Next top-level changes to the project

[prasadtalasila]

1. Fix top-level README.md
2. Remove data directory
3. Explore the possibility of bring the following config files to top-level?
   a. tsconfig.json
   b. .eslintrc
   c. .eslintignore
   d. husky --> might require a top-level script/git-hooks.bash in script
4. Add docs/INSTALL.md

[prasadtalasila]

Add ports. env information of all the project packages here.

[prasadtalasila]

Notes on Husky:

We also added husky to our pipeline so it runs pre-commit and pre-push:

    Commit: Will format all ts,tsx,css,scss files with Prettier and run eslint. If eslint fails, the commit won't go through. Override in terminal with git commit -m "message" --no-verify
    Push: Will run all unittest with yarn jest . --coverage=false . If any tests fails, the push will fail. Override with in terminal HUSKY=0 git push

Also update codeclimate.yml to increase the duplication line threshold

[prasadtalasila]

Fix the URL for codecov in STATUS.md

[prasadtalasila]

Update docs/LIB-MS.md with config examples from PR #23

An update to docs/LIB-MS.md is required to provide an example graphql query and response for both local and gitlab modes.

[prasadtalasila]

Rename vagrant user in install script to ``$USER`

[prasadtalasila]

Deduplicate script/install.bash and deploy/vagrant/make-boxes/provision.sh. Write a shell function and place it in script/functions.bash. See a good answer on stackoverflow.

Remember to separate out the docker pull commands into a separate function/shell script to avoid install failures.

[prasadtalasila]

Notes on Husky:

We also added husky to our pipeline so it runs pre-commit and pre-push:

    Commit: Will format all ts,tsx,css,scss files with Prettier and run eslint. If eslint fails, the commit won't go through. Override in terminal with git commit -m "message" --no-verify
    Push: Will run all unittest with yarn jest . --coverage=false . If any tests fails, the push will fail. Override with in terminal HUSKY=0 git push

Also update codeclimate.yml to increase the duplication line threshold

@SyntaxXeror , there is .husky at the top-level of the project. Do we also need client/.husky?

[prasadtalasila]

Commit ade0558 resolves some of the issues.

[prasadtalasila]

Add yarn build command to docs/LIB-MS.md

[prasadtalasila]

In deploy/vagrant/make_boxes/dtaas/README.md, after line-19, add the following two commands:

mv key vagrant
mv key.pub vagrant.pub

[prasadtalasila]

Correct build/static/env.js in docs/CLIENT.md to build/env.js

[SyntaxXeror]

I need @Omarley7 to answer this

[Omarley7]

Notes on Husky:

We also added husky to our pipeline so it runs pre-commit and pre-push:

    Commit: Will format all ts,tsx,css,scss files with Prettier and run eslint. If eslint fails, the commit won't go through. Override in terminal with git commit -m "message" --no-verify
    Push: Will run all unittest with yarn jest . --coverage=false . If any tests fails, the push will fail. Override with in terminal HUSKY=0 git push

Also update codeclimate.yml to increase the duplication line threshold

@SyntaxXeror , there is .husky at the top-level of the project. Do we also need client/.husky?

client/.husky can safely be deleted. It was an implementation mistake that I made and is currently not being used. Everything is correctly handled by <root>/.husky. @prasadtalasila do you want me to look further into the potential need of script/git-hooks.bash or is that already addressed in Commit ade0558?

[prasadtalasila]

The husky git hooks are currently working for client because of the husky npm package in client/package.json. Can we make husky work by putting it at the root level and still not having package.json at the root? One obvious solution is to copy husky.sh into .husky directory. But the versioning goes out the window. Is there a better solution? The git-hooks.bash is a good idea. We can perhaps keep them in .hooks directory and have git-hooks.bash copy them into .git/hooks on the developer machine. If this takes less than half a day, please finish it. Otherwise, let's copy husky.sh into .husky directory and move on for now. If you decide to use the husky hack, please open an issue and copy our discussion to that issue so that we don't forget it. Thanks. You are receiving this because you were mentioned.Message ID: ***@***.***>

[prasadtalasila]

In deploy/vagrant/make_boxes/dtaas/README.md, after line-19, add the following two commands:

mv key vagrant
mv key.pub vagrant.pub

done in commit 4293b8f

[Omarley7]

@prasadtalasila Sorry I missed this issue when responding in #34. I still don't fully understand what you meant by this:

One obvious solution is to copy husky.sh into .husky directory. But the versioning goes out the window. Is there a better solution?

Why does versioning go out the window, if we move husky.sh to top level?

[prasadtalasila]

Pending Tasks:

1. docs/admin/INSTALL.md. This INSTALL wiki page needs to be completed by everyone before we migrate the page into docs/admin.
2. Deduplicate script/install.bash and deploy/vagrant/make-boxes/provision.sh. Separate out the docker pull commands into a smaller shell script

[prasadtalasila]

All the changes done on release-v0.2 branch.