Re: help needed to complete vcmi packaging refresh (reprotest failure)

2025-02-07 Thread Patrice Duroux
HI, I'm facing this issue on another package (lix[1]). Any advice to resolve this issue would be greatly appreciated with probably my contribution to an addition with a section regarding the reproducibility to https://wiki.debian.org/AutomaticDebugPackages Many thanks, Patrice [1] https://salsa

help needed to complete vcmi packaging refresh (reprotest failure)

2025-01-29 Thread Patrice Duroux
Hi, My work is available here: https://salsa.debian.org/Peutch-guest/vcmi and my point is about its Salsa CI reprotest failure. It seems to be due to the following diff. I'm too junior a packager to solve this problem. Best, Patrice $ diffoscope control/source-root/vcmi_1.6.3+dfsg-1+salsaci+202

newby and refreshing the testdisk packaging

2024-09-04 Thread Patrice Duroux
Hi, After a quick private exchange with Chris Hofstaedtler following the recent update to the testdisk package, he suggested me doing a NMU and so I started to work on some other improvements to its packaging. Here is attached a diff of all my suggestions. I did a test build using sbuild. Could i

help needed about Salsa pipelines

2023-05-22 Thread Patrice Duroux
Hi, Trying to maintain pingus package using Salsa, I faced two failures with the Salsa pipelines. The first is related to test-crossbuild-arm64 and the message is: Reading package lists... Building dependency tree... Reading state information... Some packages could not be installed. This may mea

Re: Re: concern with pdebuild --pbuilder cowbuilder

2019-08-13 Thread Patrice Duroux
Ok, you mean before creating the X.debian.tar.xz? Because before running pdebuild, uscan was used to provide the X.orig.tar.xz and in the folder where I run pdebuild there is only the debian/* files (without any extra file). I was thinking that pdebuild is somehow a way to build packages from a "

concern with pdebuild --pbuilder cowbuilder

2019-08-13 Thread Patrice Duroux
Hi, It seems that pdebuild somehow calls 'debian/rules' before going to the pbuilder environment and so it fails as some build-deps are required to run this 'debian/rules' but not available in the call environment. In my case it is 'debian/rules' that includes some .mk files that are provided by

Re: Re: a proposal to update the abcm2ps package

2018-12-18 Thread Patrice Duroux
Hi, Another point is that would it be better to separate /debian from the upstream sources, I mean that providing only in each debian branches the content of /debian directory and moreover having like 5 or 6 (experimental) branches for something like the following: wheezy (o-o-stable): 6.6.17-1 j

Re: Re: a proposal to update the abcm2ps package

2018-12-15 Thread Patrice Duroux
So I started to progress on that. Currently I think master=debian and now there is an upstream. For pristine-tar, I don't know yet. Also I have restarted the project using 'gbp import-dscs --debsnap' to keep more traces of the past. Thanks!

a proposal to update the abcm2ps package

2018-12-14 Thread Patrice Duroux
Hi, This is more an exercise to teach how to deal with Debian packaging and with Salsa. I am not yet in contact with its DM, Anselm Lingnau, and neither with the upstream authors. But being now more familiar with git and git-buildpackage (thanks Guido!), I have started to work on updating the abc