[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: installation-guide is marked for autoremoval from testing

Am Sonntag, 13. Mai 2018 schrieb Cyril Brulebois:
> Heya,
> Holger Wansing <linux@wansing-online.de> (2018-05-13):
> > However, there is much more work to do here, "svn" and "alioth" is
> > often mentioned in the manual.
> That's fair. But maybe we should track this as a different RC bug (just
> to make extra sure this doesn't go unnoticed/unfixed until the release),
> which could be closed later on?

> > And:
> > maybe the workflow of xml-based translations is affected by git migration?
> > Tracking the up-to-date-status of those files is handled with svn revision
> > numbers, which are no longer available now IMHO.
> > How to deal with this?
> > Is it worse to overwork the relevant scripts, to make them work with git
> > hashes, or should we just drop xml-based translations altogether now?
> > Most translations have switched to po already, from the up-to-date
> > languages only Czech is left on xml.
> I don't have stats handy, but if there's only Czech around, I think time
> would be better spent moving it from xml to po?

Yes, that's my impression too.

> I suppose this is a
> transition that had been underway for a while and was never completed?

Yes, maybe.

> I'm not sure whether we should work on porting any other xml-based
> translations if there's nobody active on them?

Generally correct.
eu and pt_br could stay as they are now, for possible future use.
> > > > However, that would be an upload for Buster (there's "Bump release
> > > > name to buster" in the changelog).
> > > 
> > > I'm not sure why that would be an issue? The manual documents the buster
> > > installation process (which shouldn't have changed too much?), but the
> > > upload targets unstable as usual?
> > 
> > I remembered some trouble with the debian-refcard, where the
> > publication on the debian.org website was affected, because the
> > website directly uses the latest uploaded package version.
> I think we've had fixes on the website generation framework to handle
> that problem exactly. If we spot issues (again), we should work on
> fixing them, rather than being limited in our upload capacity on the
> installation guide side, I think?

Yes, right.
> > But the installation-guide is handled differently here apparently, so
> > that's most probably not a problem here...
> We'll see. :)

Ok  :-)
So let's go for an upload.
Samuel, maybe?



Sent from my Jolla phone

Reply to: