On Wed, 2009-04-29 at 16:58 +1000, Rodd Clarkson wrote:
On Tue, 2009-04-28 at 19:47 -0400, Matthias Clasen wrote:
> On Wed, 2009-04-29 at 01:11 +0200, Kevin Kofler wrote:
> > Matthias Clasen wrote:
> > > Previously, translations were merged by intltool from .po files into
> > > schemas files and then copied by gconftool from the schemas file into
> > > the database.
> >
> > Is that what caused the insanely-sized .xml files in libgweather? Or is that
> > a separate issue?
>
> No, the xml files in libgweather are just big on their own, no GConf
> involved.
I'm guessing this has to do with all the international data in
libgweather.
No, it's due to immense amounts of redundant data, (all the data is
duplicated for *every* language, none of it is shared) and quite a bit
of XML overhead.
Is it possible to build this (like we do for
openoffice.org with
it's
language packs) so that you only grab the locations you need? And
assuming your interested in the weather in some location your not, could
you automate the downloading of the location required?
Since it requires a net connection to get the weather information
*anyway*, this would not be unreasonable.
Still, as I pointed out before, lzma compressing the 76mb of data brings
it down to 1.5mb. It's read-only data, there is simply no reason for it
to be any bigger than that.