Recently I’ve made a Zig based CLI argument parser called argh I’ve even written myself a roadmap I’d like to follow with this project. And so far I have been following it pretty well. However last week I began working on the next item on my roadmap which was to add support for environment variables to the arg parser, where you could set an environment variable to set the value of a flag.
Not many use cases for it I know it’s more of an edge case but I thought it would be a fun exercise to implement that. So I started working on it. I carefully scaffolded out the feature what functions I’d like to see in it how they should behave what guard rails I’d like to set to ensure type safety, memory safety, and so on.
As I was implementing all the things I scaffolded out one by one I realized that I was basically writing a fully fledged dotenv parser. And I thought to myself “hey this is pretty cool I should probably make this a separate library so other people can use it too” and it’s not deeply coupled to the arg parser so it makes sense to do that. So I did. I extracted the code out into its own repository called zdotenv.
The Library
The dotenv parser itself is pretty simple to use and I’ve even provided a basic example and tests in my repository that you can check out to see how it works. But regardless here’s a quick example of how to use it:
const std = @import("std");
const Dotenv = @import("dotenv").Dotenv;
pub fn main() !void {
const allocator = std.heap.page_allocator;
var dotenv = Dotenv.init(allocator);
defer dotenv.deinit();
try dotenv.load(&[_][]const u8{"./examples/example.env"});
if (dotenv.get("FOO")) |val| {
std.debug.print("FOO={s}\n", .{val});
} else {
std.debug.print("FOO not found\n", .{});
}
}
This will load the environment variables from the specified file and you can then access them using the get method. The load method takes an array of file paths so you can load multiple dotenv files if you want to.
That’s it, that’s all there is to it. It’s simple and straightforward as a dotenv parser should ideally be.
I’ve also added support for specifying custom .env file names since not everyone uses the default .env file all the time, you may wanna dynamically load different env files based on the environment you’re in (development, staging, production, etc).
I’ve added support for comments, blank lines and values in double quotes. So it should cover most use cases out there.
Features
- Loads key-value pairs from
.envbasically the core feature it does that it says, it parses an.envfile and loads the key-value pairs into memory so that you can access them later via thegetmethod. - Supports comments and blank lines it ignores comments and blank lines so you can have a well formatted
.envfile without any issues. - Provides easy access to values via hash map you can access the values using the
getmethod which returns an optional value so you can handle the case where the key doesn’t exist. I picked a hash map for this since generally speaking your.envfiles won’t ever be that large so we can just store them in memory for easy and fast access. - Memory safe all allocations are freed with a
deinitmethod so you don’t have to worry about memory leaks.
Conclusion
Well that’s it for now. I’ll be adding additional functionality and patches to the library as I see fit. But for now it’s a pretty solid and usable dotenv parser that you can use in your Zig projects. I for one have already used it in my notes CLI application I made a while back to compare Zig against Rust and it works like a charm.
As always I welcome contributions and feedback so if you have any ideas or suggestions feel free to open an issue or a pull request on the repository.
And if you have any questions or would like to reach out to me you can find me on Twitter