I think you just need to break what you are trying to find down into smaller chunks. Instead of trying to find “move camera when mouse is near edge of screen in direction of mouse” - break that down into exactly what you need.
1: How to detect mouse position
— How to translate this position into a normalized “area” of the screen (top/bottom/left/right/corners)
2: How to move a game object
Both of these individually are very well documented and easily found via google/the unity documents.
For your next question, break that down into smaller bits.
Q: How do you select game objects?
A: There is no built-in way to handle this, but it’s not really difficult either. You need three components really.
1: A visual indication that a unity is selected
2: An internal list of all selected units
3: A method for selecting and de-selecting units.
For 1, that’s up to you. Personally I would probably have the “selected” graphic as a component of each unit that can be toggled on and off.
For 2, that’s just a simple management class (or a function of a larger class more likely) that can literally be as simple as a List<YourSelectionClass/Interface>.
For 3, you can approach this multiple ways.
First, you could use MonoBehaviors OnMouseDown function and have the unit handle it’s own selection/deselection.
Alternatively, and probably the better route, you can have a “selection” script which uses raycasting to find and select units. This approach would be more extensible (make it easier to add eventual marquee style selection instead of having to click each unit individually)/keep things cleaner.
Manipulating the objects is significantly more difficult, as it almost certainly involves pathfinding. In a strategy game, you’ll need well optimized pathfinding too. Honestly, with your likely level of programming knowledge this would be virtually impossible for you. Not only are you not yet capable of writing a pathfinding component yourself, but you also do not know the correct questions to ask to find the answers yet.
A RTS game involves relatively complex A.I. programming, some of the most difficult programming you can do in Unity really. I’m not saying it’s not impossible, but it will be very difficult and perhaps is not the best place to start. That said, 3.5 hopefully will have some built in pathfinding with Unity itself, so this portion of the project should potentially become significantly more approachable.
For now, I really - really suggest working through some of the Unity tutorials that do not necessarily apply to what you want to do for your own project. They will still greatly aid you in learning the correct questions to ask to find your answers more efficiently.
