Hi, I'm a frontend developer and was programming a web viewer for a proprietary, WMS-like map service. I discovered Openlayers one month ago and I'm really fascinated by its awesome capabilities, so I decided to integrate our map service in OpenLayers.
Here is a short description how the map service works: The client sends an AJAX request containing the current map bounds, the map id and the pixel size of the map. map=22&px=1720&py=833&left=2504400&right=2544000&top=5699500&bottom=5669500&type=0 The server responds with the generated map image (url) and the corrected bounds: <MapProperties> <mapLeft>2493227.6110444176</mapLeft> <mapTop>5699500</mapTop> <mapRight>2555172.3889555824</mapRight> <mapBottom>5669500</mapBottom> <mapType>0</mapType> <overviewMap>22</overviewMap> <url>http://localhost/cgmCache/41600065-c3d2-4934-9feb-78a32f6668c8.png </url> </MapProperties> The maps are always single-tiled. My first approach was to write a ProprietaryMapLayer, that executes the map request and creates an ImageLayer with the result url and its bounds. When the ImageLayer is moved, a new request is performed, the current ImageLayer is destroyed and a new one generated. The disadvantage: there are now 2 layers representing one map, which leads to confusion in the layer switcher control. I read over the code of Layer.WMS, which inherits from Layer.Grid and also makes use of Tile.Image, but it's difficult to derive a solution from it, since all the request/response handling is already integrated into the layer. Furthermore, WMS simply responds with tile images, whereas I have to deal with an xml response. Any ideas how I could implement my layer? kind regards, Franz
_______________________________________________ Users mailing list [email protected] http://openlayers.org/mailman/listinfo/users
