[aur-general] yaourt certs access problem

2016-07-11 Thread Jude DaShiell

When I try using yourt for searches I get:
curl error: Problem with the SSL CA cert (path? access rights?)
I had a network failure a week ago and ended up reinstalling 
talkingarchlinux and didn't have this problem with my earlier 
installation.


I thought adding a newcerts directory fixed this problem but that's not 
the case.



--


Re: [aur-general] yaourt can't access ssl certs

2016-07-11 Thread Jude DaShiell
In /etc/openssl/openssl.conf a newcerts variable is defined as 
$DIR/newcerts.  In checking the directory structure, no newcerts 
directory existed until I created it.  As soon as that was done, the 
problem went away.


On Mon, 11 Jul 2016, Doug Newgard wrote:


Date: Tue, 12 Jul 2016 00:11:46
From: Doug Newgard 
Reply-To: "Discussion about the Arch User Repository (AUR)"

To: aur-general@archlinux.org
Subject: Re: [aur-general] yaourt can't access ssl certs

On Tue, 12 Jul 2016 00:01:07 -0400 (EDT)
Jude DaShiell  wrote:


I installed yaourt on an earlier system and never had this problem then.
Has anyone else had this problem and if you've cleared the problem could
you explain how?



--


You'll have to be more descriptive on just what the problem is.



--


Re: [aur-general] yaourt can't access ssl certs

2016-07-11 Thread Doug Newgard
On Tue, 12 Jul 2016 00:01:07 -0400 (EDT)
Jude DaShiell  wrote:

> I installed yaourt on an earlier system and never had this problem then. 
> Has anyone else had this problem and if you've cleared the problem could 
> you explain how?
> 
> 
> 
> --

You'll have to be more descriptive on just what the problem is.


[aur-general] yaourt can't access ssl certs

2016-07-11 Thread Jude DaShiell
I installed yaourt on an earlier system and never had this problem then. 
Has anyone else had this problem and if you've cleared the problem could 
you explain how?




--


Re: [aur-general] Review request for a new PKGBUILD

2016-07-11 Thread Leonid Bloch
Hi,

Emil - thanks for your review!
After a discussion with the guys who are actively developing this app now,
I'm not going to push it to the AUR just yet, but will rather wait until it
will be more mature and will change less frequently.

Cheers,
Leonid.
___

On Thu, Jul 7, 2016 at 1:28 PM, Emil Lundberg via aur-general <
aur-general@archlinux.org> wrote:

> Hi!
>
> There's just one thing that stood out to me: Why are you putting
> configuration files in `/usr/etc/` rather than `/etc/`? (See [1])
>
> Other than that, your PKGBUILD looks just fine to me. Welcome aboard! :)
>
> [1]:
> https://wiki.archlinux.org/index.php/Arch_packaging_standards#Directories
>
> /Emil
>
> On Sun, 3 Jul 2016, 20:12 Leonid Bloch,  wrote:
>
> > Hi AUR List!
> >
> > We would like to introduce a new package to the AUR. You are welcome to
> > check out more about it on http://rbld.io
> >
> > Please notice, that this app is in the very early development stages, so
> > any feedback is welcomed! ;)
> >
> > I am a long time Arch user, but this is my first PKGBUILD. I'd love to
> hear
> > comments, if any, before I'll actually submit it to the AUR.
> >
> > Thanks a lot!
> > Leonid.
> >
> > File below:
> > ~
> > # Package maintainer: Leonid B 
> > # Upstream contact: r...@rbld.io
> > pkgname=rbld-git
> > pkgver=r4.96410ab
> > pkgrel=1
> > pkgdesc="Zero-dependency, reproducible build environments"
> > arch=('any')
> > url="http://rbld.io";
> > license=('Apache')
> > depends=('bash' 'docker')
> > makedepends=('git')
> > provides=('rbld')
> > conflicts=('rbld')
> > backup=('usr/etc/rebuild.conf')
> > source=("${pkgname%-git}::git+https://github.com/daynix/rebuild";)
> > md5sums=('SKIP')
> >
> > pkgver() {
> >   cd "$srcdir/${pkgname%-git}"
> >   printf "r%s.%s" "$(git rev-list --count HEAD)" "$(git rev-parse --short
> > HEAD)"
> > }
> >
> > package() {
> >   cd "$srcdir/${pkgname%-git}/cli"
> >   make prefix="$pkgdir/usr/" install
> > }
> >
>


Re: [aur-general] PKGBUILD: depends vs. makedepends and namcap warning

2016-07-11 Thread Levente Polyak
Hey, 


On July 11, 2016 12:25:08 PM GMT+02:00, "Lukas Böger"  wrote:
>The individual PKDBUILDs don't differ much, so if someone has a look at
>one of them, I can probably transfer hints and corrections to the other
>ones.

DESTDIR=${pkgdir}

Actually you want to put references to ${pkgdir} into quotes. It comes from 
outside of the PKGBUILD and may contain  spaces. 

Cheers, 
Levente 


Re: [aur-general] PKGBUILD: depends vs. makedepends and namcap warning

2016-07-11 Thread Lukas Böger
Right :) After struggling with the gcc version on some older Mint
system, I obviously failed to clear my mind towards rolling release^^

Now, here are my first PKGBUILDs together with a review request.
Unfortunately, there are 11 of them at once, because the upstream
project is subdivided into modules and they optionally depend on
external libraries which also weren't in the AUR.

https://gist.github.com/lubgr/d864d95b0db323da69624ac28a89efd9

The individual PKDBUILDs don't differ much, so if someone has a look at
one of them, I can probably transfer hints and corrections to the other
ones.

Thanks in advance,
Lukas

On 11.07.2016 11:38, Baptiste Jonglez wrote:
> On Mon, Jul 11, 2016 at 11:28:25AM +0200, Lukas Böger wrote:
>> Next question: should compiler requirements be part of a PKGBUILD?
> 
> No, all packages in base-devel are assumed to be already installed when
> using makepkg.
> 
>> In this case, when compiling with gcc, version >=4.9 is necessary.
> 
> gcc is currently at 6.1.1, so version>=4.9 is already statisfied.  We are
> a rolling release, remember :)
> 
> Baptiste
> 


Re: [aur-general] PKGBUILD: depends vs. makedepends and namcap warning

2016-07-11 Thread Baptiste Jonglez
On Mon, Jul 11, 2016 at 11:28:25AM +0200, Lukas Böger wrote:
> Next question: should compiler requirements be part of a PKGBUILD?

No, all packages in base-devel are assumed to be already installed when
using makepkg.

> In this case, when compiling with gcc, version >=4.9 is necessary.

gcc is currently at 6.1.1, so version>=4.9 is already statisfied.  We are
a rolling release, remember :)

Baptiste


signature.asc
Description: PGP signature


Re: [aur-general] PKGBUILD: depends vs. makedepends and namcap warning

2016-07-11 Thread Lukas Böger
On 11.07.2016 10:38, Baptiste Jonglez wrote:
> On Mon, Jul 11, 2016 at 10:18:32AM +0200, Lukas Böger wrote:
>> Dear AUR list,
>>
>> simple question about dependencies in PKGBUILDs:
>>
>> Package A provides some header files and build scripts. Package B
>> provides additional headers, and their usage only work out if A's
>> headers are in place. Additionaly, package B can only be compiled using
>> A's build scripts.
>>
>> 1) Is it correct to have package A in B's depends AND makedepends arrays?
> 
> In general, no.  If you need a dependency both at compile time and at
> runtime, you should put it in depends only.
> 
>> 2) Am I supposed to ignore namcap's warning 'Dependency included and not
>> neede' because it doesn't take header dependencies into account?
> 
> It seems similar to this situation: https://bugs.archlinux.org/task/48277
> 
> Basically, your package B looks like a library.  B should only makedepends
> on A, and any package depending on B should also makedepends on A.
> 
> The rationale is that header files and build scripts are not needed to run
> a program, so they should not be installed as dependencies.  They should
> only be installed as makedepends, for every package that needs them (even
> indirectly).
> 
> One such example is boost:
> 
>   https://www.archlinux.org/packages/extra/x86_64/boost/
> 
> Baptiste
> 

Ok, thanks for the quick reply, that makes sense! And you're right,
these packages are libraries.

Next question: should compiler requirements be part of a PKGBUILD? In
this case, when compiling with gcc, version >=4.9 is necessary.

Adding it to makedepends seems like a reasonable check - on the other
hand, someone who prefers clang won't have any interest in dealing with
gcc versions enforced by the PKGBUILD.

Lukas


Re: [aur-general] PKGBUILD: depends vs. makedepends and namcap warning

2016-07-11 Thread Baptiste Jonglez
On Mon, Jul 11, 2016 at 10:18:32AM +0200, Lukas Böger wrote:
> Dear AUR list,
> 
> simple question about dependencies in PKGBUILDs:
> 
> Package A provides some header files and build scripts. Package B
> provides additional headers, and their usage only work out if A's
> headers are in place. Additionaly, package B can only be compiled using
> A's build scripts.
> 
> 1) Is it correct to have package A in B's depends AND makedepends arrays?

In general, no.  If you need a dependency both at compile time and at
runtime, you should put it in depends only.

> 2) Am I supposed to ignore namcap's warning 'Dependency included and not
> neede' because it doesn't take header dependencies into account?

It seems similar to this situation: https://bugs.archlinux.org/task/48277

Basically, your package B looks like a library.  B should only makedepends
on A, and any package depending on B should also makedepends on A.

The rationale is that header files and build scripts are not needed to run
a program, so they should not be installed as dependencies.  They should
only be installed as makedepends, for every package that needs them (even
indirectly).

One such example is boost:

  https://www.archlinux.org/packages/extra/x86_64/boost/

Baptiste


signature.asc
Description: PGP signature


Re: [aur-general] PKGBUILD: depends vs. makedepends and namcap warning

2016-07-11 Thread Jan Alexander Steffens via aur-general
On Mon, Jul 11, 2016, 10:18 Lukas Böger  wrote:

> Dear AUR list,
>
> simple question about dependencies in PKGBUILDs:
>
> Package A provides some header files and build scripts. Package B
> provides additional headers, and their usage only work out if A's
> headers are in place. Additionaly, package B can only be compiled using
> A's build scripts.
>
> 1) Is it correct to have package A in B's depends AND makedepends arrays?
>
> 2) Am I supposed to ignore namcap's warning 'Dependency included and not
> neede' because it doesn't take header dependencies into account?
>
> Thanks in advance for clarification,
> Lukas
>

1) No, depends is enough. Unlike e.g. debian, the depends you declare are
implicitly part of the makedepends.
2) Yes, namcap is dumb and can only detect some dependencies, like dynamic
links.

>


[aur-general] PKGBUILD: depends vs. makedepends and namcap warning

2016-07-11 Thread Lukas Böger
Dear AUR list,

simple question about dependencies in PKGBUILDs:

Package A provides some header files and build scripts. Package B
provides additional headers, and their usage only work out if A's
headers are in place. Additionaly, package B can only be compiled using
A's build scripts.

1) Is it correct to have package A in B's depends AND makedepends arrays?

2) Am I supposed to ignore namcap's warning 'Dependency included and not
neede' because it doesn't take header dependencies into account?

Thanks in advance for clarification,
Lukas


[aur-general] Signoff report for [community-testing]

2016-07-11 Thread Arch Website Notification
=== Signoff report for [community-testing] ===
https://www.archlinux.org/packages/signoffs/

There are currently:
* 4 new packages in last 24 hours
* 0 known bad packages
* 0 packages not accepting signoffs
* 0 fully signed off packages
* 6 packages missing signoffs
* 0 packages older than 14 days

(Note: the word 'package' as used here refers to packages as grouped by
pkgbase, architecture, and repository; e.g., one PKGBUILD produces one
package per architecture, even if it is a split package.)


== New packages in [community-testing] in last 24 hours (4 total) ==

* iscan-2.30.2-1 (i686)
* mpv-1:0.18.1-1 (i686)
* iscan-2.30.2-1 (x86_64)
* mpv-1:0.18.1-1 (x86_64)


== Incomplete signoffs for [community] (6 total) ==

* bluegriffon-2.1.1-1 (i686)
0/1 signoffs
* iscan-2.30.2-1 (i686)
0/1 signoffs
* mpv-1:0.18.1-1 (i686)
0/1 signoffs
* bluegriffon-2.1.1-1 (x86_64)
0/2 signoffs
* iscan-2.30.2-1 (x86_64)
0/2 signoffs
* mpv-1:0.18.1-1 (x86_64)
0/2 signoffs


== Top five in signoffs in last 24 hours ==

1. heftig - 4 signoffs