Python Forum

Hi,

I need to get rid of " from a JSON file

I have a JSON file like this:

{
    "type": "FeatureCollection",
    "features": [
        {
            "type": "Feature",
            "properties": {
                "Ville": "OZAN"
            },
            "geometry": {
                "type": "Point",
                "coordinates": [
                    "4.91667, 46.3833"
                ]
            }
        },
        {
            "type": "Feature",
            "properties": {
                "Ville": "CORMORANCHE-SUR-SAONE"
            },
            "geometry": {
                "type": "Point",
                "coordinates": [
                    "4.83333, 46.2333"
                ]
            }
        },

I need to remove the " before and after the coordinates in order to get this.

{
    "type": "FeatureCollection",
    "features": [
        {
            "type": "Feature",
            "properties": {
                "Ville": "OZAN"
            },
            "geometry": {
                "type": "Point",
                "coordinates": [
                    4.91667, 46.3833
                ]
            }
        },
        {
            "type": "Feature",
            "properties": {
                "Ville": "CORMORANCHE-SUR-SAONE"
            },
            "geometry": {
                "type": "Point",
                "coordinates": [
                    4.83333, 46.2333
                ]
            }
        },

I can't do it manually because i about 35k cities.

I can't find any viable solution, because I can't get rid of all the " from my file and i can't convert coordniates into float directly because of the [] .. such as this solution : https://stackoverflow.com/questions/3667...rom-python

Do you have any idea ? thank you in advance.

1. load the json into json_data
2. iterate over features in json_data['features']
3. read the feature['geometry']['coordinates'][0] - this will give you a string, e.g."4.91667, 46.3833"
4. convert coordinates by splitting at comma, then convert individual values to float
5. assign coordinates from 4 as list to feature['geometry']['coordinates']
6. dump json data back to a file

Hi, thank you for clues, it helped me,

the code below worked:

with open(filepath, 'r+', encoding='utf-8') as data_jason:

 data = json.load(data_jason)

 
 for feature in data['features']:
     print (feature['geometry']['coordinates'][0])
     splitted = feature['geometry']['coordinates'][0].split(',')
     print(splitted)
     floated = [float(i) for i in splitted]
     print(floated)
     del feature['geometry']['coordinates']
     feature['geometry']['coordinates'] = []
     feature['geometry']['coordinates'].extend(floated)
     print(feature['geometry']['coordinates'])

 json.dump(data, data_jason,  indent=4)

Dumping write at the end of the file how could I make it overwrite it instead ?

don't open it in r+ mode. first open it for reading. After you process data, open it again for writing and dump data.
There are some redundant steps in what you do, e.g. no need to delete the coordinates node first

import json

json_file = 'features.json'

with open(json_file) as f:
    json_data = json.load(f)
    for feature in json_data['features']:
        coordinates = feature['geometry']['coordinates'][0].split(',')
        feature['geometry']['coordinates'] = [float(coord.strip()) for coord in coordinates]

with open(json_file, 'w') as f:
    json.dump(json_data, f, indent=4)

You are so fast :) :) :), I was about to edit because I found out the 2 steps "with open" way.

Thank you for optimization also, I'd rather make more steps in order to understand better what I do...

When you convert to float, what does strip() does (which I don't do in my version?)

Regards,

(May-06-2020, 11:36 AM)WBPYTHON Wrote: [ -> ]When you convert to float, what does strip() does (which I don't do in my version?)

it strips the leading/trailing whitespace if present after split at comma,
But you are right - it's not really necessary. it can be removed

feature['geometry']['coordinates'] = [float(coord) for coord in coordinates]

or using map if you prefer

feature['geometry']['coordinates'] = list(map(float, coordinates))

Understood, thank you again