xpe VS structured-text-tools

xpe - a commandline xpath parser. xpaths are better than css queries for getting at specific html elements in the DOM. Compared to other parsers, this one is easier to use, and supports html.

I use the heck out of xpe. It's a super simple command-line xpath parser using lxml in python.

If you like xmllint, you might like xpe. It's more user-friendly.

xpe - a commandline xpath parser. I made this after trying to use xpaths for web automation in bash, and not finding anything that worked.

For commandline xpath parsing for simple commandline web automation, xpe is pretty handy. It's a real simple python script, but it scratches that itch, and it's only a pip install away.

This works well in scripts and logs of all the commands you need to do to reproduce the current state of the system from a scratch install. Also can be used with diff -u and patch, sed, perl, and awk oneliners and structured text tools. You can also capture most of the commands using sudo logging feature but it won't capture the here documents. But for modest size files you can use newlines in echo commands. Note that commands which use redrection should use something like ~~~~ sudo bash -c "echo 'foo' >>file.txt" ~~~~ instead of "sudo echo foo >>file.txt" or "echo foo | sudo tee -a file.txt

TFA is about how to handle csv files with awk. This might be useful in straightforward cases.

For all others I’d recommend to have a look at

https://github.com/dbohdan/structured-text-tools

which lists tools to handle structure text formats

in general, I'd pick something from https://github.com/dbohdan/structured-text-tools

Also, here is a list of structured text tools. You may find some tools there that are helpful in editing configuration files from the command line. Or you can use "diff -u" to create a patch file (you need to save the patch files along with sudo.log) to recreate. Also, use sfdisk --dump and sfdisk --backup to save partition information in a form that can be used to recreate backups.