How to Use loco

Table of Contents

• Introduction
• Installing
• Quick tutorial
• Understanding less
• Running Shell Commands from less
• Updating the Database

Introduction

loco is a shell script for harnessing the output of the locate command. locate is an easy and powerful command to use for finding files quickly. Its raw output, however, is not particularly useful. The analogy is like trying to drink water from a fire hydrant. The results burst forth and scroll off your screen. Try this:

locate bash

What the hell are you supposed to do with that?

loco captures that output in a temporary text file and gives you a way to navigate the results and view, compare, or edit them. loco is also good for finding files that live in different parts of your filesystem and contain your search term in their names. If you narrow the search enough, the results can provide you with context and insight as to how the files and software are set up and where to look for things like scripts and settings.

Installing

Copy and paste this into your favo[u]rite terminal:

sh -c "$(curl -fsSL https://raw.githubusercontent.com/tristanchase/loco/main/install.sh)"

Quick tutorial

Let’s say you want to check your .bashrc file against the default file. The .bashrc file is sourced each time you start a new bash shell. It sets up several things that make the shell behave a certain way. If you’ve modified it heavily, you may want to check it against the default bashrc file to see what has been changed, especially if something doesn’t seem to be working quite right.

COMMAND:

loco search_term

Try this:

loco bash

What you see (the line numbers may be different for your output):

1       /bin/bash       [*]
2       /bin/rbash      [*]
3       /etc/bash.bashrc
4       /etc/bash_completion
5       /etc/bash_completion.d  [/]
6       /etc/apparmor.d/abstractions/bash
7       /etc/bash_completion.d/acroread.sh      [@]
8       /etc/bash_completion.d/apport_completion
...
Bunch of lines
...
49      /home/nicky/.bash_history
50      /home/nicky/.bash_logout
51      /home/nicky/.bashrc

Each line reads like this:

line_number /path/to/file_name [file_type]

You will use line_number when you use the row command (later on). The file_type is a symbol which tells you what type of file you are looking at:

/ directory
* executable
@ link (shortcut to another file)
| pipe
= socket
? unknown

If there is no file_type symbol after the file_name then it is a regular file.

Understanding less

As stated in the introduction, loco creates a text file in the "$HOME"/tmp folder and opens it for viewing in less. If you’re not familiar with less, it’s a program called a pager that displays the contents of a file. less is not an editor.

Movement in less is similar to vi:

j to move down one line
k to move up one line
<Space> to move down one screenful
g to go to the top
G to go to the bottom
/ to search (/search_term <Enter>)
n to jump to next search_term
N to jump to previous search_term
q to quit

Running Shell Commands from less

You can also run shell commands from less. This is how you use the output from loco. To do this you start the command with a bang (!).

Let’s look at our example output from above. Line 51 contains our .bashrc file. Line 3 contains the default bashrc file. To see our file we would type:

!less `row 51`

This opens a new instance of less with the contents of our .bashrc file. Notice that row 51 above is enclosed in backticks (located on the same key as the ~). This is how we get row to work from less. The backticks tell the shell to take the output of the script row with the argument 51 (our line_number). The result of that is the text /home/nicky/.bashrc. That text is then used by less as its argument. It’s kind of like the Order of Operations in math. The bang (!) at the beginning tells less to run a shell command.

When you are done exploring the file, you can quit with a q. You will be presented with a line that says:

!done (press RETURN)

at which point you may press Return to continue working with your search or press q to quit the active loco process. This will delete the file created in the "$HOME"/tmp directory and clean up any other odds and ends from the search. If you weren’t actually done with the search, you can hit Ctrl-p to get the last command and do the search over again.

If you wanted to look at the default bashrc file you would type:

!less `row 3`

If you wanted to edit our .bashrc file with nano, you would type:

!nano `row 51`

The idea here is that the commands look like this:

!command `row n`

where command is the command you want to run and n is the line_number.

Try it now in your actual search. Remember, the line numbers will most likely differ from the ones in this example.

Updating the Database

locate (and thus loco) works by looking for your search term in a database of the filesystem. The database is not updated instantaneously, however. If you create a file and then try to use loco to find it, it will not show up in the results. The database in managed by updatedb. It is normally set up to run once a day to comb the filesystem for changes (new, moved, or deleted files). I have found that this doesn’t seem to run automatically in the crouton environment. You can run it manually by issuing:

sudo updatedb

and waiting a few minutes for it to run. I don’t find that it’s necessary to do this more than once a day.