Data Demon Systems

/tools/jsonl-viewer

JSONL & NDJSON Viewer Online

Validate and inspect newline-delimited JSON record by record, with bounded type inference and local redaction.

jsonl-viewer://local

1. Source

Choose one UTF-8 JSONL or NDJSON file, or paste a small example up to 10 MB.

Processed locally · source remains unchanged

2. Search and filters

Load a source to filter
Level
Advanced evidence filters

Evidence filters

Likely exposure

3. Records

4. JSON tree

Findings

No sampled signals

5. Filtered and redacted export

Unchanged records preserve their original bytes. Select categories manually, or apply the guided sharing-safe policy.

Redact

No redaction categories are selected automatically.

Choose a JSONL file or load a small sample. Processing stays in this browser.

/structure

Inspect JSON Lines without one giant parse

Independent records

Each non-terminated or newline-terminated line is parsed independently. JSON objects, arrays, strings, numbers, booleans and null are all valid roots.

Incremental shape evidence

A bounded accumulator merges field types and optionality while indexing, without retaining the parsed records.

Malformed rows remain visible

A malformed record is reported rather than terminating the session. Filter valid or malformed records for focused inspection and export.

Local forensic workflow

Masked likely-exposure findings can become explicit, category-scoped redactions only when you choose them.

/evidence

Reproducible JSONL and NDJSON evidence

The built-in records.jsonl example is identical to the repository fixture. Data Demon Systems maintains it alongside deliberate malformed-row and synthetic exposure-review inputs for repeatable validation, filtering and redaction checks.

The public JSONL fixtures and methodology document the supported cases and deliberate 10 MiB paste, 4 GiB file and 1 MiB streaming boundaries. Each line follows the JSON grammar in RFC 8259, using the newline-delimited convention described by the NDJSON specification.

For delimited tables use the CSV Viewer Online for Large Files; for mixed diagnostic formats use the Log Viewer Online & Log Analyzer.

/faq

JSONL viewer questions

Are malformed JSONL rows skipped?

Each line is parsed independently. Malformed rows are counted and inspectable without stopping the rest of the file.

Does JSONL data leave my browser?

No. Indexing, validation, search, inspection and export run locally.

What is the difference between JSONL and NDJSON?

Both names commonly describe one complete JSON value per line. This viewer treats them equivalently.

Will export change untouched records?

No. Unchanged records are copied byte-for-byte. A record is reserialised only when a selected structured value is redacted.