DevOps Classroom Series – 25/Nov/2021

How Git Works

• Git is a stupid content tracker
• To understand git lets first understand hashing.
• Hashing is the process of converting a given key/text (any length) into another smaller value (fixed size string or number).
• Hashing algorithms
  • md5
  • SHA1
  • SHA256
• Git Use SHA1 hashing to track contents
• Lets go to .git/HEAD which will point towards a branch (.git/refs/heads/<branch>)
• In the branch file we will have the latest commit id of the branch
• The plumbing command git cat-file -t <id> will show the type and git cat-file -p <id> will print the contents
• In Git we have following types
  • tree:
    • This represents a folder or directory
  • blob:
    • This represents a file
  • commit:
    • This represents a commit
• Commit is SHA-1 Hash of
  • contents (Changes)
  • author
  • message
  • parent commit
• Lets use some plumbing commands on master and try to build the tree
• Since we have a basic idea, lets create a new git repository and then create three commits
• For commands used refer classroom video
• Lets try to create a branch ‘develop’ make one commit in it and also one commit in master branch. then merge the chagnes from develop to master branch
• As of now we are good with three areas of git
• Next Steps:
  • Lets focus on fourth area of git i.e. Remote Repositories