[Amavisd-new-debian-devel] the debian l10n template feature

Harald Jenny harald at a-little-linux-box.at
Fri Sep 9 09:23:28 UTC 2011


Hi Alex

On Fri, Sep 09, 2011 at 10:32:20AM +0200, Alexander Wirt wrote:
> Harald Jenny schrieb am Donnerstag, den 08. September 2011:
> 
> > Hi Alex
> > 
> > On Tue, Aug 09, 2011 at 10:06:51PM +0200, Alexander Wirt wrote:
> > > Harald Jenny schrieb am Tuesday, den 09. August 2011:
> > > 
> > > > On Tue, Aug 09, 2011 at 09:46:57PM +0200, Alexander Wirt wrote:
> > > > > Henrique de Moraes Holschuh schrieb am Tuesday, den 09. August 2011:
> > > > > 
> > > > > > On Tue, 09 Aug 2011, Alexander Wirt wrote:
> > > > > > > during the work on 2.7.0 I again stumbled about the feature of
> > > > > > > /etc/amavis/<lang>/. I would have to provide a patch for 2.7.0 for getting
> > > > > > > the feature working again, as there are several new templates in the code.
> > > > > > > Additionally I have to say that I dislike that diverging from upstream thingy
> > > > > > > as it is error proven.
> > > > > > 
> > > > > > We implemented it before upstream did, I think.  That caused the divergence.
> > > > > > 
> > > > > > > I am not sure if somebody is really using that "feature", so I would opt in
> > > > > > > removing it. Any oppinions here?
> > > > > > 
> > > > > > I used to ship pt_BR translations in the package, and our l10n team could
> > > > > > translate the templates if we asked them to and made them aware of the
> > > > > > templates (since they're not gettext/autopoint stuff).  The lang/ way of
> > > > > > doing things used to make it MUCH easier to handle l10n updates.
> > > > > > 
> > > > > > But I certainly don't object to switching to whatever upstream does in
> > > > > > 2.7.0, as long as it doesn't make it impossible to deal with when you
> > > > > > actually use it or distribute l10n files.
> > > > > > 
> > > > > > l10n'd notifications are a killer feture.
> > > > > unfortunatly there is no real solution inside amavisd. Just the l10n snippet
> > > > > from debian. Marc added several templates over the last years but he didn't
> > > > > updated that l10n snippet so not every of the template can be accessed via
> > > > > the l10n version.
> > > > > 
> > > > > So it seems I have to deal with that patch, update it and give marc a patch.
> > > > 
> > > > As our company also uses the l10 feature I would be more that willing to help
> > > > you with this work - how can I help you?
> > > Get extract-upstream-en_US-templates.pl (should be done) in sync with the
> > > template in __DATA__
> > 
> > Done - capitalization of spam changed, pushed the change to master repository.
> > 
> > > and sub read_l10n_templates($;$) from amavisd.
> > 
> > You mean adding the new template files to read_l10n_templates or what exactly?
> > 
> > > All
> > > templates should get extracted with useful names
> > 
> > They seem not too bad:
> > template-auto-response.txt
> > template-dsn.txt
> > template-problem-feedback.txt
> > template-release-quarantine.txt
> > template-spam-admin.txt
> > template-spam-sender.txt
> > template-virus-admin.txt
> > template-virus-recipient.txt
> > template-virus-sender.txt
> > 
> > > and amavisd should catch
> > > them up in read_l10n_templates.
> > 
> > Attached a patch proposal, is this sufficiant?
> > 
> > > If this is done create a patch a forward it
> > > to the amavis mailinglist.
> > 
> > Can you confirm that what I've done is correct and complete?
> Narf. I haven't reviewed this yet. But please don't commit into master if is
> not your package and you want a review.

My commit just extends your work on the extract-upstream-en_US-templates.pl
script and applies the results from it to the two old template-spam-* files,
IMHO there is no real need to review these changes as there basically update
some wordings in the two files but if you disagree theses changes can be
reverted. The patch I attached was produced in a local temporary respository.
But if you feel more comfortable I can in the future just send patches in to
allow for previous review by others.

> 
> Alex

Kind regards
Harald

> -- 
> Alexander Wirt, formorer at formorer.de 
> CC99 2DDD D39E 75B0 B0AA  B25C D35B BC99 BC7D 020A



More information about the Amavisd-new-debian-devel mailing list