Logical Approach on bash - bash

Question

#!/bin/bash
echo -n "Enter the domain name > "
read name
dig -t ns "$name" | cut -d ";" -f3 | cut -d ":" -f2| grep ns | cut -f6 > registerfile.txt;
cat registerfile.txt | while read line; do dig axfr "#$line" "$name"; done | cut -d"." -f-4 > nmap.txt
It is done till this section. Below, it could be not matched the line and name parameters. How should be changed?
cat nmap.txt | while read line; do if [ "$line" == "$name" ]; then host "$line"; fi; done > ping.txt
cat ping.txt | cut -d " " -f4 | while read line; do if [[ "$line" =~ ^[0-9]+$ ]]; then nmap -sS "$line";fi ;done

Answer 1

It's not clear where exactly things are going wrong, but here is a refactoring which might hopefully at least nudge you in the right direction.
#!/bin/bash
read -p "Enter the domain name > " name
dig +short -t ns "$name" |
tee registerfile.txt |
while read line; do
dig axfr "#$line" "$name"
done |
cut -d"." -f-4 |
tee nmap.txt |
while read line; do
if [ "$line" = "$name" ]; then
host "$line"
fi
done > ping.txt
cut -d " " -f4 ping.txt |
grep -E '^[0-9]+$' |
xargs -r -n 1 nmap -sS
Your remark in comments that if [ "$line" = "$name" ]; then host "$line"; fi isn't working suggests that the logic there is somehow wrong. It currently checks whether each line is identical to the original domain name, and then looks it up over and over again in those cases, which seems like a curious thing to do; but given only the code and the "does not work", it's hard to say what it's really supposed to accomplish. If you actually want something else, you need to be more specific about what you require. Perhaps you are actually looking for something like
... tee nmap.txt |
# Extract the lines which contain $name at the end
grep "\.$name\$" |
xargs -n 1 dig +short |
tee ping.txt |
grep -E '^[0-9]+$' ...
The use of multiple statically-named files is an antipattern; obviously, if these files serve no external purpose, just take out the tee commands and run the entire pipeline with no in-between output files. If you do need these files, having them overwritten on each run seems problematic -- maybe add a unique date stamp suffix to the file names?