[Featureserver] Featureserver caching

[Christopher Schmidt]

On Fri, Oct 05, 2007 at 06:46:56PM +0100, david creane wrote:
> yes thanks Christopher that did work, just one question. 
> 
> Is there any way of caching the featureserver data that is sent to the 
> browser, as in even the xml that is sent to the browser, could it be cached 
> by squid, or some other from of caching. Or is that nonsense seeing as 
> because someone can just keep panning to the left or right, that youd need an 
> infinite number of cached files, unless you have a small number of panned 
> step sizes if you know what i mean. Im just wondering if theres a way that i 
> can be done, of is there even a point.

Given the current way the requests work in OpenLayers, this probably
wouldn't make sense. For the next release, there was some interest
expressed at the code sprint in 'tiled WFS' -- which would cause this to
make more sense -- but no work has been done on that front yet.

> On a side note: This issue was annoying me for a bit and in case it comes up 
> again. When the sql statement is generated from 
> sql = "SELECT AsBinary(%s) as fs_binary_geom_col, * FROM \"%s\" WHERE %s = 
> %%(%s)d" % (self.geom_col, self.table, self.fid_col, self.fid_col )
> in PostGIS.py file, line 146. If i was to ask for a table called osi.roads  
> the sql ends up as 

crschmidt=# create table osi.roads (id int);
ERROR:  schema "osi" does not exist

I don't know how I'd create a table like that. Do you have an 'osi'
schema or something?

Regards,
-- 
Christopher Schmidt
MetaCarta