[Opensim-dev] Organization and naming patterns of commands on the console

[krtaylor]

Justin, I like this idea - it also greatly simplifies the initial help, 
which can simply refer the user to the sub command helps - e.g. user 
help  These new sub command helps can then be much more lengthy if need 
be...

Justin Clark-Casey wrote:
> Hello,
>
> I want to add more commands to the archiving module but I want to do this within the module itself rather than adding 
> any specific logic to core.
>
> The terrain module already has a 'module commander' to help do this, but this prefixes all commands with the word 
> terrain (e.g. terrain fill, terrain load, etc.)
>
> For consistency purposes, eventually I think that all console commands should largely follow the same pattern.  If we 
> the terrain module scheme, then we get the following kinds of changes (if we assume that commands from different modules 
> can actually share the same initial 'keyword')
>
> backup 		=> region backup
> create user	=> user create
> create-region 	=> region create
> help users	=> user help (for user commands)
> show assets	=> assets show
> show users 	=> user show
> load-xml	=> region xml load (?)
> load-oar	=> region oar load (?)
> kickuser	=> user kick
>
> I think that the pros of this approach are that it makes it easy to avoid name collisions and easy to distribute 
> commands to the right module.  The console seem to include the fact that it's not very intuitive (e.g. show users seems 
> easier to remember than user show) and it involves more typing in some cases (e.g. region xml load rather than load-xml 
> or load xml).
>
> Any thoughts on this?  Just kind of thinking about it at this stage - part of me prefers typing things like "show users" 
> or "help terrain" while another part of my mind likes the 'tidiness' of prefixing every module command with the same 
> word (e.g. terrain fill or region oar load).
>   
-- 
Kurt Taylor (Kurt Stringer)