[cdt-dev] How are actions in CDT contributed?

[Tomasz Wesołowski <kosashi@xxxxxxxxx>]

Hello,

I'm a bit confused with all the terminology there is in the Eclipse framework w/r/t the UI features... well, all the things the user is able to click, shortly speaking. So: there are Actions, Contributions, Contributors, Handlers, Commands, Retargetable Actions and probably more which I can't think of right now. I understand that the confusion may come from the fact that Eclipse used to have only Actions, then later a Command framework was invented on top of that and the exact relation between two APIs (whether and how they depend on each other) is the thing most unclear to me at the moment.

Before, I was able to add a couple new features to the CDT editor, mostly by finding another feature in similar place, and then finding all references to it and just duplicating them with modifications. I recognise from last year that for one menu option I had to create an Action entry, a Command entry (and a Binding entry for the key shortcut IIRC), plus some code in CEditorActionContributor (and probably another places I can't really remember now).

Long story short:

Before I've started working on CDT, I did a few experiments with Eclipse RCP and adding a new command to the menu only required a single entry in plugin.xml, plus the class itself, nothing more (so only declaratively). Yet, I understand that the approach used in CDT (declaratively in plugin.xml + procedurally in an action contributor class), despite being more complicated than that, has its own reasons.

Could someone give me a brief description how exactly the UI features (menu/popup actions, main toolbar and per-view toolbar actions) are currently contributed into CDT, and why?

Best regards,

-- Tomasz