Latest Contributions by JohnGravois

it makes sense to me, but i have absolutely no experience doing anything remotely like that...:) ... View more

1. pass the coordinate values in when instantiating a new polyline geometry object 2. pass the geometry to a new graphic (which also has a symbol defined) 3. add it to the map ... View more

if the only people that can access the app have already authenticated, why not just store your token in the proxy which is used to access the secure services and call it a day? you could use the app itself as the http referrer, since the app spawns the requests for secure services, or use the IP address of the web server which is hosting the proxy, because that is the physical machine which will always be making the requests. ... View more

in Flex and Silverlight, a proxy is only useful for working with secure services.  in JavaScript it is also necessary if you'd like to POST across domains.  no need to change anything in your map services.  just check out the conceptual article about proxies that i linked to previously. ... View more

its not necessary to declare something explicitly in code to ensure that a POST is used for a request, the API does this automatically when appropriate. the caveat is that posting across domains requires a proxy (unless both the browser and web server support CORS). do you have a valid proxy referenced in your application? ... View more

please don't hesitate to post a simplified sample, i'd be happy to take a look. 1. It appears that featureLayer perfers a feature service (which I do not have the ability to create yet), but can use a map service. Please correct me if this isn't right. they are identical, with the exception that feature services give you the opportunity to expose feature editing in your app (which you aren't currently doing). 2. I haven't seen a sample with a ArcGISTiledMapServiceLayer loaded before the featureLayer, so I am worried about some kind of conflict between the two. I need the tiled layer, though, because of custom zoom levels. its standard practice to load a tiled map service first.  (if the feature layer uses a different coordinate system it will preproject on the fly).  check out this sample.  (it loads our world streetmap as the tiled map service layer using the new technique of specifying the name of the basemap directly in the map constructor. I would really like to read some kind of extended description about capabilities and strategies with feature layers, but haven't found anything. check out this article ... View more

a 403 is telling you that your current proxy security settings don't allow you to forward traffic from that specific address.  while developing it is safe to set the global "mustMatch" tag in proxy.config to false so that any page is forwarded, but on a live public site, you would want to set the value back to false and ensure that all servers which need to be communicated with are listed explicitly. <ProxyConfig mustMatch="false"> <serverUrl url="http://TestServer:6080/arcgis/rest/services/"                matchAll="true"> </serverUrl> ... View more

i think the problem you're having is caused by the hierarchy of events. 1.  clicking on the map fires an identify task 2.  clicking on client side graphics (like the invisible building graphics) shows a popup without making another server call (because of event wiring for map.graphics) 3.  hovering over a building graphic also displays its popup. the reason why you can't fire an identify to populate the popup for a restroom when you click on one inside a building is because the graphic mouse event takes priority over the map click event. heres my recommendation for solving this problem.. load each layer people need to interact with as an ON_DEMAND or SNAPSHOT featureLayer).  This way you can let our API ensure that all relevant features (there aren't really a lot of them) are downloaded as client graphics automatically (you don't even have to worry about writing out a QueryTask. The benefit of this approach is that you can get control over the order in which the graphics are drawn in the map (by the order you add them in programmatically) to ensure that the point features draw "on top" of the building polygon features. This way clicking on the points will return information about the points, and you can skip the identify task entirely... ... View more

thanks for letting us know.  please use this feedback web map which was set up for customers to point out issues in the basemap.  Our world topo (and others) have feedback web maps as well. you can even use them to track the progress of fixes! ... View more

1.  our API doesn't know anything about the TOC widget, but you should definitely be able to instantiate an array of legendLayers and pass it to the print widget to populate the legend in your printout. 2.  i don't know of a way to remove credits from the printout (perhaps this is on purpose?) 3.  because the dimensions of the printout are typically different than your layout template, my understanding is that you have to choose whether to attempt to preserve scale or extent at the expense of the other.  are you seeing something else?  if so, can you post a sample that references public data? ... View more

please check out the "Referencing Objects and Properties in the esri Namespace" section of the article Migrating to 3.0. this version of our API (and subsequent versions) have more strict requirements when it comes to code that references the esri namespace which are called before all modules loaded. ... View more

i definitely think the JavaScript API expects the values of the system managed ObjectID field to be unique, because our software ensures that the field is present and can act as a primary key for publishing both data in flat files and SDE. i wish i had more insight or advice into what to do in your situation. ... View more