[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: 13-m4-undivert-culprit.patch
From: |
Akim Demaille |
Subject: |
Re: 13-m4-undivert-culprit.patch |
Date: |
31 Aug 2001 16:13:45 +0200 |
User-agent: |
Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.4 (Artificial Intelligence) |
>>>>> "Gary" == Gary V Vaughan <address@hidden> writes:
>> I understand this can be unwelcome, please tell me. But my
>> experience with Autoconf gave me some simple rules, and one of the
>> main conclusion is really `never compromise useful future extension
>> of macro when it can be done straightforwardly with existing
>> macros'.
>>
>> And the 1-ary -> n-ary as a simple for-loop extension is the most
>> attractive of these, IMHO, extremely useless extensions. Beware of
>> the dark side.
Gary> It took me a while to absorb this. But, having dwelt on it for
Gary> a while, I agree with you.
BTW, any reason that if we have undivert(FILE) we don't have
divert(FILE)? (Not that I find that this overload of undivert was
intelligent BTW, I would much prefer another set of macro for
undivert(UNEXPANDED BY ACCIDENT) will complain MACRO does not exist,
while divert(UNEXPANDED BY ACCIDENT) will just populate with junk
files instead of complain `not a number').
There is already so little type checking, corrupting this even more
would be bad.
I'm in favor of deprecating this `````feature''''''. divert/undivert
take numbers, period.