• Home
• Blog

Every time I create new blog these are the steps involved. (until I realised easy way - see below for update)

• gatsby build
• Remove content of Github Pages repo and push the commit, this ensures unused files are removed.
• Copy content of public folder from source repo to GitHub Pages repo, push the commit.
• Finally source repo needs to be commited and pushed.

As I was doing it manually, this is against DRY (Don’t Repeat Yourself) principle.

This is how I come up with automating this task with these bash scripts:

1. One repo requires cleaning old content or else unused files will live on.

    #!/bin/bash
    
    REPO_DIR=/Users/vinkrish/Documents/workspace/javascript/vinkrish.github.io
    cd $REPO_DIR
    rm -r *
    
    git add .
    git commit -m "deleting cache and previous files"
    git push origin master

2. Updating two repos with newly added blog post

    #!/bin/bash
    
    read blog_name
    REPO_DIR=/Users/vinkrish/Documents/workspace/javascript/vinkrish.github.io
    SRC_DIR=/Users/vinkrish/Documents/workspace/javascript/my-blog-starter
    
    cp -R $SRC_DIR/public/* $REPO_DIR
    
    cd $REPO_DIR
    git add .
    git commit -m "added $blog_name"
    git push origin master
    
    cd $SRC_DIR
    git add .
    git commit -m "added $blog_name"
    git push origin master

You need to update .bash_profile:

export PATH=$PATH:/Users/vinay/bin

Your script will be in bin directory:

cd ~
mkdir bin

Give permission for the script to be executable:

chmod u+x file_name

Now you can call file_name from terminal.

Update: I realised gh-pages allowes public or docs folder of a app to be github pages, so I created another branch which has gatsby code and master branch now has generated public files, the above steps are unnecessary if you decide to go with this approach.

Hi, I'm Vinay - a Programmer who "Read-Evaluate-Loop". You can find me on GitHub LinkedIn Twitter