fix(maps): Move Overseas department and regions closer to France mainland

[qleroy]

feat(maps): add drom closer to mainland France

SUMMARY

The notebook repositions the overseas departments and regions of France (DROM) closer to mainland.
The file france.geojson is generated from the notebook.

BEFORE/AFTER SCREENSHOTS OR ANIMATED GIF

Before (only France metropolitan departments):

After (DROM included):

TESTING INSTRUCTIONS

Execute the notebook to observe that France DROM are effectively moved closer to mainland so that it can be display in a single map.

ADDITIONAL INFORMATION

• Has associated issue:
• 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

[qleroy]

The relevant bit of code because there is no diff for notebook on github

france_copy = france.copy()
reposition(france_copy, france.name=='Guadeloupe', 57.4, 25.4, 1.5, 1.5)
reposition(france_copy, france.name=='Martinique', 58.4, 27.1, 1.5, 1.5)
reposition(france_copy, france.name=='Guyane française', 52, 37.7, 0.35, 0.35)
reposition(france_copy, france.name=='La Réunion', -55, 62.8, 1.5, 1.5)
reposition(france_copy, france.name=='Mayotte', -43, 54.3, 1.5, 1.5)

[qleroy]

I also fixed two metropolitan names (errors natural earth data),
that were fixed by hand in the superset repo in #22999 so a re-run of the notebook would have overwritten the fix.

• Seien-et-Marne => Seine-et-Marne
• Haute-Rhin => Haut-Rhin

 def replace_name(df, old, new):
    if old in list(df.name):  
        index = df[df.name == old].index[0]
        df.at[index, 'name'] = new
replace_name(france, 'Seien-et-Marne', 'Seine-et-Marne')
replace_name(france, 'Haute-Rhin', 'Haut-Rhin')

[rusackas]

This is awesome! So happy we're fixing maps the right way now!

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (db79c37) 67.20% compared to head (866eb9a) 67.20%.
Report is 4 commits behind head on master.

Additional details and impacted files

@@           Coverage Diff           @@
##           master   #26995   +/-   ##
=======================================
  Coverage   67.20%   67.20%           
=======================================
  Files        1899     1899           
  Lines       74346    74346           
  Branches     8263     8263           
=======================================
  Hits        49963    49963           
  Misses      22334    22334           
  Partials     2049     2049           

Flag	Coverage Δ
javascript	56.93% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[qleroy]

With the geopandas dissolve function,
I can get the France regions from the France departments !,
bypassing the resort to external sources like IGN like it was done in #25676 and #26848

we would have france_regions.geojson generated by the notebook !

@rusackas do I include this in this PR or in a separate one ?

france_regions = france.copy()
# source : https://www.iso.org/obp/ui/#iso:code:3166:FR
dep_to_reg = {
"FR-01": "FR-ARA",
"FR-02": "FR-HDF",
"FR-03": "FR-ARA",
"FR-04": "FR-PAC",
"FR-05": "FR-PAC",
"FR-06": "FR-PAC",
"FR-07": "FR-ARA",
"FR-08": "FR-GES",
"FR-09": "FR-OCC",
"FR-10": "FR-GES",
"FR-11": "FR-OCC",
"FR-12": "FR-OCC",
"FR-13": "FR-PAC",
"FR-14": "FR-NOR",
"FR-15": "FR-ARA",
"FR-16": "FR-NAQ",
"FR-17": "FR-NAQ",
"FR-18": "FR-CVL",
"FR-19": "FR-NAQ",
"FR-21": "FR-BFC",
"FR-22": "FR-BRE",
"FR-23": "FR-NAQ",
"FR-24": "FR-NAQ",
"FR-25": "FR-BFC",
"FR-26": "FR-ARA",
"FR-27": "FR-NOR",
"FR-28": "FR-CVL",
"FR-29": "FR-BRE",
"FR-2A": "FR-20R",
"FR-2B": "FR-20R",
"FR-30": "FR-OCC",
"FR-31": "FR-OCC",
"FR-32": "FR-OCC",
"FR-33": "FR-NAQ",
"FR-34": "FR-OCC",
"FR-35": "FR-BRE",
"FR-36": "FR-CVL",
"FR-37": "FR-CVL",
"FR-38": "FR-ARA",
"FR-39": "FR-BFC",
"FR-40": "FR-NAQ",
"FR-41": "FR-CVL",
"FR-42": "FR-ARA",
"FR-43": "FR-ARA",
"FR-44": "FR-PDL",
"FR-45": "FR-CVL",
"FR-46": "FR-OCC",
"FR-47": "FR-NAQ",
"FR-48": "FR-OCC",
"FR-49": "FR-PDL",
"FR-50": "FR-NOR",
"FR-51": "FR-GES",
"FR-52": "FR-GES",
"FR-53": "FR-PDL",
"FR-54": "FR-GES",
"FR-55": "FR-GES",
"FR-56": "FR-BRE",
"FR-57": "FR-GES",
"FR-58": "FR-BFC",
"FR-59": "FR-HDF",
"FR-60": "FR-HDF",
"FR-61": "FR-NOR",
"FR-62": "FR-HDF",
"FR-63": "FR-ARA",
"FR-64": "FR-NAQ",
"FR-65": "FR-OCC",
"FR-66": "FR-OCC",
"FR-67": "FR-GES",
"FR-68": "FR-GES",
"FR-69": "FR-ARA",
"FR-70": "FR-BFC",
"FR-71": "FR-BFC",
"FR-72": "FR-PDL",
"FR-73": "FR-ARA",
"FR-74": "FR-ARA",
"FR-75": "FR-IDF",
"FR-76": "FR-NOR",
"FR-77": "FR-IDF",
"FR-78": "FR-IDF",
"FR-79": "FR-NAQ",
"FR-80": "FR-HDF",
"FR-81": "FR-OCC",
"FR-82": "FR-OCC",
"FR-83": "FR-PAC",
"FR-84": "FR-PAC",
"FR-85": "FR-PDL",
"FR-86": "FR-NAQ",
"FR-87": "FR-NAQ",
"FR-88": "FR-GES",
"FR-89": "FR-BFC",
"FR-90": "FR-BFC",
"FR-91": "FR-IDF",
"FR-92": "FR-IDF",
"FR-93": "FR-IDF",
"FR-94": "FR-IDF",
"FR-95": "FR-IDF",
"FR-GF": "FR-GF",
"FR-MQ": "FR-MQ",
"FR-GP": "FR-GP",
"FR-RE": "FR-RE",
"FR-YT": "FR-YT",
}
france_regions['iso_3166_2_fr'] = france_regions['iso_3166_2'].apply(lambda x: dep_to_reg[x])
reposition(france_regions, france.name=='Guadeloupe', 57.4, 25.4, 1.5, 1.5)
reposition(france_regions, france.name=='Martinique', 58.4, 27.1, 1.5, 1.5)
reposition(france_regions, france.name=='Guyane française', 52, 37.7, 0.35, 0.35)
reposition(france_regions, france.name=='La Réunion', -55, 62.8, 1.5, 1.5)
reposition(france_regions, france.name=='Mayotte', -43, 54.3, 1.5, 1.5)
france_regions = france_regions.dissolve(by='iso_3166_2_reg')
france_regions['iso_3166_2'] = france_regions['iso_3166_2'].index
france_regions.plot()

[mistercrunch]

LGTM

[rusackas]

@rusackas do I include this in this PR or in a separate one ?

A separate one would be ideal, I think :) Then after I see how you sorted that out, maybe one of us can do the same for the Italy Regions... then there will be no exceptions!

[rusackas]

Finally got a chance to pull and run the notebook... everything looks great!

[qleroy]

@rusackas I did the follow up here #27014

so we have only countries_custom/italy_regions.geojson remaining

[ThomasG77]

FYI, similar recipe using mapshaper (pure CLI approach) https://gist.github.com/ThomasG77/e20c52a39c581bab6a9e42e87d15b9e2 (in French, can translate if needed).