Re: tail --recursive

[Jim Meyering]

Simon Windows <address@hidden> wrote:
> Matthew Woehlke wrote:
>> Simon Windows wrote:
>>> I've attached a modified version of tail.c from coreutils-6.12.The
>>> changes are fairly small but will look a lot bigger in a diff as I've
>>> split tail_forever() into two functions so it can be called recursively.
>>
>> Your attachment seems to be missing?
>
> Let's try that again.
>
> modified tail.c from coreutils-6.12 is actually attached this time.
> /* tail -- output the last part of file(s)
>    Copyright (C) 1989, 90, 91, 1995-2006, 2008 Free Software Foundation, Inc.

Thanks for the contribution.
I haven't looked at it at all yet,
but wonder if you've considered using inotify?
That seems like that mechanism would be much more appropriate,
when tailing many files in an arbitrary hierarchy.

There are already a couple of inotify(7)-based versions
of tail.  I see one that's called inotail.

One of the most important parts of a new-feature suggestion
is the justification for it.  Why is your feature needed?
And can it be made robust?  I.e., what if there are more files
in the hierarchy than the maximum number of open file descriptors?