auctex-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [AUCTeX-devel] Style system Was: Inconsistency in loading/saving .el


From: Pierre Lorenzon
Subject: Re: [AUCTeX-devel] Style system Was: Inconsistency in loading/saving .el files
Date: Sat, 17 May 2008 22:35:56 +0200 (CEST)

Hi again Ralf,

From: Ralf Angeli <address@hidden>
Subject: Re: [AUCTeX-devel] Style system Was: Inconsistency in loading/saving 
.el files
Date: Sat, 17 May 2008 19:02:52 +0200

> * Pierre Lorenzon (2008-05-17) writes:
> 
> > From: David Kastrup <address@hidden>
> >
> >> So at the current point of time, I would prefer not having this
> >> dependency.
> 
> AOL
> 
> >   OK clear ! But I don't know if I will be courageous enough to
> >   write a reimplementation of the style system without
> >   eieio. If you tell me in few months that eieio is definitely
> >   not acceptable I'll probably try to convert my code. 
> 
> I can tell you now that at least for me it is definitely not acceptable.
> The code in AUCTeX should be easily maintainable by anyone familiar with
> plain Emacs Lisp.  Having to learn the workings of something like eieio
> would add another barrier for understanding the code.
> 
> You haven't explained for what purpose you need and object-oriented
> approach.  If it is only for data structures you could use
> plists.

  A object oriented code is more or less reimplentable without
  objects. So I think I'll implement both versions of the
  code. But I still think that a :name tag in a defclass is
  more readable than a (defconst name 0) and then (nth name
  wathever). You might say as well that common lisp is not
  really emacs lisp and that all the dolist in the code might
  be implemented in terms of mapcar except if did not read the
  code carefully enough (wath is always possible !).

  Once more I think that object oriented code is easier to read
  and lighter and then to maintain but once more it might be
  only my own opinion.

  Anyway a non eieio version of the stuff will exist !

  Regards

  Pierre





reply via email to

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