feat: add France's regions to country map visualization

[dmeaux]

SUMMARY

France's Regions have been added as a selection to the Country Map visualization. The geometries were taken from IGN's (France's National Geographic Institute) BD TOPO (R) administrative layer, and the ISO 3166-02 codes for the regions were added as an attribute.

BEFORE/AFTER SCREENSHOTS

France's Departments:

France's Regions:

TESTING INSTRUCTIONS

ADDITIONAL INFORMATION

• Has associated issue: feat: add France's regions to country map visualization #25673
• Required feature flags:
• Changes UI
• Includes DB Migration (follow approval process in SIP-59)
  • Migration is atomic, supports rollback & is backwards-compatible
  • Confirm DB migration upgrade and downgrade tested
  • Runtime estimates and downtime expectations provided
• Introduces new feature or API
• Removes existing feature or API

[villebro]

LGTM, thanks for adding the regional granularity!

[qleroy]

@dmeaux Does it work on your system ?

The file france_regions.geojson appear to be huge compared to the other ones (12 MB).

It does not display correctly

But it can confirm it displays correctly with matplotlib, using the Country Map GEOJSON Generator notebook.

What could be the problem ?

[dmeaux]

Yes, It does. The geometries are much more detailed than the other layers. David *************************************************************** DAVID MEAUX

…

________________________________ From: Quentin Leroy ***@***.***> Sent: Monday, January 29, 2024 10:03 To: apache/superset ***@***.***> Cc: David Meaux ***@***.***>; Mention ***@***.***> Subject: Re: [apache/superset] feat: add France's regions to country map visualization (PR #25676) @dmeaux<https://github.com/dmeaux> Does it work on your system ? The file france_regions.geojson appear to be huge compared to the other ones (12 MB). It does not display correctly france.regions.broken.png (view on web)<https://github.com/apache/superset/assets/9881262/8158b978-9e10-4cc1-a46f-a548a3a9d50b> But it can confirm it displays correctly with matplotlib, using the Country Map GEOJSON Generator notebook. What could be the problem ? — Reply to this email directly, view it on GitHub<#25676 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ALUBDUEXWTXWI37RRA3DGSTYQ5QVRAVCNFSM6AAAAAA6DR4HRSVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSMJUGI2DQNJSGI>. You are receiving this because you were mentioned.Message ID: ***@***.***>

[qleroy]

@dmeaux

I think the geojson file expects latitude (range -90 to +90) and longitude (-180 to +180) values,

but the coordinates in france_regions.json are in the range 1e+6.

excerpt

{ "type": "Feature", "...n", "coordinates": [ [ [ [ 654467.0, 6522631.600000000558794 ], [ 654463.90...

[dmeaux]

@qleroy
Yeah, usually geojson is in WGS84 with longitude and latitude as the X/Y coordinates, but you can embed projected data as well.

Check out this line at the top of the file:
`"crs": { "type": "name", "properties": { "name": "urn:ogc:def:crs:EPSG::2154" } }
This refers to the Coordinate Reference System, which in this case is EPSG:2154, or Lambert 93.

I will reproject the data to WGS 84 and simplify the geometries to decrease the file size and then recommit the file. This should fix the issues you're seeing.

[qleroy]

thank you @dmeaux

maybe d3 does not take the coordinates system into account I don't know.

I was considering including also DOM/TOM with cartiflette
https://observablehq.com/@linogaliana/cartiflette-demo?collection=@linogaliana/cartiflette

[qleroy]

I double checked,
indeed france_regions.geojson is the only file exceeding the bounds

According to d3 docs and referring to the plugin CountryMap.js, all *.geojson files are expected to be projected with the default projection.

All geojson files are in the urn:ogc:def:crs:OGC:1.3:CRS84 projection except for france_regions.geojson !

So that is why it cannot by displayed properly in Superset with the plugin code as is.

for f in $(ls superset/superset-frontend/plugins/legacy-chart-country-map/src/countries)
do
echo $f
jq '.crs' $f
done

france_regions.json urn:ogc:def:crs:OGC:1.3:CRS84

{
  "type": "name",
  "properties": {
    "name": "urn:ogc:def:crs:EPSG::2154"
  }
}

...

france.geojson
{
  "type": "name",
  "properties": {
    "name": "urn:ogc:def:crs:OGC:1.3:CRS84"
  }
}
...

import geopandas as gpd

df = gpd.read_file('superset-frontend/plugins/legacy-chart-country-map/src/countries/france_regions.geojson')
print(df.bounds)

france_regionsgeojson
         minx       miny       maxx       maxy
0    428145.7  6137101.2   848021.3  6439668.0
1    476864.7  6584220.1   709599.9  6872271.0
2    342953.8  6788980.8   612865.2  6998209.9
3    311617.1  6193024.1   669901.0  6679340.2
4    728354.5  6710828.6  1082838.3  7009414.0
5    799585.9  6214487.5  1077719.3  6453457.6
6    626126.3  6338318.9  1027292.6  6633794.6
7    688333.0  6562601.2  1011638.9  6811153.5
8    276123.7  6582709.1   545050.0  6834664.4
9    583916.5  6859783.5   790288.4  7110478.8
10    99040.0  6704161.1   400665.5  6886001.0
11  1156227.0  6046546.3  1242440.3  6235448.1
12   586421.7  6780020.5   741205.6  6905411.1

[dmeaux]

I've created a pull request with the updated geojson file. Until it gets merged you can use the new GeoJSON file.

[xavier-GitHub76]

Hello,

I use superset 3.1.1 but I have the same issue :

Best resgards

[xavier-GitHub76]

It's OK with 4.0.2 version 👍