JSON Document ⇨ Crystal type definitions
MIT License
Jump-start your next JSON API client by generating types from a sample API response!
The results this outputs are meant as a starting off point, to eliminate having
to write a lot of the boilerplate involved in creating a new project to parse
the response from an API, without resorting to the runtime overhead of relying
on JSON.parse
. You will probably want to review the generated code and make
some revisions, especially if your data sample includes arrays of objects (see
Known Issues).
Add the dependency to your shard.yml
:
dependencies:
jenerator:
github: dscottboggs/jenerator
Run shards install
require "jenerator"
sample = %[{"data": "just some text", "listOfData": ["one", 2, 3.0]}]
puts Jenerator.process sample
The above code sample would output this Crystal code:
require "json"
class Document
include JSON::Serializable
@data : String
@[JSON::Field(key: "listOfData")]
@list_of_data : Array(String | Int64 | Float64)
end
There's also a compiled script which you can download as a static binary from
the "releases"
page which will generate new Crystal source files for each .json
file in a
directory.
When generating code for an array of objects, a new class or struct is generated for each object member of that array. The resulting type eliminates redundant types from the union, but the types are still declared.
See jenerator_spec.cr for an example.
git checkout -b my-new-feature
)git commit -am 'Add some feature'
)git push origin my-new-feature
)