Re: [Findutils-patches] [bug #37683] -inum doesn't work on ext4 file systems with find 4.5.10

[Bernhard Voelker]

On 11/16/2012 10:46 PM, James Youngman wrote:
> On Tue, Nov 13, 2012 at 6:07 AM, Bernhard Voelker
> <address@hidden> wrote:
>>
>> On 11/13/2012 01:20 AM, James Youngman wrote:
>>> On Mon, Nov 12, 2012 at 6:35 PM, Bernhard Voelker
>>> <address@hidden> wrote:
>>>>
>>>> BTW: Why is the latest version on ftp.gnu.org still 4.4.2
>>>> (3.5 years old) while distros already use newer versions?
>>>
>>> The short answer is that the 4.5.x code has release-blocking bugs:
>>>
>>> http://savannah.gnu.org/bugs/?group=findutils&func=browse&set=custom&msort=0&assigned_to[]=0&resolution_id[]=0&severity[]=0&status_id[]=1&vote[]=&date[]=&date_op[]==&advsrch=0&msort=0&chunksz=50&spamscore=5&report_id=206&sumORdet=&morder=bug_id%3C&order=severity#results
>>>
>>
>> Hi James,
>>
>> I understand, ...
>> ... but huh? OpenSuSE-12.2 and Fedora 16+ are already on 4.5.10
> 
> If the above bugs aren't sufficiently serious for those distributions
> to avoid a move to 4.5.x, it's their choice.   I have no problem with
> it.

I read the list, and at first glance it seems that almost all of the
bug have been opened against the 4.4.x series, e.g #30526, so it doesn't
seem to be much lost to go to 4.5.x. ;-)

>> (Ubuntu and Debian are still on 4.4.2), and there's only been
>> 1 single commit in 2012:
>>
>>   $ git log | grep ^Date: | head -n 2
>>   Date:   Mon Jun 4 12:30:52 2012 +0200
>>   Date:   Fri Dec 30 12:16:09 2011 +0000
> 
> Yes, during 2012 I have moved house, had problems with typing (partly
> with physiotherapy, this has been resolved now) and I need to spend
> time looking after my daughter (she's just over 1).  It's not been
> feasible to find time to work on findutils (it's not been easy to find
> time to work on my day job either, but that's difficult to
> de-prioritise).

You've chosen the right priority - family is much more important,
you're right.

>> Any plans how to get out of this?
> 
> Certainly!  The findutils project can always use more volunteers and
> as you can see the number of remaining known bugs in 4.5.x is quite
> small.   The findutils code is pretty readable and has fairly
> extensive tests.   All that's needed is for someone to pick up the
> bugs, write the necessary changes and contribute the code to GNU.
> 
> I'm not saying that I _won't_ work on these bugs, but as your git log
> output shows us, recent history demonstrates that having more
> contributors is a surer way to make sure that findutils progresses
> faster.
> 
> You already suggested one patch above.  Could this be interpreted as a
> sign of possible ongoing interest, or would you say it's a one-off?

Well, why not? I personally see find+xargs like the other basic utilities
in coreutils, and others like sed and grep. I'm using all of them on a
daily basis. I'm subscribed now. ;-)

Re. faster progress: maybe this idea is totally crazy. Maybe it's
too much for some folks to follow 2 mailing lists just for just
2-4 programs, i.e. the group of volunteers following find/xargs
problems is smaller than it could be. Wouldn't the time be ideal
to merge findutils into coreutils? I mean there have been days
with separate textutils, sh-utils and fileutils ...
Just a little thought.

Have a nice day,
Berny