date:20161203

On 12/03/2016 07:00 AM, William L. Thomson Jr. wrote:
> On Saturday, December 3, 2016 8:59:09 AM EST Michał Górny wrote:
>> On Fri, 2 Dec 2016 23:26:53 -0800
>>
>> Daniel Campbell  wrote:
>>> On 12/02/2016 10:47 AM, Michał Górny wrote:

 I'd say keeping things lowercase makes sense for end user packages. For
 pure dependencies with consistent conventions (e.g. perl), it makes
 sense to keep upstream's naming.
>>>
>>> What is a pure dependency? Do we handle those differently than the
>>> garden-variety dependencies in other packages?
>>
>> It is a package that is rarely installed directly, and rather commonly
>> taken as a dependency of another package. For example, packages that
>> install no programs and just Perl/Python/... modules.
> 
> Keep in mind some will emerge libraries dependencies for their own projects 
> and development. They do not always have to be merged as a dependency of 
> another package.
> 
> It might be confusing to know when it is acceptable to use mixed case and not.
> 
I think Michał was talking strictly in the case of a library being
pulled in as a dependency, e.g. program A is depending on library B, but
library B is so specialized that it doesn't really get pulled in
manually. When emerging program A, library B is pure. When emerging
library B deliberately, it becomes the target package.

(If I have this wrong please correct me, Michał. Also forgive me if the
glyph for the last letter in your name is wrong.)

-- 
Daniel Campbell - Gentoo Developer
OpenPGP Key: 0x1EA055D6 @ hkp://keys.gnupg.net
fpr: AE03 9064 AE00 053C 270C  1DE4 6F7A 9091 1EA0 55D6



signature.asc
Description: OpenPGP digital signature

[gentoo-dev] Review request: Ruby 2.0 removal news item

2016-12-03 Thread Hans de Graaff

Title: Ruby 2.0 removal; Ruby 2.1 default
Author: Hans de Graaff 
Content-Type: text/plain
Posted: 2016-12-04
Revision: 1
News-Item-Format: 1.0
Display-If-Installed: https://www.ruby-lang.org/en/news/2016/02/24/support-plan-of-ruby-2
-0-0-and-2-1/

signature.asc
Description: This is a digitally signed message part

Re: [gentoo-dev] OT Who runs Gentoo was -> RFC: Userkit.eclass

On 12/03/2016 03:01 PM, Robin H. Johnson wrote:
> On Sat, Dec 03, 2016 at 04:49:19PM -0500, William L. Thomson Jr. wrote:
>> No matter what terms you use, Google uses Gentoo to build products it makes 
>> money off. In that sense I think it could give back allot. If not in 
>> donations, resources (tinderbox), etc.
> Here's the financial parts of what Google has given us, that I can
> quantify. It DOESN'T include anything of the 20% time that might have
> been used in Gentoo's favour [some past Google-employed developers have
> specifically said they were spending Friday afternoons doing Gentoo
> dev], or paid Gentoo stuff that overlapped their actual job needs
> (ChromeOS-related).
> 
> Non-GSOC:
> -
> 2011: Google donated brand-new Dell servers, that with their volume
> discount, had an invoice price of $4331.60, and a Dell list price over
> $6k. Those servers are hosted at OSUOSL, and are still in active use
> (dipper, blackcap). 
> Net Subtotal: $4,331.60
> 
> GSOC:
> -
> Gross Payments
> (including reimbursement for mentor summit travel expenses)
> 2009:  $5,151.59
> 2006:  $7,000.00
> 2007:  $4,500.00
> 2008:  $3,000.00
> 2010: $11,001.25
> 2011:  $9,891.77
> 2012:  $7,000.00
> 2013:  $0.00 [1]
> 2014:  $4,200.00
> 2015:  $0.00 - did not participate
> 2016:  $4,700.00 [2] 
> Gross Subtotal: $56,444.61
> Less reimbursed travel expenses: $9,852.02
> Net Subtotal: $46,592.59
> 
> 
> Net Total: $50,924.19
> 
> 
> [1] 
> - Details in bug 488142. Not locked, no personal information in it
>   (other year GSOC bugs are locked due to containing mentor personal
>   information).
> - No record could be found for a 2013 invoice, in our bank records OR
>   Google's invoice archive [MANY thanks to Antarus and the Google
>   accounting department who did pull all historical invoices submitted
>   by Gentoo].
> - presumed we did not submit before the deadline and thus forfeited the
>   payment & reimbursement.
> - Net Amount if we had filed: $2591.55
> -- 6 students * $500/student: $3000
> -- Less $2608.45 in actual travel expenses (max $2200 reimbursement)
> - Net for non-filing: $-2608.45
> - NOT included in reimbursement subtotal above.
> 
> [2] Invoice submitted 2012/11/29, payment NET30 period ends 2016/12/29
> 
I just wanted to point this e-mail out and thank you for the effort
spent to share information like this. This is a great step, and once we
get the books in order, sharing this information using automated means
could get us part of the way to 501(c)3 status.

-- 
Daniel Campbell - Gentoo Developer
OpenPGP Key: 0x1EA055D6 @ hkp://keys.gnupg.net
fpr: AE03 9064 AE00 053C 270C  1DE4 6F7A 9091 1EA0 55D6



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-dev] OT Who runs Gentoo was -> RFC: Userkit.eclass

2016-12-03 Thread Robin H. Johnson

On Sat, Dec 03, 2016 at 06:30:29PM -0500, William L. Thomson Jr. wrote:
> > 
> > Net Total: $50,924.19
> > 
> So from 09-16 avg of ~$4.6k per year over 11 years. 
10 years of participation, 9 of which we got paid for. So ~$5.7k/year.
If we got paid for 2013: ~$5.4k/year over 10 years.

> With that really being earned by people doing GSoC. Not the same as if
> Google donated a lump sum of money to further development per say the
> Councils plans. Only 1 hardware donation.
That's the payment to the organization for mentoring and managing the
students, separate from what the students doing GSoC earned.

If the student's work was of use to Gentoo, then it's ALSO $5000-$5500
per student that we've had in man-hours. I do use that disclaimer,
because I know the integration rate for Gentoo students much lower than
it should be.

2006: 10 students
2007: 8 students
2008: 5 students
2009: 6 students
2010: 16 students
2011: 14 students
2012: 8 students
2013: 6 students
2014: 3 students
2016: 5 students

Total: 81 students.
Assuming $5k/student: $405,000 in student payments, over 11 years.

I don't know how many students we've failed: I do know it's been at
least one (I failed them. Their original mentor had medical issues, I
took over, and they provided a mocked video of their work and no code by
midterm).

> I believe past sponsors such as GNi incurred costs in the ~$5k range monthly. 
> I would assume some hosting sponsors to be averaging a few thousand at 
> minimum 
> per year.
The cost to GNi was much closer to $1k/month, mostly in potential lost
revenue if the hardware COULD be used for income (it was already a sunk
cost, and didn't have other users). For our present major hosting
sponsors, I believe we're more in line with $250-$400/month, but again
mostly older hardware that isn't of much other salable use.

> Just as an example. FreeBSD is seeking $1.25 Million in a fundraiser with 
> $882k thus far.
> https://www.freebsdfoundation.org/
$1.25M is their annual fund-raising target for this year and last. Not a
specific fund-raiser, but their annual target.
For 2016 Q1-Q3, on the $1.25M, they report $293k in contributions.
For 2015, on a $1.25M target, they reported $657k in contributions.
For 2014, on a $1M target, they reported $2.4M in contributions.

> They seem to average in the hundreds of thousands every year in contributions
> https://www.freebsdfoundation.org/about/financials/
They're also got a good few years on us (as do Apache).

> Always looked at FreeBSD when I was a Gentoo Trustee. Great foundation! 
> Passed 
> the 5 year probation period with IRS, and other stuff.
The Apache Foundation was very beneficial to look at I found, because
they kept superb public records, but also were not hampered by some of
our restrictions about depending on non-open software (they & the perl
foundation BOTH use QuickBooks on Windows for their accounting).

https://www.apache.org/foundation/records/

I draw your attention to their last 990 filing:
https://www.apache.org/foundation/records/990-2014.pdf
- $1.2M in annual income
- $858k spend on infrastructure, 
  of which >$400k was marked directly as IT spending.
- $1.8M in net assets

-- 
Robin Hugh Johnson
Gentoo Linux: Dev, Infra Lead, Foundation Trustee & Treasurer
E-Mail   : robb...@gentoo.org
GnuPG FP : 11ACBA4F 4778E3F6 E4EDF38E B27B944E 34884E85
GnuPG FP : 7D0B3CEB E9B85B1F 825BCECF EE05E6F6 A48F6136

Re: [gentoo-dev] Tinderboxing efforts in Gentoo

2016-12-03 Thread Hans de Graaff

On Fri, 2016-12-02 at 14:10 +0100, Michał Górny wrote:
> Hi, everyone.
> 
> I've heard multiple times about various tinderbox projects being
> started by individuals in Gentoo. In fact, so many different projects
> that I've forgotten who was working on most of them.
> 
> I know that Toralf is doing tinderboxing for most of the stuff.
> What other projects do we have there? What is their status?

For ruby we have https://github.com/gentoo/ruby-tinderbox which uses
Docker to run tests not just for the current USE_RUBY targets of the
package but also the next target, so that it can generate a list of
packages that are ready to get their next target. I expect this could
be useful for Python as well.

Current status is that it works but it is a bit fiddly to get going,
and currently we don't run this.

Hans

signature.asc
Description: This is a digitally signed message part

[gentoo-dev] Last rites: dev-python/pygtkspellcheck

2016-12-03 Thread Mike Gilbert

# Mike Gilbert  (03 Dec 2016)
# Sandbox failure on install. No maintainer.
# https://bugs.gentoo.org/597146
# https://bugs.gentoo.org/597188
# Removal in 30 days.
dev-python/pygtkspellcheck

Re: [gentoo-dev] OT Who runs Gentoo was -> RFC: Userkit.eclass

OT SNR Sorry but not sorry :) 

On Saturday, December 3, 2016 11:01:56 PM EST Robin H. Johnson wrote:
> On Sat, Dec 03, 2016 at 04:49:19PM -0500, William L. Thomson Jr. wrote:
> > No matter what terms you use, Google uses Gentoo to build products it
> > makes
> > money off. In that sense I think it could give back allot. If not in
> > donations, resources (tinderbox), etc.
> 
> Here's the financial parts of what Google has given us, that I can
> quantify. It DOESN'T include anything of the 20% time that might have
> been used in Gentoo's favour [some past Google-employed developers have
> specifically said they were spending Friday afternoons doing Gentoo
> dev], or paid Gentoo stuff that overlapped their actual job needs
> (ChromeOS-related).

Not sure how to measure time spent on job hours in contribution dollars.

> 
> Net Total: $50,924.19
> 

So from 09-16 avg of ~$4.6k per year over 11 years. With that really being 
earned by people doing GSoC. Not the same as if Google donated a lump sum of 
money to further development per say the Councils plans. Only 1 hardware 
donation.

I believe past sponsors such as GNi incurred costs in the ~$5k range monthly. 
I would assume some hosting sponsors to be averaging a few thousand at minimum 
per year.

Just as an example. FreeBSD is seeking $1.25 Million in a fundraiser with 
$882k thus far.
https://www.freebsdfoundation.org/

They seem to average in the hundreds of thousands every year in contributions
https://www.freebsdfoundation.org/about/financials/

Always looked at FreeBSD when I was a Gentoo Trustee. Great foundation! Passed 
the 5 year probation period with IRS, and other stuff.

-- 
William L. Thomson Jr.

signature.asc
Description: This is a digitally signed message part.

Re: [gentoo-dev] OT Who runs Gentoo was -> RFC: Userkit.eclass

2016-12-03 Thread Robin H. Johnson

On Sat, Dec 03, 2016 at 04:49:19PM -0500, William L. Thomson Jr. wrote:
> No matter what terms you use, Google uses Gentoo to build products it makes 
> money off. In that sense I think it could give back allot. If not in 
> donations, resources (tinderbox), etc.
Here's the financial parts of what Google has given us, that I can
quantify. It DOESN'T include anything of the 20% time that might have
been used in Gentoo's favour [some past Google-employed developers have
specifically said they were spending Friday afternoons doing Gentoo
dev], or paid Gentoo stuff that overlapped their actual job needs
(ChromeOS-related).

Non-GSOC:
-
2011: Google donated brand-new Dell servers, that with their volume
discount, had an invoice price of $4331.60, and a Dell list price over
$6k. Those servers are hosted at OSUOSL, and are still in active use
(dipper, blackcap). 
Net Subtotal: $4,331.60

GSOC:
-
Gross Payments
(including reimbursement for mentor summit travel expenses)
2009:  $5,151.59
2006:  $7,000.00
2007:  $4,500.00
2008:  $3,000.00
2010: $11,001.25
2011:  $9,891.77
2012:  $7,000.00
2013:  $0.00 [1]
2014:  $4,200.00
2015:  $0.00 - did not participate
2016:  $4,700.00 [2] 
Gross Subtotal: $56,444.61
Less reimbursed travel expenses: $9,852.02
Net Subtotal: $46,592.59


Net Total: $50,924.19


[1] 
- Details in bug 488142. Not locked, no personal information in it
  (other year GSOC bugs are locked due to containing mentor personal
  information).
- No record could be found for a 2013 invoice, in our bank records OR
  Google's invoice archive [MANY thanks to Antarus and the Google
  accounting department who did pull all historical invoices submitted
  by Gentoo].
- presumed we did not submit before the deadline and thus forfeited the
  payment & reimbursement.
- Net Amount if we had filed: $2591.55
-- 6 students * $500/student: $3000
-- Less $2608.45 in actual travel expenses (max $2200 reimbursement)
- Net for non-filing: $-2608.45
- NOT included in reimbursement subtotal above.

[2] Invoice submitted 2012/11/29, payment NET30 period ends 2016/12/29

-- 
Robin Hugh Johnson
Gentoo Linux: Dev, Infra Lead, Foundation Trustee & Treasurer
E-Mail   : robb...@gentoo.org
GnuPG FP : 11ACBA4F 4778E3F6 E4EDF38E B27B944E 34884E85
GnuPG FP : 7D0B3CEB E9B85B1F 825BCECF EE05E6F6 A48F6136


signature.asc
Description: Digital signature

Re: [gentoo-dev] OT Who runs Gentoo was -> RFC: Userkit.eclass

On Saturday, December 3, 2016 2:40:43 PM EST Alec Warner wrote:
>
> I don't think making headlines is a goal for the project. 

Making headlines and having attraction is usually something that matters in 
technology. Lots of obscure tech out there, no one writes about or cares 
about. Which leads to less usage, less contributions, less interest, etc.

Its not a goal, its just a sign of relevance.

-- 
William L. Thomson Jr.

signature.asc
Description: This is a digitally signed message part.

Re: [gentoo-dev] OT Who runs Gentoo was -> RFC: Userkit.eclass

WA OT :(

On Saturday, December 3, 2016 5:34:56 PM EST Rich Freeman wrote:
> On Sat, Dec 3, 2016 at 5:09 PM, William L. Thomson Jr.
>
> Actually, it is the opposite.  When you spend money as an individual
> in the US it normally doesn't lower your taxes unless there is some
> special tax deduction for it, such as the mortgage interest deduction.
> One of those deductions is donations to 501c3/etc status.  So, there
> is a benefit to an individual when donating to a 501c3 organization
> (or other deductible classes) because it lowers their tax burden.

Gentoo has no official status with the IRS, 501c6, or 501c3, etc. Any write off 
would have to be categorized otherwise.

> So, money given to Gentoo by a private business is the same as money
> spent on toilet paper or money thrown in the furnace as far as tax
> liability goes.  It increases expenses which means it decreases
> profits.

It depends on how a business or anyone wants to categorize. You can only 
deduct so much in any category.

-- 
William L. Thomson Jr.


signature.asc
Description: This is a digitally signed message part.

Re: [gentoo-dev] OT Who runs Gentoo was -> RFC: Userkit.eclass

2016-12-03 Thread Alec Warner

On Sat, Dec 3, 2016 at 6:00 AM, William L. Thomson Jr. 
wrote:

> OT, who runs Gentoo
>
> On Saturday, December 3, 2016 12:21:55 AM EST Daniel Campbell wrote:
> >
> > Unless something's changed in the past year or two, iirc Sony uses
> > Gentoo as part of the backend of Gaikai, Google's used it for the base
> > of ChromeOS...
>
> Also is the base of CoreOS, widely popular. Google uses it in more than
> ChromOS it was used in the OnHub router. I suspect Google has wider uses.
>
> Google has hired a few core developers as has Gaikai. Both seem to be good,
> though not sure Google is giving back as much given their financial
> benefit.
> Gaikai isn't selling an OS, but Google is based on Gentoo...


> > I can't speak for other 'big names', but Gentoo's not
> > quite as niche as the small, active userbase has most of us believing.
>
> FYI Network Solutions, which I believe at a time had a root DNS server
> running
> Gentoo. Due to their parent company Web.com is, or has by now, moving from
> Gentoo to RHEL. I was told they do not want to be in the operating system
> development business.
>
> Meetup.com runs Gentoo, not sure how widely but it is in use. Likely more
> out
> there, but I see far more RHEL and CentOS. I do not even discuss Gentoo,
> as I
> have to many say it can't be used in production, etc. Things I disagree
> with.


> > There's also our downstream neighbors: Funtoo, Pentoo, Sabayon,
> > Calculate, Exherbo, etc
>
> Two of those are more of a splinter of the Gentoo community, Exherbo and
> Funtoo. IMHO those communities should be reunited into Gentoo. Allot of
> talent
> got driven way that is still out there working on other stuff.
>
> > As for communities, lots of places from 4chan to lainchan, various mesh
> > network users, security-conscious communities, OCD support groups
> > (kidding), etc.
>
> My concern is lack of any mention in the news or tech articles. When was
> the
> last time Gentoo was making headlines and not for drama related stuff.
>

I don't think making headlines is a goal for the project. Maybe you think
it should be; but I don't see leadership (on the Council or the Trustees)
pushing in that direction. So while you can have the opinion that the
distro is dead (or going downhill or whatnot) many people do not think this.

-A


>
> Gentoo News articles
> https://www.google.com/search?hl=en=us=nws=
> 0=gentoo+linux
>
> Arch News articles for example, other distros Ubuntu and Debian have allot
> https://www.google.com/search?hl=en=us=nws=0=arch+linux
>
> Not my words but like say Infoworld articles, not that they are any
> authority
>
> http://www.infoworld.com/article/3109830/linux/why-did-
> gentoo-linux-fade-into-obscurity.html
>
> http://www.infoworld.com/article/3137969/linux/arch-
> linux-the-last-refuge-for-purists.html
>
>
> --
> William L. Thomson Jr.
>

Re: [gentoo-dev] OT Who runs Gentoo was -> RFC: Userkit.eclass

On Sat, Dec 3, 2016 at 5:09 PM, William L. Thomson Jr.
 wrote:
> There is also the charitable donation and write off aspect. Which they may be
> able to do. But since Gentoo has never received official 501c6 status or any
> from the IRS. I am not sure if companies or anyone can actually write off a
> Gentoo donation. May be moot for individuals, but not for large businesses
> with stringent accounts and standards to meet.
>

Actually, it is the opposite.  When you spend money as an individual
in the US it normally doesn't lower your taxes unless there is some
special tax deduction for it, such as the mortgage interest deduction.
One of those deductions is donations to 501c3/etc status.  So, there
is a benefit to an individual when donating to a 501c3 organization
(or other deductible classes) because it lowers their tax burden.

On the other hand, businesses are only taxed on their profits at the
federal level.  So, if a business takes in $500 and spends $400 then
it is taxed on $100.  That $400 could be spent on almost anything as
far as I'm aware.

So, money given to Gentoo by a private business is the same as money
spent on toilet paper or money thrown in the furnace as far as tax
liability goes.  It increases expenses which means it decreases
profits.

Now, where 501c3/etc status does start mattering for businesses is
internal compliance controls.  Most publicly traded companies have
standards for how money can be spent, because that money belongs to
the shareholders.  I work for a publicly traded company and I can't
just treat myself to a new car and expense it, because that deprives
the shareholders of their profits, even if the US government wouldn't
otherwise have a problem with it from a tax perspective (as long as I
declare the value of that car on my own taxes as income).  In order to
keep things simple companies often use 501c3 status as a requirement
for donations.  This eliminates debates about whether a particular
cause is or isn't a valid charity to donate to for the purposes of
goodwill/etc because the IRS acts as an unbiased filter.  501c3 also
implies financial controls on how the money gets spent, so there is
less of a risk that somebody is directing money towards a recipient
who ultimately is going to offer some kind of kickback, because that
would be illegal for the 501c3 and the IRS would enforce that (from a
tax perspective the kickback probably isn't illegal for the original
donor company, but from a shareholder responsibility standpoint it is
a misuse of funds for employees to basically be giving money to
themselves).

So, if your goal is to be the beneficiary of corporate philanthropy,
then I'm sure 501c3 status will help.

Another source of donations might be other 501c3 foundations.  The FSF
might give money to a FOSS-only linux distro, for example.  In such
situations they're almost always going to donate purely to other 501c3
organizations, because they need to ensure the money is spent on
charitable purposes to meet their own IRS requirements.

Now, companies probably also make investments that aren't intended to
be philanthropic.  A company might give money to a trade association
in exchange for some kind of benefit, or it might just give money to
an association to support their which which somehow benefits the
company.  I suspect a business that benefits from Gentoo more directly
probably wouldn't care so much about the tax-exempt status because the
donations are being justified on the basis of being a business
investment of sorts.

So, yes, the status matters, but not actually for tax reasons
themselves in most cases.  It is more of a marker of how the money
gets spent.

I used the term 501c3 just to keep this simple, but there are other
classifications in the tax code which could also apply to an
organization like Gentoo and generally be treated similarly.

If somebody is a corporate tax accountant and wants to offer a finer
explanation it is welcome, but this is the gist of it as I understand
things.

-- 
Rich

Re: [gentoo-portage-dev] [PATCH] _emerge/depgraph.py: Autounmask-write fails when there isn't a file in package.*/ bug 598116

2016-12-03 Thread Zac Medico

On 12/03/2016 01:48 PM, Brian Dolbec wrote:
> 
> From c4a61bebca1cfeb0833cefb2c64be6156bdb8e8d Mon Sep 17 00:00:00 2001
> From: "hackers.terabit" 
> Date: Thu, 27 Oct 2016 03:29:16 +
> Subject: [PATCH] _emerge/depgraph.py: Autounmask-write fails when there isn't
>  a file in package.*/ bug 598116
> 
> Instead of outputting "!!! No file to write for ..." error message,
> Use a sane default filename.
> Add zz- prefix to ensure it remains the last file in sorted order.
> 
> X-Gentoo-bug: 598116
> X-Gentoo-bug-url: https://bugs.gentoo.org/598116
> ---
>  pym/_emerge/depgraph.py | 6 ++
>  1 file changed, 6 insertions(+)
> 
> diff --git a/pym/_emerge/depgraph.py b/pym/_emerge/depgraph.py
> index 9161914..7dc1f55 100644
> --- a/pym/_emerge/depgraph.py
> +++ b/pym/_emerge/depgraph.py
> @@ -8240,6 +8240,12 @@ class depgraph(object):
>   
> child.endswith("~"):
>   continue
>   
> stack.append(os.path.join(p, child))
> + # If the directory is empty add a file with name
> + # pattern file_name.default
> + if last_file_path == None:
> + last_file_path=os.path.join(file_path, 
> file_path, "zz-autounmask")
> + with open(last_file_path,"a+") as default:
> + default.write("# " + file_name)
>  
>   return last_file_path
>  
> 

Patch looks good except it could use these style tweaks:

-if last_file_path == None:
+if last_file_path is None:

-last_file_path=os.path.join(file_path, file_path, "zz-autounmask")
+last_file_path = os.path.join(file_path, file_path, "zz-autounmask")

-with open(last_file_path,"a+") as default:
+with open(last_file_path, "a+") as default:
-- 
Thanks,
Zac

Re: [gentoo-portage-dev] [PATCH] egencache: Migrate _special_filename class to portage.utils.changelog for api use

2016-12-03 Thread Zac Medico

On 12/03/2016 01:49 PM, Brian Dolbec wrote:
> + if self.file_type_lt(self, other):
> + return True
> + elif self.file_type_lt(other, self):
> + return False

Patch looks good, except there are 2 missed renames from file_type_lt to
_file_type_lt above.
-- 
Thanks,
Zac

Re: [gentoo-dev] OT Who runs Gentoo was -> RFC: Userkit.eclass

On Saturday, December 3, 2016 1:56:40 PM EST Brian Dolbec wrote:
>
> Well, It does give back some.  I know there are times patches are
> pushed to gentoo from them.  

Yes, and they employ a couple current and former devs. At least one quite core

> Plus we get financial support from them
> via the Google Summer Of Code.  It has been pointed out recently that
> the majority of our available funds have come from the GSOC.  Not only
> that, but it also helps us get new developers to work with and further
> Gentoo projects.

GSoC is controversial. They did not do it for Gentoo 1 year, I think 2015. 
Also it has lead to a bunch of projects started that are not used. One I am 
not sure why is not used, euscan, as its highly beneficial. Others seemed to 
have fallen off rather than become integrated into Gentoo or lived on as 
projects of their own. I could be wrong on a few, but last I checked that was 
the case for most.

As for the main financial resource for Gentoo that is another thing semi off 
topic for this list. I may have brought it up on -nfp years ago. In a nutshell 
unless Gentoo spends money, and has plans to use money to further development. 
Which in turn benefits others. There is not much motivation for people to 
donate to Gentoo.

There is also the charitable donation and write off aspect. Which they may be 
able to do. But since Gentoo has never received official 501c6 status or any 
from the IRS. I am not sure if companies or anyone can actually write off a 
Gentoo donation. May be moot for individuals, but not for large businesses 
with stringent accounts and standards to meet.

-- 
William L. Thomson Jr.

signature.asc
Description: This is a digitally signed message part.

Re: [gentoo-dev] OT Who runs Gentoo was -> RFC: Userkit.eclass

2016-12-03 Thread Brian Dolbec

On Sat, 03 Dec 2016 16:49:19 -0500
"William L. Thomson Jr."  wrote:

> On Saturday, December 3, 2016 11:27:20 AM EST Brian Dolbec wrote:
> >
> > I think you will find that ChromeOS is not really based on Gentoo,
> > but built using Gentoo as the base os it IS built from.  It is
> > still Linux based, but ChromeOS doe not use ebuilds and such for
> > normal operations.  But the binary pkgs that build it are custom
> > built using Gentoo's ebuild system and modified package management
> > systems.  
> 
> All that sounds like it is based on Gentoo. Its packages before they
> turn them into binaries are ebuilds.
> 
> > The
> > final resulting OS has it's own methods of updating itself and
> > installing the binaries.  (At least that is the way it was some
> > years ago, last I had looked)  
> 
> Yes, but if you want to modify any binaries etc, you are back to
> ebuilds. If you want to add a new package not available its the same
> thing. They use overlay concepts etc.
> 
> I do get ChromeOS is very different, kinda like Android, you must
> flash images you build. No installing it in the normal sense.
> 
> No matter what terms you use, Google uses Gentoo to build products it
> makes money off. In that sense I think it could give back allot. If
> not in donations, resources (tinderbox), etc.
> 

Well, It does give back some.  I know there are times patches are
pushed to gentoo from them.  Plus we get financial support from them
via the Google Summer Of Code.  It has been pointed out recently that
the majority of our available funds have come from the GSOC.  Not only
that, but it also helps us get new developers to work with and further
Gentoo projects.

-- 
Brian Dolbec 



pgppzGjJj1BIz.pgp
Description: OpenPGP digital signature

Re: [gentoo-dev] Tinderboxing efforts in Gentoo

On Saturday, December 3, 2016 6:30:08 PM EST Michał Górny wrote:
>
> "William L. Thomson Jr."  wrote:
> >
> > Did any of the tinderboxes test all the various USE flag combinations or
> > just the defaults?
>
> I'm sorry but you are going off-topic here. I'm trying to gather
> information on any of the existing works. It's not yet time to
> discuss hypothetical problems with how they operate.

That question is not off topic and it goes in line with gathering information. 
There might be existing work on such which is why I was asking that.

-- 
William L. Thomson Jr.

signature.asc
Description: This is a digitally signed message part.

Re: [gentoo-dev] OT Who runs Gentoo was -> RFC: Userkit.eclass

On Saturday, December 3, 2016 11:27:20 AM EST Brian Dolbec wrote:
>
> I think you will find that ChromeOS is not really based on Gentoo, but
> built using Gentoo as the base os it IS built from.  It is still Linux
> based, but ChromeOS doe not use ebuilds and such for normal
> operations.  But the binary pkgs that build it are custom built using
> Gentoo's ebuild system and modified package management systems.

All that sounds like it is based on Gentoo. Its packages before they turn them 
into binaries are ebuilds.

> The
> final resulting OS has it's own methods of updating itself and
> installing the binaries.  (At least that is the way it was some years
> ago, last I had looked)

Yes, but if you want to modify any binaries etc, you are back to ebuilds. If 
you want to add a new package not available its the same thing. They use 
overlay concepts etc.

I do get ChromeOS is very different, kinda like Android, you must flash images 
you build. No installing it in the normal sense.

No matter what terms you use, Google uses Gentoo to build products it makes 
money off. In that sense I think it could give back allot. If not in 
donations, resources (tinderbox), etc.

-- 
William L. Thomson Jr.

signature.asc
Description: This is a digitally signed message part.

[gentoo-portage-dev] [PATCH] _emerge/depgraph.py: Autounmask-write fails when there isn't a file in package.*/ bug 598116

2016-12-03 Thread Brian Dolbec


From c4a61bebca1cfeb0833cefb2c64be6156bdb8e8d Mon Sep 17 00:00:00 2001
From: "hackers.terabit" 
Date: Thu, 27 Oct 2016 03:29:16 +
Subject: [PATCH] _emerge/depgraph.py: Autounmask-write fails when there isn't
 a file in package.*/ bug 598116

Instead of outputting "!!! No file to write for ..." error message,
Use a sane default filename.
Add zz- prefix to ensure it remains the last file in sorted order.

X-Gentoo-bug: 598116
X-Gentoo-bug-url: https://bugs.gentoo.org/598116
---
 pym/_emerge/depgraph.py | 6 ++
 1 file changed, 6 insertions(+)

diff --git a/pym/_emerge/depgraph.py b/pym/_emerge/depgraph.py
index 9161914..7dc1f55 100644
--- a/pym/_emerge/depgraph.py
+++ b/pym/_emerge/depgraph.py
@@ -8240,6 +8240,12 @@ class depgraph(object):

child.endswith("~"):
continue

stack.append(os.path.join(p, child))
+   # If the directory is empty add a file with name
+   # pattern file_name.default
+   if last_file_path == None:
+   last_file_path=os.path.join(file_path, 
file_path, "zz-autounmask")
+   with open(last_file_path,"a+") as default:
+   default.write("# " + file_name)
 
return last_file_path
 
-- 
2.9.3


-- 
Brian Dolbec

[gentoo-dev] [PATCH 1/1] kernel-2.eclass: Convert eclass to use documentation standards.

2016-12-03 Thread Mike Pagano

This patch modifies, removes and adds comments only. No code was harmed
in the creation of this patch.
This is to conform with ECLASS documentation standards.

---
 eclass/kernel-2.eclass | 442
+
 1 file changed, 333 insertions(+), 109 deletions(-)

diff --git a/eclass/kernel-2.eclass b/eclass/kernel-2.eclass
index 91a24e9..b19a396 100644
--- a/eclass/kernel-2.eclass
+++ b/eclass/kernel-2.eclass
@@ -2,89 +2,199 @@
 # Distributed under the terms of the GNU General Public License v2
 # $Id$

-# Description: kernel.eclass rewrite for a clean base regarding the 2.6
-#  series of kernel with back-compatibility for 2.4
-#
-# Original author: John Mylchreest 
-# Maintainer: ker...@gentoo.org
-#
+# @ECLASS: kernel-2.eclass
+# @MAINTAINER:
+# Gentoo Kernel project 
+# @AUTHOR:
+# John Mylchreest 
+# Mike Pagano 
+# 
+# @BLURB: Eclass for kernel packages
+# @DESCRIPTION:
+# This ia the kernel.eclass rewrite for a clean base regarding the 2.6
+# series of kernel with back-compatibility for 2.4
 # Please direct your bugs to the current eclass maintainer :)
-
 # added functionality:
 # unipatch - a flexible, singular method to extract, add and remove
patches.

-# A Couple of env vars are available to effect usage of this eclass
-# These are as follows:
-#
-# K_USEPV  - When setting the EXTRAVERSION 
variable, it should
-#add PV to the end.
-#this is useful for thigns 
like wolk. IE:
-#EXTRAVERSION would be 
something like : -wolk-4.19-r1
-# K_NOSETEXTRAVERSION  - if this is set then EXTRAVERSION will not be
-#automatically set within the 
kernel Makefile
-# K_NOUSENAME  - if this is set then EXTRAVERSION will not 
include the
-#first part of ${PN} in 
EXTRAVERSION
-# K_NOUSEPR- if this is set then EXTRAVERSION will 
not include the
-#anything based on ${PR}.
-# K_PREPATCHED - if the patchset is prepatched (ie: mm-sources,
-#ck-sources, ac-sources) it 
will use PR (ie: -r5) as
-#the patchset version for
-#and not use it as a true 
package revision
-# K_EXTRAEINFO - this is a new-line seperated list of einfo 
displays in
-#postinst and can be used to 
carry additional postinst
-#messages
-# K_EXTRAELOG  - same as K_EXTRAEINFO except using elog 
instead of einfo
-# K_EXTRAEWARN - same as K_EXTRAEINFO except using ewarn 
instead of einfo
-# K_SYMLINK- if this is set, then forcably create 
symlink anyway
-#
-# K_BASE_VER   - for git-sources, declare the base version 
this patch is
-#based off of.
-# K_DEFCONFIG  - Allow specifying a different defconfig target.
-#If length zero, defaults to 
"defconfig".
-# K_WANT_GENPATCHES- Apply genpatches to kernel source. Provide any
-#combination of "base", 
"extras" or "experimental".
-# K_EXP_GENPATCHES_PULL- If set, we pull "experimental" regardless of
the USE FLAG
-#but expect the ebuild 
maintainer to use K_EXP_GENPATCHES_LIST.
-# K_EXP_GENPATCHES_NOUSE   - If set, no USE flag will be provided for
"experimental";
-#as a result the user cannot 
choose to apply those patches.
-# K_EXP_GENPATCHES_LIST- A list of patches to pick from "experimental"
to apply when
-#the USE flag is unset and 
K_EXP_GENPATCHES_PULL is set.
-# K_FROM_GIT - If set, this variable signals that the kernel sources
derives from a git tree and special
-#  handling will be applied so that any patches that are applied
will actually apply.
-#
-# K_GENPATCHES_VER - The version of the genpatches tarball(s) to 
apply.
-#A value of "5" would apply 
genpatches-2.6.12-5 to
-#my-sources-2.6.12.ebuild
-# K_SECURITY_UNSUPPORTED- If set, this kernel is unsupported by Gentoo
Security
-# K_DEBLOB_AVAILABLE   - A value of "0" will disable all of the optional
deblob
-#code. If empty, will be set 
to

Re: [gentoo-dev] Uppercase characters in package names

On Sat, 3 Dec 2016 21:16:05 +0100
Ulrich Mueller  wrote:

> > On Sat, 3 Dec 2016, Michał Górny wrote:  
> 
> > Thanks, Kent, I think this is the best way to put it so far.
> > Could you try to fit it into a devmanual patch?  
> 
> There is no consensus about such a change of policy.

Does that prohibit us from preparing a patch and encouraging a good
idea?

-- 
Best regards,
Michał Górny



pgpUHCRz7N7je.pgp
Description: OpenPGP digital signature

Re: [gentoo-portage-dev] [PATCH] bin/ebuild: fix EBUILD_FORCE_TEST / RESTRICT interaction (bug 601466)

2016-12-03 Thread Zac Medico

On 12/03/2016 07:30 AM, Brian Dolbec wrote:
> On Fri,  2 Dec 2016 22:07:39 -0800
> Zac Medico  wrote:
> 
>> Ensure that config adjustments are applied to all relevant
>> config instances, in order to prevent inconsistencies like
>> the one that triggered bug 601466.
>>
>> X-Gentoo-Bug: 601466
>> X-Gentoo-Bug-URL: https://bugs.gentoo.org/show_bug.cgi?id=601466
>> ---
>>  bin/ebuild | 23 ---
>>  1 file changed, 16 insertions(+), 7 deletions(-)
>>
>> diff --git a/bin/ebuild b/bin/ebuild
>> index 1f99177..e8b8bae 100755
>> --- a/bin/ebuild
>> +++ b/bin/ebuild
>> @@ -244,7 +244,11 @@ if mytree == "porttree" and
>> build_dir_phases.intersection(pargs): ebuild_changed = \
>>  portage.portdb._pull_valid_cache(cpv, ebuild,
>> ebuild_portdir)[0] is None 
>> -tmpsettings = portage.config(clone=portage.settings)
>> +# Make configuration adjustments to portage.portdb.doebuild_settings,
>> +# in order to enforce consistency for EBUILD_FORCE_TEST support
>> +# (see bug 601466).
>> +tmpsettings = portage.portdb.doebuild_settings
>> +
>>  tmpsettings["PORTAGE_VERBOSE"] = "1"
>>  tmpsettings.backup_changes("PORTAGE_VERBOSE")
>>  
>> @@ -272,6 +276,17 @@ if "merge" in pargs and "noauto" in
>> tmpsettings.features: print("Disabling noauto in features... merge
>> disables it. (qmerge doesn't)") tmpsettings.features.discard("noauto")
>>  
>> +if 'digest' in tmpsettings.features:
>> +if pargs and pargs[0] not in ("digest", "manifest"):
>> +pargs = ['digest'] + pargs
>> +# We only need to build digests on the first pass.
>> +tmpsettings.features.discard('digest')
>> +
>> +# Now that configuration adjustments are complete, create a clone of
>> +# tmpsettings. The current instance refers to
>> portdb.doebuild_settings, +# and we want to avoid the possibility of
>> unintended side-effects. +tmpsettings =
>> portage.config(clone=tmpsettings) +
>>  try:
>>  metadata = dict(zip(Package.metadata_keys,
>>  portage.db[portage.settings['EROOT']][mytree].dbapi.aux_get(
>> @@ -315,12 +330,6 @@ def stale_env_warning():
>>  
>> open(os.path.join(tmpsettings['PORTAGE_BUILDDIR'],
>>  '.ebuild_changed'),
>> 'w').close() 
>> -if 'digest' in tmpsettings.features:
>> -if pargs and pargs[0] not in ("digest", "manifest"):
>> -pargs = ['digest'] + pargs
>> -# We only need to build digests on the first pass.
>> -tmpsettings.features.discard('digest')
>> -
>>  checked_for_stale_env = False
>>  
>>  for arg in pargs:
> 
> looks fine to me
> 

Thanks, pushed (with fixup to eliminate duplicate "Forcing Test" messages):

https://gitweb.gentoo.org/proj/portage.git/commit/?id=0c06ff5f8f3e86592bbaeb38f274797505c45b2a
-- 
Thanks,
Zac

Re: [gentoo-dev] Uppercase characters in package names

2016-12-03 Thread Ulrich Mueller

> On Sat, 3 Dec 2016, Michał Górny wrote:

> Thanks, Kent, I think this is the best way to put it so far.
> Could you try to fit it into a devmanual patch?

There is no consensus about such a change of policy.

Ulrich


pgp6TzQtyFO0X.pgp
Description: PGP signature

Re: [gentoo-dev] Uppercase characters in package names

2016-12-03 Thread William Hubbs

On Fri, Dec 02, 2016 at 08:09:31PM -0800, Nick Vinson wrote:
> 
> On 12/02/2016 10:28 AM, Ulrich Mueller wrote:
> >> On Fri, 2 Dec 2016, Mike Gilbert wrote:
> > 
> >> The devmanual states:
> >> The name section should contain only lowercase non-accented letters,
> >> the digits 0-9, hyphens, underscores and plus characters. Uppercase
> >> characters are strongly discouraged, but technically valid.
> > 
> >> https://devmanual.gentoo.org/ebuild-writing/file-format/index.html
> > 
> > 
> >> Why are uppercase characters strongly discouraged?
> > 
> >> Wouldn't it make sense to follow upstream's naming convention?
> > 
> > No, because even for the most common packages it would be hard to
> > guess what the actual convention is. For example, is it GCC (used on
> > its web page and in documentation) or gcc (name of the command and
> > displayed by gcc --version)?
> > 
> > If we allow uppercase, then should we also allow two packages in the
> > tree whose names differ only in character case?
> 
> If Gentoo chose to perfectly match GNU's naming with GCC, then the
> ebuild should be GCC-.ebuild.

I'm thinking about it more from the development standpoint, so that we
don't need things like the MY_Pn, MY_P and S manipulation you see in
dev-python/configargparse.

William



signature.asc
Description: Digital signature

Re: [gentoo-dev] Uppercase characters in package names

On Sun, 4 Dec 2016 04:50:49 +1300
Kent Fredric  wrote:

> On Sat, 3 Dec 2016 08:14:26 +0100
> Ulrich Mueller  wrote:
> 
> > The ebuild has to be written only once, while users will type the
> > package name many times.  
> 
> An ebuild will require maintaining 10x the number of times
> users need to specify the cat/pn of that package.
> 
> An ebuild will also need depending on quite frequently.
> 
> Its also obviously a question of "what would users expect the first time"
> 
> and that probably is somewhat based on "What scope of permissible characters 
> are used in primary atoms"
> 
> I suspect people expect "firefox" to just be called "firefox", because that's 
> what its called in /usr/bin/
> 
> I suspect people expect "gcc" to be called "gcc" because that's what its 
> called in /usr/bin/
> 
> I'd however be more understanding that Xorg be called Xorg 
> 
> However, when you talk about "ecosystems" like Perl/Python, the "primary 
> point of entry" 
> is not in /usr/bin/, its not "what will I type when I invoke the program".
> 
> Its "What will I type in code to use this"
> 
> And given that's case sensitive in Perl, it makes sense that people wanting 
> "Foo" would
> type "use Foo" and "emerge Foo" 
> 
> And I'd imagine similar reasons exist in python/ruby.
> 
> In short, my argument is not so much that /they should be named after what 
> they'll install/,
> but /named after how end users consume it/
> 
> That is: Whatever spelling we use, it should be consistent with the spelling 
> they use the most
> in a *non* gentoo context.

Thanks, Kent, I think this is the best way to put it so far. Could you
try to fit it into a devmanual patch?

-- 
Best regards,
Michał Górny



pgpqc4sryTtP0.pgp
Description: OpenPGP digital signature

Re: [gentoo-dev] Uppercase characters in package names

On Sat, 3 Dec 2016 11:01:32 -0500
Mike Gilbert  wrote:

> On Sat, Dec 3, 2016 at 1:48 AM, Robin H. Johnson  wrote:
> > On Sat, Dec 03, 2016 at 01:04:17AM -0500, Walter Dnes wrote:  
> >> On Fri, Dec 02, 2016 at 09:49:59PM -0600, William Hubbs wrote  
> >> >
> >> > So, should we clean up / rename packages that do not follow this as we
> >> > find them, like my ConfigArgParse example?  
> >>
> >> find /usr/portage -name *.ebuild | grep [A-Z] | sed "s/-[0-9].*//" | sort 
> >> -u > upper.txt  
> >
> > There's a double-packaged case that stands out:
> > dev-python/shapely
> > sci-libs/Shapely
> > (Both use the same pypi source)  
> 
> Thanks for pointing that out.
> 
> It looks like sci-libs/Shapely has been around longer, so I have
> converted reverse deps on dev-python/shapely over to it and masked the
> latter.

There is a bug open for 3 weeks now [1]. I'm surprised that nobody
fixed it before. Could you update it, please?

[1]:https://bugs.gentoo.org/599306

-- 
Best regards,
Michał Górny



pgpsEzUds5ylr.pgp
Description: OpenPGP digital signature

Re: [gentoo-dev] Tinderboxing efforts in Gentoo

On Sat, 3 Dec 2016 13:13:36 +
Markos Chandras  wrote:

> On 12/03/2016 10:41 AM, Michał Górny wrote:
> > On Sat, 3 Dec 2016 10:35:32 +0100
> > Patrice Clement  wrote:
> >   
> >> Friday 02 Dec 2016 14:10:27, Michał Górny wrote :  
> >>> Hi, everyone.
> >>>
> >>> I've heard multiple times about various tinderbox projects being
> >>> started by individuals in Gentoo. In fact, so many different projects
> >>> that I've forgotten who was working on most of them.
> >>>
> >>> I know that Toralf is doing tinderboxing for most of the stuff.
> >>> What other projects do we have there? What is their status?
> >>>
> >>> Is there anything we could try to integrate with pull requests to get
> >>> a better testing?
> >>>
> >>> -- 
> >>> Best regards,
> >>> Michał Górny
> >>> 
> >>
> >> Continuous integration is all the rage these days and tinderboxing is the
> >> obvious way to go concerning Gentoo. AFAIK, Toralf is the only contributor
> >> doing tinderboxing out of his own will. In reality, we should have a team 
> >> of
> >> devs looking after our own tinderboxes instead of relying on the community.
> >>
> >> I'm wondering if we could start a donation campain for this project and ask
> >> people if they've got spare machines laying around. I know a lot of folks 
> >> are
> >> reading this mailing list so maybe asking on gentoo-dev first for a start 
> >> would
> >> be appropriate.  
> > 
> > Hardware is not the problem. Lack of software is.
> >   
> 
> Have you considered using openQA[1] like openSUSE[2] and Fedora[3] do
> instead of reinventing the wheel?
> 
> [1] http://open.qa/
> [2] https://openqa.opensuse.org/
> [3] https://openqa.fedoraproject.org/

Do you by any chance happen to know how it maps to our needs?
At a first glance it seems quite tangential.

-- 
Best regards,
Michał Górny



pgpowTWA2cKQS.pgp
Description: OpenPGP digital signature

Re: [gentoo-dev] Tinderboxing efforts in Gentoo

On Sat, 03 Dec 2016 09:08:42 -0500
"William L. Thomson Jr."  wrote:

> On Friday, December 2, 2016 2:10:27 PM EST Michał Górny wrote:
> > Hi, everyone.
> > 
> > I've heard multiple times about various tinderbox projects being
> > started by individuals in Gentoo. In fact, so many different projects
> > that I've forgotten who was working on most of them.  
> 
> Did any of the tinderboxes test all the various USE flag combinations or just 
> the defaults?
> 
> Automating builds has always been questionable to me due to the USE flag 
> issue. 
> I have things auto-building, but it just runs with what ever the USE defaults 
> are at the time. In addition to any I have set. I have cron email me emerge 
> output so I can review use flags. But that doesn't really fit into the 
> automated 
> building aspect.
> 
> Like if a new use flag is added I may want to enable but is not by default. 
> Automation doesn't really help there without a human making the call to add 
> new USE flag or not.

I'm sorry but you are going off-topic here. I'm trying to gather
information on any of the existing works. It's not yet time to
discuss hypothetical problems with how they operate.

-- 
Best regards,
Michał Górny



pgpt5Zczn6CES.pgp
Description: OpenPGP digital signature

Re: [gentoo-dev] Developer GitHub usernames

2016-12-03 Thread Kent Fredric

On Thu, 1 Dec 2016 02:30:20 +0300
Andrew Savchenko  wrote:

> Error 404 here.

https://marc.info/?l=gentoo-dev=147974883628079=2


FYI, the file's removed now as I've found out that we already have
an LDAP field for that. So if you're not on Gentoo developers team on
GitHub, please set it:

  perl_ldap -buser -C gentooGitHubUser  

I've already set this attribute for all developers currently on GitHub
and for those who replied already.



pgp_LVg5GcmH6.pgp
Description: OpenPGP digital signature

Re: [gentoo-dev] OT Who runs Gentoo was -> RFC: Userkit.eclass

On Saturday, December 3, 2016 5:58:29 PM EST Tobias Klausmann wrote:
> Hi!
> 
> On Sat, 03 Dec 2016, William L. Thomson Jr. wrote:
> > Google has hired a few core developers as has Gaikai. Both seem
> > to be good, though not sure Google is giving back as much given
> > their financial benefit. Gaikai isn't selling an OS, but Google
> > is based on Gentoo...
> 
> That last bit is not true. While yes, Chrome OS and Core OS had a
> Gentoo base, a lot was done of top of that.

I realize I left out ChromeOS in the previous comment, last bit. I was not 
meaning to imply Google ran on Gentoo, just ChromeOS was based on it.

Thanks for some insight into Google though!

-- 
William L. Thomson Jr.

signature.asc
Description: This is a digitally signed message part.

Re: [gentoo-dev] OT Who runs Gentoo was -> RFC: Userkit.eclass

2016-12-03 Thread Tobias Klausmann

Hi! 

On Sat, 03 Dec 2016, William L. Thomson Jr. wrote:
> Google has hired a few core developers as has Gaikai. Both seem
> to be good, though not sure Google is giving back as much given
> their financial benefit. Gaikai isn't selling an OS, but Google
> is based on Gentoo...

That last bit is not true. While yes, Chrome OS and Core OS had a
Gentoo base, a lot was done of top of that.

Furthermore, most of Google's products (both shipped devices and
services) are _not_ based on Gentoo and never were. The base of
Google's production server OS is another Linux distribution, and
there, too, a lot of what makes the thing tick has no outside
equivalent.

Source: I worked on the relevant Google production teams for six
years.

Regards,
Tobias

-- 
panic("Fod fight!");
linux-2.2.16/drivers/scsi/aha1542.c

Re: [gentoo-dev] Revision bumps vs git commits atomicity

2016-12-03 Thread Kent Fredric

On Sat, 3 Dec 2016 15:06:36 +
Markos Chandras  wrote:

> That's reasonable but I also think that bumping and fixing an ebuild at
> the same time can be considered an atomic change since it's effectively
> a _new_ ebuild

One problem is that can seriously confuse git about what's happening,
given how similar our ebuilds are, and given gits only mechanism
for determining "rename" is 

+ filename/a 0xSHA1
- filename/b 0xSHA1

If "filename/a" is sufficiently different from "filename/b" and close enough
to some other non-target "filename/c", it could retroactively be confused into
thinking this happened.

copy filename/c -> filename/a (98%)
remove filename/b 

And then you go and rebase that somewhere were filename/c was actually modified,
and filename/b was modified, and you end up with merge conflicts
where git propagates changes from filename/c to filename/a 
and then makes you battle with "modified filename/b vs removed filename/b"

This is why I prefer to take the approach of:

1. Fix everything first.
2. -r1 bump in a dedicated (or commits) at the end.

Then git tends to be smarter about things, and realises the change in the ebuild
was applied to a moved file, and attempts to modify the moved file instead,
and the only headaches are the -r bumps themselves, and git itself is *usually*
more disposed to getting these right.

But uh. 

Don't look at what github likes to do: 

https://github.com/gentoo/gentoo/pull/2956/commits/52da7b3a525ff1c4f4e1ad73fe9e0ab91cf6e799

..rl-JSON-PP/perl-JSON-PP-2.273.0-r3.ebuild → 
...ive-Tar/perl-Archive-Tar-2.40.0-r4.ebuild 
..ocale-Maketext-Simple-0.210.100_rc.ebuild → 
...ar/perl-Archive-Tar-2.40.100_rc-r1.ebuild 
..ive-Tar/perl-Archive-Tar-2.40.0-r3.ebuild → 
...rl-JSON-PP/perl-JSON-PP-2.273.0-r4.ebuild 
..e-Tar/perl-Archive-Tar-2.40.100_rc.ebuild → 
...le-Maketext-Simple-0.210.100_rc-r1.ebuild 

Its totally understandable why it gets confused:

md5sum perl-Archive-Tar/perl-Archive-Tar-2.40.0-r4.ebuild \
   perl-Archive-Tar/perl-Archive-Tar-2.40.100_rc-r1.ebuild \
   perl-JSON-PP/perl-JSON-PP-2.273.0-r4.ebuild \

perl-Locale-Maketext-Simple/perl-Locale-Maketext-Simple-0.210.100_rc-r1.ebuild 

1fe782df38a2c2cc01615012b9915da9  
perl-Archive-Tar/perl-Archive-Tar-2.40.0-r4.ebuild
5102f6ac13ac331465c64258d8f413c5  
perl-Archive-Tar/perl-Archive-Tar-2.40.100_rc-r1.ebuild
1fe782df38a2c2cc01615012b9915da9  perl-JSON-PP/perl-JSON-PP-2.273.0-r4.ebuild
5102f6ac13ac331465c64258d8f413c5  
perl-Locale-Maketext-Simple/perl-Locale-Maketext-Simple-0.210.100_rc-r1.ebuild

Rationale: most users are working from rsync. So for them, the -r step is 
mostly a "propagate this change" mechansim.
These users are unlikely to see the intermediate stage where you modified the 
file before the -r bump, because the -r bump
was pushed atomically with the rest.

People who clone from git are also not typically "winding back" the head, and 
so they'll get their -r bumps atomically
with the changes when they pull.

People who *are* winding back the head are likely winding back to a specific 
moment in the history:

1. If they're winding back to before your commit series, then it doesn't matter
2. If they're winding back to after your commit series, then they have the 
needed -r bump 
3. If they're winding back to some point within the commit series, things may 
be a little hairy. ( and maybe this is where
the "-r bump first" strategy is more useful? )

But then it depends what they're rewinding *For*

(prebumping) If they were trying to avoid the change you made after you -r 
bumped, rewinding to before the change won't trip portage updates.
(postbumping) If they were meaning to include the change you made before you -r 
bumped, rewinding to after the change won't trip portage updates.

It seems in either case, post/pre bumping to within the change set will have an 
edge case somewhere
where a user has to be assumed to explicitly tell portage to reinstall the 
target package instead of relying
on -r bumps to propagate it.

pgpDOM2vPVkCC.pgp
Description: OpenPGP digital signature

Re: [gentoo-dev] Uppercase characters in package names

2016-12-03 Thread Mike Gilbert

On Sat, Dec 3, 2016 at 1:48 AM, Robin H. Johnson  wrote:
> On Sat, Dec 03, 2016 at 01:04:17AM -0500, Walter Dnes wrote:
>> On Fri, Dec 02, 2016 at 09:49:59PM -0600, William Hubbs wrote
>> >
>> > So, should we clean up / rename packages that do not follow this as we
>> > find them, like my ConfigArgParse example?
>>
>> find /usr/portage -name *.ebuild | grep [A-Z] | sed "s/-[0-9].*//" | sort -u 
>> > upper.txt
>
> There's a double-packaged case that stands out:
> dev-python/shapely
> sci-libs/Shapely
> (Both use the same pypi source)

Thanks for pointing that out.

It looks like sci-libs/Shapely has been around longer, so I have
converted reverse deps on dev-python/shapely over to it and masked the
latter.

Re: [gentoo-dev] Uppercase characters in package names

2016-12-03 Thread Kent Fredric

On Sat, 3 Dec 2016 08:14:26 +0100
Ulrich Mueller  wrote:

> The ebuild has to be written only once, while users will type the
> package name many times.

An ebuild will require maintaining 10x the number of times
users need to specify the cat/pn of that package.

An ebuild will also need depending on quite frequently.

Its also obviously a question of "what would users expect the first time"

and that probably is somewhat based on "What scope of permissible characters 
are used in primary atoms"

I suspect people expect "firefox" to just be called "firefox", because that's 
what its called in /usr/bin/

I suspect people expect "gcc" to be called "gcc" because that's what its called 
in /usr/bin/

I'd however be more understanding that Xorg be called Xorg 

However, when you talk about "ecosystems" like Perl/Python, the "primary point 
of entry" 
is not in /usr/bin/, its not "what will I type when I invoke the program".

Its "What will I type in code to use this"

And given that's case sensitive in Perl, it makes sense that people wanting 
"Foo" would
type "use Foo" and "emerge Foo" 

And I'd imagine similar reasons exist in python/ruby.

In short, my argument is not so much that /they should be named after what 
they'll install/,
but /named after how end users consume it/

That is: Whatever spelling we use, it should be consistent with the spelling 
they use the most
in a *non* gentoo context.

pgpYXihiDxFWV.pgp
Description: OpenPGP digital signature

Re: [gentoo-dev] Tinderboxing efforts in Gentoo

On Sat, Dec 3, 2016 at 9:47 AM, William L. Thomson Jr.
 wrote:
> On Saturday, December 3, 2016 9:33:00 AM EST Michael Orlitzky wrote:
>> On 12/03/2016 09:25 AM, William L. Thomson Jr. wrote:
>> >> This is generally considered infeasible:
>> > I would not think such, just need a wrapper to run around each package
>> > that
>> > would get its USE flags and re-emerge it a few times.
>>
>> If a package has 10 USE flags, and if each can be set on/off with no
>> constraints, then that gives you 2^10 different ways to emerge it.
>
> May make the requirements of the host system larger or take more time.  I am
> sure processing power could handle that load.
>
> Would be nice if someone like Google would sponsor such efforts. They have
> enough hardware and cloud services to make such feasible.
>

Have you given thought to how long it would take the largest
supercomputer in the world to rebuild libreoffice once for each of the
2^28 USE flag combinations it has (not including USE_EXPAND)?

It is certainly possible, but I doubt that you're going to get it
dedicated to Gentoo for a few weeks whenever one of its deps changes.

This is not a reason to give up on tinderboxing.  This is just a
reason to be realistic about just what it will do.

-- 
Rich

Re: [gentoo-dev] Tinderboxing efforts in Gentoo

2016-12-03 Thread james


On 12/03/2016 09:33 AM, Michael Orlitzky wrote:

On 12/03/2016 09:25 AM, William L. Thomson Jr. wrote:


This is generally considered infeasible:


I would not think such, just need a wrapper to run around each package that
would get its USE flags and re-emerge it a few times.


If a package has 10 USE flags, and if each can be set on/off with no
constraints, then that gives you 2^10 different ways to emerge it.



Perhaps the default (or a minimal) set of flags chosen by the ebuild 
maintainer, appropriate project or QA (as a last result) could be tested 
on an official gentoo tinderbox/CI cluster.



Then a stage-4 image could be made available so anyone can install 
gentoo cluster nodes quickly and a bit of ansible code to load the 
framework onto a openstack-gentoo-cluster, for example. This would 
streamline the task-set for  others to self-test any ebuild with a 
unique set of flags and then upload those results or make those results 
available via an overlay or github mechanism.



If you automate it for a few gentoo devs, putting out a stage (4)
for specific hardware (like amd-64) would pretty much make it so
hundreds or thousands of tinder-CI gentoo-centric efforts could exist
for the users and proxy folks.


Once folks see that "power" of a gentoo cluster,  then they'll likely 
use gentoo-clusters for a myriad of needs. That will motivate them to 
take a 'deeper-dive' into gentoo internals, culminating in many more 
gentoo devs, thus growing the technical talent base of gentoo.




hth,
James

Re: [gentoo-portage-dev] [PATCH] bin/ebuild: fix EBUILD_FORCE_TEST / RESTRICT interaction (bug 601466)

2016-12-03 Thread Brian Dolbec

On Fri,  2 Dec 2016 22:07:39 -0800
Zac Medico  wrote:

> Ensure that config adjustments are applied to all relevant
> config instances, in order to prevent inconsistencies like
> the one that triggered bug 601466.
> 
> X-Gentoo-Bug: 601466
> X-Gentoo-Bug-URL: https://bugs.gentoo.org/show_bug.cgi?id=601466
> ---
>  bin/ebuild | 23 ---
>  1 file changed, 16 insertions(+), 7 deletions(-)
> 
> diff --git a/bin/ebuild b/bin/ebuild
> index 1f99177..e8b8bae 100755
> --- a/bin/ebuild
> +++ b/bin/ebuild
> @@ -244,7 +244,11 @@ if mytree == "porttree" and
> build_dir_phases.intersection(pargs): ebuild_changed = \
>   portage.portdb._pull_valid_cache(cpv, ebuild,
> ebuild_portdir)[0] is None 
> -tmpsettings = portage.config(clone=portage.settings)
> +# Make configuration adjustments to portage.portdb.doebuild_settings,
> +# in order to enforce consistency for EBUILD_FORCE_TEST support
> +# (see bug 601466).
> +tmpsettings = portage.portdb.doebuild_settings
> +
>  tmpsettings["PORTAGE_VERBOSE"] = "1"
>  tmpsettings.backup_changes("PORTAGE_VERBOSE")
>  
> @@ -272,6 +276,17 @@ if "merge" in pargs and "noauto" in
> tmpsettings.features: print("Disabling noauto in features... merge
> disables it. (qmerge doesn't)") tmpsettings.features.discard("noauto")
>  
> +if 'digest' in tmpsettings.features:
> + if pargs and pargs[0] not in ("digest", "manifest"):
> + pargs = ['digest'] + pargs
> + # We only need to build digests on the first pass.
> + tmpsettings.features.discard('digest')
> +
> +# Now that configuration adjustments are complete, create a clone of
> +# tmpsettings. The current instance refers to
> portdb.doebuild_settings, +# and we want to avoid the possibility of
> unintended side-effects. +tmpsettings =
> portage.config(clone=tmpsettings) +
>  try:
>   metadata = dict(zip(Package.metadata_keys,
>   portage.db[portage.settings['EROOT']][mytree].dbapi.aux_get(
> @@ -315,12 +330,6 @@ def stale_env_warning():
>   
> open(os.path.join(tmpsettings['PORTAGE_BUILDDIR'],
>   '.ebuild_changed'),
> 'w').close() 
> -if 'digest' in tmpsettings.features:
> - if pargs and pargs[0] not in ("digest", "manifest"):
> - pargs = ['digest'] + pargs
> - # We only need to build digests on the first pass.
> - tmpsettings.features.discard('digest')
> -
>  checked_for_stale_env = False
>  
>  for arg in pargs:

looks fine to me

-- 
Brian Dolbec

Re: [gentoo-dev] Uppercase characters in package names

2016-12-03 Thread Mike Gilbert

On Sat, Dec 3, 2016 at 10:02 AM, Ben Kohler  wrote:
>>
>> Keep in mind some will emerge libraries dependencies for their own
>> projects
>> and development. They do not always have to be merged as a dependency of
>> another package.
>>
>> It might be confusing to know when it is acceptable to use mixed case and
>> not.
>>
>> --
>> William L. Thomson Jr.
>
> It's really not confusing, you're making up issues just to hear yourself
> talk.

Let's not start this here. I agree that there is no clear cut rule
here, and that could be confusing.

Re: [gentoo-dev] Revision bumps vs git commits atomicity

2016-12-03 Thread Markos Chandras

On 12/02/2016 03:14 PM, Andrew Savchenko wrote:
> 
> What about the following forkflow:
> - version bump first with minimal changes required, but without
> pushing commit to the tree;
> - make each logical change as a separate commit without revision
> bumps and without pushing stuff to the tree (of course repoman
> scan/full is required as usual for each commit);
> - well test package after the last commit (that it builds with
> various USE flag combinations, old and new functionality works fine
> and so on);
> - fix any problems found and only afterwards push changes to the
> tree.
> 
> This way users will see only foo-1.0 -> foo-1.1 change in the tree,
> while git will still retain each logical change as a separate
> commit, which will make future maintenance and debugging much
> easier.

That's reasonable but I also think that bumping and fixing an ebuild at
the same time can be considered an atomic change since it's effectively
a _new_ ebuild

-- 
Regards,
Markos Chandras

Re: [gentoo-dev] Uppercase characters in package names

2016-12-03 Thread Ben Kohler

>
>
> Keep in mind some will emerge libraries dependencies for their own projects
> and development. They do not always have to be merged as a dependency of
> another package.
>
> It might be confusing to know when it is acceptable to use mixed case and
> not.
>
> --
> William L. Thomson Jr.
>
It's really not confusing, you're making up issues just to hear yourself
talk.

-Ben

Re: [gentoo-dev] Uppercase characters in package names

On Saturday, December 3, 2016 8:59:09 AM EST Michał Górny wrote:
> On Fri, 2 Dec 2016 23:26:53 -0800
> 
> Daniel Campbell  wrote:
> > On 12/02/2016 10:47 AM, Michał Górny wrote:
> > >
> > > I'd say keeping things lowercase makes sense for end user packages. For
> > > pure dependencies with consistent conventions (e.g. perl), it makes
> > > sense to keep upstream's naming.
> > 
> > What is a pure dependency? Do we handle those differently than the
> > garden-variety dependencies in other packages?
> 
> It is a package that is rarely installed directly, and rather commonly
> taken as a dependency of another package. For example, packages that
> install no programs and just Perl/Python/... modules.

Keep in mind some will emerge libraries dependencies for their own projects 
and development. They do not always have to be merged as a dependency of 
another package.

It might be confusing to know when it is acceptable to use mixed case and not.

-- 
William L. Thomson Jr.


signature.asc
Description: This is a digitally signed message part.

Re: [gentoo-dev] OT Who runs Gentoo was -> RFC: Userkit.eclass

Getting further off topic, rather not create noise to bother others, minimal 
below.

On Saturday, December 3, 2016 9:36:47 AM EST Rich Freeman wrote:
> 
> Sure, and it probably will be the state of things 20 years from now,
> with Gentoo still having "little chance that even the minimum of
> release and bug-fixing goals will be met" and suffering a "rapid
> downfall of the distribution" :)
>
> The predictions of those paragraphs have not in fact come to pass.

The article is no spot on to things today, but does have lots of similarities.

> Would you agree that "if a person who repeatedly engages in personal
> attacks against other developers is permitted to remain with the
> project, then there is something wrong with the way the distribution
> is managed?"

I do not see how Gentoo the project as a whole is managed

> I find it a bit interesting that half of this article is about a
> failure to enforce a Code of Conduct that you don't actually think we
> ought to have, and that drobbins left in part because it wasn't being
> enforced.

That article is not correct on the Daniel Robbins aspect. I can have Daniel 
comment if you like. It had more to do with leading Gentoo, resuming his 
previous role etc. Nothing relating to CoC or individuals.

> Sometimes forks exist because individuals don't get along or have
> strong ideas for how things should work to the exclusion of other
> ideas of how things should work.  That's fine, there is nothing wrong
> with forks.

It does not help the main community. My favorite story of such is Firebird and 
its Vulcan fork. Which long story short was merged back into firebird and 
became Firebird 3 :)

Good stuff can happen when people reunite. Not always the case with forks. 
Some forks die. Not implying that in either case but historically that is the 
case.  Like XFree86 for example, over license changes.

> The current meta-structure of Gentoo is structured around the vision

https://archives.gentoo.org/gentoo-project/message/
3ac5418dd061fc53f4b8d55a99773f4c

Been here before, said it all before... No need to repeat.

-- 
William L. Thomson Jr.

signature.asc
Description: This is a digitally signed message part.

Re: [gentoo-dev] Tinderboxing efforts in Gentoo

2016-12-03 Thread james

On 12/03/2016 08:13 AM, Markos Chandras wrote:

On 12/03/2016 10:41 AM, Michał Górny wrote:

On Sat, 3 Dec 2016 10:35:32 +0100
Patrice Clement wrote:

Friday 02 Dec 2016 14:10:27, Michał Górny wrote :

Hi, everyone.

I've heard multiple times about various tinderbox projects being
started by individuals in Gentoo. In fact, so many different projects
that I've forgotten who was working on most of them.

I know that Toralf is doing tinderboxing for most of the stuff.
What other projects do we have there? What is their status?

Is there anything we could try to integrate with pull requests to get
a better testing?

--
Best regards,
Michał Górny

Continuous integration is all the rage these days and tinderboxing is the
obvious way to go concerning Gentoo. AFAIK, Toralf is the only contributor
doing tinderboxing out of his own will. In reality, we should have a team of
devs looking after our own tinderboxes instead of relying on the community.

If these tinderbox/CI projects where documented and a few devs availed
themselves to proxy style support, I'd make a 3-7 server (8-core 32G)
cluster available for long term usage. There are quite a few folks
interesting in CI on clusters, but things are a wee bit challenging to
setup; that's the biggest obstacle I see. I would think a variety of
small clusters running different frameworks would be a keen idea for
tinder/CI needs.

I'm wondering if we could start a donation campain for this project and ask
people if they've got spare machines laying around. I know a lot of folks are
reading this mailing list so maybe asking on gentoo-dev first for a start would
be appropriate.

Hardware is not the problem. Lack of software is.

+++

Have you considered using openQA[1] like openSUSE[2] and Fedora[3] do
instead of reinventing the wheel?

[1] http://open.qa/
[2] https://openqa.opensuse.org/
[3] https://openqa.fedoraproject.org/

This openqa project states:: "can be installed in any Linux
distribution"; so maybe several different CI/tinderbox solutions just
need a wiki/channel of support and a dev or 2 to help others get small
clusters running these codes? I'm deeply interested in both short term
and long term operations of (gentoo) clusters for these sorts of
projects/frameworks. In fact a hardened, minimized kernel that only
supports the codes (framework) for such a cluster, should make things
run very, very fast. I'd also put up a (gentoo) arm-64 cluster of
embedded SBC for this sort of project, as many are already in existence
across the net as PoC projects. I'd propose to use SBC with at least 4G
of ram each::

http://www.96boards.org/product/cello/

http://www.96boards.org/product/poplar/

or even on R.pi running gentoo

https://www.linux-toys.com/?p=7

http://antonylees.blogspot.com/2012/08/running-jenkins-ci-on-raspberry-pi.html

https://medium.com/@bossjones/how-i-setup-a-raspberry-pi-3-cluster-using-the-new-docker-swarm-mode-in-29-minutes-aa0e4f3b1768#.xbji209mb

Package up the project/team-need into something *FUN*, and lots of
gentooers will jump all over this tinderbos/CI opportunity. Sadly, there
are quite a few folks on gentoo-embedded that could do this entire
project, while blind folded... or already have such running.

The best way to recruit lots of smart kids to gentoo, is to put up a few
keenly attractive cluster projects, and have a few devs interact and
support folks in these sorts of projects. The 'kid' in me is very
interested in gentoo clusters (expecially without systemd) and there are
a myriad of gentoo dev needs that could be solve if there was a
companion (gentoo-cluster) solution that was simultaneously support.

(gentoo?) CLUSTERS are where it's at! So get hip, get real and get
current. (this comes from an 'old_fart)

hth,
James

Re: [gentoo-dev] Tinderboxing efforts in Gentoo

On Saturday, December 3, 2016 9:33:00 AM EST Michael Orlitzky wrote:
> On 12/03/2016 09:25 AM, William L. Thomson Jr. wrote:
> >> This is generally considered infeasible:
> > I would not think such, just need a wrapper to run around each package
> > that
> > would get its USE flags and re-emerge it a few times.
> 
> If a package has 10 USE flags, and if each can be set on/off with no
> constraints, then that gives you 2^10 different ways to emerge it.

May make the requirements of the host system larger or take more time.  I am 
sure processing power could handle that load.

Would be nice if someone like Google would sponsor such efforts. They have 
enough hardware and cloud services to make such feasible.

-- 
William L. Thomson Jr.

signature.asc
Description: This is a digitally signed message part.

Re: [gentoo-dev] OT Who runs Gentoo was -> RFC: Userkit.eclass

On Sat, Dec 3, 2016 at 9:20 AM, William L. Thomson Jr.
 wrote:
>
> Which was one of the last articles Gentoo mentioned in on Distro watch, till I
> believe the OnHub router. Based around that topic, quoting Ciaran.
>
> http://distrowatch.com/weekly.php?issue=20070312#future
>
> Most interesting about that article. If you read the last two paragraphs. I
> think some of that could be said about the state of things still.
>

Sure, and it probably will be the state of things 20 years from now,
with Gentoo still having "little chance that even the minimum of
release and bug-fixing goals will be met" and suffering a "rapid
downfall of the distribution" :)

The predictions of those paragraphs have not in fact come to pass.

Would you agree that "if a person who repeatedly engages in personal
attacks against other developers is permitted to remain with the
project, then there is something wrong with the way the distribution
is managed?"

I find it a bit interesting that half of this article is about a
failure to enforce a Code of Conduct that you don't actually think we
ought to have, and that drobbins left in part because it wasn't being
enforced.

Sometimes forks exist because individuals don't get along or have
strong ideas for how things should work to the exclusion of other
ideas of how things should work.  That's fine, there is nothing wrong
with forks.

The current meta-structure of Gentoo is structured around the vision
that Gentoo is a place where people can make what they want of it, and
the governance bodies of Gentoo are mostly about dealing with
conflicting goals, not picking winners.  Sure, the Council could take
a much more pro-active stance and say "Gentoo needs to be the best
distro for xyz so we should get rid of all this Java crap" but that
would be silly because the two aren't mutually exclusive and telling
people to not work on Java isn't going to magically inspire them to
work on something else instead.

-- 
Rich

Re: [gentoo-dev] Tinderboxing efforts in Gentoo

2016-12-03 Thread Michael Orlitzky

On 12/03/2016 09:25 AM, William L. Thomson Jr. wrote:
>>
>> This is generally considered infeasible:
> 
> I would not think such, just need a wrapper to run around each package that 
> would get its USE flags and re-emerge it a few times.

If a package has 10 USE flags, and if each can be set on/off with no
constraints, then that gives you 2^10 different ways to emerge it.

Re: [gentoo-dev] Tinderboxing efforts in Gentoo

On Saturday, December 3, 2016 9:14:32 AM EST Rich Freeman wrote:
> On Sat, Dec 3, 2016 at 9:08 AM, William L. Thomson Jr.
> 
>  wrote:
> > On Friday, December 2, 2016 2:10:27 PM EST Michał Górny wrote:
> >> Hi, everyone.
> >> 
> >> I've heard multiple times about various tinderbox projects being
> >> started by individuals in Gentoo. In fact, so many different projects
> >> that I've forgotten who was working on most of them.
> > 
> > Did any of the tinderboxes test all the various USE flag combinations or
> > just the defaults?
> 
> This is generally considered infeasible:

I would not think such, just need a wrapper to run around each package that 
would get its USE flags and re-emerge it a few times. I had a crude one for 
alsa-lib long ago. But it just added flags, it did not test various combos.

Anyway if any future tinderboxing efforts could take into USE flag combos into 
the testing, that would be AWESOME!

I would be interested in participating in such. May play with making a basic 
wrapper to test use flag combos.

-- 
William L. Thomson Jr.

signature.asc
Description: This is a digitally signed message part.

Re: [gentoo-dev] OT Who runs Gentoo was -> RFC: Userkit.eclass

On Saturday, December 3, 2016 9:09:30 AM EST Rich Freeman wrote:
> On Sat, Dec 3, 2016 at 9:00 AM, William L. Thomson Jr.
> 
>  wrote:
> > OT, who runs Gentoo
> > 
> > On Saturday, December 3, 2016 12:21:55 AM EST Daniel Campbell wrote:
> >> There's also our downstream neighbors: Funtoo, Pentoo, Sabayon,
> >> Calculate, Exherbo, etc
> > 
> > Two of those are more of a splinter of the Gentoo community, Exherbo and
> > Funtoo. IMHO those communities should be reunited into Gentoo. Allot of
> > talent got driven way that is still out there working on other stuff.
> 
> Uh, you do realize that the main force behind one of those projects
> left because of the main force behind the other project?

I think Exherbo was more of a loss, as my old recruiter went there and others 
I had worked with in the past.

> https://lwn.net/Articles/225060/

I hardly recall things as being that simple from that article over banning 
Ciaran. I spoke to Daniel then and since. It was more over not being able to 
lead the project. This was when the foundation needed to be reinstated and 
other things.

Since most those links are broken here is one that is not, I was very close to 
that matter at the time.
https://marc.info/?l=gentoo-dev=117303590903513=2

Which was one of the last articles Gentoo mentioned in on Distro watch, till I 
believe the OnHub router. Based around that topic, quoting Ciaran.

http://distrowatch.com/weekly.php?issue=20070312#future

Most interesting about that article. If you read the last two paragraphs. I 
think some of that could be said about the state of things still.

-- 
William L. Thomson Jr.

signature.asc
Description: This is a digitally signed message part.

Re: [gentoo-dev] Tinderboxing efforts in Gentoo

On Sat, Dec 3, 2016 at 9:08 AM, William L. Thomson Jr.
 wrote:
> On Friday, December 2, 2016 2:10:27 PM EST Michał Górny wrote:
>> Hi, everyone.
>>
>> I've heard multiple times about various tinderbox projects being
>> started by individuals in Gentoo. In fact, so many different projects
>> that I've forgotten who was working on most of them.
>
> Did any of the tinderboxes test all the various USE flag combinations or just
> the defaults?
>

This is generally considered infeasible:

https://blog.flameeyes.eu/2009/06/a-tinderbox-is-not-enough-reprise/

I think continuous integration with tinderboxing is a really solid
idea.  However, I'm not under any illusions that it will involve
testing every package with every possible set of USE flags.  IMO
testing with the defaults of a few common profiles is probably
sufficient, and mgorny's CI approach of just doing repoman checks is
already a value add.  I can't remember the last time I ran into a
silly lack-of-repoman issue because they don't exist in the tree I
sync from, but it is obvious that many users still run into them
because this is not present in the official rsync mirrors (at least
the manifest issues have been sorted out).

-- 
Rich

Re: [gentoo-dev] OT Who runs Gentoo was -> RFC: Userkit.eclass

On Sat, Dec 3, 2016 at 9:00 AM, William L. Thomson Jr.
 wrote:
> OT, who runs Gentoo
> On Saturday, December 3, 2016 12:21:55 AM EST Daniel Campbell wrote:
>
>> There's also our downstream neighbors: Funtoo, Pentoo, Sabayon,
>> Calculate, Exherbo, etc
>
> Two of those are more of a splinter of the Gentoo community, Exherbo and
> Funtoo. IMHO those communities should be reunited into Gentoo. Allot of talent
> got driven way that is still out there working on other stuff.
>

Uh, you do realize that the main force behind one of those projects
left because of the main force behind the other project?

https://lwn.net/Articles/225060/

-- 
Rich

Re: [gentoo-dev] Tinderboxing efforts in Gentoo

On Friday, December 2, 2016 2:10:27 PM EST Michał Górny wrote:
> Hi, everyone.
> 
> I've heard multiple times about various tinderbox projects being
> started by individuals in Gentoo. In fact, so many different projects
> that I've forgotten who was working on most of them.

Did any of the tinderboxes test all the various USE flag combinations or just 
the defaults?

Automating builds has always been questionable to me due to the USE flag issue. 
I have things auto-building, but it just runs with what ever the USE defaults 
are at the time. In addition to any I have set. I have cron email me emerge 
output so I can review use flags. But that doesn't really fit into the 
automated 
building aspect.

Like if a new use flag is added I may want to enable but is not by default. 
Automation doesn't really help there without a human making the call to add 
new USE flag or not.

-- 
William L. Thomson Jr.

signature.asc
Description: This is a digitally signed message part.

Re: [gentoo-dev] OT Who runs Gentoo was -> RFC: Userkit.eclass

OT, who runs Gentoo

On Saturday, December 3, 2016 12:21:55 AM EST Daniel Campbell wrote:
> 
> Unless something's changed in the past year or two, iirc Sony uses
> Gentoo as part of the backend of Gaikai, Google's used it for the base
> of ChromeOS...

Also is the base of CoreOS, widely popular. Google uses it in more than 
ChromOS it was used in the OnHub router. I suspect Google has wider uses.

Google has hired a few core developers as has Gaikai. Both seem to be good, 
though not sure Google is giving back as much given their financial benefit. 
Gaikai isn't selling an OS, but Google is based on Gentoo...

> I can't speak for other 'big names', but Gentoo's not
> quite as niche as the small, active userbase has most of us believing.

FYI Network Solutions, which I believe at a time had a root DNS server running 
Gentoo. Due to their parent company Web.com is, or has by now, moving from 
Gentoo to RHEL. I was told they do not want to be in the operating system 
development business.

Meetup.com runs Gentoo, not sure how widely but it is in use. Likely more out 
there, but I see far more RHEL and CentOS. I do not even discuss Gentoo, as I 
have to many say it can't be used in production, etc. Things I disagree with.

> There's also our downstream neighbors: Funtoo, Pentoo, Sabayon,
> Calculate, Exherbo, etc

Two of those are more of a splinter of the Gentoo community, Exherbo and 
Funtoo. IMHO those communities should be reunited into Gentoo. Allot of talent 
got driven way that is still out there working on other stuff.

> As for communities, lots of places from 4chan to lainchan, various mesh
> network users, security-conscious communities, OCD support groups
> (kidding), etc.

My concern is lack of any mention in the news or tech articles. When was the 
last time Gentoo was making headlines and not for drama related stuff.

Gentoo News articles
https://www.google.com/search?hl=en=us=nws=0=gentoo+linux

Arch News articles for example, other distros Ubuntu and Debian have allot
https://www.google.com/search?hl=en=us=nws=0=arch+linux

Not my words but like say Infoworld articles, not that they are any authority

http://www.infoworld.com/article/3109830/linux/why-did-gentoo-linux-fade-into-obscurity.html

http://www.infoworld.com/article/3137969/linux/arch-linux-the-last-refuge-for-purists.html

-- 
William L. Thomson Jr.

signature.asc
Description: This is a digitally signed message part.

Re: [gentoo-dev] Tinderboxing efforts in Gentoo

2016-12-03 Thread Markos Chandras

On 12/03/2016 10:41 AM, Michał Górny wrote:
> On Sat, 3 Dec 2016 10:35:32 +0100
> Patrice Clement  wrote:
> 
>> Friday 02 Dec 2016 14:10:27, Michał Górny wrote :
>>> Hi, everyone.
>>>
>>> I've heard multiple times about various tinderbox projects being
>>> started by individuals in Gentoo. In fact, so many different projects
>>> that I've forgotten who was working on most of them.
>>>
>>> I know that Toralf is doing tinderboxing for most of the stuff.
>>> What other projects do we have there? What is their status?
>>>
>>> Is there anything we could try to integrate with pull requests to get
>>> a better testing?
>>>
>>> -- 
>>> Best regards,
>>> Michał Górny
>>>   
>>
>> Continuous integration is all the rage these days and tinderboxing is the
>> obvious way to go concerning Gentoo. AFAIK, Toralf is the only contributor
>> doing tinderboxing out of his own will. In reality, we should have a team of
>> devs looking after our own tinderboxes instead of relying on the community.
>>
>> I'm wondering if we could start a donation campain for this project and ask
>> people if they've got spare machines laying around. I know a lot of folks are
>> reading this mailing list so maybe asking on gentoo-dev first for a start 
>> would
>> be appropriate.
> 
> Hardware is not the problem. Lack of software is.
> 

Have you considered using openQA[1] like openSUSE[2] and Fedora[3] do
instead of reinventing the wheel?

[1] http://open.qa/
[2] https://openqa.opensuse.org/
[3] https://openqa.fedoraproject.org/

-- 
Regards,
Markos Chandras

Re: [gentoo-dev] Tinderboxing efforts in Gentoo

On Sat, 3 Dec 2016 10:35:32 +0100
Patrice Clement  wrote:

> Friday 02 Dec 2016 14:10:27, Michał Górny wrote :
> > Hi, everyone.
> > 
> > I've heard multiple times about various tinderbox projects being
> > started by individuals in Gentoo. In fact, so many different projects
> > that I've forgotten who was working on most of them.
> > 
> > I know that Toralf is doing tinderboxing for most of the stuff.
> > What other projects do we have there? What is their status?
> > 
> > Is there anything we could try to integrate with pull requests to get
> > a better testing?
> > 
> > -- 
> > Best regards,
> > Michał Górny
> >   
> 
> Continuous integration is all the rage these days and tinderboxing is the
> obvious way to go concerning Gentoo. AFAIK, Toralf is the only contributor
> doing tinderboxing out of his own will. In reality, we should have a team of
> devs looking after our own tinderboxes instead of relying on the community.
> 
> I'm wondering if we could start a donation campain for this project and ask
> people if they've got spare machines laying around. I know a lot of folks are
> reading this mailing list so maybe asking on gentoo-dev first for a start 
> would
> be appropriate.

Hardware is not the problem. Lack of software is.

-- 
Best regards,
Michał Górny



pgpGmeTWa02gj.pgp
Description: OpenPGP digital signature

Re: [gentoo-dev] Tinderboxing efforts in Gentoo

2016-12-03 Thread Patrice Clement

Friday 02 Dec 2016 14:10:27, Michał Górny wrote :
> Hi, everyone.
> 
> I've heard multiple times about various tinderbox projects being
> started by individuals in Gentoo. In fact, so many different projects
> that I've forgotten who was working on most of them.
> 
> I know that Toralf is doing tinderboxing for most of the stuff.
> What other projects do we have there? What is their status?
> 
> Is there anything we could try to integrate with pull requests to get
> a better testing?
> 
> -- 
> Best regards,
> Michał Górny
> 

Continuous integration is all the rage these days and tinderboxing is the
obvious way to go concerning Gentoo. AFAIK, Toralf is the only contributor
doing tinderboxing out of his own will. In reality, we should have a team of
devs looking after our own tinderboxes instead of relying on the community.

I'm wondering if we could start a donation campain for this project and ask
people if they've got spare machines laying around. I know a lot of folks are
reading this mailing list so maybe asking on gentoo-dev first for a start would
be appropriate.

Otherwise, can the Foundation sponsor the purchase of a couple of servers?

-- 
Patrice Clement
Gentoo Linux developer
http://www.gentoo.org


signature.asc
Description: PGP signature

Re: [gentoo-dev] Uppercase characters in package names

On Sat, 3 Dec 2016 00:12:53 -0800
Daniel Campbell  wrote:

> On 12/02/2016 11:55 PM, Michał Górny wrote:
> > On Fri, 2 Dec 2016 23:21:34 -0800
> > Daniel Campbell  wrote:
> >   
> >> On 12/02/2016 10:45 AM, Ian Stakenvicius wrote:  
> >>> On 02/12/16 01:31 PM, Ciaran McCreesh wrote:
>  On Fri, 2 Dec 2016 13:24:29 -0500
>  Mike Gilbert  wrote:
> > On Fri, Dec 2, 2016 at 1:10 PM, Ciaran McCreesh
> >  wrote:
> >> On Fri, 2 Dec 2016 13:02:48 -0500
> >> Mike Gilbert  wrote:  
> >>> The devmanual states:
> >>>
> >>> The name section should contain only lowercase non-accented
> >>> letters, the digits 0-9, hyphens, underscores and plus characters.
> >>> Uppercase characters are strongly discouraged, but technically
> >>> valid.
> >>>
> >>> https://devmanual.gentoo.org/ebuild-writing/file-format/index.html
> >>>
> >>>
> >>> Why are uppercase characters strongly discouraged?
> >>>
> >>> Wouldn't it make sense to follow upstream's naming convention?  
> >>
> >> What's upstream's naming convention for Firefox?
> >
> > I have no idea. What's your point?
> 
>  That naming conventions are generally complicated and a mess, and that
>  no-one wants to have to remember whether it's firefox, Firefox, or
>  FireFox.
> 
> >>>
> >>> It's also more convenient at the consone to just type everything
> >>> lowercase.  I expect that's the primary reason it's discouraged.
> >>>
> >>>
> >>>
> >>> 
> >> That seems the most likely to me as well.
> >>
> >> We could make a more "user friendly" feature by setting up bash
> >> completion for package names, but that sounds a) daunting, b)
> >> error-prone, and c) probably not worth the time spent writing the
> >> script(s) necessary.  
> > 
> > There is a bash completion script for that for a long time now.
> > However, it no longer works correctly with new bash-completion versions
> > and it seems that nobody cares enough to fix it.
> >   
> Oh, that's good to know. I didn't find anything relevant with
> 'bash-completion' in its name in the tree. Where should I look for this
> script?

gentoo-bashcomp

-- 
Best regards,
Michał Górny



pgpqMJgma0PmT.pgp
Description: OpenPGP digital signature

Re: [gentoo-dev] Tinderboxing efforts in Gentoo

On Sat, 03 Dec 2016 09:03:11 +0100
Magnus Granberg  wrote:

> fredag 2 december 2016 kl. 23:32:37 CET skrev  Daniel Campbell:
> > On 12/02/2016 06:09 AM, Michael Mol wrote:  
> > > On Friday, December 02, 2016 02:10:27 PM Michał Górny wrote:  
> > >> Hi, everyone.
> > >> 
> > >> I've heard multiple times about various tinderbox projects being
> > >> started by individuals in Gentoo. In fact, so many different projects
> > >> that I've forgotten who was working on most of them.
> > >> 
> > >> I know that Toralf is doing tinderboxing for most of the stuff.
> > >> What other projects do we have there? What is their status?
> > >> 
> > >> Is there anything we could try to integrate with pull requests to get
> > >> a better testing?  
> > > 
> > > If there's a mostly-turnkey VM I can run to contribute to Tinderboxing, I
> > > have one or two systems I could benefit from some heat from over the
> > > winter. It's that or bring out the electric space heater. Was talking
> > > with my wife about mining Doge on one of them last night...  
> > 
> > I second that. I have a hexcore CPU and 16 GB of RAM, most of which I
> > don't use unless I'm compiling. If there's a guide that can get me up
> > and running with a VM within an hour or so, I'd be more than willing to
> > pitch in some cycles.
> > 
> > mgorny mentioned PRs, however... are such efforts moot if I don't have a
> > GitHub account?  
> I run tinderbox-cluster [1] with 4-7 WM's but it have been down for some time 
> now.  The web frontend need alot of work. It still miss bugreporting and 
> build 
> requests and the grafic need  work to. It use django.

Oh, that's the project I was thinking about. So.. what does it have
since it seems to miss almost everything? ;-)

> [1] https://wiki.gentoo.org/wiki/Project:Tinderbox-cluster


-- 
Best regards,
Michał Górny



pgpctZZlcTL7M.pgp
Description: OpenPGP digital signature

Re: [gentoo-dev] RFC: Userkit.eclass

On 11/28/2016 03:26 PM, M. J. Everitt wrote:
> On 28/11/16 19:39, William L. Thomson Jr. wrote:
>> For now who cares about other OS or distros. If Gentoo gets its house in 
>> order 
>> others may follow.
>>
> At the risk of a huge flame, remind me, who uses Gentoo again?!
> 
Unless something's changed in the past year or two, iirc Sony uses
Gentoo as part of the backend of Gaikai, Google's used it for the base
of ChromeOS... I can't speak for other 'big names', but Gentoo's not
quite as niche as the small, active userbase has most of us believing.

There's also our downstream neighbors: Funtoo, Pentoo, Sabayon,
Calculate, Exherbo, etc

As for communities, lots of places from 4chan to lainchan, various mesh
network users, security-conscious communities, OCD support groups
(kidding), etc.

I'm sure I'm missing some mentions here; this is just off the top of my
head.
-- 
Daniel Campbell - Gentoo Developer
OpenPGP Key: 0x1EA055D6 @ hkp://keys.gnupg.net
fpr: AE03 9064 AE00 053C 270C  1DE4 6F7A 9091 1EA0 55D6



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-dev] Uppercase characters in package names

On 12/02/2016 11:55 PM, Michał Górny wrote:
> On Fri, 2 Dec 2016 23:21:34 -0800
> Daniel Campbell  wrote:
> 
>> On 12/02/2016 10:45 AM, Ian Stakenvicius wrote:
>>> On 02/12/16 01:31 PM, Ciaran McCreesh wrote:  
 On Fri, 2 Dec 2016 13:24:29 -0500
 Mike Gilbert  wrote:  
> On Fri, Dec 2, 2016 at 1:10 PM, Ciaran McCreesh
>  wrote:  
>> On Fri, 2 Dec 2016 13:02:48 -0500
>> Mike Gilbert  wrote:
>>> The devmanual states:
>>>
>>> The name section should contain only lowercase non-accented
>>> letters, the digits 0-9, hyphens, underscores and plus characters.
>>> Uppercase characters are strongly discouraged, but technically
>>> valid.
>>>
>>> https://devmanual.gentoo.org/ebuild-writing/file-format/index.html
>>>
>>>
>>> Why are uppercase characters strongly discouraged?
>>>
>>> Wouldn't it make sense to follow upstream's naming convention?
>>
>> What's upstream's naming convention for Firefox?  
>
> I have no idea. What's your point?  

 That naming conventions are generally complicated and a mess, and that
 no-one wants to have to remember whether it's firefox, Firefox, or
 FireFox.
  
>>>
>>> It's also more convenient at the consone to just type everything
>>> lowercase.  I expect that's the primary reason it's discouraged.
>>>
>>>
>>>
>>>   
>> That seems the most likely to me as well.
>>
>> We could make a more "user friendly" feature by setting up bash
>> completion for package names, but that sounds a) daunting, b)
>> error-prone, and c) probably not worth the time spent writing the
>> script(s) necessary.
> 
> There is a bash completion script for that for a long time now.
> However, it no longer works correctly with new bash-completion versions
> and it seems that nobody cares enough to fix it.
> 
Oh, that's good to know. I didn't find anything relevant with
'bash-completion' in its name in the tree. Where should I look for this
script?

-- 
Daniel Campbell - Gentoo Developer
OpenPGP Key: 0x1EA055D6 @ hkp://keys.gnupg.net
fpr: AE03 9064 AE00 053C 270C  1DE4 6F7A 9091 1EA0 55D6



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-dev] Uppercase characters in package names

On 12/02/2016 11:59 PM, Michał Górny wrote:
> On Fri, 2 Dec 2016 23:26:53 -0800
> Daniel Campbell  wrote:
> 
>> On 12/02/2016 10:47 AM, Michał Górny wrote:
>>> On Fri, 2 Dec 2016 13:02:48 -0500
>>> Mike Gilbert  wrote:
>>>   
 The devmanual states:

 The name section should contain only lowercase non-accented letters,
 the digits 0-9, hyphens, underscores and plus characters. Uppercase
 characters are strongly discouraged, but technically valid.

 https://devmanual.gentoo.org/ebuild-writing/file-format/index.html


 Why are uppercase characters strongly discouraged?

 Wouldn't it make sense to follow upstream's naming convention?  
>>>
>>> I'd say keeping things lowercase makes sense for end user packages. For
>>> pure dependencies with consistent conventions (e.g. perl), it makes
>>> sense to keep upstream's naming.
>>>   
>> What is a pure dependency? Do we handle those differently than the
>> garden-variety dependencies in other packages?
> 
> It is a package that is rarely installed directly, and rather commonly
> taken as a dependency of another package. For example, packages that
> install no programs and just Perl/Python/... modules.
> 
Ah, thanks for explaining that. Makes a lot more sense.

-- 
Daniel Campbell - Gentoo Developer
OpenPGP Key: 0x1EA055D6 @ hkp://keys.gnupg.net
fpr: AE03 9064 AE00 053C 270C  1DE4 6F7A 9091 1EA0 55D6



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-dev] Tinderboxing efforts in Gentoo

2016-12-03 Thread Magnus Granberg

fredag 2 december 2016 kl. 23:32:37 CET skrev  Daniel Campbell:
> On 12/02/2016 06:09 AM, Michael Mol wrote:
> > On Friday, December 02, 2016 02:10:27 PM Michał Górny wrote:
> >> Hi, everyone.
> >> 
> >> I've heard multiple times about various tinderbox projects being
> >> started by individuals in Gentoo. In fact, so many different projects
> >> that I've forgotten who was working on most of them.
> >> 
> >> I know that Toralf is doing tinderboxing for most of the stuff.
> >> What other projects do we have there? What is their status?
> >> 
> >> Is there anything we could try to integrate with pull requests to get
> >> a better testing?
> > 
> > If there's a mostly-turnkey VM I can run to contribute to Tinderboxing, I
> > have one or two systems I could benefit from some heat from over the
> > winter. It's that or bring out the electric space heater. Was talking
> > with my wife about mining Doge on one of them last night...
> 
> I second that. I have a hexcore CPU and 16 GB of RAM, most of which I
> don't use unless I'm compiling. If there's a guide that can get me up
> and running with a VM within an hour or so, I'd be more than willing to
> pitch in some cycles.
> 
> mgorny mentioned PRs, however... are such efforts moot if I don't have a
> GitHub account?
I run tinderbox-cluster [1] with 4-7 WM's but it have been down for some time 
now.  The web frontend need alot of work. It still miss bugreporting and build 
requests and the grafic need  work to. It use django.
/Magnus

[1] https://wiki.gentoo.org/wiki/Project:Tinderbox-cluster

Re: [gentoo-dev] Uppercase characters in package names