Re: RFS: dirsum
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256
re all,
On Tue, May 25, 2010 at 12:04:34AM +0900, Ansgar Burchardt wrote:
> Hi,
>
> jaromil <jaromil@dyne.org> writes:
>
> > I am looking for a sponsor for my package "dirsum".
>
> There was some discussion [1] on debian-devel@l.d.o what dirsum
> offers that du and sort do not. You might have missed it if you are
> not subscribed to the list.
thanks for the pointer Ansgar.
I was not subscribed to the devel mailinglist at the time the thread
was posted and now i'm not sure i should argument on my reasons to
package dirsum, rather than to learn how this process unfolds.
however, here below are my replies to the concerns raised on that
thread, i'll wait a bit for sponsoring or proceed packaging something
else.
Joey said:
> I'm unsure if this ITP was filed as a reaction to my closure of
> #571575, or without knowledge of it.
it wasn't a reaction, in fact i wasn't aware of that bug at all.
> If you do care which "1.0G" file is biggest, then an option to have
> du include some additional precision in its -h output would easily
> solve your problem. Or there's the small modification I suggest to
> du at <http://bugs.debian.org/563118#70> would allow sorting its
> output exactly by size, without the ugliness of needing a sort -h,
> and while still providing human-readable file sizes.
is this a task for debian, that of modifying upstream applications
like 'du' to fit extra functionalities of other software like dirsum?
i'm just a noob and would like to understand this point better.
agreeing that the output of dirsum could be integrated in du, but
isn't this a task for upstream developers?
ultimately, i've choosen dirsum to learn debian packaging because,
besides being simple, it ended up unmaintained for years (now it's
author Dirk agreed that i take over maintainance) and its sourcecode
was no more available online. i guess its preservation is kind of
precious even for those who, in future, would like to integrate the
visualization offered by dirsum into something like 'du' or 'ncdu'.
weasel said:
> Or, to compare it to something that my users can use too, what does
> it offer over ncdu.
less clicks to get to the results and no ncurses environment fired up,
which still many people would appreciate (piping dirsum output, old
terminals, quick action etc.)
Ron Johnson said:
> $ du -Sk | sort -nr | head -n10
> 131960 ./.Newsletters.Washington_Post/cur
dirsum's output is different and cannot be compared to any output
obtained piping du in any way, IMHO. see for instance:
<pre>
jaromil@spino:/usr/lib$ dirsum
Recursion into new partitions is off
/usr/lib has 521646930 Bytes in 1470 Files, with 304 Subdirectories
which is 27.32 percent of total
Subdirs This Directory 1 Deeper 2 - 45 Deeper Totals
Name Bytes Files Dirs Bytes Files Dirs Bytes Files Files Bytes %
openoffice 0 0 6 754.3K 30 438 315.0M 3299 3329 315.7M 17.34
jvm 2.4K 1 1 0 0 119 108.4M 700 701 108.4M 5.95
git-core 90.8M 135 135 90.8M 4.99
python2.6 7.4M 419 21 15.2M 623 294 45.4M 2555 3597 68.0M 3.74
python2.5 6.0M 385 19 9.8M 677 291 44.5M 2446 3508 60.3M 3.31
pyshared 0 0 2 2.7M 23 47 47.6M 158 181 50.2M 2.76
wine 47.2M 491 491 47.2M 2.59
[...]
Totals 465.7M 4832 364 152.4M 4233 3107 705.3M 17355 27890 1.8G
</pre>
at last, for those who like to support me uploading this package, the
mentors entry is up on the web repository on
http://mentors.debian.net/cgi-bin/sponsor-pkglist?action=details;package=dirsum
thanks for your patience,
ciao
- --
jaromil, dyne.org developer, http://jaromil.dyne.org
GPG: B2D9 9376 BFB2 60B7 601F 5B62 F6D3 FBD9 C2B6 8E39
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)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=
=5hAg
-----END PGP SIGNATURE-----
Reply to: