If you have a list of file names separated by new lines, and you want to turn it into a list of file names separated by null characters, for instance for use as input to xargs, then you can use the tr command, like this:
$ cat /path/to/new-lines | tr '\n' '\0' > /path/to/null-separated
I found a whole article on various ways to replace text on *nix: rmnl — remove new line characters with tr, awk, perl, sed or c/c++.
You can use a parameter to the -x command line switch to tell the tar command that you just want to extract one particular file from the archive. For example:
$ tar -x filename/to/extract -f tarfile.tar
You can use the -O command line switch in conjunction with the above to have the file’s contents printed on stdout rather than created in the file system.
I read Chapter 21. Subshells of The Linux Documentation Project‘s Advanced Bash-Scripting Guide. One fun trick I learned was using a subshell to test if a variable is set:
if (set -u; : $variable) 2> /dev/null then echo "Variable is set." fi
I was trying to figure out how to have a child process change the current directory of a parent process. Turns out you can’t quite do that. One thing I guess you could do is export a global variable and then have the parent check that when the child returns, but I decided on another approach. What I do is write a file to ~/bin/errdir that contains a single line in the format:
pushd "/path/to/directory"
I can then source this as a command file after my script has run to change directory into the directory where the last erroneous file was encountered with the command:
$ . errdir
When I’m finished I can just type ‘popd’ to return myself to wherever I was before I changed into the error directory.
I’ve been learning about approaches to error handling in bash. I found Error handling in BASH, Exit Shell Script Based on Process Exit Code, Writing Robust Bash Shell Scripts and Bash: Error handling all of which contained a nugget of useful information.
Particularly I learned about $LINENO, set -e (equiv: set -o errexit), set -o pipefail, set -E, set -u (equiv: set -o nounset), set -C (equiv: set -o noclobber), #!/bin/bash -eEu, trap, shopt -s expand_aliases, alias, $PIPESTATUS, subshells, unset, and more.
If you call exit without specifying a return code then $? is used, I didn’t know that.
I found this code snippet that demos a relatively race-condition free method of acquiring a lock file:
if ( set -o noclobber; echo "$$" > "$lockfile") 2> /dev/null; then trap 'rm -f "$lockfile"; exit $?' INT TERM EXIT critical-section rm -f "$lockfile" trap - INT TERM EXIT else echo "Failed to acquire lockfile: $lockfile." echo "Held by $(cat $lockfile)" fi
If you want to change the modification time of a file on Linux, the command you’re looking for is touch. You can use touch with the -r parameter to specify a reference file who’s date and time information will be used as the basis for a new (or existing) file.
It took me a while, but I finally figured out how to print a number from a bash script properly formatted with commas as thousand’s separators. For those like me who weren’t in the know, the magical incantation is:
printf "%'d" 123456
That will format 123456 as 123,456. Much easier to read when you’re dealing with large numbers such as the sizes of files in gigabytes.
So now if I could only find the Unix command that took a number of bytes and turned it into an approximate value with GB or MB suffixes.
I have a directory structure with archived files, many of which are zipped or tarballed up. So if I want to search for a file, I can’t really be sure that the file I’m looking for isn’t in a compressed file. So I wrote some scripts to automatically run over a directory tree and automatically extract any compressed files that it finds there. There are a few extra features, such as if the file is larger than 10MB then it will prompt for whether to extract it or not. I have a few other features for handling errors to add in, but I’m happy to post this version up now.
#!/bin/bash
err=$1
err=${err:-1}
search() {
find -iname "*$1" -print0 | xargs -i -0 `dirname "$0"`/extract-file.sh "$1" "$2" "$err" "{}"
}
search ".tar.gz" "tar xf"
search ".tgz" "tar xf"
search ".zip" "unzip -q"
#!/bin/bash exec 0< /dev/tty path="$4" file_name=`basename "$path"` dir_name=`dirname "$path"` new_name=`basename "$path" "$1"` new_path="$dir_name/$new_name" [ -e "$new_path" ] && exit 0 echo $dir_name/$file_name file_size=`stat -c %s "$path"` #echo -n "File is $file_size bytes." printf "File is %'d bytes." $file_size check=`echo "$file_size < 10485760" | bc` if [ "$check" = "1" ]; then answer="y"; fi while [ "$answer" != "y" ]; do read -n 1 -s -p " Extract? " answer [ "$answer" = "n" ] && echo && echo && exit 0 done echo mkdir "$new_path" pushd "$new_path" > /dev/null 2>&1 $2 "../$file_name" if [ "$?" != "0" ]; then popd > /dev/null 2>&1 rm -rf "$new_path" echo answer="" while [ "$answer" != "y" ]; do read -n 1 -s -p "Do you want to ignore this file? " answer [ "$answer" = "n" ] && exit $3 done fi popd > /dev/null 2>&1 echo
When you’re at a Unix terminal and your terminal goes screwy, try typing “stty sane” to reset it. Handy!
It took me a while, but I finally found this documentation which said that you use the single quote to flag formatted output as having the thousands separator when using printf.