use pydantic.conlist to define coordinate arrays

[geospatial-jeff]

What I am changing

Use pydantic.conlist to type coordinate arrays because it allows specifying min_items. The arguments to conlist appear in the auto-generated JSON schema which makes it a better representation of the geojson spec. Only LineString implemented this before this change, now all the geometry types have length constraints on their coordinate arrays.

I noticed this while using hypothesis-jsonschema to autogenerate fake data against the JSON schema created by geojson-pydantic. Hypothesis would generate coordinates arrays with invalid number of members (ex. a linear ring with one coordinate).

How I did it

Because the coordinates members for different geometries build off of eachother I decided to define types for each coordinates type in types.py and reference those types in the pydantic models.

How you can test it

Run pytest!

[codecov-commenter]

Codecov Report

Merging #44 (12cf99b) into master (1302fb5) will increase coverage by 0.05%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master      #44      +/-   ##
==========================================
+ Coverage   98.05%   98.11%   +0.05%     
==========================================
  Files           4        4              
  Lines         103      106       +3     
==========================================
+ Hits          101      104       +3     
  Misses          2        2              

Flag	Coverage Δ
unittests	98.11% <100.00%> (+0.05%)	⬆️

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

Impacted Files	Coverage Δ
geojson_pydantic/geometries.py	98.38% <100.00%> (-0.10%)	⬇️
geojson_pydantic/types.py	100.00% <100.00%> (ø)

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 1302fb5...12cf99b. Read the comment docs.

[vincentsarago]

🙏 thanks @geospatial-jeff