Swagger spec to include operationId ?

[MarkEdmondson1234]

Hi, I used the swagger.json from the Plumber app to generate a Google Cloud Endpoint which is neat, but I had to add a operationId to the swagger spec to get it to work. Here is my horrible hack:

library(yaml)
library(jsonlite)

make_openapi <- function(projectId){

  ## this calls the Plumber apps swagger.json
  json <- jsonlite::fromJSON(sprintf("https://%s.appspot.com/swagger.json", projectId))

  ## change host from 127.0.0.1 to the appengine project
  json$host <- sprintf("%s.appspot.com", projectId)
  
  ## add operationId to each endpoint
  ohgod <- lapply(names(json$paths), 
                  function(x) {lapply(json$paths[[x]], 
                                      function(verb) {verb$operationId <- basename(x);verb})})
  json$paths <- setNames(ohgod, names(json$paths))

  # silly formatting to make these entries arrays and not characters
  yaml <- gsub("application/json", "[application/json]", yaml::as.yaml(json))
  yaml <- gsub("schemes: http", "schemes: [http]", yaml)
  
  writeLines(yaml, con = "openapi.yaml")
}

make_openapi("your-project-id")

[trestletech]

Thanks for the suggestion. Do you have any suggestions for how this field should be populated in the Swagger doc? I could just concatenate the verb with the path like @get /my/path becomes operationId: get-my-path? It gets a little awkward with endpoints that support multiple verbs, but I think the idea would still work.

Also would it be useful to include a plumber.yaml for your usecase?

[MarkEdmondson1234]

operationId: get-my-path would be perfect, I forgot to add the verb in my example. It just needs to be unique.

My use case supports JSON or YAML files, but to modify it (such as the hostname) I turned it into YAML so a user can edit the final result easier. Its not a big deal to convert between them, although annoying things like the array vs character would be easier to deal with if you have yaml available in the correct format. So yes, it would be useful :)

[MarkEdmondson1234]

Actually what would be most useful would be if the swagger.json or yaml file could be generated just from the code, without needing to launch the plumber server.

[alexhallam]

Hi, I wanted to see if there was an update on this issue?

[schloerke]

@alexhallam Coming soon! (<= a few weeks)

[schloerke]

This can be solved in plumber v1.0.0 using

pr %>%
  pr_set_api_spec(function(spec) {
    # update spec object here
    # return spec object
    spec
  })