# 如何偷懶給AI寫測試 AI prompt for writing tests in Golang
ChatGPT已經問世一陣子，開始把ChatGPT融入我們日常的開發
今天會提到讓ChatGPT幫你產生測試，並用最便宜的方法去使用它，以及額外推薦一些小工具。
在進入讓AI寫測試之前，先來了解可以用哪些形式的Prompt(咒語)讓ChatGPT幫你做事
## Prmpting principles
### Principle 1: 寫下明確的指示 Write clear and specific instructions
```
You will be provided with text delimited by triple quotes.
If it contains a sequence of instructions, \
re-write those instructions in the following format:
Step 1 - ...
Step 2 - …
…
Step N - …
If the text does not contain a sequence of instructions, \
then simply write \"No steps provided.\
```
#### 少量樣本提示 "Few-shot" prompting
[少量樣本提示](https://promptingguide.azurewebsites.net/techniques/fewshot)可以用作一種技術，以實現上下文學習，其中我們在提示中提供範例以引導模型實現更好的結果
```python
prompt = f"""
Your task is to answer in a consistent style.
<child>: Teach me about patience.
<grandparent>: The river that carves the deepest \
valley flows from a modest spring; the \
grandest symphony originates from a single note; \
the most intricate tapestry begins with a solitary thread.
<child>: Teach me about resilience.
"""
response = get_completion(prompt)
print(response)
```
### Principle 2: 讓模型去思考 Give the model time to “think”**
#### Tactic 1: 明確指定步驟，去完成任務 Specify the steps required to complete a task
```python
text = f"""
In a charming village, siblings Jack and Jill set out on \
a quest to fetch water from a hilltop \
well. As they climbed, singing joyfully, misfortune \
struck—Jack tripped on a stone and tumbled \
down the hill, with Jill following suit. \
Though slightly battered, the pair returned home to \
comforting embraces. Despite the mishap, \
their adventurous spirits remained undimmed, and they \
continued exploring with delight.
"""
# example 1
prompt_1 = f"""
Perform the following actions:
1 - Summarize the following text delimited by triple \
backticks with 1 sentence.
2 - Translate the summary into French.
3 - List each name in the French summary.
4 - Output a json object that contains the following \
keys: french_summary, num_names.
Separate your answers with line breaks.
Text:
```
"""
response = get_completion(prompt_1)
print("Completion for prompt 1:")
print(response)
```
#### 以制定的格式輸出 Ask for output in a specified format
```python
prompt_2 = f"""
Your task is to perform the following actions:
1 - Summarize the following text delimited by
<> with 1 sentence.
2 - Translate the summary into French.
3 - List each name in the French summary.
4 - Output a json object that contains the
following keys: french_summary, num_names.
Use the following format:
Text: <text to summarize>
Summary: <summary>
Translation: <summary translation>
Names: <list of names in Italian summary>
Output JSON: <json with summary and num_names>
Text: <{text}>
"""
response = get_completion(prompt_2)
print("\nCompletion for prompt 2:")
print(response)
```
## ChatGPT generates test
使用上面的技巧，可以寫出template讓ChatGPT學習，並指定格式輸出
就可以在短時間內，讓ChatGPT幫我們產生立即可以使用的測試，通常都只需要做微調即可
### Table driven tests template in Golang
#### 為何要寫 Table-Driven test?
那為什麼這邊要寫table driven test?
~~因為VScode按下右鍵產生Golang test就是table driven~~
Table-Driven test的目的是擷取出function內部的模式（pattern）
**好處**是之後要加上新的測試用例很方便，減少測試程式碼的重複性
例如，要測試 `sum.go`
```go
package sum
func Sum(a, b int) int {
return a + b
}
```
傳統的測試會這樣寫（`sum_test.go`)
根據不同test case 會新增新的test function
（Golang的test function要用 `Test` 作為前綴（prefix））
```go
package sum
import (
"testing"
)
func TestSum_One(t *testing.T) {
result := Sum(1, 2)
if result != 3 {
t.Errorf("expected 3, but got %d", result)
}
}
func TestSum_Two(t *testing.T) {
result := Sum(3, 4)
if result != 7 {
t.Errorf("expected 7, but got %d", result)
}
}
```
Table-driven test則是可以這樣寫，把測試用例寫進slice
未來要新增用例的話，可以很方便就依序加在下面
```go
package sum
import (
"testing"
)
func TestSum(t *testing.T) {
cases := []struct{
description string
num1 int
num2 int
expected int
}{
{
description: "1 + 2",
num1: 1,
num2: 2,
expected: 3,
},
{
description: "3 + 4",
num1: 3,
num2: 4,
expected: 7,
},
}
for _, tt := range cases {
t.Run(tt.description, func(t *testing.T){
result := Sum(tt.num1, tt.num2)
if result != tt.expected {
t.Errorf("expected %d, but got %d", tt.expected, result)
}
}
}
}
```
當然這樣的寫法還是有人會質疑，例如：
Q: 你的測試用例太多的話，不就列超長一串，無法見底？
A: 真的太多太長的話，可以用[golden file](https://gitlab.smart-aging.tech/go/go-go-monorepo/-/merge_requests/120)的方法，把test case的input/output，都放到`.golden`檔案當中
Q: 這樣我不就要在一開始寫測試前花很多時間，去設計思考table內要放哪些參數？
A: 這就是今天的主題，請ChatGPT幫我們處理的部分，節省時間
### Table-Driven unit testing prompt for Golang
原本寫測試要多久？這部分因人而異，也要看測試程式碼的複雜程度，少則10分鐘，多至1小時
也因為是使用table driven的template，所以ChatGPT產出蠻一致的
基本上，要填寫很多input/output的測試，節省的時間蠻多的
#### 咒語呢？
咒語包含兩部分，第一個是Table-Driven 的 template
template:
```go
func TestConvertToFHIR(t *testing.T) {
type args struct {
req *CreateVitalSignsRequest
references []repo_fhir.Reference
}
// Define the test cases
testCases := []struct {
name string
args args
want repo_fhir.Observation
wantErr error
}{
{
name: "Empty CreateVitalSignsRequest",
args: args{
req: nil,
references: emptyReferences,
},
want: repo_fhir.Observation{},
wantErr: fmt.Errorf(REQUEST_REQUIRED),
},
{
name: "With Note",
args: args{
req: vitalsSignsRequestWithNoteAndID,
references: emptyReferences,
},
want: expectedVitalSignsWithoutMembersWithNote,
wantErr: nil,
},
// Add more test cases...
}
// Iterate through the test cases
for _, tc := range testCases {
t.Run(tc.name, func(t *testing.T) {
// Call the function with the test case inputs
give := &repo_fhir.Observation{}
err := tc.args.req.ConvertToFHIR(give, tc.args.references)
// Verify the output matches the expected result
if tc.wantErr != nil && err.Error() != tc.wantErr.Error() {
t.Errorf("Expected error to be '%v', but got '%v'", tc.wantErr, err)
}
if diff := cmp.Diff(*give, tc.want); diff != "" {
t.Errorf("[%s] expected result (-got +want):\n%s", tc.name, diff)
}
})
}
}
```
prompt就會是：請ChatGPT用template的形式，將input的code產生table-driven的測試
```
template:
```go func TestConvertToFHIR(t *testing.T) { type args struct { req *CreateVitalSignsRequest references []repo_fhir.Reference } // Define the test cases testCases := []struct { name string args args want repo_fhir.Observation wantErr error }{ { name: "Empty CreateVitalSignsRequest", args: args{ req: nil, references: emptyReferences, }, want: repo_fhir.Observation{}, wantErr: fmt.Errorf(REQUEST_REQUIRED), }, { name: "With Note", args: args{ req: vitalsSignsRequestWithNoteAndID, references: emptyReferences, }, want: expectedVitalSignsWithoutMembersWithNote, wantErr: nil, }, // Add more test cases... } // Iterate through the test cases for _, tc := range testCases { t.Run(tc.name, func(t *testing.T) { // Call the function with the test case inputs give := &repo_fhir.Observation{} err := tc.args.req.ConvertToFHIR(give, tc.args.references) // Verify the output matches the expected result if tc.wantErr != nil && err.Error() != tc.wantErr.Error() { t.Errorf("Expected error to be '%v', but got '%v'", tc.wantErr, err) } if diff := cmp.Diff(*give, tc.want); diff != "" { t.Errorf("[%s] expected result (-got +want):\n%s", tc.name, diff) } }) } ```
\n
Implement table driven tests with the template for the following code
```
#### 實際操作一遍
### Template in Javascript/TypeScript
[prompt for unit test](https://gist.github.com/kostysh/dbd1dfb2181b96563754222903bf67e7)
這個我沒用過，想用的可以試試看
但我覺得不一定要用別人定好的，可以自己要使用的template需要是什麼樣子的
### 還有什麼用途
[example: test cases with API document](https://chat.openai.com/share/f1232bd7-795b-4c56-a1d5-67b4acf39a75)
#### test case的產生？
沒問題，很簡單
```
Write a test cases for the API contract given below
https://xxx.xxx.xx/swagger.json
```
沒公開swagger文件？直接貼上的swagger.json也可以（前提是不要太多字）
```
show results in the table
```
```
create more coverage test cases in table
```
```
write a edge cases and security cases in table
```
#### Test data的產生？
已經都用ChatGPT幫我產生test了，那為什麼還要自己寫，在那邊對格式真的很累，也可以請GPT順便幫你產一下
這句咒語也可接在我們原本產生的test咒語之後唸出
```
provide test data
```
### 客家ㄉ使用ChatGPT
Use ChatGPT with 1/10 of monthly billing
* [Genie AI in Vscode](https://marketplace.visualstudio.com/items?itemName=genieai.chatgpt-vscode)
* [api key](https://help.openai.com/en/articles/4936850-where-do-i-find-my-api-key)
#### 可以做到什麼
* 把咒語存在你的VSCode設定裡，不用每次都重打
* 月費只要繳不到1元美金
* 可以選擇你要用哪個model（gpt-4肯定是比較貴的，可以用3.5就好）
有時用移用會秀抖，沒辦法使用，這時只要去open AI設定中把token revoke，再重新註冊就好
VSCode裡面 `Command+Shift+P` 選擇**Genie: Clear API Key**
#### 缺點
以下需求只能乖乖付300元台幣了
* 沒辦法存下你在ChatGPT裡面的對話
* 沒辦法分享對話
### Extra to recommend-[CodeGPT](https://github.com/appleboy/CodeGPT)
* Help you to write the [commit message](https://github.com/appleboy/CodeGPT?tab=readme-ov-file#cli-mode)
* [Code review](https://github.com/appleboy/CodeGPT?tab=readme-ov-file#code-review)w
## References
- [ChatGPT Prompt Engineering for Developers](https://www.deeplearning.ai/short-courses/chatgpt-prompt-engineering-for-developers/)
- [Prompt Engineering Guide](https://promptingguide.azurewebsites.net/)
Sign in via Facebook
Sign in via X(Twitter)
Sign in via GitHub
Sign in via Dropbox
Sign in with Wallet
Connect another wallet