Go for writing web servers
Summary: in my experience Go is a good language for building websites/web servers.
It’s easy to get excited about new technology like Go. The question is: how does it stand up to scrutiny after daily use?
I’ve written 4 web applications in Go, they’ve been running in production for several months so I feel justified in publishing my opinion.
In the past I wrote web applications in Perl, PHP, Python (web.py, Tornado, App Engine) so those are the technologies I campare Go to.
Of the 4 websites, AppTranslator is a web service for crowd-sourcing translations for software. It was written completely from scratch.
QuickNotes is a note-taking application. Also written from scratch.
One reason to migrate from App Engine to my own server was to save money. At my levels of traffic (~3 requests per second) I was paying ~$80/month, mostly for frontend instance hours.
Another reason was to do more complex processing. App Engine is great but being a fully managed service, it limits what you can do.
Finally, I wanted to see how Go will handle a real project.
Things you need for building a web application
What functionality is typically needed for writing a web application and how does Go support it?
A very capable http server is part of standard library in package (net/http).
It parses incoming http requests and provides a way to send http responses back.
It supports http/2 out of the box.
High-level web frameworks provide url routing i.e. a way to say “call this function to handle this url”. Go has a built-in simple router. I also use mux, which is built on top of the built-in router.
A lot of what web server does is returning html which is greatly simplified by using templates. Go has a powerful html/template library for that. It’s roughly equivalent of Django or Tornado templates from Python world.
There are other templating libraries for Go but I found html/template good enough for my needs so I didn’t try the alternatives.
To generate cookies that cannot be spoofed or hi-jacked, I used securecookie library
Generating atom (rss) feeds
One cannot respect a blog engine that doesn’t provide full-text rss feed. I couldn’t find an existing package so I build a simple (and small) library for generating atom feeds
JSON and XML support
Modern web application are often implemented as JSON API server on the backend and Single Page Application on the front-end that generates HTML based on JSON returned from the server.
Returning JSON responses based on data from Go structs is very easy.
S3 access, support for zip files
This one is not universally needed. My web apps do backups by uploading data to s3, often as .zip archives.
Go has support for creating and decompressing zip and tar archives in the standard library. For s3 support I use official AWS SDK.
This is not specific to writing web server - all your important code should have unit tests.
Go has a built-in API and tool support for writing and running tests.
Deploying a new version of your code to the server is a pain in the ass regardless of the language used but Go makes it relatively easy.
I test my code locally on Mac but deploy to Linux-based server.
Thanks to Go’s support for cross-compilation I can build a linux binary on mac.
In the past I would copy the binary and other necesssary files to Linux server using Fabric or Ansible scripts.
Those days I pacakge them as Docker containers and deploy that using simple shell scripts.
I find it crucial for productivity to quickly deploy new versions. THat’s why I write deploy scripts early in the process.
Server config and misc thoughts
In the past I would run multiple services on a single Linux server, using nginx to multiplex.
Those days for simplicity I use a single server for each program. That makes setup and deployment easier (no need for nginx anymore).
Show me the code
Feel free to learn from the code or use it in your own projects.
I think writing non-trivial web services is a sweet spot for Go.
Most of the needed functionality is part of standard library and 3rdp party libraries are plantiful for more esoteric functionality.
Writing in Go is almost as fast and fluent as writing in Python but the code is order of magnitude faster and uses less memory.