Generically across programming languages...
- a private method can only me invoked by an object of the same class (or subclass) i.e. it can only be invoked by itself.
- a public method can be invoked from any object.
In Pharo, variables are private and methods are public.
The advantage of private methods is to avoid users calling utility methods that only do half the work to leave data in a consistent state.
Also its clear which methods are outside you public API so you are free you change those without impacting users.
Rather than such a forced constrint, Pharo deals with private methods by convention:
- prepending "private" or "basic" to the front of the method name
- placing methods into a protocol (or method category) named "private".
- having a protocol "api" for the public api and everything else to be considered private
Note that "protocols" have no impact on program execution - they are just an organisational tool.
All reflects Pharo philosophy...
Reduced forced constraints making it quicker and easier to program "in the flow"
i.e. you "can" break convention and call "private" methods if that is pragmatic and you are willing
to accept the risk that you don't properly comprehend the behavior or the behaviour changes later without warning.
cheers -ben
Locked
