使用属性从JSON对象创建新的GeoJSON LineString

Question

我想将几个Json对象与LineStrings组合成一个LineStrings特性集合。

例如，我有以下格式错误的json对象：

 {"lat":16.0269337,"lon":40.073042,"score":1,"ID":"13800006252028","TYPES":"Regional","N2C":"2","NAME":"Strada Statale della Val Sinni"}
{"lat":16.0316522,"lon":41.6808931,"score":1,"ID":"13800006260148","TYPES":"Scenic","N2C":"2"}
{"lat":16.0316523,"lon":41.6808932,"score":1,"ID":"13800006260148","TYPES":"Scenic","N2C":"2","NAME":"Giro del Gargano"}
{"lat":16.0316522,"lon":41.6808931,"score":1,"ID":"13800006203620","TYPES":"Scenic","N2C":"2"}
{"lat":16.0316522,"lon":41.6808931,"score":1,"ID":"13800006203620","TYPES":"Regional","N2C":"2","NAME":"Giro del Gargano"}
{"lat":16.032322,"lon":40.6097567,"score":1,"ID":"13800005963252","TYPES":"Scenic","N2C":"2"}
{"lat":16.032322,"lon":40.6097567,"score":1,"ID":"13800005963252","TYPES":"Regional","N2C":"2","NAME":"Via Appia"}

我打算做两件事

1. 如果ID &相同类型相同，则使用ID创建GeoJson行字符串
2. 根据score属性将匹配的json文件的得分添加到上面。

Answer 1

我不是专家，但这是第一批。This post有助于获得基本知识。对于每个"Geo“、"JSON”和熊猫操作片段，可能有比我更优雅的方法。

目前还不清楚您的数据的确切格式是什么，但是如果它确实是使用pd.DataFrame的字典列表，或者直接使用pd.from_json从JSON文件中导入它，那么您可以直接导入它。我只是和前者一起工作，但这很容易改变。

json_objs = [{"lat":16.0269337,"lon":40.073042,"score":1,"ID":"13800006252028","TYPES":"Regional","N2C":"2","NAME":"Strada Statale della Val Sinni"},
{"lat":16.0316522,"lon":41.6808931,"score":1,"ID":"13800006260148","TYPES":"Scenic","N2C":"2"},
{"lat":16.0316523,"lon":41.6808932,"score":1,"ID":"13800006260148","TYPES":"Scenic","N2C":"2","NAME":"Giro del Gargano"},
{"lat":16.0316522,"lon":41.6808931,"score":1,"ID":"13800006203620","TYPES":"Scenic","N2C":"2"},
{"lat":16.0316522,"lon":41.6808931,"score":1,"ID":"13800006203620","TYPES":"Regional","N2C":"2","NAME":"Giro del Gargano"},
{"lat":16.032322,"lon":40.6097567,"score":1,"ID":"13800005963252","TYPES":"Scenic","N2C":"2"},
{"lat":16.032322,"lon":40.6097567,"score":1,"ID":"13800005963252","TYPES":"Regional","N2C":"2","NAME":"Via Appia"}

然后与熊猫一起操作数据，并使用geojson包创建LineString对象。

import geojson
import pandas as pd

df = pd.DataFrame(json_objs)

grouped = df.groupby(["ID","TYPES"])
features = []

for name,indices in grouped.groups.items():
    group = df.iloc[indices]

    score = group["score"].sum()   
    points = list(zip(group["lat"].values,group["lon"].values))    
    properties = {"ID": name[0], "record_type": name[1], "score": score}

    ls = geojson.LineString(coordinates = points, properties = properties)
    print(ls)

    features.append(ls)

输出：

{"coordinates": [[16.032322, 40.6097567]], "properties": {"ID": "13800005963252", "record_type": "Scenic", "score": 1}, "type": "LineString"}
{"coordinates": [[16.0269337, 40.073042]], "properties": {"ID": "13800006252028", "record_type": "Regional", "score": 1}, "type": "LineString"}
{"coordinates": [[16.032322, 40.6097567]], "properties": {"ID": "13800005963252", "record_type": "Regional", "score": 1}, "type": "LineString"}
{"coordinates": [[16.0316522, 41.6808931], [16.0316523, 41.6808932]], "properties": {"ID": "13800006260148", "record_type": "Scenic", "score": 2}, "type": "LineString"}
{"coordinates": [[16.0316522, 41.6808931]], "properties": {"ID": "13800006203620", "record_type": "Scenic", "score": 1}, "type": "LineString"}
{"coordinates": [[16.0316522, 41.6808931]], "properties": {"ID": "13800006203620", "record_type": "Regional", "score": 1}, "type": "LineString"}

然后将这些LineStrings转换为FeatureCollection

feature = geojson.FeatureCollection(features)

显然，您也可以手动构造这种格式，但我希望geojson提供了更多的灵活性。显然，也有一个geopandas包，我无法安装。