TL;DR requirements.txt中不要使用非ASCII编码的字符，否则会造成字符集错误，无法解析内容

实验设计

  1. 两份requirements.txt文件，包含相同的依赖，但不同的地方是：一份是ASCII编码（或者说只包含英文字符）；另一份包含非ASCII字符（比如中文注释之类的）
  2. 分别安装两份requirements.txt文件，观察现象

实验材料

ASCII编码的requirements.txt

命名为：all_ascii_requirements.txt

具体内容如下：

dummy

非ASCII编码的requirements.txt

命名为：non_ascii_requirements.txt

具体内容如下：

# 这个是中文注释
dummy

实验过程

安装ASCII编码的requirements.txt

执行命令 pip install -r all_ascii_requirements.txt

命令工作正常，输入内容如下：

Collecting dummy (from -r all_ascii_requirements.txt (line 1))
  Downloading dummy-0.1.0.zip
Requirement already up-to-date: jinja2>=2.0.0 in /usr/local/lib/python2.7/dist-packages (from dummy->-r all_ascii_requirements.txt (line 1))
Collecting numpy>=1.0.0 (from dummy->-r all_ascii_requirements.txt (line 1))
  Downloading numpy-1.13.1-cp27-cp27mu-manylinux1_x86_64.whl (16.6MB)
    100% |################################| 16.6MB 66kB/s 
Requirement already up-to-date: mock>=1.0.0 in /usr/local/lib/python2.7/dist-packages (from dummy->-r all_ascii_requirements.txt (line 1))
Requirement already up-to-date: MarkupSafe>=0.23 in /usr/local/lib/python2.7/dist-packages (from jinja2>=2.0.0->dummy->-r all_ascii_requirements.txt (line 1))
Requirement already up-to-date: six>=1.9 in /usr/local/lib/python2.7/dist-packages (from mock>=1.0.0->dummy->-r all_ascii_requirements.txt (line 1))
Requirement already up-to-date: funcsigs>=1; python_version < "3.3" in /usr/local/lib/python2.7/dist-packages (from mock>=1.0.0->dummy->-r all_ascii_requirements.txt (line 1))
Requirement already up-to-date: pbr>=0.11 in /usr/local/lib/python2.7/dist-packages (from mock>=1.0.0->dummy->-r all_ascii_requirements.txt (line 1))
Building wheels for collected packages: dummy
  Running setup.py bdist_wheel for dummy ... done
  Stored in directory: /root/.cache/pip/wheels/fb/72/de/c12e171be0c7bff52d4bcebf680bd3b012203c68b8372b02a5
Successfully built dummy
Installing collected packages: numpy, dummy
  Found existing installation: numpy 1.11.0
    Uninstalling numpy-1.11.0:
      Successfully uninstalled numpy-1.11.0
Successfully installed dummy-0.1.0 numpy-1.13.1

安装非ASCII编码的requirements.txt

执行命令 pip install -r non_ascii_requirements.txt

命令工作出现异常，输出内容如下：

Exception:
Traceback (most recent call last):
  File "/usr/local/lib/python2.7/dist-packages/pip/basecommand.py", line 215, in main
    status = self.run(options, args)
  File "/usr/local/lib/python2.7/dist-packages/pip/commands/install.py", line 312, in run
    wheel_cache
  File "/usr/local/lib/python2.7/dist-packages/pip/basecommand.py", line 295, in populate_requirement_set
    wheel_cache=wheel_cache):
  File "/usr/local/lib/python2.7/dist-packages/pip/req/req_file.py", line 84, in parse_requirements
    filename, comes_from=comes_from, session=session
  File "/usr/local/lib/python2.7/dist-packages/pip/download.py", line 422, in get_file_content
    content = auto_decode(f.read())
  File "/usr/local/lib/python2.7/dist-packages/pip/utils/encoding.py", line 31, in auto_decode
    return data.decode(locale.getpreferredencoding(False))
UnicodeDecodeError: 'ascii' codec can't decode byte 0xe8 in position 2: ordinal not in range(128)

实验结论

pip在解析requirements.txt，只能处理ASCII编码的文件，否则会出现Unicode错误。在编写requirements.txt时，切记使用ASCII编码，不要夹杂中文等非ASCII字符