在 Python 中复制特定的 C# 脚本

Question

Apologies in advance for any formatting issues; first question posted here.

I have a C# script that consists of 3 lines of code:

var json = File.ReadAllText("input.json");
var jsonObject = JsonSerializer.Deserialize<object>(json);
var newJson = JsonSerializer.Serialize(jsonObject);

This code is being used to take a JSON file of the form:

[
{
    "fileName": "File Name",
    "productType": "custom-csv",
    "requestSchema": {
        "includeTags": true,
        "fields": [
            {
                "header": "header",
                "valueType": "static",
                "value": "value"
            },
            ...
        ]
    }
},
{
}, ...
]

flatten it, and create a string that escapes the quotes. The output looks like this:

"[{\"fileName\":\"File Name\",\"productType\":\"custom-csv\",\"requestSchema\":{\"includeTags\":true,\"fields\":[{\"header\":\"Header\",\"valueType\":\"static\",\"value\":\"Value\"},...}]}},{}...]"

The way it is used is running the C# project in Debug with a break after newJson, copying the value from newJson when it hits the break, and then pasting it into a Seed file in another C# project to create a migration using Entity Framework to a SQL DB.

This is obviously clunky, and in order to share this with other people on my team, I want to reproduce this effect in a Python Script, which we would normally use for tasks like this. I cannot seem to reproduce the output here so that the Seed file will be properly formatted.

After multiple attempts, my Python code looks like this:

def parse_json(inputFile):
    with open(inputFile, 'r') as input:
        object = input.readlines()
    jsonData = json.dumps(object)
    outputFile = r"test.txt"
    with open(outputFile, 'w') as output:
        output.write(jsonData)

and the output looks like this:

["[\n", "    {\n", "        \"fileName\": ...

I'm fairly new to writing Python, and I've done lots of searches both here and elsewhere to see what I'm missing, but I haven't been able to make much headway.

I'm open to alternative solutions, but I cannot change the end goal: obtain a string that I can past into the Seed File in a C# project for creating a migration. It needs to be formatted properly to do that, and I'd like to avoid my current solution. It seems like this should be possible with Python, but I cannot figure it out.

Thanks in advance.

Answer 1

Once the json is loaded you can cast it to string and use a str.replace .

# j from JSON
j = [{'fileName': 'File Name', 'productType': 'custom-csv', 'requestSchema': {'includeTags': True, 'fields': [{'header': 'header', 'valueType': 'static', 'value': 'value'}]}}, {}]

# cast to string
json_as_string = str(j)
# return: "[{'fileName': 'File Name', 'productType': 'custom-csv', 'requestSchema': {'includeTags': True, 'fields': [{'header': 'header', 'valueType': 'static', 'value': 'value'}]}}, {}]"

# replace ' with \". Remark: Python escapes with \ thats why the result has double backslashes.
json_as_string.replace("'", r'\"')
# dump new string to json will add another backslash
with open(r"C:\Temp\test.json", "w") as t:
    json.dump(json_as_string.replace("'", r'\"'), t)

Are additional backslashes a problem for the SQL DB?

Side note: overwriting a reserved name might cause problems if you plan using it after re-assigning it. Like if you overwrite input and want to use it later to read user input it will not work as expected.