CentOS下完全离线安装python3.6.8、pip3和numpy

CentOS下完全离线安装python3.6.8、pip3、numpy、flask

本文是基于ubuntu系统环境，安装和使用python3.6.8、pip3、numpy、flask

• CentOS 7
• python 3.6.8
• http://rpm.pbone.net/ 下载各种安装包的地址

---

(1) 安装python3.6.8

• 下载并安装依赖包 zlib-devel
  zlib-devel-1.2.7-17.el7.x86_64.rpm

rpm -ivh zlib-devel-1.2.7-17.el7.x86_64.rpm

• 下载并安装依赖包 bzip2-devel
  bzip2-devel-1.0.6-13.el7.x86_64.rpm

rpm -ivh bzip2-devel-1.0.6-13.el7.x86_64.rpm

• 下载并安装依赖包 epel-release
  epel-release-7-11.noarch.rpm

rpm -ivh epel-release-7-11.noarch.rpm

• 下载并安装依赖包 ncurses-devel
  ncurses-devel-5.9-13.20130511.el7.x86_64.rpm

rpm -ivh ncurses-devel-5.9-13.20130511.el7.x86_64.rpm 

• 下载并安装依赖包 gcc
  mpfr-3.1.1-4.el7.x86_64.rpm
  libmpc-1.0.1-3.el7.x86_64.rpm
  kernel-headers-3.10.0-514.el7.x86_64.rpm
  glibc-headers-2.17-157.el7.x86_64.rpm
  glibc-devel-2.17-157.el7.x86_64.rpm
  cpp-4.8.5-11.el7.x86_64.rpm
  gcc-4.8.5-11.el7.x86_64.rpm

# 一定要按照以下的顺序进行安装
rpm -ivh mpfr-3.1.1-4.el7.x86_64.rpm
rpm -ivh libmpc-1.0.1-3.el7.x86_64.rpm
rpm -ivh kernel-headers-3.10.0-514.el7.x86_64.rpm 
rpm -ivh glibc-headers-2.17-157.el7.x86_64.rpm 
rpm -ivh glibc-devel-2.17-157.el7.x86_64.rpm 
rpm -ivh cpp-4.8.5-11.el7.x86_64.rpm 
rpm -ivh gcc-4.8.5-11.el7.x86_64.rpm

• 下载并安装依赖包 gcc-c++
  libstdc+±4.8.5-11.el7.x86_64.rpm
  libstdc+±devel-4.8.5-11.el7.x86_64.rpm
  gcc-c+±4.8.5-11.el7.x86_64.rpm

rpm -ivh libstdc++-4.8.5-11.el7.x86_64.rpm
rpm -ivh libstdc++-devel-4.8.5-11.el7.x86_64.rpm 
rpm -ivh gcc-c++-4.8.5-11.el7.x86_64.rpm 

• 官网下载python 3.6.8
  Python-3.6.8.tgz
• 进入下载目录，进行解压

tar -xzvf Python-3.6.8.tgz

• 进入解压目录

cd Python-3.6.8

• 设置安装路径并安装

./configure --prefix=/usr/local/python3.6 --enable-shared
make
make install
ln -s /usr/local/python3.6/bin/python3 /usr/bin/python3
echo "/usr/local/python3.6/lib" > /etc/ld.so.conf.d/python3.6.conf
ldconfig

• 检查python 3.6.8是否安装成功

python3 -V

(2) 安装pip

• python 3.6.8中自带了pip，所以只需要建立链接就行

ln -s /usr/local/python3.6/bin/pip3.6 /usr/bin/pip

• 检查pip是否安装成功

pip -V

(3) 安装第三方包numpy

使用pip安装numpy

pip install numpy

离线安装numpy

1. 第一种方式

• 去官网下载numpy安装包.whl文件
  numpy-1.16.4-cp36-cp36m-manylinux1_x86_64.whl
• 打开命令终端

pip install numpy-1.16.4-cp36-cp36m-manylinux1_x86_64.whl

2. 第二种方式

• 去官网下载numpy安装包.tgz文件

# ".tar.gz"文件解压
tar -xzvf numpy-1.15.0.tar.gz 
# 进入目录"numpy-1.15.0"
cd numpy-1.15.0
# 找到“setup.py”文件，然后安装
python3 setup.py install

(4) pip 命令的使用方法

• 查看已经安装的包

pip list

(5) pip 离线安装flask

• 下载flask的安装包
  Flask-1.1.1-py2.py3-none-any.whl
• 查看flask的依赖包

Click>=5.1
Werkzeug>=0.15
Jinja2>=2.10.1 # Jinja2 依赖 MarkupSafe>=0.23
itsdangerous>=0.24

• 下载flask各种依赖的安装包
  Click-7.0-py2.py3-none-any.whl
  Werkzeug-0.15.5-py2.py3-none-any.whl
  MarkupSafe-1.1.1-cp36-cp36m-manylinux1_x86_64.whl
  Jinja2-2.10.1-py2.py3-none-any.whl
  itsdangerous-1.1.0-py2.py3-none-any.whl

• 按照下面的顺序，pip安装

pip install Click-7.0-py2.py3-none-any.whl
pip install Werkzeug-0.15.5-py2.py3-none-any.whl
pip install MarkupSafe-1.1.1-cp36-cp36m-manylinux1_x86_64.whl
pip install Jinja2-2.10.1-py2.py3-none-any.whl
pip install itsdangerous-1.1.0-py2.py3-none-any.whl
pip install Flask-1.1.1-py2.py3-none-any.whl

• 使用pip list命令，查看flask是否已安装

pip list

---