如何使用pytest测试PyQt5视频转文字应用的类方法？

我是Python新手，正在开发一个基于PyQt5的视频转文字语音识别项目。之前用pytest测试过带返回值的普通函数，但现在不知道怎么测试这个项目里UI类的各种方法。已经把目标方法导入测试文件了，但搞不清要测什么、怎么测。目前写了测试主窗口构造函数和非MP4文件选择场景的测试用例，具体代码如下：

项目核心代码 #

#Imported modules
import sys
import wave, contextlib, math, time
import speech_recognition as sr
from moviepy.editor import AudioFileClip
from PyQt5 import QtCore, QtGui, QtWidgets
from PyQt5.QtWidgets import QFileDialog, QMessageBox
from PyQt5.QtCore import QThread, pyqtSignal

class Ui_MainWindow(object):
    """Main window GUI."""
    def __init__(self):
        """Initialisation function."""
        self.mp4_file_name = ""
        self.output_file = ""
        self.audio_file = "speech.wav"
    def setupUi(self, MainWindow):
        """Define visual components and positions."""
        # Main window
        MainWindow.setObjectName("MainWindow")
        MainWindow.resize(653, 836)
        self.centralwidget = QtWidgets.QWidget(MainWindow)
        self.centralwidget.setObjectName("centralwidget")
        self.label = QtWidgets.QLabel(self.centralwidget)
        self.label.setGeometry(QtCore.QRect(50, 20, 161, 41))
        # Selected video file label
        font = QtGui.QFont()
        font.setPointSize(14)
        self.label.setFont(font)
        self.label.setObjectName("label")
        self.selected_video_label = QtWidgets.QLabel(self.centralwidget)
        self.selected_video_label.setGeometry(QtCore.QRect(230, 20, 371, 41))
        font = QtGui.QFont()
        font.setPointSize(8)
        self.selected_video_label.setFont(font)
        self.selected_video_label.setFrameShape(QtWidgets.QFrame.Box)
        self.selected_video_label.setText("")
        self.selected_video_label.setObjectName("selected_video_label")
        self.label_3 = QtWidgets.QLabel(self.centralwidget)
        self.label_3.setGeometry(QtCore.QRect(50, 90, 161, 41))
        # Transcribed text box
        font = QtGui.QFont()
        font.setPointSize(14)
        self.label_3.setFont(font)
        self.label_3.setObjectName("label_3")
        self.transcribed_text = QtWidgets.QTextBrowser(self.centralwidget)
        self.transcribed_text.setGeometry(QtCore.QRect(230, 320, 381, 431))
        self.transcribed_text.setObjectName("transcribed_text")
        self.label_5 = QtWidgets.QLabel(self.centralwidget)
        self.label_5.setGeometry(QtCore.QRect(230, 280, 161, 41))
        font = QtGui.QFont()
        font.setPointSize(14)
        self.label_5.setFont(font)
        self.label_5.setObjectName("label_5")
        self.transcribe_button = QtWidgets.QPushButton(self.centralwidget)
        self.transcribe_button.setEnabled(False)
        self.transcribe_button.setGeometry(QtCore.QRect(230, 150, 221, 81))
        # Transcribe button
        font = QtGui.QFont()
        font.setPointSize(14)
        self.transcribe_button.setFont(font)
        self.transcribe_button.setObjectName("transcribe_button")
        self.transcribe_button.clicked.connect(self.process_and_transcribe_audio)
        # progeress bar
        self.progress_bar = QtWidgets.QProgressBar(self.centralwidget)
        self.progress_bar.setGeometry(QtCore.QRect(230, 250, 381, 23))
        self.progress_bar.setProperty("value", 0)
        self.progress_bar.setObjectName("progress_bar")
        self.message_label = QtWidgets.QLabel(self.centralwidget)
        self.message_label.setGeometry(QtCore.QRect(0, 760, 651, 21))
        # Message label (for errors and warnings)
        font = QtGui.QFont()
        font.setPointSize(8)
        self.message_label.setFont(font)
        self.message_label.setFrameShape(QtWidgets.QFrame.Box)
        self.message_label.setText("")
        self.message_label.setObjectName("message_label")
        self.output_file_name = QtWidgets.QPlainTextEdit(self.centralwidget)
        self.output_file_name.setGeometry(QtCore.QRect(230, 90, 371, 41))
        # Output file name
        font = QtGui.QFont()
        font.setPointSize(14)
        self.output_file_name.setFont(font)
        self.output_file_name.setObjectName("output_file_name")
        # Menubar options
        MainWindow.setCentralWidget(self.centralwidget)
        self.menubar = QtWidgets.QMenuBar(MainWindow)
        self.menubar.setGeometry(QtCore.QRect(0, 0, 653, 21))
        self.menubar.setObjectName("menubar")
        self.menuFile = QtWidgets.QMenu(self.menubar)
        self.menuFile.setObjectName("menuFile")
        self.menuAbout = QtWidgets.QMenu(self.menubar)
        self.menuAbout.setObjectName("menuAbout")
        MainWindow.setMenuBar(self.menubar)
        self.statusbar = QtWidgets.QStatusBar(MainWindow)
        self.statusbar.setObjectName("statusbar")
        MainWindow.setStatusBar(self.statusbar)
        self.actionOpen_mp4_video_recording = QtWidgets.QAction(MainWindow)
        self.actionOpen_mp4_video_recording.setObjectName("actionOpen_mp4_video_recording")
        self.actionOpen_mp4_video_recording.triggered.connect(self.open_audio_file)
        self.actionAbout_vid2text = QtWidgets.QAction(MainWindow)
        self.actionAbout_vid2text.setObjectName("actionAbout_vid2text")
        self.actionAbout_vid2text.triggered.connect(self.show_about)
        self.actionNew = QtWidgets.QAction(MainWindow)
        self.actionNew.setObjectName("actionNew")
        self.actionNew.triggered.connect(self.new_project)
        self.menuFile.addAction(self.actionOpen_mp4_video_recording)
        self.menuFile.addAction(self.actionNew)
        self.menuAbout.addAction(self.actionAbout_vid2text)
        self.menubar.addAction(self.menuFile.menuAction())
        self.menubar.addAction(self.menuAbout.menuAction())
        self.retranslateUi(MainWindow)
        QtCore.QMetaObject.connectSlotsByName(MainWindow)
    def retranslateUi(self, MainWindow):
        """Translate UI method."""
        _translate = QtCore.QCoreApplication.translate
        MainWindow.setWindowTitle(_translate("MainWindow", "MainWindow"))
        self.label.setText(_translate("MainWindow", "Selected video file:"))
        self.label_3.setText(_translate("MainWindow", "Output file name:"))
        self.label_5.setText(_translate("MainWindow", "Transcribed text:"))
        self.transcribe_button.setText(_translate("MainWindow", "Transcribe"))
        self.output_file_name.setPlaceholderText(_translate("MainWindow", "interview1.txt"))
        self.menuFile.setTitle(_translate("MainWindow", "File"))
        self.menuAbout.setTitle(_translate("MainWindow", "About"))
        self.actionOpen_mp4_video_recording.setText(_translate("MainWindow", "Open mp4 video recording"))
        self.actionAbout_vid2text.setText(_translate("MainWindow", "About video to speech"))
        self.actionNew.setText(_translate("MainWindow", "New"))
    def open_audio_file(self):
        """Open the audio (*.mp4) file."""
        file_name = QFileDialog.getOpenFileName()
        if file_name[0][-3:] == "mp4":
            self.transcribe_button.setEnabled(True)
            self.mp4_file_name = file_name[0]
            self.message_label.setText("")
            self.selected_video_label.setText(file_name[0])
        else:
            self.message_label.setText("Please select an *.mp4 file")
    def convert_mp4_to_wav(self):
        """Convert the mp4 video file into an audio file."""
        self.message_label.setText("Converting mp4 to audio (*.wav)...")
        self.convert_thread = convertVideoToAudioThread(self.mp4_file_name, self.audio_file)
        self.convert_thread.finished.connect(self.finished_converting)
        self.convert_thread.start()
    def get_audio_duration(self, audio_file):
        """Determine the length of the audio file."""
        with contextlib.closing(wave.open(audio_file,'r')) as f:
            frames = f.getnframes()
            rate = f.getframerate()
            duration = frames / float(rate)
            return duration
    def transcribe_audio(self, audio_file):
        """Transcribe the audio file."""
        total_duration = self.get_audio_duration(audio_file) / 10
        total_duration = math.ceil(total_duration)
        self.td = total_duration
        if len(self.output_file_name.toPlainText()) > 0:
            self.output_file = self.output_file_name.toPlainText()
        else:
            self.output_file = "my_speech_file.txt"
        # Use thread to process in the background and avoid freezing the GUI
        self.thread = transcriptionThread(total_duration, audio_file, self.output_file)
        self.thread.finished.connect(self.finished_transcribing)
        self.thread.change_value.connect(self.set_progress_value)
        self.thread.start()
    def finished_converting(self):
        """Reset message text when conversion is finished."""
        self.message_label.setText("Transcribing file...")
        self.transcribe_audio(self.audio_file)
    def finished_transcribing(self):
        """This run when transcription finished to tidy up UI."""
        self.progress_bar.setValue(100)
        self.transcribe_button.setEnabled(True)
        self.message_label.setText("")
        self.update_text_output()
    def set_progress_value(self, val):
        """Update progress bar value."""
        increment = int(math.floor(100*(float(val)/self.td)))
        self.progress_bar.setValue(increment)
    def process_and_transcribe_audio(self):
        """Process the audio into a textual transcription."""
        self.transcribe_button.setEnabled(False)
        self.message_label.setText("Converting mp4 to audio (*.wav)...")
        self.convert_mp4_to_wav()
    def update_text_output(self):
        """Update the text box with the transcribed file."""
        f = open(self.output_file, "r")
        self.transcribed_text.setText(f.read())
        f.close()
    def new_project(self):
        """Clear existing fields of data."""
        self.message_label.setText("")
        self.transcribed_text.setText("")
        self.selected_video_label.setText("")
        self.output_file_name.document().setPlainText("")
        self.progress_bar.setValue(0)
    def show_about(self):
        """Show about message box."""
        msg = QMessageBox()
        msg.setWindowTitle("About video to Speech")
        msg.setText("Nelson Petro - CSE 111")
        msg.setIcon(QMessageBox.Information)
        msg.exec_()
class convertVideoToAudioThread(QThread):
    """Thread to convert mp4 video file to wav file."""
    def __init__(self, mp4_file_name, audio_file):
        """Initialization function."""
        QThread.__init__(self)
        self.mp4_file_name = mp4_file_name
        self.audio_file = audio_file
    def __del__(self):
        """Destructor."""
        self.wait()
    def run(self):
        """Run video conversion task."""
        audio_clip = AudioFileClip(self.mp4_file_name)
        audio_clip.write_audiofile(self.audio_file)
class transcriptionThread(QThread):
    """Thread to transcribe file from audio to text."""
    change_value = pyqtSignal(int)
    def __init__(self, total_duration, audio_file, output_file):
        """Initialization function."""
        QThread.__init__(self)
        self.total_duration = total_duration
        self.audio_file = audio_file
        self.output_file = output_file
    def __del__(self):
        """Destructor."""
        self.wait()
    def run(self):
        """Run transcription, audio to text."""
        r = sr.Recognizer()
        for i in range(0, self.total_duration):
            try:
                with sr.AudioFile(self.audio_file) as source:
                    audio = r.record(source, offset=i*10, duration=10)
                    f = open(self.output_file, "a")
                    f.write(r.recognize_google(audio))
                    f.write(" ")
                self.change_value.emit(i)
            except:
                print("Unknown word detected...")
                continue
            f.close()
            
def main():
    app = QtWidgets.QApplication(sys.argv)
    MainWindow = QtWidgets.QMainWindow()
    ui = Ui_MainWindow()
    ui.setupUi(MainWindow)
    MainWindow.show()
    sys.exit(app.exec_())

if __name__ == "__main__":
    main()

已编写的测试代码 #

def test_MainWindowConstructor():
  ui = Ui_MainWindow()
  assert ui.mp4_file_name == ""
  assert ui.output_file == ""
  assert ui.audio_file == "speech.wav"

def test_that_providing_a_raw_file_is_rejected():
  app = QtWidgets.QApplication(sys.argv)
  MainWindow = QtWidgets.QMainWindow()
  ui = Ui_MainWindow()
  ui.setupUi(MainWindow)
  ui.actionOpen_mp4_video_recording.triggered
  QFileDialog.getOpenFileName = "video.raw"  
  ui.open_audio_file
  assert ui.message_label.text == "Please select an *.mp4 file"

UI测试修正与扩展建议 #

1. 修复现有测试的问题 #

你写的第二个测试存在几个关键错误：

• 没有调用方法：ui.open_audio_file只是引用方法，需要加()执行
• 错误修改QFileDialog.getOpenFileName：这是类方法，不能直接赋值字符串，需要用mock模拟返回值
• 错误访问控件属性：ui.message_label.text是方法，要写成ui.message_label.text()

修正后的测试（使用pytest-mock模拟Qt组件）：

import pytest
from PyQt5.QtWidgets import QApplication, QMainWindow
from your_module import Ui_MainWindow  # 替换为你的实际模块名

def test_MainWindowConstructor():
    ui = Ui_MainWindow()
    assert ui.mp4_file_name == ""
    assert ui.output_file == ""
    assert ui.audio_file == "speech.wav"

def test_that_providing_a_raw_file_is_rejected(mocker):
    app = QApplication([])
    MainWindow = QMainWindow()
    ui = Ui_MainWindow()
    ui.setupUi(MainWindow)
    
    # 模拟文件选择对话框返回非MP4文件
    mocker.patch('PyQt5.QtWidgets.QFileDialog.getOpenFileName', return_value=("video.raw", ""))
    
    ui.open_audio_file()  # 执行方法
    
    assert ui.message_label.text() == "Please select an *.mp4 file"
    assert ui.transcribe_button.isEnabled() is False  # 额外验证按钮状态

def test_valid_mp4_file_selection(mocker):
    app = QApplication([])
    MainWindow = QMainWindow()
    ui = Ui_MainWindow()
    ui.setupUi(MainWindow)
    
    test_file = "test_video.mp4"
    mocker.patch('PyQt5.QtWidgets.QFileDialog.getOpenFileName', return_value=(test_file, ""))
    
    ui.open_audio_file()
    
    assert ui.mp4_file_name == test_file
    assert ui.selected_video_label.text() == test_file
    assert ui.message_label.text() == ""
    assert ui.transcribe_button.isEnabled() is True

2. 可扩展的测试场景 #

针对你的项目，还可以覆盖这些测试点：

• new_project方法：验证所有UI元素是否被清空（文本框、进度条、标签等）
• update_text_output方法：临时创建测试输出文件，验证文本框是否正确加载内容
• 进度条计算：测试set_progress_value是否根据输入值正确更新进度条
• 线程信号触发：模拟线程的finished信号，验证后续UI更新逻辑是否执行

3. Qt UI测试注意事项 #

• 用pytest-mock或unittest.mock模拟对话框、文件读写、外部API调用，避免测试依赖真实环境
• 测试时无需显示窗口，避免弹出UI干扰测试流程
• 线程相关逻辑重点测试信号连接后的UI行为，线程内部的业务逻辑可以单独抽离测试

内容的提问来源于stack exchange，提问作者Nelson Petro