Complete.Org: Mailing Lists: Archives: freeciv-ai: November 2002:
[freeciv-ai] Re: patch/rfc: wrap activity

[freeciv-ai] Re: patch/rfc: wrap activity

[Top] [All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index] [Thread Index]
To: "Per I. Mathisen" <per@xxxxxxxxxxx>
Cc: "Ross W. Wetmore" <rwetmore@xxxxxxxxxxxx>, freeciv-ai@xxxxxxxxxxx
Subject: [freeciv-ai] Re: patch/rfc: wrap activity
From: "Ross W. Wetmore" <rwetmore@xxxxxxxxxxxx>
Date: Mon, 25 Nov 2002 23:15:22 -0500

Wrappers are fine to provide a consistent interface layer.

The problem is when you start fixing underlying code problems in the 
wrappers, or introduce multiple subtly differnt interfaces because the
underlying code is broken and they all patch differently.

If you have access to the underlying code, fix it and make the interface
clean and compact.

This has nothing to do with moving broken code to the client. But that
is equally a bad way to go :-).


At 12:38 PM 02/11/24 +0000, Per I. Mathisen wrote:
>On Sat, 23 Nov 2002, Ross W. Wetmore wrote:
>> If the checks make sense, then push them down to the appropriate level
>> and do them there.
>> Do not clutter the code with Tom, Dick and Harry functions that all
>> patch up the underlying core routine in subtly different ways at the
>> expense of layered function calls and code duplication.
>This might make sense, I agree. The original reason for making this
>wrapper was to packetify the activity changes, but if we decide not to
>move this AI code to the client anyway, as Greg has suggested, then we
>don't need it.
>So maybe we should take the move-to-client discussion first.
>  - Per

[Prev in Thread] Current Thread [Next in Thread]