Involved Source Filesdecode_hooks.goerror.go
Package mapstructure exposes functionality to convert one arbitrary
Go type into another, typically to convert a map[string]interface{}
into a native Go structure.
The Go structure can be arbitrarily complex, containing slices,
other structs, etc. and the decoder will properly decode nested
maps and so on into the proper structures in the native Go struct.
See the examples to see what the decoder is capable of.
The simplest function to start with is Decode.
Field Tags
When decoding to a struct, mapstructure will use the field name by
default to perform the mapping. For example, if a struct has a field
"Username" then mapstructure will look for a key in the source value
of "username" (case insensitive).
type User struct {
Username string
}
You can change the behavior of mapstructure by using struct tags.
The default struct tag that mapstructure looks for is "mapstructure"
but you can customize it using DecoderConfig.
Renaming Fields
To rename the key that mapstructure looks for, use the "mapstructure"
tag and set a value directly. For example, to change the "username" example
above to "user":
type User struct {
Username string `mapstructure:"user"`
}
Embedded Structs and Squashing
Embedded structs are treated as if they're another field with that name.
By default, the two structs below are equivalent when decoding with
mapstructure:
type Person struct {
Name string
}
type Friend struct {
Person
}
type Friend struct {
Person Person
}
This would require an input that looks like below:
map[string]interface{}{
"person": map[string]interface{}{"name": "alice"},
}
If your "person" value is NOT nested, then you can append ",squash" to
your tag value and mapstructure will treat it as if the embedded struct
were part of the struct directly. Example:
type Friend struct {
Person `mapstructure:",squash"`
}
Now the following input would be accepted:
map[string]interface{}{
"name": "alice",
}
When decoding from a struct to a map, the squash tag squashes the struct
fields into a single map. Using the example structs from above:
Friend{Person: Person{Name: "alice"}}
Will be decoded into a map:
map[string]interface{}{
"name": "alice",
}
DecoderConfig has a field that changes the behavior of mapstructure
to always squash embedded structs.
Remainder Values
If there are any unmapped keys in the source value, mapstructure by
default will silently ignore them. You can error by setting ErrorUnused
in DecoderConfig. If you're using Metadata you can also maintain a slice
of the unused keys.
You can also use the ",remain" suffix on your tag to collect all unused
values in a map. The field with this tag MUST be a map type and should
probably be a "map[string]interface{}" or "map[interface{}]interface{}".
See example below:
type Friend struct {
Name string
Other map[string]interface{} `mapstructure:",remain"`
}
Given the input below, Other would be populated with the other
values that weren't used (everything but "name"):
map[string]interface{}{
"name": "bob",
"address": "123 Maple St.",
}
Omit Empty Values
When decoding from a struct to any other value, you may use the
",omitempty" suffix on your tag to omit that value if it equates to
the zero value. The zero value of all types is specified in the Go
specification.
For example, the zero type of a numeric type is zero ("0"). If the struct
field value is zero and a numeric type, the field is empty, and it won't
be encoded into the destination type.
type Source {
Age int `mapstructure:",omitempty"`
}
Unexported fields
Since unexported (private) struct fields cannot be set outside the package
where they are defined, the decoder will simply skip them.
For this output type definition:
type Exported struct {
private string // this unexported field will be skipped
Public string
}
Using this map as input:
map[string]interface{}{
"private": "I will be ignored",
"Public": "I made it through!",
}
The following struct will be decoded:
type Exported struct {
private: "" // field is left with an empty string (zero value)
Public: "I made it through!"
}
Other Configuration
mapstructure is highly configurable. See the DecoderConfig struct
for other features and options that are supported.
Package-Level Type Names (total 8, all are exported)
/* sort exporteds by: | */
DecodeHookFunc is the callback function that can be used for
data transformations. See "DecodeHook" in the DecoderConfig
struct.
The type must be one of DecodeHookFuncType, DecodeHookFuncKind, or
DecodeHookFuncValue.
Values are a superset of Types (Values can return types), and Types are a
superset of Kinds (Types can return Kinds) and are generally a richer thing
to use, but Kinds are simpler if you only need those.
The reason DecodeHookFunc is multi-typed is for backwards compatibility:
we started with Kinds and then realized Types were the better solution,
but have a promise to not break backwards compat so we now support
both.
func ComposeDecodeHookFunc(fs ...DecodeHookFunc) DecodeHookFunc
func RecursiveStructToMapHookFunc() DecodeHookFunc
func StringToIPHookFunc() DecodeHookFunc
func StringToIPNetHookFunc() DecodeHookFunc
func StringToSliceHookFunc(sep string) DecodeHookFunc
func StringToTimeDurationHookFunc() DecodeHookFunc
func StringToTimeHookFunc(layout string) DecodeHookFunc
func typedDecodeHook(h DecodeHookFunc) DecodeHookFunc
func ComposeDecodeHookFunc(fs ...DecodeHookFunc) DecodeHookFunc
func DecodeHookExec(raw DecodeHookFunc, from reflect.Value, to reflect.Value) (interface{}, error)
func github.com/spf13/viper.DecodeHook(hook DecodeHookFunc) viper.DecoderConfigOption
func typedDecodeHook(h DecodeHookFunc) DecodeHookFunc
DecodeHookFuncKind is a DecodeHookFunc which knows only the Kinds of the
source and target types.
DecodeHookFuncType is a DecodeHookFunc which has complete information about
the source and target types.
func TextUnmarshallerHookFunc() DecodeHookFuncType
DecodeHookFuncRaw is a DecodeHookFunc which has complete access to both the source and target
values.
DecoderConfig is the configuration that is used to create a new decoder
and allows customization of various aspects of decoding.
DecodeHook, if set, will be called before any decoding and any
type conversion (if WeaklyTypedInput is on). This lets you modify
the values before they're set down onto the resulting struct. The
DecodeHook is called for every map and value in the input. This means
that if a struct has embedded fields with squash tags the decode hook
is called only once with all of the input data, not once for each
embedded struct.
If an error is returned, the entire decode will fail with that error.
If ErrorUnused is true, then it is an error for there to exist
keys in the original map that were unused in the decoding process
(extra keys).
Metadata is the struct that will contain extra metadata about
the decoding. If this is nil, then no metadata will be tracked.
Result is a pointer to the struct that will contain the decoded
value.
Squash will squash embedded structs. A squash tag may also be
added to an individual struct field using a tag. For example:
type Parent struct {
Child `mapstructure:",squash"`
}
The tag name that mapstructure reads for field names. This
defaults to "mapstructure"
If WeaklyTypedInput is true, the decoder will make the following
"weak" conversions:
- bools to string (true = "1", false = "0")
- numbers to string (base 10)
- bools to int/uint (true = 1, false = 0)
- strings to int/uint (base implied by prefix)
- int to bool (true if value != 0)
- string to bool (accepts: 1, t, T, TRUE, true, True, 0, f, F,
FALSE, false, False. Anything else is an error)
- empty array = empty map and vice versa
- negative numbers to overflowed uint values (base 10)
- slice of maps to a merged map
- single values are converted to slices if required. Each
element is weakly decoded. For example: "4" can become []int{4}
if the target type is an int slice.
ZeroFields, if set to true, will zero fields before writing them.
For example, a map will be emptied before decoded values are put in
it. If this is false, a map will be merged.
func github.com/spf13/viper.defaultDecoderConfig(output interface{}, opts ...viper.DecoderConfigOption) *DecoderConfig
func NewDecoder(config *DecoderConfig) (*Decoder, error)
func github.com/spf13/viper.decode(input interface{}, config *DecoderConfig) error
Error implements the error interface and can represents multiple
errors that occur in the course of a single decode.
Errors[]string(*T) Error() string
WrappedErrors implements the errwrap.Wrapper interface to make this
return value more useful with the errwrap and go-multierror libraries.
*T : error
Metadata contains information about decoding a structure that
is tedious or difficult to get otherwise.
Keys are the keys of the structure which were successfully decoded
Unused is a slice of keys that were found in the raw value but
weren't decoded since there was no matching field in the result interface
func DecodeMetadata(input interface{}, output interface{}, metadata *Metadata) error
func WeakDecodeMetadata(input interface{}, output interface{}, metadata *Metadata) error
Package-Level Functions (total 19, in which 15 are exported)
ComposeDecodeHookFunc creates a single DecodeHookFunc that
automatically composes multiple DecodeHookFuncs.
The composed funcs are called in order, with the result of the
previous transformation.
Decode takes an input structure and uses reflection to translate it to
the output structure. output must be a pointer to a map or struct.
DecodeHookExec executes the given decode hook. This should be used
since it'll naturally degrade to the older backwards compatible DecodeHookFunc
that took reflect.Kind instead of reflect.Type.
DecodeMetadata is the same as Decode, but is shorthand to
enable metadata collection. See DecoderConfig for more info.
NewDecoder returns a new decoder for the given configuration. Once
a decoder has been returned, the same configuration must not be used
again.
StringToIPHookFunc returns a DecodeHookFunc that converts
strings to net.IP
StringToIPNetHookFunc returns a DecodeHookFunc that converts
strings to net.IPNet
StringToSliceHookFunc returns a DecodeHookFunc that converts
string to []string by splitting on the given sep.
StringToTimeDurationHookFunc returns a DecodeHookFunc that converts
strings to time.Duration.
StringToTimeHookFunc returns a DecodeHookFunc that converts
strings to time.Time.
TextUnmarshallerHookFunc returns a DecodeHookFunc that applies
strings to the UnmarshalText function, when the target type
implements the encoding.TextUnmarshaler interface
WeakDecode is the same as Decode but is shorthand to enable
WeaklyTypedInput. See DecoderConfig for more info.
WeakDecodeMetadata is the same as Decode, but is shorthand to
enable both WeaklyTypedInput and metadata collection. See
DecoderConfig for more info.
WeaklyTypedHook is a DecodeHookFunc which adds support for weak typing to
the decoder.
Note that this is significantly different from the WeaklyTypedInput option
of the DecoderConfig.
typedDecodeHook takes a raw DecodeHookFunc (an interface{}) and turns
it into the proper DecodeHookFunc type, such as DecodeHookFuncType.
The pages are generated with Goldsv0.3.2. (GOOS=linux GOARCH=amd64)
Golds is a Go 101 project developed by Tapir Liu.
PR and bug reports are welcome and can be submitted to the issue list.
Please follow @Go100and1 (reachable from the left QR code) to get the latest news of Golds.