如何使用appium为ui选择器正确安装测试-ai-分类器插件？

Question

遇到的问题是在ubuntu 20.04。

遵循https://github.com/testdotai/appium-classifier-plugin的指示

我在appium路径下安装了测试-ai-分类器，并在全局上使用npm安装-g测试-ai-分类器。

我确保appium医生没有在命令行appium- not中给我任何错误。我访问了appium路径，并从命令行执行了以下操作：

1. node
2. require('test-ai-classifier');

总是有相同的错误：

Thrown:
Error: Cannot find module 'test-ai-classifier'
Require stack:

at Function.Module._resolveFilename (internal/modules/cjs/loader.js:797:15)
at Function.Module._load (internal/modules/cjs/loader.js:690:27)
at Module.require (internal/modules/cjs/loader.js:852:19)
at require (internal/modules/cjs/helpers.js:74:18) { code: 'MODULE_NOT_FOUND', requireStack: [ '' ]

你们中有人遇到过这个问题吗？如果是，你是怎么解决的？你用的是哪种环境？您是在全球范围内还是在appium的路径内安装了测试-ai-分类器？

Answer 1

环境：

Windows 10
Tensorflow=1.1.0
python=2.7
global modules path = C:\Users\natha\AppData\Roaming\npm
appium = 1.19.1
node = 14.15.0
npm=6.14.9
test-ai-classifies=2.1.1

先决条件:安装Android并配置ANDROID_HOME、安装Java和配置JAVA_HOME

Appium安装步骤：

npm install -g appium --drivers=xuitest,uiautomator2
npm install -g wd
appium -a 127.0.0.1 -p 4723
npm install -g appium-doctor
appium-doctor
(appium 1.19.1 is installed)

测试-ai-分类器安装-全局：

npm install -g node-pre-gyp@0.11.0
npm install --unsafe-perm --verbose -g node-sass
npm install -g node-gyp --unsafe-perm=true
//NODE_GYP_FORCE_PYTHON=python
//NODE_GYP_FORCE_PYTHON=python2
npm install -g test-ai-classifier@2.1.1 --unsafe-perm=true --build-from-source --python=python2.7

测试-ai-分类器安装-在appium路径：

where appium
C:\Users\natha\AppData\Roaming\npm\appium
cd C:\Users\natha\AppData\Roaming\npm\node_modules\appium\
npm install node-pre-gyp@0.11.0
npm install --unsafe-perm --verbose node-sass
npm install node-gyp --unsafe-perm=true
npm install @tensorflow/tfjs-node@1.2.3
npm install test-ai-classifier@2.1.1

全局节点模块配置:如果要更改全局模块路径，请使用Where does npm install packages?配置编辑，并在文件中放置前缀=/my/ npm /全局/模块/前缀

change from
; prefix=C:\Program Files\nodejs
to
; prefix=C:\Users\natha\AppData\Roaming\npm

如何验证测试-ai-分类器的安装？

[https://github.com/cloudgrey-io/appiumpro/issues/9][2]

If you go into the place where appium is installed, and run node, you'll get an interactive node.js prompt. From there, run require("test-ai-classifier");
If you get an error, it means the classifier module is not correctly installed and accessible to appium.
If not, it means everything is good, just make sure you restarted the appium server after installing it.

在InteeliJ中创建一个java类，并添加以下代码：

package com.example.appium;

import io.appium.java_client.MobileBy;
import org.openqa.selenium.remote.CapabilityType;
import org.openqa.selenium.remote.DesiredCapabilities;
import org.openqa.selenium.support.ui.ExpectedConditions;
import org.testng.annotations.Test;
import io.appium.java_client.android.AndroidDriver;


import java.io.File;
import java.net.MalformedURLException;
import java.net.URL;
import java.util.HashMap;
import java.util.concurrent.TimeUnit;

public class TestAppiumAI {


    private static AndroidDriver driver;

    @Test
    public void installTest() throws MalformedURLException, InterruptedException {

        DesiredCapabilities capabilities = new DesiredCapabilities();
        capabilities.setCapability("deviceName", "MotoG");
        capabilities.setCapability("platformVersion", "7.1.1");
        capabilities.setCapability("platformName","Android");
        capabilities.setCapability("udid", "ZH33L2Z6KL");
        capabilities.setCapability("automationName", "UiAutomator2");
        capabilities.setCapability("skipServerInstallation", "true");
        capabilities.setCapability("autoGrantPermissions", "true");
        capabilities.setCapability("noReset", "false");


        capabilities.setCapability("appPackage", "com.android.settings");
        capabilities.setCapability("appActivity","com.android.settings.Settings");

        HashMap<String, String> customFindModules = new HashMap<String, String>();
        customFindModules.put("ai", "test-ai-classifier");

        capabilities.setCapability("customFindModules", customFindModules);
        capabilities.setCapability("shouldUseCompactResponses", false);


        driver = new AndroidDriver(new URL("http://127.0.0.1:4723/wd/hub"), capabilities);
        driver.manage().timeouts().implicitlyWait(5, TimeUnit.SECONDS);

        //driver.findElement(MobileBy.custom("ai:search")).click();
        driver.findElement(MobileBy.custom("ai:bluetooth")).click();
        driver.quit();

    }
}

创建TestNG.xml，如下所示：

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE suite SYSTEM "http://testng.org/testng-1.0.dtd">
<suite name="Regression" thread-count="2" parallel="tests" preserve-order="true" verbose="2">

    <test name="Real Device1">
        <classes>
            <class name="com.example.appium.TestAppiumAI"/>
        </classes>
    </test>
</suite>

从命令行启动Appium服务器：

appium -a 127.0.0.1 -p 4723

右键单击TestNG.xml并执行。观察蓝牙下的设置应用程序是被选中的。请随意获取源代码/项目

[https://github.com/anbunathan/appium-examples/blob/master/Template1/src/test/java/com/example/appium/TestAppiumAI.java][2]

另外，我还添加了一个视频：https://youtu.be/CFDSejLJEQY

Answer 2

在我的例子中，我使用python并在Ubuntu20.04上工作。Anbunathan也为那些使用或热爱java并在Windows上工作的人提供了一个答案

先决条件：

Install Android SDK and configure ANDROID_HOME
Install Java SDK and configure JAVA_HOME

环境：

Ubuntu 20.04,
Tensorflow=2.3.1,
Tensorflow-estimator=2.3.0,
Tensorflow-gpu=2.2.0,
python=2.7.18,
appium = 1.19.1,
node = 12.14.0,
npm=6.13.4,
test-ai-classifier=1.0.0

Put the following at the end of your bashrc
#Nodejs
VERSION=v12.14.0
DISTRO=linux-x64
export PATH=/usr/local/lib/node-$VERSION-$DISTRO/bin:$PATH

Appium安装步骤(我没有在全局安装appium )：

On the command line, in your home directory, create a directory for global installations: mkdir ~/.npm-global

Configure npm to use the new directory path: npm config set prefix '~/.npm-global'

In your prefer text editor, open or create a /.profile file and add this line to it: export PATH=/.npm-globa/bin:$PATH

On the command line, update your system variables: source ~/.profile

To test your new configuration, install a package globally without using sudo: npm install -g jshint

Instead of steps 2-4, you can use the corresponding ENV variable (e.g. if you don't want to modify ~/.profile): NPM_CONFIG_PREFIX=/home/user/.npm-global

npm install appium
npm install wd --> in the command line start appium
npm install appium-doctor --> in the command line start appium-doctor

Do not forget to set the path of appium in your bashrc (export NODE_PATH=~/.npm-global/lib/node_modules/appium:$PATH)

If you want to do it globally just do the following:

npm install -g appium --drivers=xuitest,uiautomator2
npm install -g wd --> in the command line start appium
npm install -g appium-doctor --> in the command line start appium-doctor

测试-ai-分类器安装-在appium路径：

go to cd ~/.npm-global/lib/node_modules/appium
npm install test-ai-classifier@1.0.0

You might want to install this if you ever face some problems:
npm install node-pre-gyp@0.11.0
npm install --unsafe-perm --verbose node-sass
npm install node-gyp --unsafe-perm=true
npm install @tensorflow/tfjs-node@1.2.3

为了确保安装了测试-ai-分类器，执行以下操作：

Go to your command line
cd ~/.npm-global/lib/node_modules/appium
Enter node
Enter require('test-ai-classifier');

If you get an error, it means the classifier module is not correctly installed and accessible to appium.

If not, it means everything is good, just make sure you restarted the appium server after installing it.

Create your python class using your favorite IDE (Android Studio in my case).
Go to your command line and start appium(appium -a 127.0.0.1 -p 4723).
Check as well if appium-doctor does not show any error. Start it from where you installed it.
Connect your device or start your simulator from your IDE.
Go to command line and start your python script. In my case I used python3 for the shebang but working with python 2.7.

请随意查看下面的源代码项目：

https://github.com/ericnana/appium_classifier_ai_python