Analyser NDJSON (Newline-Delimited JSON)
NDJSON stores one JSON objet per ligne, enabling streaming et ligne-par-ligne processing. Commun en logs, API streaming, et big données pipelines.
- 1
NDJSON (also JSONL) puts one complete JSON object on each line, with no commas or wrapping array. Advantages: streamable (process line by line), appendable (add new lines without rewriting), fault-tolerant (one bad line does not invalidate the rest). Common in log files, Kafka, and BigQuery exports.
- 2 Validateur JSON →
Valider each line as a separate JSON object. Unlike regular JSON, you cannot parse the entire file as one value — split by newlines and validate each line independently. Skip empty lines and report invalid lines with their line number.
- 3 Formateur JSON →
Formater individual JSON objects from the NDJSON stream. Each line is a standalone JSON object — format it for inspection. This is useful for debugging log entries or API streaming responses.
- 4
Processing NDJSON: read line by line (do not load the entire file into memory), parse each line as JSON, handle parse errors per line (log and continue or abort), process the object. For large files, use a streaming reader (Node.js readline, browser ReadableStream). For output, write each result as a new NDJSON line.
Questions fréquentes
- NDJSON vs JSON array — which to use?
- Use NDJSON for streaming, logs, and large datasets (each line is independent, can be appended). Use a JSON array for small datasets that need to be loaded entirely (atomic, easier to validate). NDJSON is better for big data; JSON arrays are better for configuration.
- How do I handle invalid lines in NDJSON?
- Analyser each line independently and handle errors per line. Options: skip and log (fault-tolerant), abort on first error (strict), or collect all errors and report at end. For logs, skip and log the error line. For data import, abort to ensure data integrity.