A Codec a contain a JSON Decoder a and the corresponding a -> Value encoder.
Represents a JavaScript value.
A structured error describing exactly how the decoder failed. You can use this to create more elaborate visualizations of a decoder problem. For example, you could show the entire JSON object and show the part causing the failure in red.
A value that knows how to decode JSON values.
Parse the given string into a JSON value and then run the Codec on it.
This will fail if the string is not well-formed JSON or if the Codec
fails for some reason.
Run a Codec to decode some JSON Value. You can send these JSON values
through ports, so that is probably the main time you would use this function.
Convert a value into a prettified JSON string. The first argument specifies the amount of indentation in the result string.
Convert a value into a Javascript Value.
Codec for a fixed list of Elm values.
This can be used for custom types, if they are really simple:
type Semaphore = Red | Yellow | Green
semaphoreCodec : Codec Semaphore
semaphoreCodec =
enum Codec.string
[ ("Red", Red)
, ("Yellow", Yellow)
, ("Green", Green)
]
encodeToString 0 semaphoreCodec Red
--> "\"Red\""
decodeString semaphoreCodec "\"Red\""
--> Ok Red
type Count = One | Two | Three
countCodec : Codec Count
countCodec =
enum Codec.int
[ (1, One)
, (2, Two)
, (3, Three)
]
encodeToString 0 countCodec Two
--> "2"
decodeString countCodec "2"
--> Ok Two
incompleteCodec : Codec Count
incompleteCodec =
enum Codec.int
[ (1, One)
, (2, Two)
]
encodeToString 0 incompleteCodec Three
--> "null"
decodeString incompleteCodec "3" |> Result.mapError (\_ -> "...")
--> Err "..."Represents an optional value.
This is encoded as null when the input is Nothing, and the same as x when Just x.
If the decoding using the inner Codec fails, it will succeed with Nothing. If you want it to fail use nullable instead.
encodeToString 0 (maybe int) (Just 3)
--> "3"
encodeToString 0 (maybe int) Nothing
--> "null"
decodeString (maybe int) "3"
--> Ok (Just 3)
decodeString (maybe int) "null"
--> Ok Nothing
decodeString (maybe int) "\"hello\""
--> Ok NothingRepresents an optional value.
This is encoded as null when the input is Nothing, and the same as x when Just x.
When decoding, it decodes null to Nothing. Otherwise, if the decoding using the inner Codec fails, it will fail. If you want it to succeed with Nothing use maybe instead.
encodeToString 0 (nullable int) (Just 3)
--> "3"
encodeToString 0 (nullable int) Nothing
--> "null"
decodeString (nullable int) "3"
--> Ok (Just 3)
decodeString (nullable int) "null"
--> Ok Nothing
decodeString (nullable int) "\"hello\"" |> Result.mapError (\_ -> "...")
--> Err "..."Codec between a JSON array of length 2 and an Elm Tuple.
Codec between a JSON array of length 3 and an Elm triple.
A partially built Codec for an object.
Start creating a Codec for an object. You should pass the main constructor as argument.
If you don't have one (for example it's a simple type with no name), you should pass a function that given the field values builds an object.
Example with constructor:
type alias Point =
{ x : Float
, y : Float
}
pointCodec : Codec Point
pointCodec =
Codec.object Point
|> Codec.field "x" .x Codec.float
|> Codec.field "y" .y Codec.float
|> Codec.buildObjectExample without constructor:
pointCodec : Codec { x : Int, y : Bool }
pointCodec =
Codec.object (\x y -> { x = x, y = y })
|> Codec.field "x" .x Codec.int
|> Codec.field "y" .y Codec.bool
|> Codec.buildObjectSpecify the name, getter and Codec for a field.
The name is only used as the field name in the resulting JSON, and has no impact on the Elm side.
Specify the name getter and Codec for an optional field.
This is particularly useful for evolving your Codecs.
If the field is not present in the input then it gets decoded to Nothing. If the field cannot be decoded this will fail. If the value is null then this will fail, use optionalNullableField if you want it to succeed instad.
If the optional field's value is Nothing then the resulting object will not contain that field.
Specify the name getter and Codec for an optional field.
This is particularly useful for evolving your Codecs.
If the field is not present in the input then it gets decoded to Nothing. If the field cannot be decoded this will fail. If the value is null then this will succeed with Nothing, use optionalField if you want it to fail instad.
If the optional field's value is Nothing then the resulting object will not contain that field.
An object field that might not be present or might contain data that maps to Nothing.
This function flattens both into Nothing rather than making you deal with a Maybe (Maybe a) type.
When encoding, if the value is Nothing then the field is not included.
Create a Codec from a fully specified ObjectCodec.
A partially built Codec for a custom type.
Starts building a Codec for a custom type.
You need to pass a pattern matching function, built like this:
type Semaphore
= Red Int String
| Yellow
| Green Float
semaphoreCodec : Codec Semaphore
semaphoreCodec =
Codec.custom
(\red yellow green value ->
case value of
Red i s ->
red i s
Yellow ->
yellow
Green f ->
green f
)
|> Codec.variant2 "Red" Red Codec.int Codec.string
|> Codec.variant0 "Yellow" Yellow
|> Codec.variant1 "Green" Green Codec.float
|> Codec.buildCustomDefine a variant with 0 parameters for a custom type.
Define a variant with 1 parameters for a custom type.
Define a variant with 2 parameters for a custom type.
Define a variant with 3 parameters for a custom type.
Define a variant with 4 parameters for a custom type.
Define a variant with 5 parameters for a custom type.
Define a variant with 6 parameters for a custom type.
Define a variant with 7 parameters for a custom type.
Define a variant with 8 parameters for a custom type.
Build a Codec for a fully specified custom type.
Try a set of decoders (in order). The first argument is used for encoding and decoding, the list of other codecs is used as a fallback while decoding.
This is particularly useful for backwards compatibility. You would pass the current codec as the first argument,
and the old ones (eventually mapped) as a fallback list to use while decoding.
Create a Codec that produces null as JSON and always decodes as the same value.
Create a Codec for a recursive data structure.
The argument to the function you need to pass is the fully formed Codec.
Ignore the JSON and make the decoder fail. This is handy when used with
oneOf or andThen where you want to give a custom error message in some
case. The encoder will produce null.
Create codecs that depend on previous results.
This is useful for recursive structures that are not easily modeled with recursive.
Have a look at the Json.Decode docs for examples.
Create a Codec that doesn't transform the JSON value, just brings it to and from Elm as a Value.
Build your own custom Codec.
Useful if you have pre-existing Decoders you need to use.
Create a Codec that produces null as JSON and always decodes as the same value. Obsolete alias of succeed, will be removed in a future version.
Specify the name getter and Codec for a required field, whose value can be null.
Warning! This is a footgun and thus deprecated, you should probably use field with nullable instead.
If the field is not present in the input then the decoding fails. If the field is present but can't be decoded then the decoding succeeds with Nothing.
If the field's value is Nothing then the resulting object will contain the field with a null value.
This is a shorthand for a field having a Codec built using maybe.
Specify the name getter and Codec for an optional field.
Warning! This is a footgun and thus deprecated, you should probably use optionalField instead.
This is particularly useful for evolving your Codecs.
If the field is not present in the input then it gets decoded to Nothing. _If the field value cannot be decoded by the given Codec it also gets decoded to Nothing. Even worse, if the input is not an object, this Codec still succeeds!
When encoding, if the optional field's value is Nothing then the resulting object will not contain that field.
Get all of the fields present in an ObjectCodec. This might be useful if you are making a request to a database and you want to both decode the response and be able to include in the request a filter so you only receive the fields you intend to decode.
A value that knows how to encode and decode JSON values.