DevOps Classroomnotes 24/Apr/2022

Git Contd…

• Create the following folder structure in a folder called as qtecommerce
  
• initialize the git repository
• create a first commit with all the changes apart from bin directory
• Steps:
  • Create the files and add to the staging area
    
  • commit the changes
    
• This repository should never have history of bin folder as it is generated on developers laptop whenever you run application. Every technology has such folders and file formats
  • java => target
  • python => virtual env
  • .net => bin/debug bin/release
• Now we want to perform the following changes
  
  
• As i never want bin to part of my version control it would be better if it is ignored in all the git commands like git status
• In git we can create a file in the root folder of working tree qtecommerce called as .gitignore.
• In this file mention the folders, file etc.. which you want git to ignore
  
• Add this file to the staging area and execute git status. you should see git ignoring bin directory
  
• Now i can add all the changes at one shot
  
• We can also include some file extensions. Consider the following .pyc which we never want it to be part of version control
  
  
• Add the change to the staging area and see the status
  
• Now commit the changes
  
• Now depending on the technology of your application, there are predefined gitignore which can be generate online, but we should be able to modify them as per our custom needs
• Refer Here for generating .gitignore files:
• Lets assume your developers are working on
  • java with eclipse
    
    
  • python with vscode
• Can you represent the commits which we have done in the graph style
  
  
• The history so far looks linear
• Now lets assume we have got two customers
  • reliance
  • tatagroup
• We need to add some custom features in the code for reliance and tata
• For common features this linear approach is ok, but also need to have a way to add custom features for reliance and for tata group which includes common
• We need an approach which allows us effectively do the parallel development for the following by using the same working tree
  • different customers
  • different releases
  • and so on
    
• This can be acheived by using branching. Branches in git are very light weight and easy
• Now to bring the changes from one branch into another the following options are available.
  • merging
  • cherry-picking
  • rebase