Python + Github + GraphQL = PyGitGraph

• 
• Python + Github + GraphQL = PyGitGraph
• "PyGitGraph" is a comprehensive suite of Jupyter notebooks for efficiently working with GitHub issues via GitHub's GraphQL API
• the code in this repo allows you to analyze, visualize, and manage your Github issues like a pro!
• future roadmap involves extending this functionality to PRs, Discussions and other Github functionalities

blog post on "PyGitGraph"

• I have made a blog post on my website covering PyGitGraph. You can find it here. The blog post looks at the tool in a more non-technical way.
• Alternatively, you can continue reading up more about the tool in the sections below.

purpose of "PyGitGraph"

• GitHub's Issues feature provides robust project management tools, and its GraphQL API enables deep data access
• this repository offers Jupyter notebooks that allow you to:
  • create new GitHub issues
  • retrieve and analyze existing issues
  • analyze trends and generate reports like these ones here

use cases of "PyGitGraph"

data analysis

• export and analyze your issues to answer questions like:
  • what is the average issue open time?
  • which labels have the most associated issues?
• create interactive graphs and rich visualizations from your data e.g.
• 
• 

time tracking

• track tasks using github issues:
  • open an issue to start a task
  • close it when done
  • start new issues for new tasks
• use the notebooks to extract data, analyze trends, and generate reports e.g.
• 

data backup

• github does not provide a direct way to back up issues
• these notebooks aggregate and export issues to safeguard your data. e.g.

features of "PyGitGraph"

completed

• fetch issues - retrieve issues by various parameters
• filter issues - filter issues by labels, time, etc
• create issues - create issues with specific titles, descriptions, and labels
• close issues - close open issues
• delete issues - delete issues
• aggregate issues - combine issues into a single file/dataframe
• visualize issues - create visualizations and export data for other tools

in progress

• Pull Requests & Discussions - implement similar actions for pull requests github discussions
• Bulk Actions - Use CSV/JSON input for batch operations on several issues at the same time

folder structure of this repository

1. set parameters - configuration settings for queries and scripts. details.
2. work with issues - fetch, filter, create, close, and delete issues
   • fetch issues
   • filter issues
   • create issues
   • close issues
   • delete issues
3. aggregate data - append and aggregate data for backup and analysis. learn more
4. visualize data - generate visualizations from aggregated data. explore
5. clean up data - remove files created by notebooks. details

use "PyGitGraph"

• 4 out-of-the-box options
  • you have four setup options to work with this code
    • using Binder (no installation or registration needed)
    • local python installation WITHOUT any virtual environment
    • local python installation WITH a virtual environment
    • VSCode with Devcontainers
  • you can find more info for each of this option in the next 4 sections
• you do you
  • of course, you are not limited to these 4 options only
  • you can have your own setup approach as well
  • there is restriction set by the repo or the code

create github token

• prerequisite
  • creating a github token is prerequisite for interacting with Github GraphQL
  • you would be asked for it when executing the code in any of the other files in this repo
  • dont worry, you will be asked just once - see more details here
  • find here details on how to create a new token
• asked just once
  • you would not be asked to enter this token again and again...just once
  • you would asked for it once when you start with executing any of the jupyter notebooks in this repo
• next steps
  • assuming you now have the token, you can use any of the below 4 options to use this repo

using Binder

• just launch
  • simply click the badge below to launch this repository in Binder:
  • this will launch a live environment where you can interact with the Jupyter notebooks and try out the code without any installation
• all done
  • you are now free to execute any of the notebook in this repo

local python installation

• create a github token

  • create a token and keep it secure
  • you would be asked for it when executing the code in any of the other files in this repo
  • dont worry, you will be asked just once - see more details here

• clone the repository

  git clone https://github.com/kanad13/github_issues_explorer.git
  cd PyGitGraph

• install dependencies

  pip install -r ./requirements.txt

• all done

  • you are now free to execute any of the notebook in this repo

python virtual environment

• create a github token

  • create a token and keep it secure
  • you would be asked for it when executing the code in any of the other files in this repo
  • dont worry, you will be asked just once - see more details here

• clone the repository

  git clone https://github.com/kanad13/PyGitGraph.git
  cd PyGitGraph

• create a virtual environment

  python3 -m venv PyGitGraph_env

• activate the environment

  • windows -

    .\PyGitGraph_env\scripts\activate

  • macos/linux -

    source PyGitGraph_env/bin/activate

• install dependencies

  pip install -r ./requirements.txt

• all done

  • you are now free to execute any of the notebook in this repo

vscode devcontainer (docker)

• create a github token
  • create a token and keep it secure
  • you would be asked for it when executing the code in any of the other files in this repo
  • dont worry, you will be asked just once - see more details here
• open the repository in a container
  • open the command palette (ctrl+shift+p), and select "Remote-Containers: Open Folder in Container"
• automatic setup
  • the environment will automatically install dependencies
• customization
  • adjust these files for your environment
    • devcontainer.json
    • postStart.sh
• all done
  • you are now free to execute any of the notebook in this repo