On Mon, May 28, 2012 at 6:50 PM, Daniel Baumann
<daniel.baum...@progress-technologies.net> wrote:
> On 05/28/2012 07:28 PM, Rui Miguel P. Bernardo wrote:
>> That's why I've put these lines in lb_bootstrap_debootstrap
>>
>>
>> + export http_proxy="${PREVIOUS_PROXY_ENV}"
>> + [ "$PREVIOUS_PROXY_ENV" = "" ] && unset http_proxy
>
> which is overengineering as previously said, there's no point in
> handling this in all existing invocation of commands that should make
> use of it, and all future ones; the right way is to define it once for
> the entire time of the lb environment, so that all commands use it now
> and in future without any further changes, and let it 'die' correctly
> when lb terminates (like all LB_* variables too).
ok, understood.
> if my commit doesn't do what it's supposed to do, then that needs
> improving.
Yes, we need to use those variables you have set in lb_config in
lb_bootstrap_debootstrap and lb_binary_debian-installer. With the vars
set by lb_config I suppose that's the next step?
I'll try tomorow to use LB_APT_HTTP_PROXY instead of messing with
http_proxy. The thing is that debootstrap doesn't have proxy options.
That's why I was "faking" http_proxy.
> but yours is not a good solution to the problem, 'works for
> me'-only is not good enough here, sorry.
The "works for me" is not my philosophy too, but as I said I don't
have all possible environments here, just mine. When I proposed the
patch it was for other users with other environments to test and give
feedback. Thank you.
--
To UNSUBSCRIBE, email to debian-live-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive:
http://lists.debian.org/cap1yx5wgpo5jknw62xjljoyvh314-dtznv5ncvvdgtntv0h...@mail.gmail.com
