GET /entity/human_users/<user_id>/following

Params

No fields, no page[], no sort.

Sample requests

r = c.get("/entity/human_users/3/following")

{
  "data": [
    {"id": 346, "type": "Note", "links": {"self": "/api/v1/entity/Note/346"}},
    {"id": 360, "type": "Note", "links": {"self": "/api/v1/entity/Note/360"}},
    {"id": 391, "type": "Note", "links": {"self": "/api/v1/entity/Note/391"}}
  ]
}

On the probed site that call answered 896 rows in one body, 707 Note and 189 Task, with no page key. The two filters cut it server-side:

c.get("/entity/human_users/3/following", params={"entity": "notes"})                 # 707 rows
c.get("/entity/human_users/3/following", params={"entity": "Note"})                  # 707 rows
c.get("/entity/human_users/3/following", params={"entity": "shots"})                 # 0 rows
c.get("/entity/human_users/3/following", params={"project_id": 70})                  # 865 rows
c.get("/entity/human_users/3/following", params={"entity": "notes", "project_id": 70})  # 707 rows

Response codes

Edge cases

• An ApiUser id under /entity/human_users/ answers Couldn't find HumanUser with id="1". A script token cannot ask what it follows, only what a named person follows.
• entity takes the schema name or the snake_case plural, unlike the entity key in the follow body, which takes the schema name alone and answers 500 to the plural.
• links.self is /api/v1/entity/Note/346, singular and CamelCase, matching endpoints/get_entity_type_id_followers and nothing else in the API. It resolves at 200.
• Each row is id, type and a link. Neither the name of the followed record nor the date the follow started is returned, so a display list costs one call per row or a _search on the ids.
• Nothing pages. Filter with entity and project_id or take the whole list.