Question

I have a web server that saves the logs files of a web application numbered. A file name example for this would be:

dbsclog01s001.log
dbsclog01s002.log
dbsclog01s003.log

The last 3 digits are the counter and they can get sometime up to 100.

I usually open a web browser, browse to the file like:

http://someaddress.com/logs/dbsclog01s001.log

and save the files. This of course gets a bit annoying when you get 50 logs.
I tried to come up with a BASH script for using wget and passing

http://someaddress.com/logs/dbsclog01s*.log

but I am having problems with my the script.
Anyway, anyone has a sample on how to do this?

thanks!

Answer 1

#!/bin/sh

if [ $# -lt 3 ]; then
        echo "Usage: $0 url_format seq_start seq_end [wget_args]"
        exit
fi

url_format=$1
seq_start=$2
seq_end=$3
shift 3

printf "$url_format\\n" `seq $seq_start $seq_end` | wget -i- "$@"

Save the above as seq_wget, give it execution permission (chmod +x seq_wget), and then run, for example:

$ ./seq_wget http://someaddress.com/logs/dbsclog01s%03d.log 1 50

Or, if you have Bash 4.0, you could just type

$ wget http://someaddress.com/logs/dbsclog01s{001..050}.log

Or, if you have curl instead of wget, you could follow Dennis Williamson’s answer.