Re: "functional" programming in Ada

["Dmitry A. Kazakov" <mailbox@dmitry-kazakov.de>]

On 06/03/2018 23:16, Manuel Collado wrote:
> El 06/03/2018 a las 17:37, Dmitry A. Kazakov escribió:
>> On 06/03/2018 17:00, Manuel Collado wrote:
>>>
>>> Well, it may sound a bit heterodox, but IMHO functional programming is
>>> just a subset of imperative (= procedural) programming.
>>>
>>> - Functional programming = expressions
>>> - Imperative programming = expressions + variables + procedural actions
>>
>> Imperative is in opposition to declarative. Another axis is stateless
>> vs. stateful. Procedural is usually attributed to the method of
>> decomposition: procedural vs. object vs. functional vs. data/event
>> driven etc.
>>
>> Ada used to strictly separate imperative from declarative and took care
>> about the states, especially about the correspondence between the
>> program state and problem domain. Then Ada supported both procedural and
>> object decomposition, even Ada 83 with its abstract data types.
> 
> Did I mentioned the word "declarative"?
> 
> Functional and declarative are different axis. Fans of functional 
> programming claim it is declarative "per se". Not me.
> 
> Expressions without side-effects are functional, and not necessarily 
> declarative.

Absence of side-effects does not make it functional:

    procedure Swap (X, Y : in out Integer);

is imperative and has no side-effects.

Stateless constructs, if that fiction exists, are functional and 
necessarily declarative.

But I would attribute "functional" rather to the decomposition method. 
Other attributes are more of less consequences of choice. Standing on 
your head, changes the gait... (:-))

-- 
Regards,
Dmitry A. Kazakov
http://www.dmitry-kazakov.de