[georss] WAS: GeoRSS Validation Service? RETURNINGTO:multiplelocations and time
Carl Reed OGC Account
creed at opengeospatial.org
Fri Mar 9 16:12:37 EST 2007
Hi -
A bit of clarification as the referenced email is quite old. There is now an
GML OASIS Application Schema that is being used as the basis for all
go-forward encodings of location for OASIS specifications in both the EM and
CIQ Technical Committees. This profile uses the exact same encoding for
polygon as does the new GML PIDF Shape Application Schema used in the IETF
for encoding of non-binary location payloads and both are identical to the
georss encoding. Notice that for expression of coordinates, white space is
always used.
Carl
===============================================
GeoRSS
<georss:where>
<gml:Polygon>
<gml:exterior>
<gml:LinearRing>
<gml:posList>
45.256 -110.45 46.46 -109.48 43.84 -109.86 45.256 -110.45
</gml:posList>
</gml:LinearRing>
</gml:exterior>
</gml:Polygon>
</georss:where>
OASIS - Default CRS is WGS-84 2d, but the application schema supports
expressing the CRS as defined in GML 3.1.1. OASIS members wanted to be
compatible with GeoRSS so they determined that they wanted to use the
"where" tag.
<oasis:where> <gml:Polygon> <gml:exterior>
<gml:LinearRing> <gml:posList> 45.256 -110.45
46.46 -109.48 43.84 -109.86 45.256 -110.45 </gml:posList>
</gml:LinearRing> </gml:exterior> </gml:Polygon>
</oasis:where>
And the IETF encoding for use in a variety of internet standards. They also
have an example that uses poslist.
<gml:Polygon srsName="urn:ogc:def:crs:EPSG::4326"
xmlns:gml="http://www.opengis.net/gml"> <gml:exterior>
<gml:LinearRing> <gml:pos>42.556844 -73.248157</gml:pos>
<gml:pos>42.549631 -73.237283</gml:pos>
<gml:pos>42.539087 -73.240328</gml:pos>
<gml:pos>42.535756 -73.254242</gml:pos>
<gml:pos>42.542969 -73.265115</gml:pos>
<gml:pos>42.553513 -73.262075</gml:pos>
<gml:pos>42.556844 -73.248157</gml:pos> </gml:LinearRing>
</gml:exterior> </gml:Polygon>
----- Original Message -----
From: "Peter Borissow" <peter.borissow at yahoo.com>
To: "Ron Lake" <rlake at galdosinc.com>; "Raj Singh" <raj at rajsingh.org>
Cc: "Allan Doyle" <adoyle at eogeo.org>; <georss at lists.eogeo.org>
Sent: Friday, March 09, 2007 1:52 PM
Subject: Re: [georss] WAS: GeoRSS Validation Service?
RETURNINGTO:multiplelocations and time
> Thanks Ron-
> Coordinate encoding is indeed one of the primary differences I've seen
> between various gml representations of a given geometry (e.g. gml:Point).
> But there are others. For example, in following link Dr Reed describes 2
> different ways to define a polygon:
>
> http://www.oasis-open.org/archives/emergency-gis/200507/msg00003.html
>
> The coordinate encoding is identical in these 2 examples. I don't think
> this variation is attributable to spec differences.
>
> It would be nice to come up with a running list of gml examples that cover
> all of the variations of gml geometries across all 3(?) versions of the
> gml spec. Perhaps there's one out there already - I just haven't seen one
> yet. I'd love to expand the list I have so far to include additional gml
> geometries, 3d geometries, and gml time stamps (e.g. a GML point that's
> represented using x,y,z,t). Would you be willing to help compile such a
> document?
>
> By the way, I believe the code I posted earlier covers all of the
> coordinate encoding schemes the OGC has come up with so far. However, as
> stated on the download page, it only supports 2d tuples at the moment.
>
> Thanks again,
> Peter
>
>
>
> ----- Original Message ----
> From: Ron Lake <rlake at galdosinc.com>
> To: Peter Borissow <peter.borissow at yahoo.com>; Raj Singh
> <raj at rajsingh.org>
> Cc: Allan Doyle <adoyle at eogeo.org>; georss at lists.eogeo.org
> Sent: Friday, March 9, 2007 3:22:44 PM
> Subject: RE: [georss] WAS: GeoRSS Validation Service? RETURNING
> TO:multiplelocations and time
>
>
> Hi,
>
> Let me address the issue of coordinate encoding which I think is what
> you are referring to by GML variations.
>
> The original GML encoding (v1.0, V2.1.2) used <gml:coordinates> on
> Points, LineStrings etc. This was string valued with attributes that
> could be used to specify coordinate and n-tuple separators. Defaults
> were comma and whitespace respectively. This is the same as used in KML.
> There was another variant that used explicit x,y tags but this was
> fortunately deprecated and then deleted.
>
> GML v3.1.1 introduced the use of not string values but XML token lists
> so that an additional string parser was not required, and so that all
> coordinate values were visible in XML. This required the introduction
> of <gml:pos> and <gml:posList> elements. This is also somewhat more
> compact than is the case for the deprecated coordinates.
>
> Note that <gml:coordinates> is dropped (the XML content model will still
> exist) in GML v3.2 so only pos and posList will be supported.
>
> I hope that this helps.
>
> R
>
> -----Original Message-----
> From: Peter Borissow [mailto:peter.borissow at yahoo.com]
> Sent: March 9, 2007 11:41 AM
> To: Ron Lake; Raj Singh
> Cc: Allan Doyle; georss at lists.eogeo.org
> Subject: Re: [georss] WAS: GeoRSS Validation Service? RETURNING
> TO:multiplelocations and time
>
> I believe I used the term "variations". I'm by no means an expert on gml
> - I haven't spent the time to actually understand the differences
> between the differest versions. All I know is what I've seen - and what
> I've seen is a lot of different ways to express a point, line, polygon,
> etc. Perhaps these differences are a result of different versions of the
> spec - I just don't know.
>
> What I've tried to do in the atom feed is capture some of the different
> ways people describe a geometry using gml. I'd love to create a complete
> set of examples for all the different flavors of gml - something people
> can use to test thier parsers. I guess it's possible to automatically
> generate these examples using schemas - I just haven't spent the time to
> do something like that. In the meantime, I've employed a rather lazy
> approach - I've simply been grabbing examples off the web.
>
> Does that answer your question?
>
> Thanks,
> Peter
>
>
>
> ----- Original Message ----
> From: Ron Lake <rlake at galdosinc.com>
> To: Peter Borissow <peter.borissow at yahoo.com>; Raj Singh
> <raj at rajsingh.org>
> Cc: Allan Doyle <adoyle at eogeo.org>; georss at lists.eogeo.org
> Sent: Friday, March 9, 2007 2:14:17 PM
> Subject: RE: [georss] WAS: GeoRSS Validation Service? RETURNING
> TO:multiplelocations and time
>
>
> Hi Peter:
>
> What do you mean by variations of GML? Do you mean different versions
> (2.1.2, 3.1.1)? Otherwise I don't understand what you mean?
>
> Ron
>
> -----Original Message-----
> From: georss-bounces at lists.eogeo.org
> [mailto:georss-bounces at lists.eogeo.org] On Behalf Of Peter Borissow
> Sent: March 9, 2007 10:01 AM
> To: Peter Borissow; Raj Singh
> Cc: Allan Doyle; georss at lists.eogeo.org
> Subject: Re: [georss] WAS: GeoRSS Validation Service? RETURNING
> TO:multiplelocations and time
>
> Well I finally got a chance to knock out some code this morning - a java
> library to parse GeoRSS.
>
> You can download it here:
>
> http://www.kartographia.com/geoRSS/
>
>
> Again, my hope is that we can start generating and sharing some
> reference software to parse GeoRSS.
>
> I haven't had a lot of time to test this library. I can definately use
> some additional test data if you have any to share. I've created an atom
> feed to try to capture all of the variations of GML - even those not yet
> supported by GeoRSS.
>
> http://www.kartographia.com/geoRSS/examples/Atom.xml
>
>
> My hope is that once we have some reference software, we can start
> introducing more complex GML - like multipoint!
>
>
> Thanks,
> Peter
>
>
>
>
>
> ----- Original Message ----
> From: Peter Borissow <peter.borissow at yahoo.com>
> To: Raj Singh <raj at rajsingh.org>
> Cc: Pat Cappelaere <pat at cappelaere.com>; Allan Doyle <adoyle at eogeo.org>;
> georss at lists.eogeo.org
> Sent: Friday, March 2, 2007 5:13:53 PM
> Subject: Re: [georss] WAS: GeoRSS Validation Service? RETURNING TO:
> multiplelocations and time
>
>
> So that's a good point - the existing GeoRSS "spec" provides the option
> to specify coordinates in ANY reference frame! And you think multipoint
> geometries are going to scare users?
>
> Just kidding.
>
> Anyway I'll try to get a partial implementation of a java-based GeoRSS
> parser out sometime this weekend or early next week for you to review. I
> already wrote one a while back - I just need to clean it up. It supports
> the following features:
>
> - RSS, Atom, and RDF
> - GeoRSS "Simple" Geometries
> - A subset of GML Geometries (Point, Line, Polygon, MultiPoint,
> MultiLine, MultiPolygon, Envelope)
> - Supports the 3 different GML coordinate formats
> - It also happens to support DB2 and Sybase/SQS geometry types but I'll
> cut that code out
>
> All the code does is "normalize" the geometries into a set of points and
> extracts the srs attribute. Coordinate transformation, rendering,
> mensuration, etc. are all higher level functions that should be handled
> by other classes.
>
> So why do I feel the need to do this? Because I'm still trying to
> convince you to adopt a wider range of gml geometries into the GeoRSS
> spec.
>
>
> Have a good weekend.
>
> Peter
>
>
>
> ----- Original Message ----
> From: Raj Singh <raj at rajsingh.org>
> To: Peter Borissow <peter.borissow at yahoo.com>
> Cc: Pat Cappelaere <pat at cappelaere.com>; Allan Doyle <adoyle at eogeo.org>;
> georss at lists.eogeo.org
> Sent: Friday, March 2, 2007 3:57:55 PM
> Subject: Re: [georss] WAS: GeoRSS Validation Service? RETURNING TO:
> multiplelocations and time
>
>
> Parsing is one thing. Doing something smart with it is another. For
> example, if you get a GeoRSS GML geometry in, for example, a state
> plane coordinate system, can you either:
> a. reproject it into lat/long to show on a Google/Yahoo/MS map or
> b. reproject your map into the state plane system
>
> That's just one example of the more general point that if you support
> GeoRSS GML, you should be able to manipulate coordinate reference
> systems in your application.
> ---
> Raj
>
>
> On Mar 2, 2007, at 2:53 PM, Peter Borissow wrote:
>
>> Parsing RSS and Atom is a joke. Parsing GML geometries is not that
>> complicated either. In fact I wrote a limited GML parser that can
>> handle points, lines, and polygons + their "multi-" varients in
>> less than an hour. Adding the GeoRSS "simple" geometries took even
>> less time.
>
>
>
> ________________________________________________________________________
> ____________
> It's here! Your new message!
> Get new email alerts with the free Yahoo! Toolbar.
> http://tools.search.yahoo.com/toolbar/features/mail/
>
>
>
> ________________________________________________________________________
> ____________
> We won't tell. Get more on shows you hate to love
> (and love to hate): Yahoo! TV's Guilty Pleasures list.
> http://tv.yahoo.com/collections/265
> _______________________________________________
> georss mailing list
> georss at lists.eogeo.org
> http://lists.eogeo.org/mailman/listinfo/georss
>
>
>
> ________________________________________________________________________
> ____________
> 8:00? 8:25? 8:40? Find a flick in no time
> with the Yahoo! Search movie showtime shortcut.
> http://tools.search.yahoo.com/shortcuts/#news
>
>
>
> ____________________________________________________________________________________
> Bored stiff? Loosen up...
> Download and play hundreds of games for free on Yahoo! Games.
> http://games.yahoo.com/games/front
> _______________________________________________
> georss mailing list
> georss at lists.eogeo.org
> http://lists.eogeo.org/mailman/listinfo/georss
More information about the georss
mailing list