Add interactive console rather than global usage

[activatedgeek]

Current usage works by prepending put to every operation.
Here is a suggested usage:

$ put-cli
# this opens up the put cli shell (something like what happens on doing Python)
put-cli v0.1.4
()>>> login
Email: <user input>
Password: <user input>
# open browser, get the access token (preferably automatically)
Yay! Login Successful!
(/)>>> ls
filea fileb filec foldera
(/)>>> cd foldera
(/foldera)>>> ls
file1 file2 file3
...

and so on.

The current folder state, tokens and everything can be maintained in the .put-cli/ as it is done right now.

[AgamAgarwal]

What about command completion and history of commands?
Any easy way of doing them?

[activatedgeek]

@AgamAgarwal we can store .puthistory file inside ~/.put-cli/, that is what most programs do for history.

Completion, not sure.

[AgamAgarwal]

What I meant is that is there a way of implementing the history feature which includes capturing up and down keystroke and according changing the command?

[activatedgeek]

Again, not sure about how that would work. We can search for packages before implementing our own.

[AjayBrahmakshatriya]

In the beginning the idea was to have both the operation modes. But later I thought that it would difficult to manage the host file system when in interactive mode. Like I want to change the directory, I should not have to exit the interactive mode, change and get back in. "cd "is just an example. If we go on instrumenting each and every shell command inside the interactive console, it would be a huge coding effort. Avoiding for now. Let us keep it for later.