testspace.testit

The concept is to define specs, test cases, and status using an excel spreadsheet. A GitHub Workflow is used to generate the specs, push them to a branch, and then execute session(s) using the defined status.

Spec	Spec Description	Tag	Case	Case Description	Steps	Status	Comments
One	This is a spec	t1	case 1	Do something	do this	P
One		t1	case 2	Do something	do that	F	this is a comment

Multiple tabs within the excel spreadsheet will be represented as groups in Testspace.

Basic Usage

This repo is a template to be used on the s2testorg. All credentials (secrets) have been set up - TESTSPACE_CRED and GH_PAT_FOR_TF.

1. Create a new repo on s2testorg using this repo.
2. Create a new project using the repo on s2testorg.
3. Run the testit workflow manually.

NOTES:

• Make the new repo public, otherwise, the secrets will not be defined, a Team account is required for private org level secrets.
• Specs generated are always committed to a branch and will overwrite any existing specs if the branch already exists. The default branch used is testit. You can also create new branches.

Changes

It is simple to create new specs.

1. Modify the testit.xlsx on your desktop.
2. Use the github.dev web-based editor. Just use . on the repo.
3. Install via Extensions the Excel Viewer (takes ~2 seconds). This enables to modify existings cells.
4. Or drag and drop an updated file from your desktop and commit the changes.

Note, you can add other spreadsheets to the repo. To execute a different spreadsheet, pass in the name when running the workflow.

Technical

The following tools have been used in developing this utility:

• pandas for generating specs from excel
• Jupyter Notebook for developing the python script
• Puppeteer for automating the Testspace UI using Javascript
• GitHub Workflows - for executing automation independent of person's desktop
  • Reusable Workflows - used separate workflows
  • Git Auto Commit Action - commiting generated specs
• GitHub Codespaces - Just because I wanted to learn about it

Excelsheet

The following is an overview of the sheet format:

Spec	Spec Description	Tag	Case	Case Description	Steps	Status	Comments
One	This is a spec	t1	case 1	Do something	do this	P
One		t1	case 2	Do something	do that	F	this is a comment
Two	Another spec	t2	case 1	Basic stuff	check it

• For multiple cases repeat the Spec name
• Use new line in cel for Steps and Comments
• Status: P | p, F | f, N | n (NA)
• Status left blank is mapped to Untested
• Each Tab represents a group, if there is more than 1 tab

Setup

Secrets

• TESTSPACE_CRED - "Username:Password"; either super-admin or testuser
• GH_PAT_FOR_TF - s2-testspace GitHub Personal Access Token used for installing the private testspace test function package

Installation

pip install -r requirements.txt

npm install 

Desktop Usage

To generate Specs for a branch:

git checkout -b testit
python generate.specs.py testit.xlsx
git add .
git commit -m "auto-gen specs"
git push origin testit

To run sessions:

Onetime setup:

export TESTSPACE_CRED="USERNAME:PASSWORD"  # Using super admins credentials 
export HEADLESS=false  # not working in Codespaces

Project setup:

export GHORG="NAME"  # export GHORG=munderseth or s2testorg
export REPO="NAME"   # export REPO=testspace.testit

To run session(s):

node run.sessions.js BRANCH-NAME specs/allSessions.json  # node run.sessions.js testit specs/allSessions.json

Note, don't forget to make sure that the repo has access to the TESTSPACE_CRED secret. Refer to personal Codespaces settings.

Initial Setup

Create a .testspace.yml file:

manual:
  issues:
    provider: generic 
release:
  - "*"

Create a .gitignore file:

node_modules
_site
.jekyll-cache

Python

Required for script:

pip install pandas
pip install openpyxl

• marketplace install Excel Viewer

Required for Notebook (Codespaces)

pip install ipykernel
python -m ipykernel install --user --name=testspace.testit

Javascript

Note, don't forget to make sure that the repo has access to the GH_PAT_FOR_TF secret. Refer to personal Codespaces settings.

• npm init --yes

Production packages required for the /src/ts.js module.

npm install request
npm install request-promise
npm install @octokit/rest
npm install puppeteer

Note. Using ts.js from https://github.com/s2technologies/testspace.test