{"html_url": "https://github.com/simonw/datasette/issues/294#issuecomment-506801311", "issue_url": "https://api.github.com/repos/simonw/datasette/issues/294", "id": 506801311, "node_id": "MDEyOklzc3VlQ29tbWVudDUwNjgwMTMxMQ==", "user": {"value": 9599, "label": "simonw"}, "created_at": "2019-06-28T16:45:28Z", "updated_at": "2019-06-28T16:45:28Z", "author_association": "OWNER", "body": "This can happen as part of #531 ", "reactions": "{\"total_count\": 0, \"+1\": 0, \"-1\": 0, \"laugh\": 0, \"hooray\": 0, \"confused\": 0, \"heart\": 0, \"rocket\": 0, \"eyes\": 0}", "issue": {"value": 327365110, "label": "inspect should record column types"}, "performed_via_github_app": null} {"html_url": "https://github.com/simonw/datasette/issues/294#issuecomment-504882686", "issue_url": "https://api.github.com/repos/simonw/datasette/issues/294", "id": 504882686, "node_id": "MDEyOklzc3VlQ29tbWVudDUwNDg4MjY4Ng==", "user": {"value": 9599, "label": "simonw"}, "created_at": "2019-06-24T06:54:22Z", "updated_at": "2019-06-24T06:54:22Z", "author_association": "OWNER", "body": "Consider this when solving #465 ", "reactions": "{\"total_count\": 0, \"+1\": 0, \"-1\": 0, \"laugh\": 0, \"hooray\": 0, \"confused\": 0, \"heart\": 0, \"rocket\": 0, \"eyes\": 0}", "issue": {"value": 327365110, "label": "inspect should record column types"}, "performed_via_github_app": null} {"html_url": "https://github.com/simonw/datasette/issues/294#issuecomment-405026800", "issue_url": "https://api.github.com/repos/simonw/datasette/issues/294", "id": 405026800, "node_id": "MDEyOklzc3VlQ29tbWVudDQwNTAyNjgwMA==", "user": {"value": 45057, "label": "russss"}, "created_at": "2018-07-14T14:24:31Z", "updated_at": "2018-07-14T14:24:31Z", "author_association": "CONTRIBUTOR", "body": "I had a quick look at this in relation to #343 and I feel like it might be worth modelling the inspected table metadata internally as an object rather than a dict. (We'd still have to serialise it back to JSON.)\r\n\r\nThere are a few places where we rely on the structure of this metadata dict for various reasons, including in templates (and potentially also in user templates). It would be nice to have a reasonably well defined API for accessing metadata internally so that it's clearer what we're breaking.", "reactions": "{\"total_count\": 0, \"+1\": 0, \"-1\": 0, \"laugh\": 0, \"hooray\": 0, \"confused\": 0, \"heart\": 0, \"rocket\": 0, \"eyes\": 0}", "issue": {"value": 327365110, "label": "inspect should record column types"}, "performed_via_github_app": null} {"html_url": "https://github.com/simonw/datasette/issues/294#issuecomment-393557968", "issue_url": "https://api.github.com/repos/simonw/datasette/issues/294", "id": 393557968, "node_id": "MDEyOklzc3VlQ29tbWVudDM5MzU1Nzk2OA==", "user": {"value": 9599, "label": "simonw"}, "created_at": "2018-05-31T14:55:46Z", "updated_at": "2018-05-31T14:55:46Z", "author_association": "OWNER", "body": "I'm not sure what the best JSON shape for this would be considering the potential complexity of geospatial columns. I do think it's worth exposing these in the inspect JSON though, mainly so Datasette Registry can keep track of all of the openly available geodata out there.", "reactions": "{\"total_count\": 0, \"+1\": 0, \"-1\": 0, \"laugh\": 0, \"hooray\": 0, \"confused\": 0, \"heart\": 0, \"rocket\": 0, \"eyes\": 0}", "issue": {"value": 327365110, "label": "inspect should record column types"}, "performed_via_github_app": null} {"html_url": "https://github.com/simonw/datasette/issues/294#issuecomment-393549215", "issue_url": "https://api.github.com/repos/simonw/datasette/issues/294", "id": 393549215, "node_id": "MDEyOklzc3VlQ29tbWVudDM5MzU0OTIxNQ==", "user": {"value": 9599, "label": "simonw"}, "created_at": "2018-05-31T14:29:37Z", "updated_at": "2018-05-31T14:29:37Z", "author_association": "OWNER", "body": "Also of note: `spatialite-test` uses readable strings in the `type` column, while `timezones` has a `geometry_type` column with integers in it.\r\n\r\nThose integers are documented here: https://www.gaia-gis.it/fossil/libspatialite/wiki?name=switching-to-4.0\r\n\r\n![2018-05-31 at 7 29 am](https://user-images.githubusercontent.com/9599/40788210-5d0f0dd4-64a4-11e8-8141-0386b5c7b384.png)\r\n", "reactions": "{\"total_count\": 0, \"+1\": 0, \"-1\": 0, \"laugh\": 0, \"hooray\": 0, \"confused\": 0, \"heart\": 0, \"rocket\": 0, \"eyes\": 0}", "issue": {"value": 327365110, "label": "inspect should record column types"}, "performed_via_github_app": null} {"html_url": "https://github.com/simonw/datasette/issues/294#issuecomment-393548602", "issue_url": "https://api.github.com/repos/simonw/datasette/issues/294", "id": 393548602, "node_id": "MDEyOklzc3VlQ29tbWVudDM5MzU0ODYwMg==", "user": {"value": 9599, "label": "simonw"}, "created_at": "2018-05-31T14:27:41Z", "updated_at": "2018-05-31T14:27:56Z", "author_association": "OWNER", "body": "Presumably the difference in primary key structure between those two is caused by the fact that the `spatialite-test` database (actually https://www.gaia-gis.it/spatialite-2.3.1/test-2.3.sqlite.gz downloaded from https://www.gaia-gis.it/spatialite-2.3.1/resources.html ) was created by a much older version of SpatialLite - presumably v2.3.1", "reactions": "{\"total_count\": 0, \"+1\": 0, \"-1\": 0, \"laugh\": 0, \"hooray\": 0, \"confused\": 0, \"heart\": 0, \"rocket\": 0, \"eyes\": 0}", "issue": {"value": 327365110, "label": "inspect should record column types"}, "performed_via_github_app": null} {"html_url": "https://github.com/simonw/datasette/issues/294#issuecomment-393547960", "issue_url": "https://api.github.com/repos/simonw/datasette/issues/294", "id": 393547960, "node_id": "MDEyOklzc3VlQ29tbWVudDM5MzU0Nzk2MA==", "user": {"value": 9599, "label": "simonw"}, "created_at": "2018-05-31T14:25:43Z", "updated_at": "2018-05-31T14:25:43Z", "author_association": "OWNER", "body": "SpatialLite columns are actually quite a bit more interesting than this - they also have a `geometry_type` (point, polygon, linestring etc), a `coord_dimension` (usually 2 but can be higher) and an `srid`.\r\n\r\nFor example:\r\n\r\nhttps://datasette-publish-spatialite-demo.now.sh/spatialite-test-c88bc35/geometry_columns\r\n\r\n![2018-05-31 at 7 22 am](https://user-images.githubusercontent.com/9599/40787843-6f9600ee-64a3-11e8-84e5-64d7cc69603a.png)\r\n\r\nThe SRID here is particularly interesting, because it helps hint at the fact that the results from these queries won't be latitude/longitude co-ordinates - which means that `AsGeoJSON()` won't return results that can be easily rendered by Leaflet:\r\n\r\nhttps://datasette-publish-spatialite-demo.now.sh/spatialite-test-c88bc35?sql=select+AsGeoJSON(Geometry)+from+HighWays%20limit1\r\n\r\nCompare with https://timezones-api.now.sh/timezones-a99b2e3/geometry_columns:\r\n\r\n![2018-05-31 at 7 25 am](https://user-images.githubusercontent.com/9599/40787991-d2650756-64a3-11e8-936e-2dcce7dd1515.png)\r\n", "reactions": "{\"total_count\": 0, \"+1\": 0, \"-1\": 0, \"laugh\": 0, \"hooray\": 0, \"confused\": 0, \"heart\": 0, \"rocket\": 0, \"eyes\": 0}", "issue": {"value": 327365110, "label": "inspect should record column types"}, "performed_via_github_app": null}