I installed the treefetch binary in my system from the Releases. When I try to run treefetch I get this error:

treefetch: /lib/x86_64-linux-gnu/libc.so.6: version `GLIBC_2.33' not found (required by treefetch)
treefetch: /lib/x86_64-linux-gnu/libc.so.6: version `GLIBC_2.32' not found (required by treefetch)

I'm on Pop!_OS 20.04 (based on Ubuntu 20.04)

Use the crate clap to handle arguments. Now the help message is like:

treefetch 2.0.0

USAGE:
    treefetch [TREE]

ARGS:
    <TREE>    Which tree to display. [default: normal] [possible values: normal, xmas, bonsai]

OPTIONS:
    -h, --help       Print help information
    -V, --version    Print version information

Host ⇒ Arch and Archcraft

The displayed memory usage is higher than actual memory usage. First I tried it on Archcraft there the actual memory usage was somewhere around 800 MB but treefetch shows somewhere around a GB. Also check it on an arch vm there also the result was same, actual memory usage was around 200 MB but treefetch shows 400 MB.

VM ⇒ {archlinux + openbox)

Changes

• Created a macro to simplify the code;
• Replaced print_left_to_right() with something more concise (based on the fact that every tree is 'taller' than the right side);
• Modified README.md (options => arguments). Also please consider changing the example images;

Suggestions

• Rewrite fields.rs using third-party crates (e.g. nixinfo) so that we can support more systems and at the same time simplify the code ; (It's OK not to do so as maybe you are practicing Rust through this project, which is exactly what I'm doing :P);
• Try using copilot, RustFmt in neovim. These will certainly increase the efficiency.

First I would like to say treefetch is an excellent fetch program, but it has an issue with the displayed memory usage and total memory.

I did a little research on htop to see how they are calculating it. I have actually done this my self in the Pascal language by using the /proc/meminfo, assuming you are familiar with the /proc/meminfo fields.

Total physical memory = MemTotal
Memory Used = (MemTotal + Shmem) - MemFree - Buffers - Cached - SReclaimable

these as you know are in kb, its just a matter of

• Total phsical memory / (1024 * 1024)
• Memory Used / (1024 * 1024)

or instead of using (1024*1024) just divide by the sum which is 1048576, this will yeild GB format as you already know.

Anyway I just thought to show you the calculation used by htop and btop ect... to get used memory, as your current implementation is is not adding up to the more common program memory usage programs out there.

( Help wanted )

treefetch in its current state only supports Linux. Use Rust conditional compilation in the fields.rs file to write separate code for each OS; Windows and macOS. You may also modify the main.rs a bit if needed.

The fetch information you need to implement for either Windows or macOS are:

• username / hostname
• OS name
• kernel (if applicable)
• Shell name
• Uptime
• Memory usage

You can make a PR implementing only Windows or only macOS support for now if you want.

I suggest doing something like this:

#[cfg(target_os = "linux")]
pub fn get_memory() -> Result<String, String> {
    ...
}

#[cfg(target_os = "macos")]
pub fn get_memory() -> Result<String, String> {
    ...
}

#[cfg(target_os = "windows")]
pub fn get_memory() -> Result<String, String> {
    ...
}

Hello, here's a suggestion

Implement a caching system that caches the more "stable*" values in a file under /tmp, this method reduces file IO from 4-5 to 1, it could improve performance and make room for future expansion

~If you want, i also can implement this feature.~

Here's an example:

#cat /tmp/treecache/

hostname : ethereal
key1 : value1
key2 : value2

And also an argument to delete it:

treefetch -R :

Removes the temp file Treefetch