Bug#902631: devscripts: CI fails in unstable due to Python 3.7

2018-09-12 Thread Benjamin Drung
Hi,

On Sat, 1 Sep 2018 12:43:27 +0200 Nicolas Dandrimont 
wrote:
> > As a proof of concept I've gone and done the changes needed to pull
> this off on a fork of the git repo in my personal namespace.
> 
> https://salsa.debian.org/olasd/astroid
> 
> The astroid source package is in the master / upstream branches;
> The astroid2 source package is in the astroid2 / upstream-1.x
 > branches.
> 
> I had to do some wrangling of the astroid source package, to convert
 > it to pybuild, because the default setuptools buildsystem only
works 
> when python2 is in the build depends.
> 
> I could see a point in splitting the git repositories but I don't
 > think that's very critical.
> 
> I'm happy to do a NMU in a week so that we can move forward with the
 > split package plan.

I agree on splitting the source package into the legacy Python 2
version and a recent Python 3 version. I looked at your fork in your
personal namespace. It looks good to me.

Since this package is team maintained, will you merge the changes and
upload them (unless Sandro says something against it)? I like to open a
freeze exception bug in Ubuntu and get this solution synced to Ubuntu.

-- 
Benjamin Drung
System Developer
Debian & Ubuntu Developer

ProfitBricks GmbH
Greifswalder Str. 207
10405 Berlin

Email: benjamin.dr...@profitbricks.com
URL: https://www.profitbricks.de

Sitz der Gesellschaft: Berlin
Registergericht: Amtsgericht Charlottenburg, HRB 125506 B
Geschäftsführer: Achim Weiss, Matthias Steinberg, Christoph Steffens



Bug#902631: devscripts: CI fails in unstable due to Python 3.7

2018-09-01 Thread Nicolas Dandrimont
Control: tags -1 + patch

* Thomas Goirand  [2018-08-25 22:02:51 +0200]:

> Hi Sando,
> 
> I do support uploading 2 source packages, one for py2, and one for py3.
> This is the most reasonable solution. Worst case, if that cannot be
> done, then we got to move forward and restore Py3 support (and
> eventually drop Py2), but I prefer the former solution.
> 
> The issue has been opened for nearly 2 months now, with nearly a month
> without any news, and this makes me a bit nervous, as I need the py3
> support to be restored.
> 
> What's holding you off more?

Hi,

As a proof of concept I've gone and done the changes needed to pull this off on
a fork of the git repo in my personal namespace.

https://salsa.debian.org/olasd/astroid

The astroid source package is in the master / upstream branches;
The astroid2 source package is in the astroid2 / upstream-1.x branches.

I had to do some wrangling of the astroid source package, to convert it to
pybuild, because the default setuptools buildsystem only works when python2 is
in the build depends.

I could see a point in splitting the git repositories but I don't think that's
very critical.

I'm happy to do a NMU in a week so that we can move forward with the split
package plan.

Cheers,
-- 
Nicolas Dandrimont

We come to bury DOS, not to praise it.
(Paul Vojta, vo...@math.berkeley.edu, paraphrasing a quote of Shakespeare)


signature.asc
Description: PGP signature


Bug#902631: devscripts: CI fails in unstable due to Python 3.7

2018-08-25 Thread Thomas Goirand
Hi Sando,

I do support uploading 2 source packages, one for py2, and one for py3.
This is the most reasonable solution. Worst case, if that cannot be
done, then we got to move forward and restore Py3 support (and
eventually drop Py2), but I prefer the former solution.

The issue has been opened for nearly 2 months now, with nearly a month
without any news, and this makes me a bit nervous, as I need the py3
support to be restored.

What's holding you off more?

Cheers,

Thomas Goirand (zigo)



Bug#902631: devscripts: CI fails in unstable due to Python 3.7

2018-08-03 Thread Stuart Prescott
Hi Sandro,

thanks for this update

On Thursday, 2 August 2018 12:39:18 AEST Sandro Tosi wrote:
> there are currently 2 source packages:
> 
> astroid: on the 1.x branch, builds python-astroid and python3-astroid
> pylint: on the 1.x branch, builds pylint (depends on python-astroid),
> pylint3 (depends on python3-astroid), pylint-doc
> 
> the only way forward to keep compat with python2 while supporting 3.7 i see
> is:
> 
> astroid: uses the 2.x branch, builds only python3-astroid
> pylint: uses the 2.x branch, builds only pylint3, pylint-doc
> NEW astroid2: remains on the 1.x branch, builds only python-astroid
> NEW pylint2: remains on the 1.x branch, builds only pylint (depending
> on python-astroid)
> 
> astroid2 and pylint2 will be dropped after buster is released, keeping
> only astroid and pylint in bullseye

I think this is a perfectly sensible way of proceeding.

> adding ftpmasters in the loop as a headsup since this plan would
> require a bit of coordination to reduce users disruption during the
> migration of the bin pkgs from one src pkg to the new ones.

I don't think there is actually any assistance required from the ftpmasters 
beyond accepting the NEW package. I've not found the relevant document, but my 
understanding is that a source package can take over the binary packages from 
another source package by simply uploading the new one to NEW, and once it is 
accepted, upload the old one without the extra stanzas in d/control.
 
> what people feel about it? just to make it clear: i'm not really
> excited about the package duplication, although it's the only viable
> solution i can see for now. comments/suggestions welcome

We've got a few examples of this in the archive where there is no polyglot 
version of a module or package and so two versions are shipped. We've even had 
some modules like that since wheezy. These are all transitional in the sense 
that the extra packages will disappear from Debian soon enough.

Please let me know if I can be of assistance with this work.

cheers
Stuart

-- 
Stuart Prescotthttp://www.nanonanonano.net/   stu...@nanonanonano.net
Debian Developer   http://www.debian.org/ stu...@debian.org
GPG fingerprint90E2 D2C1 AD14 6A1B 7EBB 891D BBC1 7EBB 1396 F2F7



Bug#902631: devscripts: CI fails in unstable due to Python 3.7

2018-08-02 Thread Bastian Blank
Hi

I miss any description of the real problem.  You only describe
solutions, but you forget to mention why the easiest solution, just
upgrade both sources, does not work.  Sure, some of the points can be
found in the bug report.

On Thu, Aug 02, 2018 at 12:39:18PM -0400, Sandro Tosi wrote:
> astroid: on the 1.x branch, builds python-astroid and python3-astroid
> pylint: on the 1.x branch, builds pylint (depends on python-astroid),
> pylint3 (depends on python3-astroid), pylint-doc
> 
> the only way forward to keep compat with python2 while supporting 3.7 i see 
> is:
> 
> astroid: uses the 2.x branch, builds only python3-astroid

Why can't this build python-astroid?

> pylint: uses the 2.x branch, builds only pylint3, pylint-doc

Why can't this build pylint?

> NEW astroid2: remains on the 1.x branch, builds only python-astroid
> NEW pylint2: remains on the 1.x branch, builds only pylint (depending
> on python-astroid)

Given that pylint is a program and no library, why do we need both
pylint and pylint3?

Regards,
Bastian

-- 
We'll pivot at warp 2 and bring all tubes to bear, Mr. Sulu!



Bug#902631: devscripts: CI fails in unstable due to Python 3.7

2018-08-02 Thread Pierre-Elliott Bécue
Le jeudi 02 août 2018 à 19:32:15+0200, Bastian Blank a écrit :
> [why should one package two source packages?]

See [1] and following posts. astroid 2.0 dropped the support for Python2.

Regards,

[1] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=902631#77

-- 
Pierre-Elliott Bécue
GPG: 9AE0 4D98 6400 E3B6 7528  F493 0D44 2664 1949 74E2
It's far easier to fight for one's principles than to live up to them.


signature.asc
Description: PGP signature


Bug#902631: devscripts: CI fails in unstable due to Python 3.7

2018-08-02 Thread Pierre-Elliott Bécue
Le jeudi 02 août 2018 à 12:39:18-0400, Sandro Tosi a écrit :
> there are currently 2 source packages:
> 
> astroid: on the 1.x branch, builds python-astroid and python3-astroid
> pylint: on the 1.x branch, builds pylint (depends on python-astroid),
> pylint3 (depends on python3-astroid), pylint-doc
> 
> the only way forward to keep compat with python2 while supporting 3.7 i see 
> is:
> 
> astroid: uses the 2.x branch, builds only python3-astroid
> pylint: uses the 2.x branch, builds only pylint3, pylint-doc
> NEW astroid2: remains on the 1.x branch, builds only python-astroid
> NEW pylint2: remains on the 1.x branch, builds only pylint (depending
> on python-astroid)
> 
> astroid2 and pylint2 will be dropped after buster is released, keeping
> only astroid and pylint in bullseye
> 
> adding ftpmasters in the loop as a headsup since this plan would
> require a bit of coordination to reduce users disruption during the
> migration of the bin pkgs from one src pkg to the new ones.
> 
> what people feel about it? just to make it clear: i'm not really
> excited about the package duplication, although it's the only viable
> solution i can see for now. comments/suggestions welcome

That seems, for me, the best solution, provided it's temporary (which we all
agree on). I had a look on porting the changes to 1.x from 2.x that would
allow python3-astroid to work properly, and it seems quite a lot of work
with a mixed outcome.

If you wish, I'm eager to contribute into implementing this solution, but
keep in mind that, as a DM, I'll eventually require your help to upload
these packages.

If you prefer to handle this on your own, it's fine with me, too.

Thanks for considering, and cheers!

-- 
Pierre-Elliott Bécue
GPG: 9AE0 4D98 6400 E3B6 7528  F493 0D44 2664 1949 74E2
It's far easier to fight for one's principles than to live up to them.


signature.asc
Description: PGP signature


Bug#902631: devscripts: CI fails in unstable due to Python 3.7

2018-08-02 Thread Sandro Tosi
there are currently 2 source packages:

astroid: on the 1.x branch, builds python-astroid and python3-astroid
pylint: on the 1.x branch, builds pylint (depends on python-astroid),
pylint3 (depends on python3-astroid), pylint-doc

the only way forward to keep compat with python2 while supporting 3.7 i see is:

astroid: uses the 2.x branch, builds only python3-astroid
pylint: uses the 2.x branch, builds only pylint3, pylint-doc
NEW astroid2: remains on the 1.x branch, builds only python-astroid
NEW pylint2: remains on the 1.x branch, builds only pylint (depending
on python-astroid)

astroid2 and pylint2 will be dropped after buster is released, keeping
only astroid and pylint in bullseye

adding ftpmasters in the loop as a headsup since this plan would
require a bit of coordination to reduce users disruption during the
migration of the bin pkgs from one src pkg to the new ones.

what people feel about it? just to make it clear: i'm not really
excited about the package duplication, although it's the only viable
solution i can see for now. comments/suggestions welcome

-- 
Sandro "morph" Tosi
My website: http://sandrotosi.me/
Me at Debian: http://wiki.debian.org/SandroTosi
G+: https://plus.google.com/u/0/+SandroTosi



Bug#902631: devscripts: CI fails in unstable due to Python 3.7

2018-07-23 Thread Pierre-Elliott Bécue
Le dimanche 22 juillet 2018 à 20:43:02-0400, Sandro Tosi a écrit :
> On Sun, Jul 22, 2018 at 8:31 PM Mattia Rizzolo  wrote:
> >
> > On Sun, Jul 22, 2018 at 03:52:04PM -0400, Sandro Tosi wrote:
> > > thanks for ignoring the next paragraph in that email, so let me re-add it 
> > > here:
> >
> > I didn't ignore it, I explicitly picked on your first, fairly taunting,
> > and inappropriately passive-aggressive sentence (like this one).
> 
> good good

Mmh, I tend to admit that Sandro's answers can be blunt. I assume it may be
connected with a busy life, but sometimes it's a little awkward.

That said, I'd rather all of us to act/behave accordingly to [1], please.
Because we're all smart people, with good intentions, who share common
values (the ones that made us join Debian). :)

> > > since astroid/pylint only support python 3.4+, it is not as simple as
> > > package the new version and upload it; i'm not even (yet) sure buster
> > > should have 2.0 but maybe it should to 1.9
> >
> > This is a fairly common situation for distributors.
> >
> > Did you also evaluate the feasibility and practicality of working on a
> > patch to introduce py3.7 compatibility to the current 1.6.5?
> 
> no, i didnt look into it myself: upstream dismissed the forwarded bug
> saying to upgrade to 2.0 to get 3.7 support.
> 
> if someone wants to try and dedicate time to that effort, that'd be
> welcome and please attach a patch here when ready.

Would you consider repackaging 1.6.5 without py3 support and provide another
package with exclusive py3 support? That'd ease the drop of py2 in a few
years, and, also, avoid big horrible patches. If you agree with this kind of
idea I'd be pleased to provide the new package work and a team patch for the
current package.

If you really wish to go by a patch for 1.6.5, I'll try to do it after the
end of my PhD thesis manuscript writing.

Sincerely

[1] https://www.debian.org/code_of_conduct.en.html

-- 
Pierre-Elliott Bécue
GPG: 9AE0 4D98 6400 E3B6 7528  F493 0D44 2664 1949 74E2
It's far easier to fight for one's principles than to live up to them.


signature.asc
Description: PGP signature


Bug#902631: devscripts: CI fails in unstable due to Python 3.7

2018-07-22 Thread Sandro Tosi
On Sun, Jul 22, 2018 at 8:31 PM Mattia Rizzolo  wrote:
>
> On Sun, Jul 22, 2018 at 03:52:04PM -0400, Sandro Tosi wrote:
> > thanks for ignoring the next paragraph in that email, so let me re-add it 
> > here:
>
> I didn't ignore it, I explicitly picked on your first, fairly taunting,
> and inappropriately passive-aggressive sentence (like this one).

good good

> > since astroid/pylint only support python 3.4+, it is not as simple as
> > package the new version and upload it; i'm not even (yet) sure buster
> > should have 2.0 but maybe it should to 1.9
>
> This is a fairly common situation for distributors.
>
> Did you also evaluate the feasibility and practicality of working on a
> patch to introduce py3.7 compatibility to the current 1.6.5?

no, i didnt look into it myself: upstream dismissed the forwarded bug
saying to upgrade to 2.0 to get 3.7 support.

if someone wants to try and dedicate time to that effort, that'd be
welcome and please attach a patch here when ready.

-- 
Sandro "morph" Tosi
My website: http://sandrotosi.me/
Me at Debian: http://wiki.debian.org/SandroTosi
G+: https://plus.google.com/u/0/+SandroTosi



Bug#902631: devscripts: CI fails in unstable due to Python 3.7

2018-07-22 Thread Mattia Rizzolo
On Sun, Jul 22, 2018 at 03:52:04PM -0400, Sandro Tosi wrote:
> thanks for ignoring the next paragraph in that email, so let me re-add it 
> here:

I didn't ignore it, I explicitly picked on your first, fairly taunting,
and inappropriately passive-aggressive sentence (like this one).


> since astroid/pylint only support python 3.4+, it is not as simple as
> package the new version and upload it; i'm not even (yet) sure buster
> should have 2.0 but maybe it should to 1.9

This is a fairly common situation for distributors.

Did you also evaluate the feasibility and practicality of working on a
patch to introduce py3.7 compatibility to the current 1.6.5?

-- 
regards,
Mattia Rizzolo

GPG Key: 66AE 2B4A FCCF 3F52 DA18  4D18 4B04 3FCD B944 4540  .''`.
more about me:  https://mapreri.org : :'  :
Launchpad user: https://launchpad.net/~mapreri  `. `'`
Debian QA page: https://qa.debian.org/developer.php?login=mattia  `-


signature.asc
Description: PGP signature


Bug#902631: devscripts: CI fails in unstable due to Python 3.7

2018-07-22 Thread Sandro Tosi
> What do you mean with "support"? Astroid 2.0 is working properly with py3.7 
> according to upstream.

i mean that pylint/astroid 2.0 do not work with python 2 at all, but
only with python 3.4 onwards (so yes, also 3.7 in that regards).

-- 
Sandro "morph" Tosi
My website: http://sandrotosi.me/
Me at Debian: http://wiki.debian.org/SandroTosi
G+: https://plus.google.com/u/0/+SandroTosi



Bug#902631: devscripts: CI fails in unstable due to Python 3.7

2018-07-22 Thread Sandro Tosi
> Upstream README seems to indicate that there is no release that supports
> *both* python 2.7 and 3.7

i know, that's exactly my point

> so this is not a choice between 2.0 and 1.9.

well it is: it can be decided to ignore the 2.x branch and stick with
1.9 for buster (which support python 2 and python 3 up to 3.6) and not
build astroid/pylint for 3.7 which possibly wont be the default py3k
in buster (but it's hard to tell as common developers dont get much
insight in the decisional process)

> More likely, two source packages are required for buster. Alternatively one of

i'm not really a fan of keeping 2 packages sets around

> Python 2 / Python 3 must be dropped from pylint (presumably Python 2). If
> these are the options, which are you intending to follow?

i dont know, that's what's holding up a resolution for this bug! on
one hand we still want to support python 2 in buster, so that's a
compelling reason to keep python 2 support in pylint, OTOH we want to
introduce 3.7 as (at least) a supported version, and the only way to
make pylint available for 3.7 is to drop entirely support for python2.

few other project i maintain are planning on released a py3k-only
version, so maybe that's the direction to go with such modules?

> On Sunday, 22 July 2018 15:52:04 AEST Sandro Tosi wrote:
> > thanks for ignoring the next paragraph in that email, so let me re-add it
> > here:
> >
> > ```
> > since astroid/pylint only support python 3.4+, it is not as simple as
> > package the new version and upload it; i'm not even (yet) sure buster
> > should have 2.0 but maybe it should to 1.9
> > ```
>
> No-one is ignoring that statement, just hoping that there is progress towards
> a solution rather than statements of what the problem is. We have all read the

until your email, there werent any solutions proposed by external
parties, so thanks for that, i think we should still understand how to
approach it in the right way.

> bug log and know the problem. You've indicated that you don't want help, but
> not what the timeline to a fix is. If time is a problem to deal with RC bugs,
> perhaps you'd be happy for team mates in PAPT to deal with this?

that's correct: i dont need help to prepare a trivial patch to update
astroid/pylint to the latest version, that's something anyone can do;
what i'm concerned here is what approach to follow for packages
dropping support for python releases Debian still cares to support.


-- 
Sandro "morph" Tosi
My website: http://sandrotosi.me/
Me at Debian: http://wiki.debian.org/SandroTosi
G+: https://plus.google.com/u/0/+SandroTosi



Bug#902631: devscripts: CI fails in unstable due to Python 3.7

2018-07-22 Thread Pierre-Elliott Bécue
Le 22 juillet 2018 21:52:04 GMT+02:00, Sandro Tosi  a écrit :
>On Sat, Jul 21, 2018 at 6:58 AM Mattia Rizzolo 
>wrote:
>>
>> On Tue, Jul 17, 2018 at 03:06:14PM -0400, Sandro Tosi wrote:
>> > no need, i can upgrade the package myself when needed, but thanks
>for the offer.
>>
>> Then can you pretty please do it?  It is neeeded.
>
>thanks for ignoring the next paragraph in that email, so let me re-add
>it here:
>
>```
>since astroid/pylint only support python 3.4+, it is not as simple as
>package the new version and upload it; i'm not even (yet) sure buster
>should have 2.0 but maybe it should to 1.9
>```
What do you mean with "support"? Astroid 2.0 is working properly with py3.7 
according to upstream. 
-- 
PEB



Bug#902631: devscripts: CI fails in unstable due to Python 3.7

2018-07-22 Thread Stuart Prescott
Hi Sandro,

Upstream README seems to indicate that there is no release that supports 
*both* python 2.7 and 3.7 so this is not a choice between 2.0 and 1.9.

More likely, two source packages are required for buster. Alternatively one of 
Python 2 / Python 3 must be dropped from pylint (presumably Python 2). If 
these are the options, which are you intending to follow?

On Sunday, 22 July 2018 15:52:04 AEST Sandro Tosi wrote:
> thanks for ignoring the next paragraph in that email, so let me re-add it
> here:
> 
> ```
> since astroid/pylint only support python 3.4+, it is not as simple as
> package the new version and upload it; i'm not even (yet) sure buster
> should have 2.0 but maybe it should to 1.9
> ```

No-one is ignoring that statement, just hoping that there is progress towards 
a solution rather than statements of what the problem is. We have all read the 
bug log and know the problem. You've indicated that you don't want help, but 
not what the timeline to a fix is. If time is a problem to deal with RC bugs, 
perhaps you'd be happy for team mates in PAPT to deal with this?

For me, at least, the inability to have pylint3 installed and test out 
python3.7 on the same machine is now hindering other Debian work, including 
the work that I wanted to do this week at DebCamp.

cheers
Stuart

-- 
Stuart Prescotthttp://www.nanonanonano.net/   stu...@nanonanonano.net
Debian Developer   http://www.debian.org/ stu...@debian.org
GPG fingerprint90E2 D2C1 AD14 6A1B 7EBB 891D BBC1 7EBB 1396 F2F7



Bug#902631: devscripts: CI fails in unstable due to Python 3.7

2018-07-22 Thread Sandro Tosi
On Sat, Jul 21, 2018 at 6:58 AM Mattia Rizzolo  wrote:
>
> On Tue, Jul 17, 2018 at 03:06:14PM -0400, Sandro Tosi wrote:
> > no need, i can upgrade the package myself when needed, but thanks for the 
> > offer.
>
> Then can you pretty please do it?  It is neeeded.

thanks for ignoring the next paragraph in that email, so let me re-add it here:

```
since astroid/pylint only support python 3.4+, it is not as simple as
package the new version and upload it; i'm not even (yet) sure buster
should have 2.0 but maybe it should to 1.9
```

--
Sandro "morph" Tosi
My website: http://sandrotosi.me/
Me at Debian: http://wiki.debian.org/SandroTosi
G+: https://plus.google.com/u/0/+SandroTosi



Bug#902631: devscripts: CI fails in unstable due to Python 3.7

2018-07-21 Thread Mattia Rizzolo
On Tue, Jul 17, 2018 at 03:06:14PM -0400, Sandro Tosi wrote:
> no need, i can upgrade the package myself when needed, but thanks for the 
> offer.

Then can you pretty please do it?  It is neeeded.


signature.asc
Description: PGP signature


Bug#902631: devscripts: CI fails in unstable due to Python 3.7

2018-07-17 Thread Pierre-Elliott Bécue
Le mardi 17 juillet 2018 à 15:06:14-0400, Sandro Tosi a écrit :
> no need, i can upgrade the package myself when needed, but thanks for the 
> offer.
> 
> since astroid/pylint only support python 3.4+, it is not as simple as
> package the new version and upload it; i'm not even (yet) sure buster
> should have 2.0 but maybe it should to 1.9

I had a look at the changelog of upstream and failed to see any notice
regarding py2 drop. Thanks for your remark.

Unfortunately, 2.0 is the only py3.7 compatible release. :/

Do you wish me to change my patch to drop py2 support?

-- 
Pierre-Elliott Bécue
GPG: 9AE0 4D98 6400 E3B6 7528  F493 0D44 2664 1949 74E2
It's far easier to fight for one's principles than to live up to them.


signature.asc
Description: PGP signature


Bug#902631: devscripts: CI fails in unstable due to Python 3.7

2018-07-17 Thread Sandro Tosi
no need, i can upgrade the package myself when needed, but thanks for the offer.

since astroid/pylint only support python 3.4+, it is not as simple as
package the new version and upload it; i'm not even (yet) sure buster
should have 2.0 but maybe it should to 1.9
On Tue, Jul 17, 2018 at 2:41 PM Pierre-Elliott Bécue  wrote:
>
> Le dimanche 08 juillet 2018 à 06:44:56-0400, Sandro Tosi a écrit :
> > Why don't you just submit it?
>
> Hi,
>
> Here is the diff for astroid 2.0-1.
>
> I also made a fork on salsa, with the changes I've made. Do you wish me to
> submit 3 merge requests? (one per branch)
>
> Cheers,
>
> --
> Pierre-Elliott Bécue
> GPG: 9AE0 4D98 6400 E3B6 7528  F493 0D44 2664 1949 74E2
> It's far easier to fight for one's principles than to live up to them.



-- 
Sandro "morph" Tosi
My website: http://sandrotosi.me/
Me at Debian: http://wiki.debian.org/SandroTosi
G+: https://plus.google.com/u/0/+SandroTosi



Bug#902631: devscripts: CI fails in unstable due to Python 3.7

2018-07-08 Thread Pierre-Elliott Bécue
Le dimanche 08 juillet 2018 à 06:44:56-0400, Sandro Tosi a écrit :
> Why don't you just submit it?

Because it's not finished yet.

-- 
Pierre-Elliott Bécue
GPG: 9AE0 4D98 6400 E3B6 7528  F493 0D44 2664 1949 74E2
It's far easier to fight for one's principles than to live up to them.


signature.asc
Description: PGP signature


Bug#902631: devscripts: CI fails in unstable due to Python 3.7

2018-07-08 Thread Sandro Tosi
Why don't you just submit it?

On Sun, Jul 8, 2018, 06:24 Pierre-Elliott Bécue  wrote:

> Hi, Sandro,
>
> Would you agree to review and upload a fix I'd submit for Bug#902631?
>
> If you don't have enough time, may I have a fix uploaded myself?
>
> Cheers and wish you well! :)
>
> --
> Pierre-Elliott Bécue
> GPG: 9AE0 4D98 6400 E3B6 7528  F493 0D44 2664 1949 74E2
> It's far easier to fight for one's principles than to live up to them.
>


Bug#902631: devscripts: CI fails in unstable due to Python 3.7

2018-07-08 Thread Pierre-Elliott Bécue
Hi, Sandro,

Would you agree to review and upload a fix I'd submit for Bug#902631?

If you don't have enough time, may I have a fix uploaded myself?

Cheers and wish you well! :)

-- 
Pierre-Elliott Bécue
GPG: 9AE0 4D98 6400 E3B6 7528  F493 0D44 2664 1949 74E2
It's far easier to fight for one's principles than to live up to them.


signature.asc
Description: PGP signature


Bug#902631: devscripts: CI fails in unstable due to Python 3.7

2018-06-28 Thread James McCoy
Control: reassign -1 src:astroid
Control: retitle -1 Not compatible with Python 3.7
Control: tag -1 upstream
Control: affects -1 devscripts

On Thu, Jun 28, 2018 at 03:12:14PM -0400, Lev Lazinskiy wrote:
> The current CI process fails in unstable[1]. It appears to be due to Python
> 3.7 being released in the last few days[2].
> 
> The issue is:
> 
> 
> : generator raised StopIteration
> 
> 
> It can be reproduced by running:
> 
> python3.7 -m flake8 --max-line-length=99
> debdiff-apply reproducible-check sadt suspicious-source wrap-and-sort
> devscripts setup.py
> 
> in unstable.

I think you mean "-m pylint" here, since the astroid modules are used by
pylint.

https://github.com/PyCQA/astroid/issues/539

> This appears to be an intentional regression that is caused by python
> 3.7 enforcing PEP 479 [3].
> 
> 
> [1] https://salsa.debian.org/levlaz-guest/devscripts/-/jobs/28131
> [2] https://www.python.org/downloads/release/python-370/
> [3] https://www.python.org/dev/peps/pep-0479/

Cheers,
-- 
James
GPG Key: 4096R/91BF BF4D 6956 BD5D F7B7  2D23 DFE6 91AE 331B A3DB



Bug#902631: devscripts: CI fails in unstable due to Python 3.7

2018-06-28 Thread Lev Lazinskiy
Package: devscripts
Version: 2.18.3
Severity: normal

Dear Maintainer,

The current CI process fails in unstable[1]. It appears to be due to Python
3.7 being released in the last few days[2].

The issue is:


: generator raised StopIteration


It can be reproduced by running:

python3.7 -m flake8 --max-line-length=99
debdiff-apply reproducible-check sadt suspicious-source wrap-and-sort
devscripts setup.py

in unstable.


This appears to be an intentional regression that is caused by python
3.7 enforcing PEP 479 [3].


[1] https://salsa.debian.org/levlaz-guest/devscripts/-/jobs/28131
[2] https://www.python.org/downloads/release/python-370/
[3] https://www.python.org/dev/peps/pep-0479/

-- Package-specific info:

--- /etc/devscripts.conf ---

--- ~/.devscripts ---
DEBUILD_DPKG_BUILDPACKAGE_OPTS="-i -I -us -uc"
DEBUILD_LINTIAN_OPTS="-i -I --show-overrides"
DEBSIGN_KEYID="7F826F1C"

-- System Information:
Debian Release: buster/sid
  APT prefers unstable
  APT policy: (500, 'unstable')
Architecture: amd64 (x86_64)

Kernel: Linux 4.9.87-linuxkit-aufs (SMP w/4 CPU cores)
Locale: LANG=C, LC_CTYPE=C (charmap=ANSI_X3.4-1968), LANGUAGE=C
(charmap=ANSI_X3.4-1968)
Shell: /bin/sh linked to /usr/bin/dash
Init: unable to detect

Versions of packages devscripts depends on:
ii  dpkg-dev  1.19.0.5
ii  libc6 2.27-3
ii  libfile-homedir-perl  1.004-1
ii  perl  5.26.2-6
ii  python3   3.6.5-3
ii  sensible-utils0.0.12

Versions of packages devscripts recommends:
ii  apt 1.6.2
ii  at  3.1.20-6
ii  curl7.60.0-2
ii  dctrl-tools 2.24-2+b1
ii  debian-keyring  2018.06.24
ii  dput1.0.2
ii  equivs  2.1.0
ii  fakeroot1.22-2
ii  file1:5.33-3
ii  gnupg   2.2.8-3
ii  libdistro-info-perl 0.18
ii  libdpkg-perl1.19.0.5
ii  libencode-locale-perl   1.05-1
ii  libgit-wrapper-perl 0.048-1
ii  liblist-compare-perl0.53-1
ii  liblwp-protocol-https-perl  6.07-2
ii  libsoap-lite-perl   1.27-1
ii  libstring-shellquote-perl   1.04-1
ii  libtry-tiny-perl0.30-1
ii  liburi-perl 1.74-1
ii  libwww-perl 6.34-1
ii  licensecheck3.0.31-2
ii  lintian 2.5.91
ii  man-db  2.8.3-2
ii  patch   2.7.6-2
ii  patchutils  0.3.4-2
ii  python3-apt 1.6.1
ii  python3-debian  0.1.32
ii  python3-magic   2:0.4.15-1
ii  python3-requests2.18.4-2
ii  python3-unidiff 0.5.4-1
ii  python3-xdg 0.25-4
ii  strace  4.21-1
ii  unzip   6.0-21
ii  wdiff   1.2.2-2+b1
ii  wget1.19.5-1
ii  xz-utils5.2.2-1.3

Versions of packages devscripts suggests:
pn  adequate 
pn  autopkgtest  
pn  bls-standalone   
ii  build-essential  12.5
pn  check-all-the-things 
pn  cvs-buildpackage 
pn  devscripts-el
pn  diffoscope   
pn  disorderfs   
pn  dose-extra   
pn  duck 
pn  faketime 
pn  gnuplot  
ii  gpgv 2.2.8-3
pn  how-can-i-help   
ii  libauthen-sasl-perl  2.1600-1
pn  libfile-desktopentry-perl
pn  libnet-smtps-perl
pn  libterm-size-perl
ii  libtimedate-perl 2.3000-2
pn  libyaml-syck-perl
ii  mailutils [mailx]1:3.4-1+b1
pn  mozilla-devscripts   
pn  mutt 
ii  openssh-client [ssh-client]  1:7.7p1-2
pn  piuparts 
pn  postgresql-client
pn  quilt
pn  ratt 
pn  reprotest
pn  svn-buildpackage 
pn  w3m  

-- no debconf information

-- 
Lev Lazinskiy
e: l...@levlaz.org
w: https://levlaz.org/about-me/