Re: [firstname.lastname@example.org: Re: DDTP-Links on CDD tasks web pages]
On Mon, Jul 21, 2008 at 11:26 AM, Andreas Tille <email@example.com> wrote:
> Well, OK, I have not yet looked at DDTSS code but implementing a
> diff old new and drop the issue if there is no diff does not sound as
> hard to implement. Martijn, could you please comment on this?
I haven't had much time in the last few days but the reason it is like
this is because DDTSS started as an interface that was completely
seperate. It has its own database for tracking translations. In the
beginning there was only the mail interface so everything was
asynchronous, you couldn't even request specific packages, because the
mail interface didn't support it.
People wanted that, so I wrote some code that scraped the DDTP web
interface and stored the result into the local database. Then you
could not tell the difference between one fetched via mail or one
fetched via scraping.
What happened at the last Extremadura is that the DDTSS got installed
onto the DDTP server and the web scraping and the mail sending
replaced by direct DB access. However it still has its own DB, and
this is what you are loading the translation into when you hit the
force option. There is no way to undo this action at present (though
perhaps there should be).
So what you're asking is to change the entire interface code so it can
work with a translation that does not exist in the local DB. OTOH,
there has to be some persistant state, otherwise how would it know if
you've changed something? You're welcome to try, but I don't think its
anywhere near as easy as you think it is. Allowing people to remove
pending translations would be less work.
Have a nice day,
Martijn van Oosterhout <firstname.lastname@example.org> http://svana.org/kleptog/