Code for create application

[yawjalik]

Implementing the command deploifai application create NAME [FLAGS]

Users can specify these using flags:

• Container image name using -i --image
• Container image tag using -t or --tag
• Port -p or --port

If not provided, users will be prompted along with other details:

• cloud profile
• app plan depending on provider of cloud profile
• environment variables

[yawjalik]

@utkarsh867 this is what I got so far

[utkarsh867]

Great! Let me get through quickly.

[utkarsh867]

@yawjalik There are two scenarios here for the Docker image that we should handle.

1. The user has a Docker registry and the image url is of that registry: In this case, the Docker image will have a https:// in the URL. Hence we use this image URL as it is.
2. Image is a local image: It will be a local image. We use our servers to create the registry for the user in the push process. We can have an output in the terminal to tell the user that we created the registry.

All that happening automatic, without user prompt will be insane.

The only catch to this I see is that Dockerhub images don't need a URL. Not sure how we can handle that. @98sean98 You have any ideas how to resolve this?

[yawjalik]

Updates:

• Option for local image or publicly accessible image
• Option to create new repository or use existing repository (deploifai-managed)
• Tag and push image to the selected repository uri

TODO:

• Error handling
• Handle external repositories
• Check for invalid tags (the pushing process doesn’t throw an error)
• Potentially need to check for repository name restrictions (e.g. AWS has specific requirements for the repository name)

[98sean98]

Updates:

• Option for local image or publicly accessible image
• Option to create new repository or use existing repository (deploifai-managed)
• Tag and push image to the selected repository uri

TODO:

• Error handling
• Handle external repositories
• Check for invalid tags (the pushing process doesn’t throw an error)
• Potentially need to check for repository name restrictions (e.g. AWS has specific requirements for the repository name)

Pretty good work.

1. try to catch and handle the exceptions that you think can most likely happen
2. we can build the feature to push to an existing (non-deploifai managed) repository in the future, or maybe we don't
3. do u mean invalid local image tags? need more elaboration
4. our backend api actually demands that the container registry name to be

• between 4 and 30 characters
• must be lowercase alphanumeric and not contain spaces or special characters other than -

in my experience, these are the requirements that aws, azure, and gcp sort of demand
see backend code

[yawjalik]

1. Accept multiple -e or --env flags either with one or more KEY=VALUE values or single env filepath path/to/file
2. Call get_application to poll for deployment status every 10 seconds
3. Added click spinner for prettier loading

[98sean98]

"Deploying application (this will take a few minutes, and it's safe to CTRL+C) ..."

[98sean98]

Catch python ValueError when env_var fails to unpack into a key-value pair, this would indicate that this env_var is faulty.

Print this env_var for the user, and abort the command

[98sean98]

Add another --env-file option to read env files. This should be able to accept multiple files as well.

If the user uses both --env and --env-file options, just read everything and aggregate into one list.

[98sean98]

Choose a size of the application ({provider})

[yawjalik]

@98sean98 removed config stuff and should be ready for merge? Any other issues?