| [23:18:37] | * kunigunde[m] has quit (Ping timeout: 265 seconds) |
| [23:18:37] | * RevCornWallaceII has quit (Ping timeout: 265 seconds) |
| [23:18:37] | * MatrixTravelerb4 has quit (Ping timeout: 265 seconds) |
| [23:19:06] | * DanielAshcraft[m has quit (Ping timeout: 265 seconds) |
| [23:25:49] | * MatrixTravelerb4 has joined #farmos |
| [23:35:02] | * kunigunde[m] has joined #farmos |
| [23:35:12] | * RevCornWallaceII has joined #farmos |
| [23:39:50] | * DanielAshcraft[m has joined #farmos |
| [02:11:17] | * farmBOT has joined #farmos |
| [11:29:35] | <symbioquine[m]> | Where does this validation occur in farmOS 1.x; |
| [11:29:49] | <symbioquine[m]> | ACTION uploaded an image: (6KiB) < https://libera.ems.host/_matrix/media/r0/download/matrix.org/KcUiwRHeocJ... > |
| [11:30:20] | <mstenta[m]> | ooh 1.x? really testing me here... :-) |
| [11:30:27] | <symbioquine[m]> | :) |
| [11:31:11] | <mstenta[m]> | geofield module |
| [11:31:46] | <mstenta[m]> | `grep -irn 'specified location data is invalid' profiles/farm/modules/contrib/geofield` |
| [11:32:22] | <mstenta[m]> | `function geofield_field_validate()` |
| [11:33:14] | <mstenta[m]> | https://git.drupalcode.org/project/geofield/-/blob/7.x-2.x/geofield.modu... |
| [11:34:55] | <symbioquine[m]> | I see it now... |
| [11:35:11] | <symbioquine[m]> | I was looking in the 7.x-1.x branch before and couldn't find it :( |
| [11:35:16] | <symbioquine[m]> | Thanks! |
| [12:21:44] | <symbioquine[m]> | I figured it out... |
| [12:21:45] | <symbioquine[m]> | The problem was that [this code](https://git.drupalcode.org/project/geofield/-/blob/1a84143568c7f56632156...) doesn't surface the actual geometry problem so if - for instance - you paste a rectangle that isn't closed because it only has 4 points specified rather than having the 5th, a user can't see the error about an unclosed polygon ring. |
| [12:22:07] | <symbioquine[m]> | I haven't figured out whether this will be a problem in 2.x yet... |
| [12:22:24] | <mstenta[m]> | Ah - I think something similar came up semi-recently... |
| [12:22:30] | <mstenta[m]> | Specifically regarding closed polygons |
| [12:23:01] | <mstenta[m]> | IIRC it was something to do with whether or not GEOS was installed |
| [12:23:02] | <symbioquine[m]> | Yeah, I think OpenLayers is okay with leaving off that last closing point. |
| [12:23:32] | <mstenta[m]> | With GEOS it would flag unclosed polygons... but without it it passed validation... maybe? |
| [12:26:01] | <symbioquine[m]> | ACTION uploaded an image: (25KiB) < https://libera.ems.host/_matrix/media/r0/download/matrix.org/RfvEQVSuELT... > |
| [12:26:10] | <symbioquine[m]> | That's in 2.x |
| [12:26:22] | <mstenta[m]> | w/ or w/o GEOS? |
| [12:26:30] | <mstenta[m]> | (if using the Docker image, then it has GEOS) |
| [12:28:44] | <symbioquine[m]> | Yeah, I'm using a recent 2.x-dev image |
| [12:30:32] | <paul121[m]> | this is something we contributed to geofield for 2.x: https://www.drupal.org/project/geofield/issues/3220860#comment-14150434 |
| [12:30:57] | <paul121[m]> | just fwiw! |
| [12:35:01] | <symbioquine[m]> | It looks like the 2.x version of geofield is swallowing the exceptions somewhere... |
| [12:35:28] | <symbioquine[m]> | At least with the 1.x version it looked kinda trivial to surface the actual geometry error message. |
| [12:36:19] | <paul121[m]> | ohh I see. You want it to say "the polygon is not closed" - not just display the geom? |
| [13:05:46] | <symbioquine[m]> | Yeah 😄 |
