| [10:06:25] | * cpm has joined #farmos |
| [10:58:18] | <paul121[m]> | > QGIS also supports something called "OGC API - features"... but I haven't been able to make very much sense of it yet; https://www.ogc.org/standards/ogcapi-features |
| [10:58:18] | <paul121[m]> | I thought OGC API - features == WFS 3.0 ?? But may be wrong! |
| [11:01:46] | <paul121[m]> | > Seems like WFS 2.0 makes more sense than 3.0 for now since QGIS doesn't support WFS 3.0 yet. |
| [11:01:46] | <paul121[m]> | I was able connect QGIS (latest version) to that pygeoapi app and read features. My understanding is that is running WFS 3.0. But perhaps the larger hurdle is that 3.0 does not define POST or transformation operations, *yet* |
| [11:19:00] | <symbioquine[m]> | ACTION sent a long message: < https://matrix.org/_matrix/media/r0/download/matrix.org/EDUzcbzHDdhcRDIA... > |
| [11:20:41] | <symbioquine[m]> | ACTION sent a long message: < https://matrix.org/_matrix/media/r0/download/matrix.org/xDiByNkvSvNihrsB... > |
| [11:22:08] | <symbioquine[m]> | It's less clear to me whether OpenLayers is following along with newer WFS versions... I only see explicit support for WFS 1.0.0/1.1.0; https://github.com/openlayers/openlayers/blob/v6.4.3/src/ol/format/WFS.j... |
| [11:22:34] | <symbioquine[m]> | Though I could be missing something there too... |
| [11:23:28] | <symbioquine[m]> | Maybe that OGC API stuff is supported magically because it uses GeoJSON? |
| [11:49:15] | <paul121[m]> | hmm interesting. Since it's GeoJSON its really just a matter of knowing the full URL and being able to parse it... |
| [11:50:17] | <paul121[m]> | What would OL "supporting" WFS 3.0 mean? You give a server URL and collection name, and it knows how to traverse the `/collections`? 🤔 |
| [11:50:33] | <paul121[m]> | * What would OL "supporting" WFS 3.0 mean? You give a server URL and collection name, and it knows how to traverse the `/collections` endpoint? 🤔 |
| [11:53:49] | <symbioquine[m]> | Hmmm, I see what you mean. It looks like OpenLayers is already making the application layer do most of the work of handling the "meta" work for WFS; https://openlayers.org/en/latest/examples/vector-wfs.html |
| [11:54:09] | <symbioquine[m]> | In other words it doesn't care about the GetCapabilities stuff |
| [11:54:32] | <symbioquine[m]> | It expects you to handle that yourself and just directly add the layers via their urls |
| [11:56:05] | <symbioquine[m]> | I think that also means that it won't be able to tell what spatial operations are supported or what kinds of transactions are allowed. All that has to be either hard-coded or extracted from the capabilities document by the application layer. |
| [11:58:42] | <paul121[m]> | right |
| [11:59:58] | <paul121[m]> | it would be cool to have a farmOS-map behavior that displays everything in a given collection though! |
| [12:00:49] | <paul121[m]> | showing all "field areas" is one, but also for more custom things like "drone imagery" |
| [12:38:13] | <symbioquine[m]> | I think you'd want WMS for drone imagery |
| [12:41:39] | <symbioquine[m]> | Unless you mean something like a point feature layer with a image reference for each point instead of having the drone imagery already stitched together into a tile layer... |
| [14:43:37] | * cpm has quit (Quit: Leaving) |
| [16:23:01] | <paul121[m]> | yes I agree (although that is an interesting solution!) |
| [16:24:29] | <paul121[m]> | although I was curious why pygeoapi had a raster plugin provider option... https://docs.pygeoapi.io/en/stable/plugins.html#example-custom-pygeoapi-... |
| [16:25:11] | <paul121[m]> | I took that to mean OGC API might encompass more than just features? a little confused tho |
| [16:25:25] | <paul121[m]> | but hey symbioquine a question re: your OAuth debugging |
| [16:26:11] | <paul121[m]> | could you report which versions of `league/oauth2-server` and `lcobucci/jwt` are installed? |
| [16:26:51] | <paul121[m]> | both of those got updated recently and were not pinned down by `simple_oauth` |
| [16:27:43] | <paul121[m]> | I think you might have been facing this issue: https://github.com/thephpleague/oauth2-server/issues/1161 |
| [16:29:58] | <paul121[m]> | the latest release assumes that all OAuth clients (and thus consumers w/ simple_oauth) will have a `redirectUri` set, but I don't think that is the case |
| [16:42:02] | <paul121[m]> | * the latest release assumes that all OAuth clients (and thus consumers w/ simple_oauth) will have a `redirectUri` set, but I don't think that *should always be the case |