public inbox for cygwin-apps@cygwin.com
 help / color / mirror / Atom feed
From: Libor Ukropec <ace@seznam.cz>
To: cygwin-apps@cygwin.com
Subject: Re: [ITA] duplicity
Date: Tue, 12 Apr 2022 02:09:14 +0200	[thread overview]
Message-ID: <480ef945-55f2-b00d-6dd7-7a204acf14e9@seznam.cz> (raw)
In-Reply-To: <b1fe5275-b3f2-d0c3-6eae-e82cd9a58fb1@seznam.cz>

 > Run cygport ... all with --debug flag which enables shell tracing

I'll answer it myself. If the cygport is given the filename *without* 
".cygport" extension, it executes, but wrongly detects the PVR - 
NAME/VERSION/RELEASE. When I provided full name, it works as I'd 
expected. I think this deserves improvement.

Probably caused by TAB completion when it stopped on the first "." 
because python-fasteners.cygport file exists and also the 
python-fasteners.noarch existed already from previous run in the same 
directory.

Regards,
Libor

I'll try to forget how many hours it took.

Dne 10.04.2022 v 20:26 Libor Ukropec napsal(a):
> Hi Brian,
> 
> 1. regarding python-fasteners:
> 
> Dne 08.04.2022 v 1:44 Brian Inglis napsal(a):
>  > Run cygport ... all with --debug flag which enables shell tracing
>  > throughout and redirect all output &> debug.log for review.
> 
> Output for command `cygport --debug python-fasteners download all check 
> &> debug.log`
> 
> is here, if you can deduct from it something useful:
> 
> https://gist.github.com/cz6ace/929812203a42bd2d69506cad19385eed#file-debug-log 
> 
> 
> (it is quite long, I do not want to paste it directly into the email)
> 
> Also it is unknown to me, how the new repository can be added into the 
> https://cygwin.com/git/?a=project_list;pf=git/cygwin-packages so I can 
> execute the tests in the playground too.
> 
> 
> 2. regarding duplicity itself. My first successful build: 
> https://github.com/cygwin/scallywag/actions/runs/2144688591
> 
> Regards,
> Libor
> 
> Dne 08.04.2022 v 1:44 Brian Inglis napsal(a):
>> Run cygport ... all with --debug flag which enables shell tracing 
>> throughout and redirect all output &> debug.log for review.
>>
>> On 2022-04-07 16:26, Libor Ukropec wrote:
>>> Hi Brian,
>>>
>>> I solved the issue with Python 2.7 by adding PKG_NAMES and *_CONTENTS:
>>>
>>> inherit python-wheel
>>>
>>> PYTHON_WHEEL_VERSIONS="2.7:3.8:3.9"
>>> NAME="python-fasteners"
>>> VERSION=0.16.3
>>> RELEASE=1
>>> CATEGORY="Python"
>>> SUMMARY="Cross platform locks for threads and processes."
>>> DESCRIPTION="Python standard library provides a lock for threads 
>>> (both a reentrant one, and a non-reentrant one, see below). Fasteners 
>>> extends this, and provides a lock for processes, as well as Reader 
>>> Writer locks for both threads and processes."
>>> SRC_URI="https://github.com/harlowja/fasteners/archive/refs/tags/${VERSION}.tar.gz" 
>>>
>>> SRC_DIR="fasteners-${VERSION}"
>>> ARCH=noarch
>>> PKG_NAMES+=" python27-fasteners"
>>> python27_fasteners_CONTENTS="usr/lib/python2.7/site-packages/ 
>>> usr/share/doc/python27-fasteners/"
>>>
>>>
>>> still I'm concerned about the generated requirements, where the 
>>> package itself is referring to itself with very long name. Is that 
>>> normal?
>>>
>>>  >>> python38-fasteners requires: python38 
>>> python38-fasteners-python-fasteners-fasteners python38-six
>>>  >>> python39-fasteners requires: python39 
>>> python39-fasteners-python-fasteners-fasteners python39-six
>>>  >>> python27-fasteners requires: python27 
>>> python27-fasteners-python-fasteners-fasteners python27-six
>>>
>>>
>>> Libor
>>>
>>> Dne 07.04.2022 v 21:39 Libor Ukropec napsal(a):
>>>> Hi Brian,
>>>> Dne 07.04.2022 v 1:40 Brian Inglis napsal(a):
>>>>> On 2022-04-06 16:10, Libor Ukropec wrote:
>>>>>> I'd like to offer to adopt maintenance of duplicity (Encrypted 
>>>>>> bandwidth-efficient backup system)
>>>>>> Information from https://duplicity.gitlab.io/ - """The last stable 
>>>>>> 0.7 release is *0.7.19*, released Apr 19, 2019""", while cygwin 
>>>>>> contains 0.7.11 from 2017
>>>>>> Updated cygport:
>>>>>> https://github.com/cz6ace/cygwin-duplicity
>>>>>
>>>>> You need to define BUILD_REQUIRES and list all Cygwin packages 
>>>>> needed to build this package:
>>>>>
>>>>> https://cygwin.github.io/cygport/check_funcs_cygpart.html#robo791
>>>>>
>>>>> Use BUILD_REQUIRES+=" ..." for additional lines of packages.
>>>>>
>>>>>> Updated build:
>>>>>> https://github.com/cz6ace/cygwin-duplicity/releases
>>>>>
>>>>> See:
>>>>>
>>>>>      https://cygwin.com/git/cygwin-packages/duplicity.git
>>>>
>>>> This repository was my starting point, I just increased the version 
>>>> and prepared the package (below) for the new python dependency 
>>>> (fasterners). I did not see on the contribution page any mention to 
>>>> the `playground` thing and an automation - will try that, once my 
>>>> SSH key is added.
>>>>
>>>> As my first contribution to cygwin I wanted to start with small 
>>>> steps and stay with 0.7 duplicity, which still depends on the Python 
>>>> 2.7
>>>>
>>>>>
>>>>> You can clone the repo for the original files, checkout a 
>>>>> playground branch, commit your changes and patches (and any extra 
>>>>> source files), define the upstream playground branch, and push your 
>>>>> changes there, which will run Scallywag CI under Github Actions (or 
>>>>> Appveyor if you configure that cygport option).
>>>>>
>>>>>> Please note for successful installation the python 2.7 fasteners 
>>>>>> package is required, not yet in cygwin, I plan to offer [ITP] for it:
>>>>>> cygport:
>>>>>> https://github.com/cz6ace/cygwin-python-fasteners
>>>>>> build:
>>>>>> https://github.com/cz6ace/cygwin-python-fasteners/releases
>>>>>
>>>>> Need to support python3/39 now: see python package cygports in 
>>>>> cygwin-packages repos as above!
>>>>
>>>> Is it a must at this moment? As I stated above, duplicity 0.7.x 
>>>> requires Python 2.7
>>>>
>>>> I changed the `inherit` to python-wheel, which should support 
>>>> 2.7,3.8, 3.9:
>>>>
>>>> inherit python-wheel
>>>>
>>>> PYTHON_WHEEL_VERSIONS="2.7:3.8:3.9"
>>>>
>>>> But the cygport does not create the tar.cz archive for 2.7 and fails 
>>>> with
>>>>
>>>> ... <shortened the list>
>>>> -usr/lib/python2.7/site-packages/fasteners/version.py
>>>> ...
>>>> *** ERROR: Packages are missing files:
>>>>
>>>>
>>>> I tried previous cygport version, but it fails creating python 3.7 
>>>> package.
>>>>
>>>> I'll try to look to cygport sources if I am able to see something, 
>>>> which I doubt.
>>>>
>>>>>
>>>>>> Python standard library provides a lock for threads (both a 
>>>>>> reentrant one, and a non-reentrant one, see below). Fasteners 
>>>>>> extends this, and provides a lock for processes, as well as Reader 
>>>>>> Writer locks for both threads and processes.
>>>>>>
>>>>>> This library (in newer version) is also present in LInux 
>>>>>> distributions, e.g. Ubuntu:
>>>>>> https://packages.ubuntu.com/source/focal/s390x/python-fasteners
>>>>>
>>>>
>>>>
>>>
>>
>>
> 
> 


  reply	other threads:[~2022-04-12  0:09 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-04-06 22:10 Libor Ukropec
2022-04-06 23:40 ` Brian Inglis
2022-04-07 19:39   ` Libor Ukropec
2022-04-07 22:26     ` Libor Ukropec
2022-04-07 23:44       ` Brian Inglis
2022-04-10 18:26         ` Libor Ukropec
2022-04-12  0:09           ` Libor Ukropec [this message]
2022-04-12 13:10             ` Jon Turney
2022-04-12 19:24               ` Libor Ukropec
2022-04-12 16:56             ` Achim Gratz

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=480ef945-55f2-b00d-6dd7-7a204acf14e9@seznam.cz \
    --to=ace@seznam.cz \
    --cc=cygwin-apps@cygwin.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).