[Top][All Lists]

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

Re: GNU Fileutils and GNU/Hurd extentions.

From: Alfred M. Szmidt
Subject: Re: GNU Fileutils and GNU/Hurd extentions.
Date: Sun, 17 Mar 2002 19:16:26 +0100
User-agent: Gnus/5.090004 (Oort Gnus v0.04) Emacs/21.2.50

* Jim Meyering writes:
> Hi Alfred, Thanks for helping with GNU.  Please send future
> suggestions about the fileutils to the mailing list,
> address@hidden

>   http://mail.gnu.org/pipermail/bug-fileutils/

> If you're interested in discussion (and with new feature addition
> that's always a good idea), you'll find a wider audience at
> address@hidden

OK, subscribed to both lists.


> If you mean `detect', then please reconsider your approach.
> Instead, please look into using autoconf-style feature tests that
> detect the precise function/interface/semantics/functionality that
> you want to use.  Then write an autoconf macro that sets
> HAVE_SOME_FEATURE appropriately and make the code test that macro.

Yes, I meant detect, a bit to early in the morning. :-)

I sort of dislike having an macro for each feature, it would result in a lot
of them (unknown user bits, translators, author field, etc). But I suppose that
I could live with this, if any system ever adds these options they will
work without any weird changes.


>> Here is an small example on how `ls' will work on GNU (/dev/null
>> being an passive translator and /dev/active (fictions translator)
>> being an active one):
>> $ ls -l /dev/null /dev/active
>> crw-rw-rw-rw- 1 root mhp 0, 0 Jan 25 20:56 /dev/null => /hurd/null (17)
>> crw-r--r--r-- 1 ams ams 0, 0 Mar 8 14:38 /dev/active => unknown (760)

> This behavior should not be the default.  Many programs parse the
> output of `ls -l', and wouldn't be able to handle such a change.

I have mixed feelings about this because this is very useful information for
the user, and any program that actually parses the output of ls is broken IMHO.

Would it be possible to change the behaviour at a later point in time?
Also, would it be OK to use one switch for the showing the unknown bits
and translator information (--gnu, I know it's a bad name) or should one
use two switches (--show-translators/--show-unknown-user-bits)?

But I do think that this should be the default behaviour on GNU.

On an different note, is there a CVS repository for fileutils (or FETISH in

Sorry, for the time it took to reply have been busy with other things.

Alfred M. Szmidt

reply via email to

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