[Top][All Lists]

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

Re: Proposed change to `make-readline-port'

From: Neil Jerram
Subject: Re: Proposed change to `make-readline-port'
Date: 10 Mar 2001 11:53:17 +0000
User-agent: Gnus/5.0808 (Gnus v5.8.8) Emacs/20.5

>>>>> "Dirk" == Dirk Herrmann <address@hidden> writes:

    Dirk> On 9 Mar 2001, Neil Jerram wrote:
    >>  OK; what do you think of the diff below, where I've made the
    >> interpolation of a character such as newline optional, and off
    >> by default?  Would you still prefer to remove the interpolation
    >> option completely?

    Dirk> Well, yes, because this can easily be solved by the reader
    Dirk> function, even if it requires to do some string-append
    Dirk> stuff.  What is the problem with that?  I don't see any
    Dirk> benefits to put this into the definition of the port.
    Dirk> (Well, yes, one could argue that this is better, performance
    Dirk> wise.  But I don't buy such an argument if it means to
    Dirk> compromise an otherwise clean interface.)

OK, you've convinced me.  I'll generalize
`make-line-buffered-input-port' to `make-buffered-input-port' and then
provide a `make-line-buffered-input-port' (for convenience) that is
defined in terms of `make-buffered-input-port'.

    Dirk> But, what if 'char-ready?' was available for soft ports (as
    Dirk> you suggested)?  Then, all the special repl stuff could be
    Dirk> extracted. If we go that way, everything is nice and
    Dirk> beautiful.

I absolutely agree.  But the details of the implementation need more
thought first.  I'm away from home for a week now but will look at it
when I get back.

    Dirk> BTW: The comment for make-soft-port seems to be ahead of its
    Dirk> time: It says that a soft port is characterized by 6
    Dirk> parameters :-)



reply via email to

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