[PATCH] dpkg class sbuild allows extra arguments by vars

[PATCH] dpkg class sbuild allows extra arguments by vars

[Roberto A. Foglietta]

[-- Attachment #1: Type: text/plain, Size: 467 bytes --]

Hi all,

Sometimes it is necessary to add some extra commands or arguments to sbuild
but create a class into an upper layer just for this breaks the updates.
So, this patch allows setting extra parameters through two variables.

Signed-off-by: Roberto A. Foglietta <roberto.foglietta@gmail.com>
---
 meta/classes/dpkg.bbclass | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

https://github.com/ilbers/isar/commit/7e72921747aedf588af492be95fb0b91ad70f455

[-- Attachment #2: 0001-dpkg-class-sbuild-allows-extra-arguments-by-vars.patch --]
[-- Type: text/x-patch, Size: 2336 bytes --]

From 7e72921747aedf588af492be95fb0b91ad70f455 Mon Sep 17 00:00:00 2001
From: "Roberto A. Foglietta" <roberto.foglietta@gmail.com>
Date: Tue, 3 Jan 2023 20:53:00 +0100
Subject: [PATCH] dpkg class sbuild allows extra arguments by vars

Sometimes it is necessary to add some extra commands or arguments to sbuild
but create a class into an upper layer just for this breaks the updates.
So, this patch allows setting extra parameters through two variables.

Signed-off-by: Roberto A. Foglietta <roberto.foglietta@gmail.com>
---
 meta/classes/dpkg.bbclass | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/meta/classes/dpkg.bbclass b/meta/classes/dpkg.bbclass
index 7822b14..bf60a5b 100644
--- a/meta/classes/dpkg.bbclass
+++ b/meta/classes/dpkg.bbclass
@@ -23,6 +23,9 @@ do_prepare_build_append() {
     env > ${DPKG_PREBUILD_ENV_FILE}
 }
 
+DPKG_SBUILD_EXTRA_ARGS_PRE ?= ""
+DPKG_SBUILD_EXTRA_ARGS_POST ?= ""
+
 # Build package from sources using build script
 dpkg_runbuild[vardepsexclude] += "${SBUILD_PASSTHROUGH_ADDITIONS}"
 dpkg_runbuild() {
@@ -104,7 +107,7 @@ dpkg_runbuild() {
     sbuild -A -n -c ${SBUILD_CHROOT} --extra-repository="${ISAR_APT_REPO}" \
         --host=${PACKAGE_ARCH} --build=${SBUILD_HOST_ARCH} ${profiles} \
         --no-run-lintian --no-run-piuparts --no-run-autopkgtest --resolve-alternatives \
-        --no-apt-update \
+        --no-apt-update ${DPKG_SBUILD_EXTRA_ARGS_PRE} \
         --chroot-setup-commands="echo \"Package: *\nPin: release n=${DEBDISTRONAME}\nPin-Priority: 1000\" > /etc/apt/preferences.d/isar-apt" \
         --chroot-setup-commands="echo \"APT::Get::allow-downgrades 1;\" > /etc/apt/apt.conf.d/50isar-apt" \
         --chroot-setup-commands="rm -f /var/log/dpkg.log" \
@@ -112,7 +115,7 @@ dpkg_runbuild() {
         --finished-build-commands="rm -f ${deb_dir}/sbuild-build-depends-main-dummy_*.deb" \
         --finished-build-commands="cp -n --no-preserve=owner ${deb_dir}/*.deb -t ${ext_deb_dir}/ || :" \
         --finished-build-commands="cp /var/log/dpkg.log ${ext_root}/dpkg_partial.log" \
-        --debbuildopts="--source-option=-I" \
+        --debbuildopts="--source-option=-I" ${DPKG_SBUILD_EXTRA_ARGS_POST} \
         --build-dir=${WORKDIR} --dist="isar" ${DSC_FILE}
 
     sbuild_dpkg_log_export "${WORKDIR}/rootfs/dpkg_partial.log"
-- 
2.34.1

Re: [PATCH] dpkg class sbuild allows extra arguments by vars

[Henning Schild]

Again can you please send you patches in a way that they can actually
be reviewed in a mail based workflow.

You have conducted reviews yourself already so you should know how
things should work.

Am Tue, 3 Jan 2023 20:58:15 +0100
schrieb "Roberto A. Foglietta" <roberto.foglietta@gmail.com>:

> Hi all,
> 
> Sometimes it is necessary to add some extra commands or arguments to
> sbuild but create a class into an upper layer just for this breaks
> the updates. So, this patch allows setting extra parameters through
> two variables.
> 
> Signed-off-by: Roberto A. Foglietta <roberto.foglietta@gmail.com>
> ---
>  meta/classes/dpkg.bbclass | 7 +++++--
>  1 file changed, 5 insertions(+), 2 deletions(-)
> 
> https://github.com/ilbers/isar/commit/7e72921747aedf588af492be95fb0b91ad70f455

I can not reply to that nor to a specific line in the attachment. So
this can not be reviewed ... and therefore not approved.

I would have comments on this one that would likely require you to send
a v2. But i will just say NACK, because you deny the review again after
being asked countless times to send your patches in a proper way.

Henning

Re: [PATCH] dpkg class sbuild allows extra arguments by vars

[Roberto A. Foglietta]

On Tue, 3 Jan 2023 at 21:26, Henning Schild <henning.schild@siemens.com> wrote:
>
> Again can you please send you patches in a way that they can actually
> be reviewed in a mail based workflow.

I send the patches with git-send but they went in moderation and
usually remain there.
Tomorrow, I will search for another open SMTP server hoping that the
m-list will accept it but I am desperate to find one that suits the
stringent anti-spam rules of google.

> You have conducted reviews yourself already so you should know how
> things should work.

...and everytime, I send a patch I always let people here remember
about the problem.
For compensation, I started to add this link

> >
> > https://github.com/ilbers/isar/commit/7e72921747aedf588af492be95fb0b91ad70f455
>

Best regards, R-

Re: [PATCH] dpkg class sbuild allows extra arguments by vars

[Henning Schild]

Am Tue, 3 Jan 2023 22:39:55 +0100
schrieb "Roberto A. Foglietta" <roberto.foglietta@gmail.com>:

> On Tue, 3 Jan 2023 at 21:26, Henning Schild
> <henning.schild@siemens.com> wrote:
> >
> > Again can you please send you patches in a way that they can
> > actually be reviewed in a mail based workflow.  
> 
> I send the patches with git-send but they went in moderation and
> usually remain there.
> Tomorrow, I will search for another open SMTP server hoping that the
> m-list will accept it but I am desperate to find one that suits the
> stringent anti-spam rules of google.

Maybe we start with your problem here and help you use the SMTP servers
from google. My guess is that you might be struggling with 2FA. And
doing an internet search on how to use plain SMTP on gmail might also
give you the solution.

Using any other SMTP server you will likely get problems with SPF,
DKIM, DMARC forcing you to send emails "From" the domain of that other
server. In which case git-send-email would put a secondary "From" into
the commit message so that git-am on the other side will get the real
author email.

But best try to use googles SMTP servers, that is for sure possible and
not too hard once you found the solution.

Henning

> > You have conducted reviews yourself already so you should know how
> > things should work.  
> 
> ...and everytime, I send a patch I always let people here remember
> about the problem.
> For compensation, I started to add this link
> 
> > >
> > > https://github.com/ilbers/isar/commit/7e72921747aedf588af492be95fb0b91ad70f455
> > >  
> >  
> 
> Best regards, R-

Re: [PATCH] dpkg class sbuild allows extra arguments by vars

[Roberto A. Foglietta]

On Wed, 4 Jan 2023 at 08:54, Henning Schild <henning.schild@siemens.com> wrote:
>
> Am Tue, 3 Jan 2023 22:39:55 +0100
> schrieb "Roberto A. Foglietta" <roberto.foglietta@gmail.com>:
>
> > On Tue, 3 Jan 2023 at 21:26, Henning Schild
> > <henning.schild@siemens.com> wrote:
> > >
> > > Again can you please send you patches in a way that they can
> > > actually be reviewed in a mail based workflow.
> >
> > I send the patches with git-send but they went in moderation and
> > usually remain there.
> > Tomorrow, I will search for another open SMTP server hoping that the
> > m-list will accept it but I am desperate to find one that suits the
> > stringent anti-spam rules of google.
>
> Maybe we start with your problem here and help you use the SMTP servers
> from google. My guess is that you might be struggling with 2FA. And
> doing an internet search on how to use plain SMTP on gmail might also
> give you the solution.
>

It is not possible to authenticate on SMTP with 2FA. In the past there
was the option to create an app credential to send e-mail but this
option has been removed. So, all the how-to like this would simply not
work anymore

https://blog.revathskumar.com/2019/08/git-send-patch-using-send-email-and-gmail.html

> Using any other SMTP server you will likely get problems with SPF,
> DKIM, DMARC forcing you to send emails "From" the domain of that other
> server. In which case git-send-email would put a secondary "From" into
> the commit message so that git-am on the other side will get the real
> author email.

In fact, Google Groups intercept such a e-mail and tag it as SPAM both
in the web application (on the reader side) and in the m-list incoming
rules.

>
> But best try to use googles SMTP servers, that is for sure possible and
> not too hard once you found the solution.
>

Finally, I used a regular SMTP from an internet provider. However, It
costs me a domain name fee.

Best regards, R-

Re: [PATCH] dpkg class sbuild allows extra arguments by vars

[Henning Schild]

Am Thu, 5 Jan 2023 18:43:40 +0100
schrieb "Roberto A. Foglietta" <roberto.foglietta@gmail.com>:

> On Wed, 4 Jan 2023 at 08:54, Henning Schild
> <henning.schild@siemens.com> wrote:
> >
> > Am Tue, 3 Jan 2023 22:39:55 +0100
> > schrieb "Roberto A. Foglietta" <roberto.foglietta@gmail.com>:
> >  
> > > On Tue, 3 Jan 2023 at 21:26, Henning Schild
> > > <henning.schild@siemens.com> wrote:  
> > > >
> > > > Again can you please send you patches in a way that they can
> > > > actually be reviewed in a mail based workflow.  
> > >
> > > I send the patches with git-send but they went in moderation and
> > > usually remain there.
> > > Tomorrow, I will search for another open SMTP server hoping that
> > > the m-list will accept it but I am desperate to find one that
> > > suits the stringent anti-spam rules of google.  
> >
> > Maybe we start with your problem here and help you use the SMTP
> > servers from google. My guess is that you might be struggling with
> > 2FA. And doing an internet search on how to use plain SMTP on gmail
> > might also give you the solution.
> >  
> 
> It is not possible to authenticate on SMTP with 2FA. In the past there
> was the option to create an app credential to send e-mail but this
> option has been removed. So, all the how-to like this would simply not
> work anymore
> 
> https://blog.revathskumar.com/2019/08/git-send-patch-using-send-email-and-gmail.html
> 
> > Using any other SMTP server you will likely get problems with SPF,
> > DKIM, DMARC forcing you to send emails "From" the domain of that
> > other server. In which case git-send-email would put a secondary
> > "From" into the commit message so that git-am on the other side
> > will get the real author email.  
> 
> In fact, Google Groups intercept such a e-mail and tag it as SPAM both
> in the web application (on the reader side) and in the m-list incoming
> rules.
> 
> >
> > But best try to use googles SMTP servers, that is for sure possible
> > and not too hard once you found the solution.
> >  
> 
> Finally, I used a regular SMTP from an internet provider. However, It
> costs me a domain name fee.

you could try msmtp and google again, if application credentials do not
work anymore ... you need to dance the oauth2 dance ...

Or you stay with a provider that does not enforce so much security on
you that in the end you can not send mail from the campfire in the cave
;)

Henning

> Best regards, R-

Re: [PATCH] dpkg class sbuild allows extra arguments by vars

[Roberto A. Foglietta]

On Thu, 5 Jan 2023 at 19:31, Henning Schild <henning.schild@siemens.com> wrote:
>
> Am Thu, 5 Jan 2023 18:43:40 +0100
> schrieb "Roberto A. Foglietta" <roberto.foglietta@gmail.com>:
>
> > On Wed, 4 Jan 2023 at 08:54, Henning Schild
> > <henning.schild@siemens.com> wrote:
> > >
> > > Am Tue, 3 Jan 2023 22:39:55 +0100
> > > schrieb "Roberto A. Foglietta" <roberto.foglietta@gmail.com>:
> > >
> > > > On Tue, 3 Jan 2023 at 21:26, Henning Schild
> > > > <henning.schild@siemens.com> wrote:
> > > > >
> > > > > Again can you please send you patches in a way that they can
> > > > > actually be reviewed in a mail based workflow.
> > > >
> > > > I send the patches with git-send but they went in moderation and
> > > > usually remain there.
> > > > Tomorrow, I will search for another open SMTP server hoping that
> > > > the m-list will accept it but I am desperate to find one that
> > > > suits the stringent anti-spam rules of google.
> > >
> > > Maybe we start with your problem here and help you use the SMTP
> > > servers from google. My guess is that you might be struggling with
> > > 2FA. And doing an internet search on how to use plain SMTP on gmail
> > > might also give you the solution.
> > >
> >
> > It is not possible to authenticate on SMTP with 2FA. In the past there
> > was the option to create an app credential to send e-mail but this
> > option has been removed. So, all the how-to like this would simply not
> > work anymore
> >
> > https://blog.revathskumar.com/2019/08/git-send-patch-using-send-email-and-gmail.html
> >
> > > Using any other SMTP server you will likely get problems with SPF,
> > > DKIM, DMARC forcing you to send emails "From" the domain of that
> > > other server. In which case git-send-email would put a secondary
> > > "From" into the commit message so that git-am on the other side
> > > will get the real author email.
> >
> > In fact, Google Groups intercept such a e-mail and tag it as SPAM both
> > in the web application (on the reader side) and in the m-list incoming
> > rules.
> >
> > >
> > > But best try to use googles SMTP servers, that is for sure possible
> > > and not too hard once you found the solution.
> > >
> >
> > Finally, I used a regular SMTP from an internet provider. However, It
> > costs me a domain name fee.
>
> you could try msmtp and google again, if application credentials do not
> work anymore ... you need to dance the oauth2 dance ...
>
> Or you stay with a provider that does not enforce so much security on
> you that in the end you can not send mail from the campfire in the cave
> ;)
>

The problem has been solved as you can see here:

Message ID<20230104185604.2542129-1-roberto.foglietta@gmail.com>
Created on:4 January 2023 at 19:56 (Delivered after 5 seconds)
From:roberto.foglietta@gmail.com Using git-send-email 2.34.1
To:isar-users@googlegroups.com
Subject:[PATCH] rootfs class lock undefined in corner cases fix
DKIM:'PASS' with domain googlegroups.com Learn more

It was a very simple problem to solve but it required a smart internet
provider that gives domain names for a reasonable price and allows
users to do magic with the SMTP mangling rules.

The key information is "what is the name of that internet provider?"

If you take a look at that e-mail, can you discover it? LOL

Best regards, R-

Re: [PATCH] dpkg class sbuild allows extra arguments by vars

[Uladzimir Bely]

In mail from Tuesday, 3 January 2023 22:58:15 +03 user Roberto A. Foglietta 
wrote:
> Hi all,
> 
> Sometimes it is necessary to add some extra commands or arguments to sbuild
> but create a class into an upper layer just for this breaks the updates.
> So, this patch allows setting extra parameters through two variables.
> 
> Signed-off-by: Roberto A. Foglietta <roberto.foglietta@gmail.com>
> ---
>  meta/classes/dpkg.bbclass | 7 +++++--
>  1 file changed, 5 insertions(+), 2 deletions(-)
> 
> https://github.com/ilbers/isar/commit/7e72921747aedf588af492be95fb0b91ad70f4
> 55

Can't respond on specific line in the patch due to attachment form, but it 
would be better not to modify existing lines but just add new ones, like:

         --no-apt-update \
+        ${DPKG_SBUILD_EXTRA_ARGS_PRE} \
         --chroot-setup-commands=...

It's a bit more readable.

At first glance I wrongly tried to understand how it's related to "--no-apt-
update" and "--debbuildopts"... :)

You could send v2 or I can modify it on my side while testing / merging.

Re: [PATCH] dpkg class sbuild allows extra arguments by vars

[Roberto A. Foglietta]

On Wed, 25 Jan 2023 at 13:27, Uladzimir Bely <ubely@ilbers.de> wrote:
>
> In mail from Tuesday, 3 January 2023 22:58:15 +03 user Roberto A. Foglietta
> wrote:
> > Hi all,
> >
> > Sometimes it is necessary to add some extra commands or arguments to sbuild
> > but create a class into an upper layer just for this breaks the updates.
> > So, this patch allows setting extra parameters through two variables.
> >
> > Signed-off-by: Roberto A. Foglietta <roberto.foglietta@gmail.com>
> > ---
> >  meta/classes/dpkg.bbclass | 7 +++++--
> >  1 file changed, 5 insertions(+), 2 deletions(-)
> >
> > https://github.com/ilbers/isar/commit/7e72921747aedf588af492be95fb0b91ad70f4
> > 55
>
> Can't respond on specific line in the patch due to attachment form, but it
> would be better not to modify existing lines but just add new ones, like:
>
>          --no-apt-update \
> +        ${DPKG_SBUILD_EXTRA_ARGS_PRE} \
>          --chroot-setup-commands=...
>
> It's a bit more readable.

Two variables are useless, I realised. One in the middle is the
optimal compromise. Remove the _PRE in append, it does not make sense
anymore.

> You could send v2 or I can modify it on my side while testing / merging.

I am doing, just give me the time to rebase it on the next.

Best, R-