Re: [arch-general] iptables forward policy

2012-08-25 Thread Corrado Primier
2012/8/25 Juan Diego Tascón :
> Good day,

Hello :)

> I'm thinking of
> setting the default FORWARD policy to ACCEPT as my default INPUT
> policy is DROP and unless there is a valid FORWARD rule for a given
> port the packets wont go anywhere. I'm I right on this?

You're wrong. Either a packet goes through the INPUT chain or it goes
through the FORWARD chain, depending on its destination. Take a look
at this packet flow diagram:
http://www.linuxhomenetworking.com/wiki/images/f/f0/Iptables.gif

Corrado


Re: [arch-general] renaming files

2012-03-02 Thread Corrado Primier
Il 02 marzo 2012 00:13, pete  ha scritto:
> I have some 350 picture files  with names  along the lines of
> "IMG_7127 EOS-1D Mark III copy.jpg"    i would like to rename them all
> to more like "IMG_7127.jpg"  i have tried a few times tonight and cant
> get my head around it anyone got a script that can do it

There have been lots of good suggestion. I put on the plate "detox":
https://aur.archlinux.org/packages.php?ID=7130

Its purpose is eliminating from filenames charachters that make them
difficult to work with, so it's also good for files *beginning* with a
space, esoteric utf8 characters, escape codes, parentheses and
whatever. It's kind of overkill for this, but it deserves to be
mentioned :-)

C


Re: [arch-general] PKGBUILD question

2012-01-07 Thread Corrado Primier
2012/1/7 Allan McRae :
> I'd just go for something like this:
>
> _pkgver=2.0.1-BETA1
> pkgver=2.0.1beta1
>
> No need for all that fancy stuff...

I'd go for that too. But I find these little challenges very funny,
and I wondered if it was possible without adding another variable :)

bardo


Re: [arch-general] PKGBUILD question

2012-01-07 Thread Corrado Primier
2012/1/7 Don Juan :
> The problem I have run into and can not seem to find info about or something
> I have ever run into is that it seems bsdtar when extracting is making the
> name as name_2.0.1-BETA1 and not making it name_2.0.1-beta1 and makepkg
> fails saying it can not find the proper file of name_2.0.1-beta1.

There is a nice construct in bash (beware: version 4 only) which
allows you to make something uppercase with brace expansion.
Look at this:

[bardo@antani ~]$ pkgver=2.0.1-beta1
[bardo@antani ~]$ echo ${pkgver^^}
2.0.1-BETA1

The only problem in your specific situation is the need to also change
'-' in '_'. That would require a double brace expansion, but I'm not
sure it is possible. If it is, it seems to be undocumented. The
cleanest way I could come up with is:

$(sed 's/-/_/' <<<${pkgver^^})

Not very nice, huh? :)


bardo


Re: [arch-general] pacman "Targets" sort order

2011-06-05 Thread Corrado Primier
2011/5/4 David C. Rankin :
> bluez -> run-parts -> ca-certificates??

Targets *are* sorted alphabetically, and upgraded in that order, but
there are exceptions, namely dependencies, which then appear before
the package that depends on them and are upgraded before it, in case
it needs the dependency at update time.

C


Re: [arch-general] AUR deletion request

2011-05-14 Thread Corrado Primier
2011/5/14 |^ `/ () () | ( (-) | :
> Btw, why is it that package maintainers can't just delete their own
> packages?

Because then nothing could stop you from adopting all orphans and mass
deleting them.

Corrado


Re: [arch-general] What happened to Powerpill?

2011-03-28 Thread Corrado Primier
2011/3/28 Nicolas Bigaouette :
> I guess it would make sense if your own bandwidth is bigger then the
> mirror's...

It happens with my ISP that the single connection bandwidth looks to
be capped. I have 8Mbit but a single transfer rarely exceeds 2Mbit,
unless it is very late night or Sunday morning :) Powerpill allowed me
to fully use my bandwidth. I recognize it's a bad traffic shaping
problem (and powerpill had its shortcomings too), but it solved it for
me.

Corrado


[arch-general] i686 build machines for rent

2009-04-01 Thread Corrado Primier
Since the decision of discontinuing the i686 binary repo, many people
who can't or don't want to switch will need to recompile their
packages themselves. Then there's also the problem of devs and TUs who
don't have access to an x86_64 machine. Striving to find a solution I
came up with an idea: selling processing power!

That's right, guys. Do you own one of those slow i686-only dinosaurs?
Do you need the power of an x86_64 machine to get your packages built
quickly in i686 chroots? I'm here for you! I have a couple of machines
I can dedicate to this task. I decided for a "political" price of
20€/year; all revenues will be donated to the OxPD (One x86_64 Per
Developer) Project.

Before setting everything up I'd like to hear from the community if
this is something they're interested in, and I'm obviously also asking
all Arch contributors who have a decent machine (x86_64, that is...)
to dedicate it to this important project that will allow all of our
devs to continue their work for our favorite distro.

Corrado Primier
(Arch Linux Trusted User)