Hacker News new | past | comments | ask | show | jobs | submit login

What really convinced me about HATEOAS and links was the first time I used a HAL browser and started clicking around to discover an API using only its entry point and navigating from there.

From that point on I try to use it as much as possible. A typical API response for my projects looks like this:

  {
  "id": "3ccf0f1b-dd3f-48d9-911a-ddf479078c37",
  "name": "Quantus Tasks",
  "description": "Quantus Tasks Desktop Application",
  "license_key_type": "alphanumeric_32",
  "created_at": "2019-05-12T10:45:42.089406Z",
  "updated_at": "2019-05-12T10:45:42.089406Z",
  "_links": {
    "self": {
      "href": "http://localhost:8000/v1/applications/3ccf0f1b-dd3f-48d9-911a-ddf479078c37"
    },
    "licenses": {
      "href": "http://localhost:8000/v1/applications/3ccf0f1b-dd3f-48d9-911a-ddf479078c37/licenses"
    },
    "templates": {
      "href": "http://localhost:8000/v1/applications/3ccf0f1b-dd3f-48d9-911a-ddf479078c37/templates"
    },
    "apikeys": {
      "href": "http://localhost:8000/v1/applications/3ccf0f1b-dd3f-48d9-911a-ddf479078c37/apikeys"
    }
  }
  }
It still has the ID field in there for cases where the client needs to store the id itself but it should not be used to template URI's for related resources, the links are there for that.



I've always been a bit undecided on this. I can see some obvious upsides, but on the other hand (apart from the downsides mentioned in the article and elsewhere) you've added 500 bytes to this entity for functionality that's only useful to the developer. Is the ability to click around in a HAL browser instead of a Swagger document worth the verbosity? Is there an argument that we might produce clients that can meaningfully deal with new links being presented without a developer being involved? I'd be surprised if that was realistic.


Yes, 500 bytes extra but it's not only for the developer discovering the API in a browser. If the client developer is exploring the API and then manually implementing an API client something is not right.

Hal clients for this use case do exist. you point them to an entrypoint and they can discover and generate the necessary code/methods for interacting with the API.

For example: https://github.com/pezra/hal-client




Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: