emacs-devel
[Top][All Lists]
Advanced

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

Re: table.el


From: Stefan Monnier
Subject: Re: table.el
Date: Tue, 04 Dec 2001 10:20:49 -0500

> Tue, 04 Dec 2001 06:29:42 -0500: "Stefan Monnier" 
> <monnier+gnu/address@hidden> wrote:
> 
> > I suggest to make this FUNC into just a convention (after all, the
> > code for with-wrapper-hook does not actually use it, depend on it
> > or anything like that).
> > Also, the function does not have to call the original function.
> > It can use its own code if it wants to.
> > 
> > So I'd just say something like:
> > 
> >   WRAPPER-FORM looks like (WRAPPER-HOOK ARGS...), where ARGS is a
> >   list of arguments passed to each function in WRAPPER-HOOK.
> >   While this macro is calling....
> 
> Is it a bad idea touching upon the recommended convention (both FUNC
> as the first ARG and wrapper calling the original function) in the doc
> string?  Or do you think the example speaks for it.
> 
> A wrapper calling the original function is essential for the
> recursion, otherwise the following wrappers in the same hook variable
> will be ignored.  "While this macro is calling each wrapper ..." may
> imply this though.

Take write-file-hooks for example and imagine we redefine it
as a wrapper-hook.  Now instead of returning non-nil to indicate
"I did the save", it can simply not do the recursive call.

Or just like around-advices do not have to refer back to ad-do-it
which can also prevent other advices from being executed.


        Stefan




reply via email to

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