python project to solve sandbox dependence problem in production environment without external network

development environment

Establishing Project Development Path

 mkdir -p /data/python/project/

Let's see if there is a pip command tool.
pip's command-line installation can be seen on the official website link:
https://pip.pypa.io/en/stable/installing/#installing-with-get-pip-py

root@-dev:/data/python/project# whereis pip
pip: /usr/bin/pip /usr/local/bin/pip /usr/local/bin/pip2.7 /root/miniconda2/bin/pip /usr/share/man/man1/pip.1.gz

Installation and development of Python virtual environment sandbox development tools

pip install virtualenv #install

Enter the Development Catalogue

cd /data/python/project/

Creating virtual sandbox virtual alenv

virtualenv venv

View Sandbox Catalogue

Entering the virtual environment, we can avoid polluting the python environment of the system by entering the sandbox environment and executing the project code we developed.

root@dev:/data/python/project# source venv/bin/activate
(venv) root@dev:/data/python/project# 

When you see a sandbox environment, the command line headers with a sandbox directory name [venv]

Exit from Virtual Environment

 deactivate

View the differences between virtual python environment and system python environment

(venv) root@dev:/data/python/project# pwd
/data/python/project
(venv) root@dev:/data/python/project# which python
/data/python/project/venv/bin/python
(venv) root@dev:/data/python/project# python -V
Python 2.7.12
(venv) root@dev:/data/python/project# deactivate 
root@dev:/data/python/project# which python
/root/miniconda2/bin/python
root@dev:/data/python/project# python -V
Python 2.7.13 :: Continuum Analytics, Inc.
root@dev:/data/python/project# 

You can see that both python path calls and versions are different

We installed some development modules in the virtual python environment, such as Flask.

(venv) root@dev:/data/python/project# pip install Flask
Collecting Flask
  Using cached Flask-0.12.2-py2.py3-none-any.whl
Collecting itsdangerous>=0.21 (from Flask)
Collecting click>=2.0 (from Flask)
  Using cached click-6.7-py2.py3-none-any.whl
Collecting Werkzeug>=0.7 (from Flask)
  Using cached Werkzeug-0.12.2-py2.py3-none-any.whl
Collecting Jinja2>=2.4 (from Flask)
  Using cached Jinja2-2.9.6-py2.py3-none-any.whl
Collecting MarkupSafe>=0.23 (from Jinja2>=2.4->Flask)
Installing collected packages: itsdangerous, click, Werkzeug, MarkupSafe, Jinja2, Flask
Successfully installed Flask-0.12.2 Jinja2-2.9.6 MarkupSafe-1.0 Werkzeug-0.12.2 click-6.7 itsdangerous-0.24
(venv) root@dev:/data/python/project# 

In order to transplant these required execution modules in the non-extranet environment, the code is first developed in the virtual sandbox environment, and then packaged and deployed.

We pack the whole virtual ALV environment related projects, and generate a complete project package with execution environment. After decompression, we use it, that is, the above development project directory / data/python/project whole package compression.

Use the virtualenv --relocatable instruction to modify the ENV execution environment with variable locations.

Practical operation,
Or project directory / data/python/project

cd /data/python/project

root@dev:/data/python/project# ll
total 12
drwxr-xr-x 3 root root 4096 8 May 15:08 ./
drwxr-xr-x 3 root root 4096 8 May 15:08 ../
drwxr-xr-x 6 root root 4096 8 May 15:08 venv/
root@dev:/data/python/project# 

root@dev:/data/python/project# pwd
/data/python/project
root@dev:/data/python/project# cat venv/bin/pip 
#!/data/python/project/venv/bin/python   ###Focus on this line of code, which is the python reference path before executing virtualenv -- relocatable. / ven

# -*- coding: utf-8 -*-
import re
import sys

from pip import main

if __name__ == '__main__':
    sys.argv[0] = re.sub(r'(-script\.pyw?|\.exe)?$', '', sys.argv[0])
    sys.exit(main())
root@dev:/data/python/project# 

root@dev:/data/python/project# pwd
/data/python/project
root@dev:/data/python/project# ll
total 12
drwxr-xr-x 3 root root 4096 8 May 15:08 ./
drwxr-xr-x 3 root root 4096 8 May 15:08 ../
drwxr-xr-x 6 root root 4096 8 May 15:08 venv/
root@dev:/data/python/project# virtualenv --relocatable ./venv/
Making script /data/python/project/venv/bin/pip relative
Making script /data/python/project/venv/bin/pip2 relative
Making script /data/python/project/venv/bin/python-config relative
Making script /data/python/project/venv/bin/easy_install relative
Making script /data/python/project/venv/bin/wheel relative
Making script /data/python/project/venv/bin/pip2.7 relative
Making script /data/python/project/venv/bin/easy_install-2.7 relative
root@dev:/data/python/project# cat venv/bin/pip
#!/usr/bin/env python2.7    ###Focus on this line of code, which is the python reference path after executing virtualenv -- relocatable. / ven

import os; activate_this=os.path.join(os.path.dirname(os.path.realpath(__file__)), 'activate_this.py'); exec(compile(open(activate_this).read(), activate_this, 'exec'), dict(__file__=activate_this)); del os, activate_this


# -*- coding: utf-8 -*-
import re
import sys

from pip import main

if __name__ == '__main__':
    sys.argv[0] = re.sub(r'(-script\.pyw?|\.exe)?$', '', sys.argv[0])
    sys.exit(main())root@dev:/data/python/project# 

You can see that after virtualenv --relocatable execution, the python path invoked by the pip command in the virtual sandbox environment has changed.

Deploy the project

After the above execution of virtualenv --relocatable

1. The virtual sandbox venv in the development environment is packaged and compressed.

   tar -cvf venv.tar venv/

2. Draw project code to the online production environment directory / app/proj1

   Then throw venv.tar into the / app/proj1 directory and unzip it. Prepare for solving project dependencies.

3. Enter the / app/proj1/venv directory

   A key:
   Modify the active file in bin directory
   vim /app/proj1/venv/bin/active

   Change the variable name: VIRTUAL_ENV from the original packaged / data/python/project/venv to the actual / app/proj1/venv

*** So far, the whole project relies on the solution, and the *** python sandbox environment is transplanted from the development environment to the production environment.