Improve documentation

[jdigger]

Right now the documentation is extremely sparse.

Need examples of usage, especially of configuration files.

[ghost]

Right now the documentation says refer back to Grabbit Doc for config files. I keep getting this error

Exception in thread "main" java.lang.Exception: There are no authors or publishers defined in local-author.json

Contents of my local-author.json file
{ "serverUsername" : "admin", "serverPassword" : "admin", "serverHost" : "localhost", "serverPort" : "4502" }

I'm fairly sure its wrong, since the error code is asking for defined authors and publishers. I can't seem figure what other configuration you meant. Mind directing me to the right place?

[jdigger]

The test resources area is probably your best quick bet: https://github.com/TWCable/grabbit-cli/tree/master/src/test/resources

[jdigger]

Otherwise, the section of the Grabbit docs that you likely want to reference is https://github.com/TWCable/grabbit#configuration

[timdonovanuk]

Would it be possible to provide a simple, localhost, author to author sync example? I.e. the most basic scenario, syncing between two development AEM instances on a localhost environment. I did get the old tool working, but I'm struggling to get this new version going (not sure if its pebcak or bug!) and a KISS example would be appreciated.

On the KISS topic, I think it would be helpful if all references in documentation and configuration files were either "sourceServer" or "destinationServer".

Since there are multiple configuration files, having names in docs and configuration items like serverUsername, localhost, author is a bit confusing - I would think the tool is agnostic to all this and should work with simple a "source" and a "destination".

I'd also welcome the evolution of this tool to a more KISS orientated approach; I'm sure there are design reasons and use cases for the current approach but the decade old command line paradigm of ./command -arguments /path /path with a line by line output to stdout is tried, tested and popular for a reason :)

Looking forward to seeing where this all goes; even in its current state its super cool. Thanks!

[jdigger]

Completely agree with essentially everything you said, @timdonovanuk

The tool was built as "the simplest thing that could possibly work" within our infrastructure at Time Warner Cable, but it's clearly extremely specific to that setup. I'm currently trying to both improve the documentation as well as making the way that you interact with Grabbit (the CLI as well as its REST interfaces) more flexible.

That said, of course, doing so is a "on my own time" effort since it works well enough for work purposes. So any help (including even just small documentation changes) via pull-requests would be very appreciated.

[ghost]

@jdigger I'd be happy to help via documentation changes. I'm still trying to get the working and once I figure it out I'll submit some pull requests. I agree with @timdonovanuk with keeping all the references the same