[ZOO-Discuss] ZOO-kernel web service requirements for Geopard / indoor mapping

Jonathan-David SCHRODER jonathan.schroder at gmail.com
Wed Apr 7 14:33:24 PDT 2010


Hello,

I am writing those requirements upon Gerald Fenoy and Nicolas Bozon's
proposal.
I apologize for having waited so long before sending this.

To all the list members, Geopard is an open source indoor geopositioning
project relying on wifi technology. The students project ended end of
january... now on hold is further described at
http://geoparden.wordpress.com/ .
I had a french skype talk with Gerald Fenoy and Nicolas Bozon in march
during which we could explain ZOO-side / Geopard-side what our projects are
about and how Geopard could make use of the ZOO-project with a request from
Gerald and Nicolas to tell our services requirements, now below.

I guess that Geopard needs (for now, quite frankly only A) !)
---------------IMPORTANT-------------
A) a transparent WPS service for the OSM v0.6 API (and later), that tells
http://url:port/ for this api (so it's like an "other API URL teller").
=> thus one can know from GetCapabilities calls to ZOO/OGC servers whether
they have an OSM r/w API and what's the url & port for it.

--------------NOT SO IMPORTANT-------------

B) a logging/result WPS service : reads logs & status/results from either
WPS jobs or regular unix jobs (eg. from /var/log/someFile) and displays them
at http://domain/logs/[LOGNUMBER or WPSJOBNUMBER]/{log and/or status}

=> example 1 (WPS job): 1) launch conversion osm2sqlite server dumping job,
2) watch for job status, 3) when job is over, download .sqlite file that's
been generated for us
=> example 2 (unix cron job) : the OSM stack makes a minutely rails port =>
PostGIS data transfer, it's good have the log for this handy over http

C) a generic long duration process WPS service template/framework :
        - input :
                    i) the long process method to use (eg. in our case
dxf2osm, buildingLevel2sqlite...)
                    ii) Parameters for the process : strings (eg.
"building2") , a public file local or remote path from 2) (eg. if local:
(hidden uploads path here) /mybuilding.dxf, if remote : eg.
http://domain/filepathToDownloadAndUse)
        - output :
                    i) a JOBNUMBER
                    ii) a B)-like job status message/log (eg.
http://domain/converterService/jobs/JOBNUMBER <= job number from i) that can
be looked up frequently with AJAX/simple HTTP calls

D) a service chainer WPS service : that helps ask for chained tasks
for example : 1) hello server, here is a tasks chain for you : "download
file from http://domain/myFile with DownloadService", ii) "on success pass
this file to dxf2osm"

E) a graphical web frontend that allows to manage (read info,
install/remove, tweak settings) for a ZOO server

Thanks for your attention and hard work on the ZOO-project,

Jonathan-David Schröder
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/zoo-discuss/attachments/20100407/c799aac7/attachment.html>


More information about the Zoo-discuss mailing list