[OSM-dev] [OSM-talk] Ideas for OSM enhancements

[Brett Henderson]

Karl Newman wrote:
> No, I svn up like a nervous habit to keep the changes small, and
> really I haven't had any conflicts and very few merges (most of my
> changes have been isolated in developing the new Bounds class, but now
> I'm nearly done with incorporating it into the tasks. I just got
> nervous about some changes I was making and so I stopped to write some
> tests.
>
> I've been seeing the classnames that you've been committing--I think
> the Dataset will be a great help for writing a "tiling" task. One
> thing to think about--it might be nice to support writing out the
> Dataset (to a SQLite file or whatever) and then allow it to be reused
> on a future invocation, to avoid re-parsing a file. Maybe you're
> already planning to do this; I haven't looked closely.
>   
The dataset itself is just an interface so that differing 
implementations can be written.  I'm currently using a custom approach 
with raw data files and index files.  My intention is to allow the files 
to be written to a specific directory (as opposed to automatic temp 
files) so they can be re-used between invocations.  This should cover 
off your suggestion above.  It could take a while to build the dataset 
in the first place so allowing it to be re-used is likely to be important.

It should be fairly straightforward to write a SQLite implementation as 
well although I haven't looked into that.  Something like SQLite may 
have smarter indexing strategies than my naive implementation ...  It 
may also be possible to make a SQLite implementation that supports 
application of changesets which I've put in the too hard basket for now.

I have no idea how fast any of this will run, it is going to involve a 
lot of random disk seeking which may not perform well.

It's taken me a lot longer than I expected, but I'm getting closer :-)