forgejo/modules/util/util.go
wxiaoguang 3c4153b195
[PORT] Refactor tests to prevent from unnecessary preparations (gitea#32398)
Some preparations are only used by a few tests, so to make the tests fast, they should only be prepared when they are used.

By the way, this PR splits PrepareTestEnv into small functions to make it simple.

---

Conflict resolution: Mostly magical and just re-pasting the code into
the right places.
Done differently: use `require.NoError` instead of `assert.NoError`.

(cherry picked from commit ec2d1593c269e06655525deb96f74b8094221b6f)
2024-11-07 23:02:10 +01:00

284 lines
6.9 KiB
Go

// Copyright 2017 The Gitea Authors. All rights reserved.
// Copyright 2024 The Forgejo Authors. All rights reserved.
// SPDX-License-Identifier: MIT
package util
import (
"bytes"
"crypto/ed25519"
"crypto/rand"
"encoding/pem"
"fmt"
"math/big"
"strconv"
"strings"
"code.gitea.io/gitea/modules/optional"
"golang.org/x/crypto/ssh"
"golang.org/x/text/cases"
"golang.org/x/text/language"
)
// OptionalBoolParse get the corresponding optional.Option[bool] of a string using strconv.ParseBool
func OptionalBoolParse(s string) optional.Option[bool] {
v, e := strconv.ParseBool(s)
if e != nil {
return optional.None[bool]()
}
return optional.Some(v)
}
// IsEmptyString checks if the provided string is empty
func IsEmptyString(s string) bool {
return len(strings.TrimSpace(s)) == 0
}
// NormalizeEOL will convert Windows (CRLF) and Mac (CR) EOLs to UNIX (LF)
func NormalizeEOL(input []byte) []byte {
var right, left, pos int
if right = bytes.IndexByte(input, '\r'); right == -1 {
return input
}
length := len(input)
tmp := make([]byte, length)
// We know that left < length because otherwise right would be -1 from IndexByte.
copy(tmp[pos:pos+right], input[left:left+right])
pos += right
tmp[pos] = '\n'
left += right + 1
pos++
for left < length {
if input[left] == '\n' {
left++
}
right = bytes.IndexByte(input[left:], '\r')
if right == -1 {
copy(tmp[pos:], input[left:])
pos += length - left
break
}
copy(tmp[pos:pos+right], input[left:left+right])
pos += right
tmp[pos] = '\n'
left += right + 1
pos++
}
return tmp[:pos]
}
// CryptoRandomInt returns a crypto random integer between 0 and limit, inclusive
func CryptoRandomInt(limit int64) (int64, error) {
rInt, err := rand.Int(rand.Reader, big.NewInt(limit))
if err != nil {
return 0, err
}
return rInt.Int64(), nil
}
const alphanumericalChars = "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789"
// CryptoRandomString generates a crypto random alphanumerical string, each byte is generated by [0,61] range
func CryptoRandomString(length int64) (string, error) {
buf := make([]byte, length)
limit := int64(len(alphanumericalChars))
for i := range buf {
num, err := CryptoRandomInt(limit)
if err != nil {
return "", err
}
buf[i] = alphanumericalChars[num]
}
return string(buf), nil
}
// CryptoRandomBytes generates `length` crypto bytes
// This differs from CryptoRandomString, as each byte in CryptoRandomString is generated by [0,61] range
// This function generates totally random bytes, each byte is generated by [0,255] range
func CryptoRandomBytes(length int64) ([]byte, error) {
buf := make([]byte, length)
_, err := rand.Read(buf)
return buf, err
}
// ToUpperASCII returns s with all ASCII letters mapped to their upper case.
func ToUpperASCII(s string) string {
b := []byte(s)
for i, c := range b {
if 'a' <= c && c <= 'z' {
b[i] -= 'a' - 'A'
}
}
return string(b)
}
// ToTitleCase returns s with all english words capitalized
func ToTitleCase(s string) string {
// `cases.Title` is not thread-safe, do not use global shared variable for it
return cases.Title(language.English).String(s)
}
// ToTitleCaseNoLower returns s with all english words capitalized without lower-casing
func ToTitleCaseNoLower(s string) string {
// `cases.Title` is not thread-safe, do not use global shared variable for it
return cases.Title(language.English, cases.NoLower).String(s)
}
// ToInt64 transform a given int into int64.
func ToInt64(number any) (int64, error) {
var value int64
switch v := number.(type) {
case int:
value = int64(v)
case int8:
value = int64(v)
case int16:
value = int64(v)
case int32:
value = int64(v)
case int64:
value = v
case uint:
value = int64(v)
case uint8:
value = int64(v)
case uint16:
value = int64(v)
case uint32:
value = int64(v)
case uint64:
value = int64(v)
case float32:
value = int64(v)
case float64:
value = int64(v)
case string:
var err error
if value, err = strconv.ParseInt(v, 10, 64); err != nil {
return 0, err
}
default:
return 0, fmt.Errorf("unable to convert %v to int64", number)
}
return value, nil
}
// ToFloat64 transform a given int into float64.
func ToFloat64(number any) (float64, error) {
var value float64
switch v := number.(type) {
case int:
value = float64(v)
case int8:
value = float64(v)
case int16:
value = float64(v)
case int32:
value = float64(v)
case int64:
value = float64(v)
case uint:
value = float64(v)
case uint8:
value = float64(v)
case uint16:
value = float64(v)
case uint32:
value = float64(v)
case uint64:
value = float64(v)
case float32:
value = float64(v)
case float64:
value = v
case string:
var err error
if value, err = strconv.ParseFloat(v, 64); err != nil {
return 0, err
}
default:
return 0, fmt.Errorf("unable to convert %v to float64", number)
}
return value, nil
}
// ToPointer returns the pointer of a copy of any given value
func ToPointer[T any](val T) *T {
return &val
}
// Iif is an "inline-if", it returns "trueVal" if "condition" is true, otherwise "falseVal"
func Iif[T any](condition bool, trueVal, falseVal T) T {
if condition {
return trueVal
}
return falseVal
}
// IfZero returns "def" if "v" is a zero value, otherwise "v"
func IfZero[T comparable](v, def T) T {
var zero T
if v == zero {
return def
}
return v
}
// OptionalArg helps the "optional argument" in Golang:
//
// func foo(optArg ...int) { return OptionalArg(optArg) }
// calling `foo()` gets zero value 0, calling `foo(100)` gets 100
// func bar(optArg ...int) { return OptionalArg(optArg, 42) }
// calling `bar()` gets default value 42, calling `bar(100)` gets 100
//
// Passing more than 1 item to `optArg` or `defaultValue` is undefined behavior.
// At the moment only the first item is used.
func OptionalArg[T any](optArg []T, defaultValue ...T) (ret T) {
if len(optArg) >= 1 {
return optArg[0]
}
if len(defaultValue) >= 1 {
return defaultValue[0]
}
return ret
}
func ReserveLineBreakForTextarea(input string) string {
// Since the content is from a form which is a textarea, the line endings are \r\n.
// It's a standard behavior of HTML.
// But we want to store them as \n like what GitHub does.
// And users are unlikely to really need to keep the \r.
// Other than this, we should respect the original content, even leading or trailing spaces.
return strings.ReplaceAll(input, "\r\n", "\n")
}
// GenerateSSHKeypair generates a ed25519 SSH-compatible keypair.
func GenerateSSHKeypair() (publicKey, privateKey []byte, err error) {
public, private, err := ed25519.GenerateKey(nil)
if err != nil {
return nil, nil, fmt.Errorf("ed25519.GenerateKey: %w", err)
}
privPEM, err := ssh.MarshalPrivateKey(private, "")
if err != nil {
return nil, nil, fmt.Errorf("ssh.MarshalPrivateKey: %w", err)
}
sshPublicKey, err := ssh.NewPublicKey(public)
if err != nil {
return nil, nil, fmt.Errorf("ssh.NewPublicKey: %w", err)
}
return ssh.MarshalAuthorizedKey(sshPublicKey), pem.EncodeToMemory(privPEM), nil
}