Move JSON into deterministic CSV pipelines.

Endpoint

Post one JSON document, NDJSON stream, or multipart file upload. jsonnorm detects null and missing fields separately, but the default CSV representation renders both as empty cells. The default plan discovers a union schema, preserves field order deterministically, and writes standards-compliant CSV.

POST https://jsonnorm.com/api/v1/convert
Content-Type: application/json

[{"id":1,"name":"Ada"}]

The endpoint accepts JSON, NDJSON, or a multipart file upload. Add ?response=metadata to receive JSON with row, column, input, and output counts instead of the CSV body.

Quickstarts

curl

output="$(mktemp)"
if curl --fail-with-body --silent --show-error \
  -X POST https://jsonnorm.com/api/v1/convert \
  -H 'Content-Type: application/json' \
  --data-binary @orders.json \
  --output "$output"; then
  mv "$output" orders.csv
else
  cat "$output" >&2
  rm -f "$output"
  exit 1
fi

curl · NDJSON

curl --fail-with-body --silent --show-error \
  -X POST https://jsonnorm.com/api/v1/convert \
  -H 'Content-Type: application/x-ndjson' \
  --data-binary @orders.ndjson

curl · multipart file

curl --fail-with-body --silent --show-error \
  -X POST https://jsonnorm.com/api/v1/convert \
  -F '[email protected];type=application/json'

curl · metadata

curl --fail-with-body --silent --show-error \
  -X POST 'https://jsonnorm.com/api/v1/convert?response=metadata' \
  -H 'Content-Type: application/json' \
  --data-binary @orders.json

Python

import requests

with open("orders.json", "rb") as source, open("orders.csv", "wb") as target:
    with requests.post(
        "https://jsonnorm.com/api/v1/convert",
        headers={"Content-Type": "application/json"},
        data=source,
        stream=True,
    ) as response:
        response.raise_for_status()
        for chunk in response.iter_content(chunk_size=1024 * 1024):
            if chunk:
                target.write(chunk)

Node.js / TypeScript

import { createWriteStream } from "node:fs";
import { readFile } from "node:fs/promises";
import { Readable } from "node:stream";
import { pipeline } from "node:stream/promises";

const response = await fetch("https://jsonnorm.com/api/v1/convert", {
  method: "POST",
  headers: { "Content-Type": "application/json" },
  body: await readFile("orders.json"),
});
if (!response.ok) throw new Error(response.status + ": " + await response.text());
if (!response.body) throw new Error("The API returned no response body.");
await pipeline(
  Readable.fromWeb(response.body),
  createWriteStream("orders.csv"),
);

PowerShell 7

Invoke-WebRequest -Uri 'https://jsonnorm.com/api/v1/convert' -Method Post -ContentType 'application/json' -InFile .\orders.json -OutFile .\orders.csv

Invoke-RestMethod -Uri 'https://jsonnorm.com/api/v1/convert?response=metadata' -Method Post -ContentType 'application/json' -InFile .\orders.json

Fair-use limits

Automated abuse protection, fair-use limits, input and output budgets, and concurrency safeguards apply to every request. If you reach a limit, the API returns 429 with a Retry-After value. These safeguards protect availability and keep retries from becoming an accidental request storm.

Plan for bounded requests and retry with backoff.

If the service is temporarily unavailable, retry after the Retry-After delay and use backoff.

Errors

Failures use a JSON envelope with error.code, error.message, and an optional privacy-safe warning.

Privacy by default

Request payloads are processed temporarily. jsonnorm keeps no conversion history and does not log payload contents or CSV output.