Mac M3 Sonoma下py2app打包Azure Speech SDK程序提示找不到azure模块

环境：Mac OS M3 Pro Sonoma，使用Poetry管理依赖，Azure Speech SDK版本1.41.1，尝试用py2app将基于Tkinter的Python程序打包成.dmg文件，运行构建时报错No module named 'azure'。

原setup.py代码 #

import sys
import os
import site
from typing import List, Dict, Any

APP_NAME: str = 'app'
VERSION: str = '1.0.0'
MIN_PYTHON_VERSION: str = '>=3.13'

REQUIRED_PACKAGES: List[str] = [
    'azure-cognitiveservices-speech>=1.41.1',
    'python-dotenv>=0.19.0',
]

SETUP_REQUIRES: List[str] = [
    'py2app',
    'setuptools>=44.0.0',
    'azure-cognitiveservices-speech>=1.41.1',
]

# Add Azure SDK import at module level with error handling
try:
    import azure.cognitiveservices.speech as speech_sdk
    SPEECH_SDK_PATH = os.path.dirname(speech_sdk.__file__)
    AZURE_ROOT = os.path.dirname(os.path.dirname(os.path.dirname(SPEECH_SDK_PATH)))
except ImportError:
    SPEECH_SDK_PATH = None
    print("Warning: Azure Speech SDK not found. Build may be incomplete.")

def log_environment_info() -> None:
    """Log environment information for debugging purposes.
    
    Raises:
        ImportError: If azure speech SDK cannot be imported
    """
    print("Python interpreter being used:", sys.executable)
    print("Python path:", sys.path)
    print("Site packages:", site.getsitepackages())
    
    try:
        # List all files in the azure package directory
        azure_dir = os.path.dirname(os.path.dirname(os.path.dirname(speech_sdk.__file__)))
        print("Contents of azure package directory:")
        for root, dirs, files in os.walk(azure_dir):
            print(f"Directory: {root}")
            for file in files:
                print(f"  - {file}")
    except ImportError as e:
        print("Import Error:", str(e))
        raise

# Call this before setup()
log_environment_info()

APP = ['app.py']
DATA_FILES = [
    'requirements.txt',
    'settings.ini',
    '.venv',
    'languages.json',
]

# Get the site-packages directory and package paths
SITE_PACKAGES = site.getsitepackages()[0]
AZURE_PATH = os.path.join(SITE_PACKAGES, 'azure')
AZURE_SPEECH_PATH = os.path.join(SITE_PACKAGES, 'azure', 'cognitiveservices', 'speech')

OPTIONS: Dict[str, Any] = {
    'argv_emulation': True,
    'includes': [
        'azure',
        'azure.cognitiveservices',
        'azure.cognitiveservices.speech',
    ],
    'packages': [
        'azure',
        'azure.cognitiveservices',
        'azure.cognitiveservices.speech',
    ],
    'site_packages': True,
    'resources': [
        AZURE_ROOT,  # Include the entire azure package
        SPEECH_SDK_PATH,  # Include the speech SDK specifically
    ],
    'frameworks': [],
    'plist': {
        'CFBundleName': 'App',
        'CFBundleShortVersionString': '1.0.0',
        'CFBundleVersion': '1.0.0',
        'CFBundleIdentifier': 'com.yourdomain.domain',
        'NSMicrophoneUsageDescription': 'test app',
        'LSMinimumSystemVersion': '10.10',
    }
}

setup(
    name=APP_NAME,
    version=VERSION,
    packages=find_packages(),
    app=APP,
    data_files=DATA_FILES,
    options={'py2app': OPTIONS},
    setup_requires=SETUP_REQUIRES,
    install_requires=REQUIRED_PACKAGES,
    python_requires=MIN_PYTHON_VERSION,
    description='Kasana Application',
    long_description=open('README.md').read(),
    long_description_content_type='text/markdown',
    author='Your Name',
    author_email='your.email@example.com',
    url='https://github.com/yourusername/app',
    classifiers=[
        'Development Status :: 5 - Production/Stable',
        'Intended Audience :: Developers',
        'Programming Language :: Python :: 3',
        'Programming Language :: Python :: 3.13',
    ],
)

报错栈信息 #

1120 total
42 filtered
150 orphaned
1078 remaining
PACKAGE DIRECTORY /Users/sethhammock/Code/Repos/kasana-production/azure
PACKAGE DIRECTORY /Users/sethhammock/Code/Repos/kasana-production/.venv/lib/python3.11/site-packages/azure/azure
PACKAGE DIRECTORY /Users/sethhammock/Code/Repos/kasana-production/azure
PACKAGE DIRECTORY /Users/sethhammock/.pyenv/versions/3.11.6/lib/python311.zip/azure
PACKAGE DIRECTORY /Users/sethhammock/.pyenv/versions/3.11.6/lib/python3.11/azure
PACKAGE DIRECTORY /Users/sethhammock/.pyenv/versions/3.11.6/lib/python3.11/lib-dynload/azure
PACKAGE DIRECTORY /Users/sethhammock/Code/Repos/kasana-production/.venv/lib/python3.11/site-packages/azure
PACKAGE DIRECTORY /Users/sethhammock/Code/Repos/kasana-production/.venv/lib/python3.11/site-packages/setuptools/_vendor/azure
Traceback (most recent call last):
  File "/Users/sethhammock/Code/Repos/kasana-production/setup.py", line 135, in <module>
    setup(
  File "/Users/sethhammock/Code/Repos/kasana-production/.venv/lib/python3.11/site-packages/setuptools/__init__.py", line 117, in setup
    return distutils.core.setup(**attrs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/sethhammock/Code/Repos/kasana-production/.venv/lib/python3.11/site-packages/setuptools/_distutils/core.py", line 183, in setup
    return run_commands(dist)
           ^^^^^^^^^^^^^^^^^^
  File "/Users/sethhammock/Code/Repos/kasana-production/.venv/lib/python3.11/site-packages/setuptools/_distutils/core.py", line 199, in run_commands
    dist.run_commands()
  File "/Users/sethhammock/Code/Repos/kasana-production/.venv/lib/python3.11/site-packages/setuptools/_distutils/dist.py", line 954, in run_commands
    self.run_command(cmd)
  File "/Users/sethhammock/Code/Repos/kasana-production/.venv/lib/python3.11/site-packages/setuptools/dist.py", line 999, in run_command
    super().run_command(command)
  File "/Users/sethhammock/Code/Repos/kasana-production/.venv/lib/python3.11/site-packages/setuptools/_distutils/dist.py", line 973, in run_command
    cmd_obj.run()
  File "/Users/sethhammock/Code/Repos/kasana-production/.venv/lib/python3.11/site-packages/py2app/build_app.py", line 987, in run
    self._run()
  File "/Users/sethhammock/Code/Repos/kasana-production/.venv/lib/python3.11/site-packages/py2app/build_app.py", line 1217, in _run
    self.run_normal()
  File "/Users/sethhammock/Code/Repos/kasana-production/.venv/lib/python3.11/site-packages/py2app/build_app.py", line 1329, in run_normal
    pkgdirs = self.collect_packagedirs()
              ^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/sethhammock/Code/Repos/kasana-production/.venv/lib/python3.11/site-packages/py2app/build_app.py", line 1293, in collect_packagedirs
    [
  File "/Users/sethhammock/Code/Repos/kasana-production/.venv/lib/python3.11/site-packages/py2app/build_app.py", line 1294, in <listcomp>
    os.path.join(os.path.realpath(self.get_bootstrap(pkg)), "")
                                  ^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/sethhammock/Code/Repos/kasana-production/.venv/lib/python3.11/site-packages/py2app/build_app.py", line 2205, in get_bootstrap
    bootstrap = imp_find_module(bootstrap)[1]
                ^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/sethhammock/Code/Repos/kasana-production/.venv/lib/python3.11/site-packages/modulegraph/util.py", line 34, in imp_find_module
    result = imp.find_module(name, path)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/sethhammock/.pyenv/versions/3.11.6/lib/python3.11/imp.py", line 298, in find_module
    raise ImportError(_ERR_MSG.format(name), name=name)
ImportError: No module named 'azure'

1. 修正Python版本不一致问题 #

setup.py中指定的MIN_PYTHON_VERSION为>=3.13，但实际使用的Python版本是3.11.6，版本不匹配会导致依赖识别异常。将MIN_PYTHON_VERSION修改为>=3.11，同时更新classifiers中的Python版本：

MIN_PYTHON_VERSION: str = '>=3.11'

# ...

classifiers=[
    # ...
    'Programming Language :: Python :: 3.11',
],

2. 移除干扰性目录 #

• 从DATA_FILES中删除.venv，虚拟环境目录无需打包，反而会干扰py2app的路径识别：

DATA_FILES = [
    'requirements.txt',
    'settings.ini',
    'languages.json',
]

• 检查项目根目录是否存在名为azure的文件夹（报错中显示/Users/sethhammock/Code/Repos/kasana-production/azure），如果这不是Python包，直接删除该文件夹，避免py2app误识别。

3. 调整py2app配置，排除冲突模块 #

修改OPTIONS配置，移除冗余的resources路径，添加excludes排除干扰的azure目录，同时精简includes和packages到实际需要的模块：

OPTIONS: Dict[str, Any] = {
    'argv_emulation': True,
    'includes': [
        'azure.cognitiveservices.speech',
        'python_dotenv',
    ],
    'packages': [
        'azure.cognitiveservices.speech',
    ],
    'site_packages': True,
    'excludes': [
        'setuptools._vendor.azure',
    ],
    'plist': {
        'CFBundleName': 'App',
        'CFBundleShortVersionString': '1.0.0',
        'CFBundleVersion': '1.0.0',
        'CFBundleIdentifier': 'com.yourdomain.domain',
        'NSMicrophoneUsageDescription': 'test app',
        'LSMinimumSystemVersion': '10.10',
    }
}

4. 修正Azure路径计算错误 #

原代码中AZURE_ROOT的路径计算多了一层dirname，导致指向site-packages而非azure包目录，修正为：

try:
    import azure.cognitiveservices.speech as speech_sdk
    SPEECH_SDK_PATH = os.path.dirname(speech_sdk.__file__)
    AZURE_ROOT = os.path.dirname(os.path.dirname(SPEECH_SDK_PATH))  # 仅两次dirname，指向azure包根目录
except ImportError:
    SPEECH_SDK_PATH = None
    print("Warning: Azure Speech SDK not found. Build may be incomplete.")

5. 调整packages参数 #

如果项目根目录没有自定义Python包，将setup中的packages=find_packages()改为packages=[]，避免py2app误识别无关目录：

setup(
    # ...
    packages=[],
    # ...
)

6. 正确使用Poetry虚拟环境打包 #

运行打包命令时，确保使用Poetry的虚拟环境：

poetry run python setup.py py2app

内容的提问来源于stack exchange，提问作者serxio