How to open a large JSON file without crashing your editor

A JSON document is one value, often a single top-level array, so many tools parse the whole file into memory before they show anything. Editors add syntax highlighting and folding on top, and browser viewers render every node, so a file of a few hundred megabytes can freeze them. JSON Hero, for example, stops at 1 MB.

JSON Lines (.jsonl, .ndjson, .ldjson) is easier on tools: every line is a separate JSON object, so readers can process it line by line. If you're stuck with one huge .json array, you need a tool that reads it as a table or streams it.

1. Open https://rows.page and drop your .json, .jsonl, .ndjson or .ldjson file on the page, choose it from your disk, or click Paste data to paste JSON text.
2. Scroll the grid. There is no row limit, and DuckDB-WASM reads the file inside your browser tab, so nothing is uploaded unless you click Share.
3. Click any row to open it in a side drawer with a JSON/Fields toggle and a Copy as JSON button.
4. Filter in the search box with terms like country:NL or price:>20, or type SQL that starts with SELECT, WITH or FROM against the table data. Put dot-path column names in double quotes, like "user.city".

Nested objects flatten into dot-path columns like user.address.city, up to 8 levels deep and 200 child fields per object. Arrays stay as list columns that show the first 6 items and "+N". A wrapper object like {"data": [...]} is unwrapped to its biggest list, with the note: Showing the N items in "data"; the other top-level fields were skipped.

A .json file over 1.5 GB is refused with "Too big for this browser". A JSONL or NDJSON file over 1.5 GB loads partially instead, with the note "Loaded the first N rows: the file is X, more than the browser can hold." rows.page has no whole-file tree view, and you can't edit cells.

jq

Plain jq parses a whole top-level array before running your filter, which fails on multi-gigabyte files. With --stream, jq reads the input piece by piece, and this idiom writes each array item to its own line:

jq -cn --stream 'fromstream(1|truncate_stream(inputs))' big.json > big.jsonl

The result is JSONL, so you can split it with line-based tools. It also helps in rows.page: a JSONL file over 1.5 GB opens with its first rows, where the same data as one .json array is refused.

DuckDB

The DuckDB command-line tool reads JSON arrays and newline-delimited JSON from disk, detects the layout and infers column types. Look at the first 10 records:

SELECT * FROM read_json('big.json') LIMIT 10;

To explore the whole file in rows.page, convert it to Parquet once. Nested data stays as STRUCT and LIST columns, and rows.page doesn't hold Parquet files to the 1.5 GB check.

COPY (SELECT * FROM read_json('big.json')) TO 'big.parquet' (FORMAT parquet, COMPRESSION zstd);

Dadroit JSON Viewer

Dadroit is a desktop app for large JSON files. The free version opens files up to 50 MB, Standard costs $98 per year and opens files up to 2 GB, and Advanced goes up to 1 TB.

jless

jless is a free, MIT-licensed JSON viewer for the terminal on macOS and Linux. You expand and collapse objects and arrays and move around with vim-style keys instead of scrolling raw text.