Home  >  Article  >  Backend Development  >  How to manage tests that require API keys

How to manage tests that require API keys

WBOY
WBOYforward
2024-02-09 23:06:09813browse

如何管理需要 API 密钥的测试

php editor Apple introduces you how to manage tests that require API keys. During development, many applications require API keys to access third-party services. However, managing these keys can become complex and lengthy. To simplify this process, we can adopt some effective management strategies, such as using environment variables, key management tools, key rotation, etc. These methods not only improve security but also make testing and deployment easier for developers. In this article, we'll detail these strategies and provide some suggestions to help you better manage tests that require API keys.

Question content

I have some seal tests that work just fine. Code in foo.go is tested in foo_test.go.

But I also have some tests that require an API key and I want to keep those separate from the sealed tests since we don't run them in CI.

How to effectively isolate these tests in a way that plays well with the Go tooling and ecosystem?

Solution

There are multiple ways to skip testing. Most of them utilize testing.t.skip(now). Here are some common use cases for skipnow. However, since this is just like any old function call, you can get as creative as you want.

Use environment variables

package main

import (
    "os"
    "testing"
)

func testalways(t *testing.t) {
}

func testsometimes(t *testing.t) {
    key := os.getenv("foo_api_key")
    if key == "" {
        t.skip("foo_api_key is empty")
    }
}

Run one of the following:

go test -v
foo_api_key=bar go test -v

Use the built-in short flag

package main

import (
    "testing"
)

func testalways(t *testing.t) {
    // ...
}

func testsometimes(t *testing.t) {
    if testing.short() {
        t.skip("-short is set")
    }

    // ...
}

Run one of the following:

go test -v -short
go test -v

Use custom flags

package main

import (
    "flag"
    "testing"
)

var withfoo = false // or true to run test by default

func init() {
    flag.boolvar(&withfoo, "with-foo", withfoo, "include foo tests")
}

func testalways(t *testing.t) {
    // ...
}

func testsometimes(t *testing.t) {
    if !withfoo {
        t.skip("-with-foo is not set")
    }

    // ...
}

Run one of the following:

go test -v
go test -v -with-foo

Using Building Constraints

// main_test.go
package main

import (
    "testing"
)

func testalways(t *testing.t) {
    // ...
}
// foo_test.go
//go:build foo
package main

import (
    "testing"
)

func testsometimes(t *testing.t) {
    // ...
}

Run one of the following:

go test -v
go test -v -tags=foo

Please note that with build flags, the output will not indicate that the test has been skipped. A tagged .go file is invisible to the compiler unless the build flag is included on the command line (or in goflags).

The above is the detailed content of How to manage tests that require API keys. For more information, please follow other related articles on the PHP Chinese website!

Statement:
This article is reproduced at:stackoverflow.com. If there is any infringement, please contact admin@php.cn delete