Go supports JSON via its standard libraries. However it does not support YAML – which is interesting when you think about it. YAML is after all just JSON written in a specific way.
It is useful to have a high level understanding of how GO supports JSON before diving into YAML. The first thing you need to do is model your data as a struct:
Notice how there is a string with some meta data for each field. This tells Go what each field is named in your JSON. After this you can include the encoding/json library you get a whole host of functions including:
Adding YAML Support
It is a shame that YMAL is not supported by Go as part of the standard libraries but it can be easily added by a third party library called go-yaml. Firstly it needs to be included in your project:
You of course need the library installed locally on your machine – enter and run this at the command line:
go get gopkg.in/yaml.v2
Once you have done this you are ready to get started. Nicely the library tries to be as similar to the standard lib for JSON as possible. First update your struct:
Notice the only change is in the metadata. Now it specifies YAML and not JSON, but the intent is the same. What name does the field have in the YAML. To interact with it there is a marshal and unmarshal function:
Above is a complete example. The lib is imported and a struct called Person created. Then in the main function, a person variable is instantiated to the type Person – a struct object.
On line 20 the data variable – containing some valid YAML – is read into the person variable using the Marshal function.
On line 23 this person variable is converted back into JSON using the Unmarshal function and stored in the personJSON variable. Both functions are members of the yaml class which is introduced by the lib.
Supporting JSON and YAML
In my use case there was not a need to support both JSON and YAML. But if you need to then check out the write up The right way to handle YAML in Go by Sam Ghods which was an interesting read – it should set you on the right path.