If you’re writing a database query for your web service and there’s pagination, you should always sort by a unique field like ID. This should also be done on top of any other sorting. Otherwise you can get double entries or lose results across pages.

5 years ago I wrote Rgeo, a small Go package for reverse geocoding (you input some coordinates, it gives you information about the location). Today I am putting the projected on an indefinite hiatus.

I’ve been working on web backends in Go for the past 5 years, and on various projects of covering a range of ages from brand new greenfield projects to one built on top of a core that predates the Go language. The needs of these backends tend not to be overly complex, handling user management, payments, and a relativly simple product. They’re maintained by teams of 2-20 backend engineers and work with a SPA web frontend communicating with JSON over HTTP.

With my most recent work being mostly greenfield, I’ve been thinking more about how I can apply what I’ve learned from my colleagues working on older projects, and avoid decisions made in the name of moving fast which cause a lot of pain further down the line.

Rgeo is a fast, simple solution for local reverse geocoding, Rather than relying on external software or online APIs, rgeo packages all of the data it needs in your binary. This means it will only ever work down to the level of cities , but if that’s all you need then this is the library for you.