On Thu, 24 Apr 2008 08:27:23 +0100 Tom Hughes <[EMAIL PROTECTED]> wrote:
> In message <[EMAIL PROTECTED]> > Gora Mohanty <[EMAIL PROTECTED]> wrote: [...] > > I am attaching some snippets from a document on using Unicode with > > RoR that I wrote up for my own benefit while trying to set up a Rails > > site with a mysql backend to handle Indian language input. Hopefully > > it will be of some help. > > Most of what is listed in that document is long since done. The > database holds unicode and can return it just fine. Great. Most of that was pretty elementary stuff, though it must have taken me at least a couple days to work through all this initially. I was hoping that it might help Arindam get started on a local copy. However, as per your other message, I agree that making a new internationalisation branch is the way to go. > Only the second bit (using gettext) has not been done. > > The real question to my mind is whether gettext is the best solution > to translating the web site, or whether there is something better. > The main advantage of gettext is probably that there will more tools > for working with that than with anything else, and it looks like it > is reasonably well integrated. I am quite strongly in favour of gettext mainly because of the reason that you cite above, and also because it is well-tested, and does a good job of separating the roles of the programmer and the translator. > The disadvantage is that I suspect it will make quite a mess of our > views - the only example I can see on the cited web page of a view > being translated is for a very simple one with little translatable > text. [...] Hmm, I will have to actually try this out with a small Rails application before commenting further. Let me also try getting a local copy of the OSM site up and running over this weekend. Regards, Gora _______________________________________________ dev mailing list [email protected] http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/dev
