Importando GeoJSON com um array de objetos altamente aninhado

“Importando GeoJSON com um array de objetos altamente aninhado“

​

Pergunta

Como importar GeoJSON com um array de objetos aninhado?

​

Resposta

Neste tutorial, usaremos dados abertos disponíveis publicamente aqui. Uma cópia pode ser encontrada aqui.

1. Baixe os dados no formato GeoJSON e renomeie o arquivo para geojson.json.

2. Entenda a estrutura.

DESCRIBE TABLE file('geojson.json', 'JSON')
┌─name─────┬─type─────────────────────────────────────────────────────────────────────────────────────────┐
│ type     │ Nullable(String)                                                                             │     
│ name     │ Nullable(String)                                                                             │     
│ crs      │ Tuple( properties Tuple(name Nullable(String)),type Nullable(String))                        │     
│ features │ Array(Tuple(                                                                                 │  
│          │                  geometry Tuple(coordinates Array(Array(Array(Array(Nullable(Float64))))),   │  
│          │                  type Nullable(String)),                                                     │  
│          │                  properties Tuple(   CODIGOINE Nullable(String),                             │  
│          │                                      CODNUT1 Nullable(String),                               │  
│          │                                      CODNUT2 Nullable(String),                               │  
│          │                                      CODNUT3 Nullable(String),                               │  
│          │                                      FID Nullable(Int64),                                    │  
│          │                                      INSPIREID Nullable(String),                             │  
│          │                                      NAMEUNIT Nullable(String),                              │ 
│          │                                      NATCODE Nullable(String),                               │  
│          │                                      SHAPE_Area Nullable(Float64),                           │  
│          │                                      SHAPE_Length Nullable(Float64)                          │  
│          │                                  ),                                                          │  
│          │                  type Nullable(String)                                                       │  
│          │              )                                                                               │  
│          │      )                                                                                       │  
└──────────┴──────────────────────────────────────────────────────────────────────────────────────────────┘ 

3. Crie uma tabela para armazenar as linhas de GeoJSON.

O requisito aqui é gerar uma linha para cada objeto no array features. O tipo de dado inferido para o campo geometry sugere que ele corresponde ao MultiPolygon, tipo de dado do ClickHouse.

create table geojson 
(
    type String,
    name String,
    crsType String,
    crsName String,
    featureType String,
    id Int64,
    inspiredId String,
    natCode String,
    nameUnit String,
    codNut1 String,
    codNut2 String,
    codNut3 String,
    codigoIne String,
    shapeLength Float64,
    shapeArea Float64,
    geometryType String,
    geometry MultiPolygon
)
engine = MergeTree
order by id;

4. Prepare os dados.

O principal objetivo da consulta é verificar se obtemos uma linha para cada objeto no array features.

O campo features.geometry.coordinates foi comentado para tornar o conjunto de resultados mais legível.

SELECT
    type AS type,
    name AS name,
    crs.type AS crsType,
    crs.properties.name AS crsName,
    features.type AS featureType,
    features.properties.FID AS id,
    features.properties.INSPIREID AS inspiredId,
    features.properties.NATCODE AS natCode,
    features.properties.NAMEUNIT AS nameUnit,
    features.properties.CODNUT1 AS codNut1,
    features.properties.CODNUT2 AS codNut2,
    features.properties.CODNUT3 AS codNut3,
    features.properties.CODIGOINE AS codigoIne,
    features.properties.SHAPE_Length AS shapeLength,
    features.properties.SHAPE_Area AS shapeArea,
    features.geometry.type AS geometryType
    --,features.geometry.coordinates
FROM file('municipios_ign.geojson', 'JSON')
ARRAY JOIN features
LIMIT 5

┌─type──────────────┬─name───────────┬─crsType─┬─crsName───────────────────────┬─featureType─┬─id─┬─inspiredId───────────────┬─natCode─────┬─nameUnit──────────────┬─codNut1─┬─codNut2─┬─codNut3─┬─codigoIne─┬────────shapeLength─┬─────────────shapeArea─┬─geometryType─┐
│ FeatureCollection │ Municipios_IGN │ name    │ urn:ogc:def:crs:OGC:1.3:CRS84 │ Feature     │  1 │ ES.IGN.SIGLIM34081616266 │ 34081616266 │ Villarejo-Periesteban │ ES4     │ ES42    │ ES423   │ 16266     │ 0.2697476997304121 │ 0.0035198414406406673 │ MultiPolygon │
│ FeatureCollection │ Municipios_IGN │ name    │ urn:ogc:def:crs:OGC:1.3:CRS84 │ Feature     │  2 │ ES.IGN.SIGLIM34081616269 │ 34081616269 │ Villares del Saz      │ ES4     │ ES42    │ ES423   │ 16269     │ 0.4476083901269905 │   0.00738179315030249 │ MultiPolygon │
│ FeatureCollection │ Municipios_IGN │ name    │ urn:ogc:def:crs:OGC:1.3:CRS84 │ Feature     │  3 │ ES.IGN.SIGLIM34081616270 │ 34081616270 │ Villarrubio           │ ES4     │ ES42    │ ES423   │ 16270     │ 0.3053942273994179 │ 0.0029777582813496337 │ MultiPolygon │
│ FeatureCollection │ Municipios_IGN │ name    │ urn:ogc:def:crs:OGC:1.3:CRS84 │ Feature     │  4 │ ES.IGN.SIGLIM34081616271 │ 34081616271 │ Villarta              │ ES4     │ ES42    │ ES423   │ 16271     │ 0.2831226979821184 │  0.002680273189024594 │ MultiPolygon │
│ FeatureCollection │ Municipios_IGN │ name    │ urn:ogc:def:crs:OGC:1.3:CRS84 │ Feature     │  5 │ ES.IGN.SIGLIM34081616272 │ 34081616272 │ Villas de la Ventosa  │ ES4     │ ES42    │ ES423   │ 16272     │ 0.5958276749246777 │  0.015354885085133583 │ MultiPolygon │
└───────────────────┴────────────────┴─────────┴───────────────────────────────┴─────────────┴────┴──────────────────────────┴─────────────┴───────────────────────┴─────────┴─────────┴─────────┴───────────┴────────────────────┴───────────────────────┴──────────────┘

5. Insira os dados.

INSERT INTO geojson
SELECT
    type AS type,
    name AS name,
    crs.type AS crsType,
    crs.properties.name AS crsName,
    features.type AS featureType,
    features.properties.FID AS id,
    features.properties.INSPIREID AS inspiredId,
    features.properties.NATCODE AS natCode,
    features.properties.NAMEUNIT AS nameUnit,
    features.properties.CODNUT1 AS codNut1,
    features.properties.CODNUT2 AS codNut2,
    features.properties.CODNUT3 AS codNut3,
    features.properties.CODIGOINE AS codigoIne,
    features.properties.SHAPE_Length AS shapeLength,
    features.properties.SHAPE_Area AS shapeArea,
    features.geometry.type AS geometryType,
    features.geometry.coordinates as geometry
FROM file('municipios_ign.geojson', 'JSON')
ARRAY JOIN features

Aqui, obtemos o seguinte erro:

Code: 53. DB::Exception: Received from localhost:9000. DB::Exception: ARRAY JOIN requires array or map argument. (TYPE_MISMATCH)
Received exception from server (version 24.1.2):

Isso é causado pelo parsing de features.geometry.coordinates.

6. Vamos verificar seu tipo de dado.

SELECT DISTINCT toTypeName(features.geometry.coordinates) AS geometry
FROM file('municipios_ign.geojson', 'JSON')
ARRAY JOIN features

┌─geometry──────────────────────────────────────┐
│ Array(Array(Array(Array(Nullable(Float64))))) │
└───────────────────────────────────────────────┘

Isso pode ser corrigido convertendo multipolygon.properties.coordinates em Array(Array(Array(Tuple(Float64,Float64)))). Para isso, podemos usar a função arrayMap(func,arr1,…).

SELECT distinct
    toTypeName(
        arrayMap(features.geometry.coordinates->
                    arrayMap(features.geometry.coordinates-> 
                            arrayMap(features.geometry.coordinates-> (features.geometry.coordinates[1],features.geometry.coordinates[2]) 
                    ,features.geometry.coordinates),
                features.geometry.coordinates),
        features.geometry.coordinates)
    ) as toTypeName
FROM file('municipios_ign.geojson', 'JSON')
ARRAY JOIN features;

┌─toTypeName───────────────────────────────────────────────────────┐
│ Array(Array(Array(Tuple(Nullable(Float64), Nullable(Float64))))) │
└──────────────────────────────────────────────────────────────────┘

7. Insira os dados.

INSERT INTO geojson
SELECT
    type as type,
    name as name,
    crs.type as crsType,
    crs.properties.name as crsName,
    features.type as featureType,
    features.properties.FID id,
    features.properties.INSPIREID inspiredId,
    features.properties.NATCODE natCode,
    features.properties.NAMEUNIT nameUnit,
    features.properties.CODNUT1 codNut1,
    features.properties.CODNUT2 codNut2,
    features.properties.CODNUT3 codNut3,
    features.properties.CODIGOINE codigoIne,
    features.properties.SHAPE_Length shapeLength,
    features.properties.SHAPE_Area shapeArea,
    features.geometry.type geometryType,
    arrayMap(features.geometry.coordinates->
                arrayMap(features.geometry.coordinates-> 
                        arrayMap(features.geometry.coordinates-> (features.geometry.coordinates[1],features.geometry.coordinates[2]),features.geometry.coordinates)
                ,features.geometry.coordinates)
    ,features.geometry.coordinates) geometry
FROM file('municipios_ign.geojson', 'JSON')
ARRAY JOIN features;

SELECT count()
FROM geojson

┌─count()─┐
│    8205 │
└─────────┘

SELECT DISTINCT toTypeName(geometry)
FROM geojson

┌─toTypeName(geometry)─┐
│ MultiPolygon         │
└──────────────────────┘

​

Conclusão

Trabalhar com JSON pode ser uma tarefa complexa. Este tutorial abordou um cenário em que um array de objetos aninhados pode tornar essa tarefa ainda mais difícil. Para qualquer outra necessidade relacionada a JSON, consulte nossa documentação.