使用 PYTHON 适用于每个记录内的 JSON 格式缩进

Question

我有 2 个示例 JSON 记录，我想使用 python 删除缩进。 使用这种语法（line.replace（“”，“”））我删除了所有缩进。 但问题是所有记录都是一张记录。 期望是每条记录都应该排成一行。 请指导我

{
    "device_model": "TITAN",
    "enb_locan": {
        "coordinates": [
            -900011,
            900.87
        ],
        "crs": {
            "properties": {
                "name": "EPSG:4326"
            },
            "type": "name"
        },
        "type": "Point"
    },
    "enb_order_distance": 25,
    "trans_dt": "2024-01-11"
}
{
    "device_model": "TITAN2222",
    "enb_locan": {
        "coordinates": [
            -900011,
            900.87
        ],
        "crs": {
            "properties": {
                "name": "EPSG:4326"
            },
            "type": "name"
        },
        "type": "Point"
    },
    "enb_order_distance": 29,
    "trans_dt": "2024-01-11"
}

代码使用

  def process(self, inputStream, outputStream):
    text = IOUtils.readLines(inputStream, StandardCharsets.UTF_8)
    for line in text:
        output = line.replace(" ", "")
        outputStream.write(output)

Answer 1

听起来您想要 JSON Lines 格式，其中每个完整的 JSON 记录都位于以换行符结尾的单行上。

下面是一个粗略的嵌套花括号计数器，用于收集完整的记录并重写为单行：

import io
import json
import sys

data = '''\
{
    "device_model": "TITAN",
    "enb_locan": {
        "coordinates": [
            -900011,
            900.87
        ],
        "crs": {
            "properties": {
                "name": "EPSG:4326"
            },
            "type": "name"
        },
        "type": "Point"
    },
    "enb_order_distance": 25,
    "trans_dt": "2024-01-11"
}
{
    "device_model": "TITAN2222",
    "enb_locan": {
        "coordinates": [
            -900011,
            900.87
        ],
        "crs": {
            "properties": {
                "name": "EPSG:4326"
            },
            "type": "name"
        },
        "type": "Point"
    },
    "enb_order_distance": 29,
    "trans_dt": "2024-01-11"
}
'''

# Could read from file as well.
with io.StringIO(data) as f:
    record = []
    nested = 0
    while c := f.read(1):  # read and store characters
        record.append(c)
        if c == '{':       # keep track of nesting
            nested += 1
        if c == '}':
            nested -= 1
        if nested == 0:    # Once nested returns to zero, we have a full record
            data = ''.join(record).strip()  # build record and remove whitespace before and after
            if data:  # load the record and dump again without indentation
                print(json.dumps(json.loads(data)), file=sys.stdout)  # could print to a file.
                record = []

输出：

{"device_model": "TITAN", "enb_locan": {"coordinates": [-900011, 900.87], "crs": {"properties": {"name": "EPSG:4326"}, "type": "name"}, "type": "Point"}, "enb_order_distance": 25, "trans_dt": "2024-01-11"}
{"device_model": "TITAN2222", "enb_locan": {"coordinates": [-900011, 900.87], "crs": {"properties": {"name": "EPSG:4326"}, "type": "name"}, "type": "Point"}, "enb_order_distance": 29, "trans_dt": "2024-01-11"}