python自定义安装python包－pip的使用

pip在python3中是自带的，而在python2中需要自行安装，安装步骤见：http://yangl.net/2018/05/18/centos7_install_python2/

1. pip安装的包路径一般默认路径为：
  [python install path]/lib/python[version]/site-packages

以我的路径为例（我的python安装在/share/soft/python-2.7.15）：
  /share/soft/python-2.7.15/lib/python2.7/site-packages

以默认路径安装的python2.7环境为例，对应路径为：
/usr/local/python2.7/lib/python2.7/site-packages

如果是某些非root用户，路径可能为：
$HOME/.loca/lib/python[version]/site-packages

2. pip 查看已经安装的python包：pip list

[root@localhost pip-10.0.1]# pip list
Package Version
---------- -------
pip 10.0.1 
setuptools 39.1.0 
wheel 0.31.1

3. pip在官方python包管理网站 PyPI上搜索相关的包： pip search “包名”

[root@localhost pip-10.0.1]# pip search "biopython" 
biopython-extensions (0.18.6) - Misc utilities and definitions not included or hidden in
 BioPython
biopython (1.71) - Freely available tools for computational molecular biology.
A3MIO (0.0.1) - A3M/A2M I/O for BioPython
Bio_Eutils (1.65) - Standalone version of the Biopython Eutils modules
bioext (0.19.1) - Misc utilities and definitions not included or hidden in
 BioPython
snapgene_reader (0.1.14) - Convert Snapgene *.dna files dict/json/biopython.
bioscripts.convert (0.4) - Biopython scripts for converting molecular sequences between
 formats.
bcbio-gff (0.6.4) - Read and write Generic Feature Format (GFF) with Biopython
 integration.
SnapGeneFileReader (0.1.11) - A Python project to read and write Snapgene *.dna into dict,
 json, and biopython object.
gbgb (0.1.0) - Goodbye, GenBank is a package for use with Biopython that gives
 feature annotations from GenBank records a new and better life.

4. 从PyPI上安装包：pip install 包名

$ pip install SomePackage            # latest version
$ pip install SomePackage==1.0.4     # specific version
$ pip install 'SomePackage>=1.0.4'     # minimum version
#例子:
[root@localhost pip-10.0.1]# pip install biopython
Collecting biopython
 Downloading https://files.pythonhosted.org/packages/5f/d4/f785c7e1c08c41c1773c7bc3e9466709dd07bfc3dba27b9e82a098ad618d/biopython-1.71-cp27-cp27m-manylinux1_x86_64.whl (2.0MB)
 100% |████████████████████████████████| 2.0MB 14kB/s 
Collecting numpy (from biopython)
 Downloading https://files.pythonhosted.org/packages/ea/7d/9f99896cc3f4834871619a36da2a833c71a2178a5bdadd4fff40c261c119/numpy-1.14.3-cp27-cp27m-manylinux1_x86_64.whl (12.1MB)
 100% |████████████████████████████████| 12.1MB 12kB/s 
Installing collected packages: numpy, biopython
Successfully installed biopython-1.71 numpy-1.14.3

5. pip查看包的具体信息，包括包的安装路径 ： pip show 包名

[root@localhost pip-10.0.1]# pip show biopython
Name: biopython
Version: 1.71
Summary: Freely available tools for computational molecular biology.
Home-page: http://www.biopython.org/
Author: The Biopython Contributors
Author-email: biopython@biopython.org
License: UNKNOWN
Location: /share/soft/python-2.7.15/lib/python2.7/site-packages
Requires: numpy
Required-by:

6. 将包安装在自己指定的任意位置：有很多方式，可通过 pip install –help查看，也可查看：https://stackoverflow.com/questions/2915471/install-a-python-package-into-a-different-directory-using-pip
自行指定安装包的路径后需要将路径加入环境变量PYTHONPATH中，不然之后import模块的时候是找不到它的（https://www.jianshu.com/p/92a109ddf695）。［我总结但未尝试］

pip install --target=\somewhere\other\than\the\default package_name #列出我最认可的一种

7. 修改pip的一系列默认行为，包括默认的包安装路径等：修改pip.conf文件，因为我是自定义安装的，没有找到这个文件的位置［没找到其实可自行创建pip.conf配置文件到正确位置］。具体修改方法见官方文档：https://pip.pypa.io/en/stable/user_guide/ ［我总结但未尝试］

比如，来自：http://jyd.me/linux/pip-install-to-different-home/ 的例子

用户目录下面，.pip目录下建立pip.conf文件，文件内容如下：

[install]
install-option=--prefix=~/.local

然后pip的库就安装到用户目录下的.local下面了

参考：

https://pip.pypa.io/en/stable/user_guide/

https://www.zhihu.com/question/48330791

https://blog.csdn.net/feiniao8651/article/details/71467537

https://stackoverflow.com/questions/2915471/install-a-python-package-into-a-different-directory-using-pip

https://www.jianshu.com/p/92a109ddf695

https://segmentfault.com/q/1010000012499098

pip安装到指定目录

尊重他人劳动成果，转载请注明出处:Bluesky's blog » python自定义安装python包－pip的使用