Merge branch 'add-transformations' into 'master'

Add Transformers

See merge request go/enumer!2
This commit is contained in:
G.J.R. Timmer 2019-01-17 13:13:07 +01:00
commit 42bd8ecbf7
14 changed files with 380 additions and 60 deletions

View File

@ -1,32 +1,36 @@
# Enumer [![GoDoc](https://godoc.org/github.com/alvaroloes/enumer?status.svg)](https://godoc.org/github.com/alvaroloes/enumer) [![Go Report Card](https://goreportcard.com/badge/github.com/alvaroloes/enumer)](https://goreportcard.com/report/github.com/alvaroloes/enumer) [![cover.run go](https://cover.run/go/github.com/alvaroloes/enumer.svg?tag=golang-1.10)](https://cover.run/go/github.com/alvaroloes/enumer?tag=golang-1.10)
Enumer is a tool to generate Go code that adds useful methods to Go enums (constants with a specific type).
It started as a fork of [Rob Pikes Stringer tool](https://godoc.org/golang.org/x/tools/cmd/stringer).
## Generated functions and methods
When Enumer is applied to a type, it will generate:
* The following basic methods/functions:
- The following basic methods/functions:
* Method `String()`: returns the string representation of the enum value. This makes the enum conform
- Method `String()`: returns the string representation of the enum value. This makes the enum conform
the `Stringer` interface, so whenever you print an enum value, you'll get the string name instead of a number.
* Function `<Type>String(s string)`: returns the enum value from its string representation. This is useful
- Function `<Type>String(s string)`: returns the enum value from its string representation. This is useful
when you need to read enum values from command line arguments, from a configuration file, or
from a REST API request... In short, from those places where using the real enum value (an integer) would
be almost meaningless or hard to trace or use by a human.
* Function `<Type>Values()`: returns a slice with all the values of the enum
* Method `IsA<Type>()`: returns true only if the current value is among the values of the enum. Useful for validations.
* When the flag `json` is provided, two additional methods will be generated, `MarshalJSON()` and `UnmarshalJSON()`. These make
- Function `<Type>Values()`: returns a slice with all the values of the enum
- Method `IsA<Type>()`: returns true only if the current value is among the values of the enum. Useful for validations.
- When the flag `json` is provided, two additional methods will be generated, `MarshalJSON()` and `UnmarshalJSON()`. These make
the enum conform to the `json.Marshaler` and `json.Unmarshaler` interfaces. Very useful to use it in JSON APIs.
* When the flag `text` is provided, two additional methods will be generated, `MarshalText()` and `UnmarshalText()`. These make
- When the flag `text` is provided, two additional methods will be generated, `MarshalText()` and `UnmarshalText()`. These make
the enum conform to the `encoding.TextMarshaler` and `encoding.TextUnmarshaler` interfaces.
**Note:** If you use your enum values as keys in a map and you encode the map as _JSON_, you need this flag set to true to properly
convert the map keys to json (strings). If not, the numeric values will be used instead
* When the flag `yaml` is provided, two additional methods will be generated, `MarshalYAML()` and `UnmarshalYAML()`. These make
- When the flag `yaml` is provided, two additional methods will be generated, `MarshalYAML()` and `UnmarshalYAML()`. These make
the enum conform to the `gopkg.in/yaml.v2.Marshaler` and `gopkg.in/yaml.v2.Unmarshaler` interfaces.
* When the flag `sql` is provided, the methods for implementing the Scanner and Valuer interfaces will be also generated.
- When the flag `sql` is provided, the methods for implementing the Scanner and Valuer interfaces will be also generated.
Useful when storing the enum in a database.
For example, if we have an enum type called `Pill`,
```go
type Pill int
@ -38,7 +42,9 @@ const (
Acetaminophen = Paracetamol
)
```
executing `enumer -type=Pill -json` will generate a new file with four basic methods and two extra for JSON:
```go
func (i Pill) String() string {
//...
@ -64,7 +70,9 @@ func (i *Pill) UnmarshalJSON(data []byte) error {
//...
}
```
From now on, we can:
```go
// Convert any Pill value to string
var aspirinString string = Aspirin.String()
@ -119,15 +127,29 @@ For example, the command `enumer -type=MyType -json -transform=snake` would gene
```go
name := MyTypeValue.String() // name => "my_type_value"
```
**Note**: The transformation only works form CamelCase to snake_case or kebab-case, not the other way around.
### Transformers
- snake
- snake-upper
- kebab
- kebab-upper
- lower (Lowercase)
- upper (Uppercase)
- title (TitleCase)
- first (Use first character of string)
- first-lower (same as first only lower case)
- first-upper (same as first only upper case)
## How to use
The usage of Enumer is the same as Stringer, so you can refer to the [Stringer docs](https://godoc.org/golang.org/x/tools/cmd/stringer)
for more information.
There are four boolean flags: `json`, `text`, `yaml` and `sql`. You can use any combination of them (i.e. `enumer -type=Pill -json -text`),
For enum string representation transformation the `transform` and `trimprefix` flags
were added (i.e. `enumer -type=MyType -json -transform=snake`).
Possible transform values are `snake` and `kebab` for transformation to snake_case and kebab-case accordingly.
@ -137,5 +159,6 @@ If a prefix is provided via the `trimprefix` flag, it will be trimmed from the s
it is transformed). If a name doesn't have the prefix it will be passed unchanged.
## Inspiring projects
* [Stringer](https://godoc.org/golang.org/x/tools/cmd/stringer)
* [jsonenums](https://github.com/campoy/jsonenums)
- [Stringer](https://godoc.org/golang.org/x/tools/cmd/stringer)
- [jsonenums](https://github.com/campoy/jsonenums)

View File

@ -72,13 +72,45 @@ func TestEndToEnd(t *testing.T) {
t.Logf("cgo is no enabled for %s", name)
continue
}
// Names are known to be ASCII and long enough.
typeName := fmt.Sprintf("%c%s", name[0]+'A'-'a', name[1:len(name)-len(".go")])
transformNameMethod := "noop"
if name == "transform.go" {
typeName = "CamelCaseValue"
// Names are known to be ASCII and long enough.
var typeName string
var transformNameMethod string
switch name {
case "transform_snake.go":
typeName = "SnakeCaseValue"
transformNameMethod = "snake"
case "transform_snake_upper.go":
typeName = "SnakeUpperCaseValue"
transformNameMethod = "snake-upper"
case "transform_kebab.go":
typeName = "KebabCaseValue"
transformNameMethod = "kebab"
case "transform_kebab_upper.go":
typeName = "KebabUpperCaseValue"
transformNameMethod = "kebab-upper"
case "transform_upper.go":
typeName = "UpperCaseValue"
transformNameMethod = "upper"
case "transform_lower.go":
typeName = "LowerCaseValue"
transformNameMethod = "lower"
case "transform_title.go":
typeName = "TitleCaseValue"
transformNameMethod = "title"
case "transform_first.go":
typeName = "FirstCaseValue"
transformNameMethod = "first"
case "transform_first_upper.go":
typeName = "FirstUpperCaseValue"
transformNameMethod = "first-upper"
case "transform_first_lower.go":
typeName = "FirstLowerCaseValue"
transformNameMethod = "first-lower"
default:
typeName = fmt.Sprintf("%c%s", name[0]+'A'-'a', name[1:len(name)-len(".go")])
transformNameMethod = "noop"
}
stringerCompileAndRun(t, dir, stringer, typeName, name, transformNameMethod)

View File

@ -27,6 +27,7 @@ import (
"path/filepath"
"sort"
"strings"
"unicode/utf8"
"github.com/pascaldekloe/name"
)
@ -240,18 +241,57 @@ func (pkg *Package) check(fs *token.FileSet, astFiles []*ast.File) {
}
func (g *Generator) transformValueNames(values []Value, transformMethod string) {
var sep rune
var fn func(src string) string
switch transformMethod {
case "snake":
sep = '_'
fn = func(s string) string {
return strings.ToLower(name.Delimit(s, '_'))
}
case "snake_upper", "snake-upper":
fn = func(s string) string {
return strings.ToUpper(name.Delimit(s, '_'))
}
case "kebab":
sep = '-'
fn = func(s string) string {
return strings.ToLower(name.Delimit(s, '-'))
}
case "kebab_upper", "kebab-upper":
fn = func(s string) string {
return strings.ToUpper(name.Delimit(s, '-'))
}
case "upper":
fn = func(s string) string {
return strings.ToUpper(s)
}
case "lower":
fn = func(s string) string {
return strings.ToLower(s)
}
case "title":
fn = func(s string) string {
return strings.Title(s)
}
case "first":
fn = func(s string) string {
r, _ := utf8.DecodeRuneInString(s)
return string(r)
}
case "first_upper", "first-upper":
fn = func(s string) string {
r, _ := utf8.DecodeRuneInString(s)
return strings.ToUpper(string(r))
}
case "first_lower", "first-lower":
fn = func(s string) string {
r, _ := utf8.DecodeRuneInString(s)
return strings.ToLower(string(r))
}
default:
return
}
for i := range values {
values[i].name = strings.ToLower(name.Delimit(values[i].name, sep))
values[i].name = fn(values[i].name)
}
}

25
testdata/transform.go vendored
View File

@ -1,25 +0,0 @@
package main
import "fmt"
type CamelCaseValue int
const (
CamelCaseValueOne CamelCaseValue = iota
CamelCaseValueTwo
CamelCaseValueThree
)
func main() {
ck(CamelCaseValueOne, "camel_case_value_one")
ck(CamelCaseValueTwo, "camel_case_value_two")
ck(CamelCaseValueThree, "camel_case_value_three")
ck(-127, "CamelCaseValue(-127)")
ck(127, "CamelCaseValue(127)")
}
func ck(value CamelCaseValue, str string) {
if fmt.Sprint(value) != str {
panic("transform.go: " + str)
}
}

25
testdata/transform_first.go vendored Normal file
View File

@ -0,0 +1,25 @@
package main
import "fmt"
type FirstCaseValue int
const (
Male FirstCaseValue = iota
Female
unknown
)
func main() {
ck(Male, "M")
ck(Female, "F")
ck(unknown, "u")
ck(-127, "FirstCaseValue(-127)")
ck(127, "FirstCaseValue(127)")
}
func ck(value FirstCaseValue, str string) {
if fmt.Sprint(value) != str {
panic("transform_first.go: " + str)
}
}

25
testdata/transform_first_lower.go vendored Normal file
View File

@ -0,0 +1,25 @@
package main
import "fmt"
type FirstLowerCaseValue int
const (
Male FirstLowerCaseValue = iota
Female
Unknown
)
func main() {
ck(Male, "m")
ck(Female, "f")
ck(Unknown, "u")
ck(-127, "FirstLowerCaseValue(-127)")
ck(127, "FirstLowerCaseValue(127)")
}
func ck(value FirstLowerCaseValue, str string) {
if fmt.Sprint(value) != str {
panic("transform_first_lower.go: " + str)
}
}

25
testdata/transform_first_upper.go vendored Normal file
View File

@ -0,0 +1,25 @@
package main
import "fmt"
type FirstUpperCaseValue int
const (
male FirstUpperCaseValue = iota
female
unknown
)
func main() {
ck(male, "M")
ck(female, "F")
ck(unknown, "U")
ck(-127, "FirstUpperCaseValue(-127)")
ck(127, "FirstUpperCaseValue(127)")
}
func ck(value FirstUpperCaseValue, str string) {
if fmt.Sprint(value) != str {
panic("transform_first_upper.go: " + str)
}
}

25
testdata/transform_kebab.go vendored Normal file
View File

@ -0,0 +1,25 @@
package main
import "fmt"
type KebabCaseValue int
const (
KebabCaseValueOne KebabCaseValue = iota
KebabCaseValueTwo
KebabCaseValueThree
)
func main() {
ck(KebabCaseValueOne, "kebab-case-value-one")
ck(KebabCaseValueTwo, "kebab-case-value-two")
ck(KebabCaseValueThree, "kebab-case-value-three")
ck(-127, "KebabCaseValue(-127)")
ck(127, "KebabCaseValue(127)")
}
func ck(value KebabCaseValue, str string) {
if fmt.Sprint(value) != str {
panic("transform_kebab.go: " + str)
}
}

25
testdata/transform_kebab_upper.go vendored Normal file
View File

@ -0,0 +1,25 @@
package main
import "fmt"
type KebabUpperCaseValue int
const (
KebabUpperCaseValueOne KebabUpperCaseValue = iota
KebabUpperCaseValueTwo
KebabUpperCaseValueThree
)
func main() {
ck(KebabUpperCaseValueOne, "KEBAB-UPPER-CASE-VALUE-ONE")
ck(KebabUpperCaseValueTwo, "KEBAB-UPPER-CASE-VALUE-TWO")
ck(KebabUpperCaseValueThree, "KEBAB-UPPER-CASE-VALUE-THREE")
ck(-127, "KebabUpperCaseValue(-127)")
ck(127, "KebabUpperCaseValue(127)")
}
func ck(value KebabUpperCaseValue, str string) {
if fmt.Sprint(value) != str {
panic("transform_kebab_upper.go: " + str)
}
}

25
testdata/transform_lower.go vendored Normal file
View File

@ -0,0 +1,25 @@
package main
import "fmt"
type LowerCaseValue int
const (
LowerCaseValueOne LowerCaseValue = iota
LowerCaseValueTwo
LowerCaseValueThree
)
func main() {
ck(LowerCaseValueOne, "lowercasevalueone")
ck(LowerCaseValueTwo, "lowercasevaluetwo")
ck(LowerCaseValueThree, "lowercasevaluethree")
ck(-127, "LowerCaseValue(-127)")
ck(127, "LowerCaseValue(127)")
}
func ck(value LowerCaseValue, str string) {
if fmt.Sprint(value) != str {
panic("transform_lower.go: " + str)
}
}

25
testdata/transform_snake.go vendored Normal file
View File

@ -0,0 +1,25 @@
package main
import "fmt"
type SnakeCaseValue int
const (
SnakeCaseValueOne SnakeCaseValue = iota
SnakeCaseValueTwo
SnakeCaseValueThree
)
func main() {
ck(SnakeCaseValueOne, "snake_case_value_one")
ck(SnakeCaseValueTwo, "snake_case_value_two")
ck(SnakeCaseValueThree, "snake_case_value_three")
ck(-127, "SnakeCaseValue(-127)")
ck(127, "SnakeCaseValue(127)")
}
func ck(value SnakeCaseValue, str string) {
if fmt.Sprint(value) != str {
panic("transform_snake.go: " + str)
}
}

25
testdata/transform_snake_upper.go vendored Normal file
View File

@ -0,0 +1,25 @@
package main
import "fmt"
type SnakeUpperCaseValue int
const (
SnakeUpperCaseValueOne SnakeUpperCaseValue = iota
SnakeUpperCaseValueTwo
SnakeUpperCaseValueThree
)
func main() {
ck(SnakeUpperCaseValueOne, "SNAKE_UPPER_CASE_VALUE_ONE")
ck(SnakeUpperCaseValueTwo, "SNAKE_UPPER_CASE_VALUE_TWO")
ck(SnakeUpperCaseValueThree, "SNAKE_UPPER_CASE_VALUE_THREE")
ck(-127, "SnakeUpperCaseValue(-127)")
ck(127, "SnakeUpperCaseValue(127)")
}
func ck(value SnakeUpperCaseValue, str string) {
if fmt.Sprint(value) != str {
panic("transform_snake_upper.go: " + str)
}
}

25
testdata/transform_title.go vendored Normal file
View File

@ -0,0 +1,25 @@
package main
import "fmt"
type TitleCaseValue int
const (
titlecasevalueone TitleCaseValue = iota
titlecasevaluetwo
titlecasevaluethree
)
func main() {
ck(titlecasevalueone, "Titlecasevalueone")
ck(titlecasevaluetwo, "Titlecasevaluetwo")
ck(titlecasevaluethree, "Titlecasevaluethree")
ck(-127, "TitleCaseValue(-127)")
ck(127, "TitleCaseValue(127)")
}
func ck(value TitleCaseValue, str string) {
if fmt.Sprint(value) != str {
panic("transform_title.go: " + str)
}
}

25
testdata/transform_upper.go vendored Normal file
View File

@ -0,0 +1,25 @@
package main
import "fmt"
type UpperCaseValue int
const (
UpperCaseValueOne UpperCaseValue = iota
UpperCaseValueTwo
UpperCaseValueThree
)
func main() {
ck(UpperCaseValueOne, "UPPERCASEVALUEONE")
ck(UpperCaseValueTwo, "UPPERCASEVALUETWO")
ck(UpperCaseValueThree, "UPPERCASEVALUETHREE")
ck(-127, "UpperCaseValue(-127)")
ck(127, "UpperCaseValue(127)")
}
func ck(value UpperCaseValue, str string) {
if fmt.Sprint(value) != str {
panic("transform_upper.go: " + str)
}
}