cross-posted from: https://programming.dev/post/58898830

There has been some QoL improvements added recently. Notable ones are:

1. Support for OIDC login
2. Batched inserts
3. Temporary link disabling
4. Faster reply and redirection due to async handling of some less-important tasks

There have also been small changes to the backend and UI that improve the user experience overall.

If you're looking for a selfhosted URL shortener, please check it out.

Extra links: Codeberg, Official website

There has been some QoL improvements added recently. Notable ones are:

1. Support for OIDC login
2. Batched inserts
3. Temporary link disabling
4. Faster reply and redirection due to async handling of some less-important tasks

There have also been small changes to the backend and UI that improve the user experience overall.

If you're looking for a selfhosted URL shortener, please check it out.

Extra links: Codeberg, Official website

It's a tiny selfhosted service that determines the client's IP address or accepts an IP address through a query parameter, then returns geolocation and ASN information using the MaxMind GeoLite2 databases.

I often used services likr ipinfo.io for checking my IP, location etc. e.g. to make sure a VPN was working properly. These are also handy for figuring out which timezone to use etc. I wanted to selfhost it, but couldn't find one that satisfied my needs. So I decided to write one.

Check out the README for details.

It's a tiny selfhosted service that determines the client's IP address or accepts an IP address through a query parameter, then returns geolocation and ASN information using the MaxMind GeoLite2 databases.

I often used services like ipinfo.io for checking my IP, location etc. e.g. to make sure a VPN was working properly. These are also handy for figuring out which timezone to use etc. I wanted to selfhost it, but couldn't find one that satisfied my needs. So I decided to write one.

Check out the README for details.

This release brings three main changes.

1. The ability to filter links.
2. Support for an optional notes field.
3. Ability to edit expiry time and notes.

I try not to too many new features to avoid bloat, but it seemed like these were pretty useful for a link shortener, especially when managing thousands of short links. (To my surprise, some people even use it to manage millions of links.)

Please take a look at the release notes for a complete list of changes.

P.S. The next thing I'll be focusing on is improving throughput under sustained load. If anyone has experience with SQLite, feel free to drop any tips. All the db related code is here. I'm mostly interested in improving insert speeds when 1000s of inserts are done per second.

Edit: There's a Codeberg mirror as well.

Hi, I'm a beginner when it comes to Haskell, but I guess I'm somewhat comfortable with the language itself. I'd love to try building some stuff using it. However, I'm finding it very hard to properly use external libraries/packages.

I'm familiar with Rust's cargo, and go's package management system, and feel very comfortable using them. Are there similar ones for Haskell? Basically, I want them defined per project, and not have to install everything system-wide. I'd like to be able to provide users with simple instructions for replicating the whole build setup.

What's the idiomatic way? Are there any example packages that I can look at that are not too complicated?

As the title says, it's a rather simple golang script for creating memories in Immich based on certain criteria, since there aren't any customization options in the official app.

As you may gather from the README, I created this for very personal reasons. But I wanted to share it here in case someone else finds it useful as well.

Currently it can do filtering based on people, and tags. But I'm more than happy to add more options, if requested.

I had looked for selfhosted recipe management apps for a while, but none really fit my need. I liked WikiJS a lot, but it was a hassle to get recipes on to it since there's no tooling for scraping from sites.

So I decided to create a python tool that does the scraping. It's finally in a state where I can share it with others. Of course, it's meant to be used only for archival purposes.

You can see some recipes created with it on my wiki.

Any feedback is welcome. I hope that this is the correct community for this. Given how pro-selfhosting lemmy is in general, I thought this might be welcome here.

cross-posted from: https://programming.dev/post/39212874

I recently migrated my services from rootful docker to rootless podman quadlets. It went smoothly, since nothing I use actually needs to be rootful. Well, except for caddy. It needs to be able to attach to privileged ports 80 and 443.

My current way to bypass it is using HAProxy running as root and forwarding connections using proxy protocol. (Tried to use firewalld but that makes the client IP opaque to caddy.) But that adds an extra layer, which means extra latency. It's perfectly usable, but I'd like to get rid of it, if possible.

I'm willing to run caddy in rootful podman if needed. But from what I understand, that means I can't have it in the same rootless network as my other containers. I really don't wanna open most of my containers' ports, so that's not an option.

So, I'm asking whether any of these three things are possible.

1. Use firewalld to forward ports to caddy without obscuring the client's IP.
2. Make rootful caddy share a network with other rootless containers.
3. Assign privileged ports to caddy somehow, in rootless mode. (I know there's a way to make all these ports unprivileged, but is it possible to only assign these 2 ports as unprivileged?)

Or maybe there's a fourth way that I'm missing. I feel like this is a common enough setup, that there must be a way to do it. Any pointers are appreciated, thanks.

I recently migrated my services from rootful docker to rootless podman quadlets. It went smoothly, since nothing I use actually needs to be rootful. Well, except for caddy. It needs to be able to attach to privileged ports 80 and 443.

My current way to bypass it is using HAProxy running as root and forwarding connections using proxy protocol. (Tried to use firewalld but that makes the client IP opaque to caddy.) But that adds an extra layer, which means extra latency. It's perfectly usable, but I'd like to get rid of it, if possible.

I'm willing to run caddy in rootful podman if needed. But from what I understand, that means I can't have it in the same rootless network as my other containers. I really don't wanna open most of my containers' ports, so that's not an option.

So, I'm asking whether any of these three things are possible.

1. Use firewalld to forward ports to caddy without obscuring the client's IP.
2. Make rootful caddy share a network with other rootless containers.
3. Assign privileged ports to caddy somehow, in rootless mode. (I know there's a way to make all these ports unprivileged, but is it possible to only assign these 2 ports as unprivileged?)

Or maybe there's a fourth way that I'm missing. I feel like this is a common enough setup, that there must be a way to do it. Any pointers are appreciated, thanks.

This release adds the ability to edit existing links, show and download QR codes for easy sharing, and various improvements in the frontend. Check out the release note for a list of all changes.

I simply want to emulate the effect of -p by default i.e. open all files in tabs when multiple files are supplied. I wrote the following autocommand to make it work.

-- Open files in tabs by default
vim.api.nvim_create_autocmd("VimEnter", {
    callback = function()
        if not vim.opt.diff:get() and #vim.fn.argv() > 1 then
            vim.cmd("tab sball")
            vim.cmd("tabfirst")
        end
    end,
})

But it seems to bork the colorscheme for all but the first tab. It's weird since Running the same commands manually after neovim is loaded works perfectly. I may have something to do with the order in which things are run. Is it possible to run this command as late as possible?

I'm open to alternative approaches that gets the job done.