JSON

Because of the ubiquitous nature of JSON, support for it is built directly into V.

V generates code for JSON encoding and decoding. No runtime reflection is used. This results in much better performance.

Decoding JSON

  1. import json
  3. struct Foo {
  4.     x int
  5. }
  7. struct User {
  8.     // Adding a [required] attribute will make decoding fail, if that
  9.     // field is not present in the input.
  10.     // If a field is not [required], but is missing, it will be assumed
  11.     // to have its default value, like 0 for numbers, or '' for strings,
  12.     // and decoding will not fail.
  13.     name string [required]
  14.     age  int
  15.     // Use the `skip` attribute to skip certain fields
  16.     foo Foo [skip]
  17.     // If the field name is different in JSON, it can be specified
  18.     last_name string [json: lastName]
  19. }
  21. data := '{ "name": "Frodo", "lastName": "Baggins", "age": 25 }'
  22. user := json.decode(User, data) or {
  23.     eprintln('Failed to decode json, error: $err')
  24.     return
  25. }
  26. println(user.name)
  27. println(user.last_name)
  28. println(user.age)
  29. // You can also decode JSON arrays:
  30. sfoos := '[{"x":123},{"x":456}]'
  31. foos := json.decode([]Foo, sfoos)?
  32. println(foos[0].x)
  33. println(foos[1].x)

The json.decode function takes two arguments: the first is the type into which the JSON value should be decoded and the second is a string containing the JSON data.

Encoding JSON

  1. import json
  3. struct User {
  4.     name  string
  5.     score i64
  6. }
  8. mut data := map[string]int{}
  9. user := &User{
  10.     name: 'Pierre'
  11.     score: 1024
  12. }
  14. data['x'] = 42
  15. data['y'] = 360
  17. println(json.encode(data)) // {"x":42,"y":360}
  18. println(json.encode(user)) // {"name":"Pierre","score":1024}