GO TESTING;HOW,WHAT,WHY

Introduction

關於 Go 測試，咱們應該知道測試方式（或者說測試手段）、測試對象及測試緣由。

How 測試方式

測試實現

舉個例子。針對字符串分割函數（以下），實現單元測試。

package goTest

import "strings"

// Split slices s into all substrings separated by sep and
// returns a slice of the substrings between those separators.
func Split(s, sep string) []string {
    var result []string
    i := strings.Index(s, sep)
    for i > -1 {
        result = append(result, s[:i])
        s = s[i+len(sep):]
        i = strings.Index(s, sep)
    }
    return append(result, s)
}

在當前目錄下且同樣的包名 goTest ，寫一個簡單的 go 測試函數，以下：

package goTest

import (
    "reflect"
    "testing"
)

func TestSplit(t *testing.T) {
    got := Split("a/b/c", "/")
    want := []string{"a", "b", "c"}
    if !reflect.DeepEqual(want, got) {
        t.Fatalf("expected: %v, got: %v", want, got)
    }
}

測試函數必須以 Test 開頭， 且必須攜帶一個 *testing.T 參數。 t *testing.T 提供改測試函數的打印、跳過、失敗功能。

測試執行

當前目錄下，執行 go test ，輸出以下：

> go test
PASS
ok      goTest  0.005s

若是項目中存在多個 package ，若要執行全部包的測試能夠在項目根目錄下使用 go test ./... ，輸出以下（例子：github.com/mattn/go-sqlite3）：

> go test ./...
ok      github.com/mattn/go-sqlite3    14.693s
?       github.com/mattn/go-sqlite3/upgrade    [no test files]

代碼測試覆蓋率

仍是以字符串分割函數爲例， 獲取當前代碼測試覆蓋率方式以下：

> go test -coverprofile=c.out
PASS
coverage: 100.0% of statements
ok      goTest    0.005s

數據顯示覆蓋率爲 100% 。若要以 HTML 方式顯示可使用命令 go tool cover -html=c.out 。

【tip】 一行命令 cover 獲取當前目錄下的代碼測試覆蓋度。 在 ~/.bashrc 中添加以下命令：

cover () {
    local t=$(mktemp -t cover)
    go test $COVERFLAGS -coverprofile=$t $@ \
      && go tool cover -func=$t \
      && unlink $t
}

執行後獲取的測試覆蓋度結果以下：

> cover
PASS
coverage: 100.0% of statements
ok      goTest    0.008s
goTest/wwg_split.go:7:    Split         100.0%
total:            (statements)     100.0%

問題：測試覆蓋率 100% ，結束了？

多個測試用例的狀況下，使用表組測試用例裝填。更改 TestSplit 以下：

func TestSplit(t *testing.T) {
    tests := []struct{
        input string
        sep string
        want []string
    }{
        {input: "a/b/c", sep: "/", want: []string{"a", "b", "c"}},
        {input: "a/b/c", sep: ",", want: []string{"a/b/c"}},
        {input: "a/b/c/", sep: "/", want: []string{"a", "b", "c"}},  // trailing sep
        {input: "abc", sep: "/", want: []string{"abc"}},
    }

    for _, tc := range tests {
        got := Split(tc.input, tc.sep)
        if !reflect.DeepEqual(tc.want, got) {
            t.Fatalf("expected: %v, got: %v", tc.want, got)
        }
    }
}

增長測試用例 trailing sep 後，執行測試，結果以下：

> go test
--- FAIL: TestSplit (0.00s)
    wwg_split_test.go:23: expected: [a b c], got: [a b c ]
FAIL
exit status 1
FAIL    goTest    0.005s

根據該結果很難一會兒在表組測試用例中查出是哪條。能夠將 表組測試用例實現改成 map 形式 ，具體以下：

func TestSplit(t *testing.T) {
    tests := map[string]struct{
        input string
        sep string
        want []string
    }{
        "simple": {input: "a/b/c", sep: "/", want: []string{"a", "b", "c"}},
        "wrong sep": {input: "a/b/c", sep: ",", want: []string{"a/b/c"}},
        "trailing sep": {input: "a/b/c/", sep: "/", want: []string{"a", "b", "c"}},
        "no sep": {input: "abc", sep: "/", want: []string{"abc"}},
    }

    for name, tc := range tests {
        got := Split(tc.input, tc.sep)
        if !reflect.DeepEqual(tc.want, got) {
            t.Errorf("%s expected: %v, got: %v", name, tc.want, got)
        }
    }
}

執行測試結果以下：

> go test
--- FAIL: TestSplit (0.00s)
    wwg_split_test.go:23: trailing sep expected: [a b c], got: [a b c ]
FAIL
exit status 1
FAIL    goTest    0.005s

Sub tests 使用，及 '%#v' format 使用，更改 TestSplit 以下：

func TestSplit(t *testing.T) {
    tests := map[string]struct{
        input string
        sep string
        want []string
    }{
        "simple": {input: "a/b/c", sep: "/", want: []string{"a", "b", "c"}},
        "wrong sep": {input: "a/b/c", sep: ",", want: []string{"a/b/c"}},
        "trailing sep": {input: "a/b/c/", sep: "/", want: []string{"a", "b", "c"}},
        "no sep": {input: "abc", sep: "/", want: []string{"abc"}},
    }

    for name, tc := range tests {
        t.Run(name, func(t *testing.T) {
            got := Split(tc.input, tc.sep)
            if !reflect.DeepEqual(tc.want, got) {
                t.Fatalf("%s expected: %#v, got: %#v", name, tc.want, got)
            }
        })
    }
}

測試結果以下：

> go test
--- FAIL: TestSplit (0.00s)
    --- FAIL: TestSplit/trailing_sep (0.00s)
        wwg_split_test.go:24: trailing sep expected: []string{"a", "b", "c"}, got: []string{"a", "b", "c", ""}
FAIL
exit status 1
FAIL    goTest    0.005s

更好的打印格式，能夠訪問：

• https://github.com/k0kubun/pp
• https://github.com/davecgh/go...
• https://github.com/google/go-cmp

使用 google/go-cmp 優化打印， 更改 TestSplit 以下：

for name, tc := range tests {
        t.Run(name, func(t *testing.T) {
            got := Split(tc.input, tc.sep)
            diff := cmp.Diff(tc.want, got)
            if diff != "" {
                t.Fatalf(diff)
            }
        })
    }

執行測試結果以下：

> go test
--- FAIL: TestSplit (0.00s)
    --- FAIL: TestSplit/trailing_sep (0.00s)
        wwg_split_test.go:29:   []string{
                  "a",
                  "b",
                  "c",
            +     "",
              }
FAIL
exit status 1
FAIL    goTest    0.005s

修復bug後 Split 代碼以下：

// Split slices s into all substrings separated by sep and
// returns a slice of the substrings between those separators.
func Split(s, sep string) []string {
    var result []string
    i := strings.Index(s, sep)
    for i > -1 {
        result = append(result, s[:i])
        s = s[i+len(sep):]
        i = strings.Index(s, sep)
    }
    **if len(s) > 0 {**
        result = append(result, s)
    }
    return result**
}

執行測試，結果以下：

> go test
PASS
ok      goTest    0.006s

> cover
PASS
coverage: 100.0% of statements
ok      goTest    0.006s
goTest/wwg_split.go:7:    Split        100.0%
total:                  (statements)    100.0%

What 測試對象

Q_1：Go 應該測試全部因子嗎？

A_1：顯然不是。

Q_2：什麼時候編寫測試? 1.編碼完成後？ 2.編碼前？ 3.其餘人遍寫測試，像QA、TE？ 4.項目設計人員編寫測試？

A_2：編碼的同時編寫測試代碼（TDD）Article TheThreeRulesOfTdd

Q_3：C 單元測試對象是 function ，Java 單元測試對象是 Class ，類內部的方法， Go 的單元測試對象是？

A_3：package 。測試行爲，而非實施。 "The public API of a package declare this is what(行爲) I do, not this is how(實施) I do it."

Why 測試緣由

即便你不作代碼測試，別人也會作。本身發現 issues 總比別人發現來得好，不是嗎？

1. 大部分的測試(自動化)應該是開發人員本身作。
2. 手工測試不該該是你測試的主體部分，由於手工測試的複雜度爲O(n)
3. 測試能夠確保您始終能夠運送主分支
4. 測試肯定軟件行爲（作什麼、不作什麼）
5. 測試讓你有信心修改他人的代碼

總結

• You should write tests.
• You should write tests at the same time as you write your code.
• Each Go package is a self contained unit.
• Your tests should assert the observable behaviour of your package, not its implementation.
• You should design your packages around their behaviour, not their implementation.

【注】部分資料源於GopherChina 2019 - 'How to write testable code'