$1==search { .. }
# When true, store second field of line in array and increment index
a[count] = $2; count++;
# Additional comands after processing everything
END {..}
# Print random index from array
srand(); print a[int(rand()*(count-1))+1]
# Use file as input for awk
$tag_dir/$file
# Stop when no new_tag has been found
[ -z "$new_tag" ] && break
# I would have preferred the syntax
test -z "${new_tag}" && break
With the sample input and old_tag="apple", the code will find the lines with apple as the first word
apple x1
apple x5
apple x3
The words x1 x5 x3 are stored in array a and randomly one of these 3 is assigned to new_tag.