JSON Array Deduplicator

What this tool does

JSON Array Deduplicator removes repeated elements from a JSON array. Paste an array straight out of an API response, an export, or a merge of two datasets, and get back the same array with the duplicates gone — original order intact, nothing uploaded.

By default two elements are duplicates when they are structurally equal: same values all the way down through nested objects and arrays. Object key order is ignored when comparing but preserved in the output, so {"a":1,"b":2} and {"b":2,"a":1} collapse into one while the survivor keeps the shape you pasted. When "same record" means something narrower, name the fields to compare — id, or user.email,country — and everything else is ignored.

You choose whether the first or last occurrence survives, whether matching ignores letter case, and whether the result is the de-duplicated array, only the duplicates that were removed, or a report of counts and duplicate groups. If the array is wrapped inside an object ({"data":{"items":[…]}}), point Path to the array at it and the wrapper comes back untouched.

Worked example

Input

[
  { "id": 1, "email": "[email protected]" },
  { "id": 2, "email": "[email protected]" },
  { "id": 1, "email": "[email protected]" }
]

With Compare only these fields left blank (whole-element comparison) and indent = 2, the output is:

[
  {
    "id": 1,
    "email": "[email protected]"
  },
  {
    "id": 2,
    "email": "[email protected]"
  }
]

Switch Show to Counts + duplicate groups on the same input and you get the audit trail instead — which positions collided, and which one was kept:

{
  "total": 3,
  "unique": 2,
  "removed": 1,
  "duplicate_groups": [
    {
      "count": 2,
      "indexes": [0, 2],
      "kept_index": 0,
      "value": { "id": 1, "email": "[email protected]" }
    }
  ]
}

How to use it

1. Paste a JSON array — of objects, strings, numbers, or nested values.
2. Leave Compare only these fields blank for whole-element matching, or list fields comma-separated (id, user.email,country) to match on just those.
3. If the array sits inside a wrapper object, set Path to the array to its dot-path, e.g. data.items.
4. Pick the occurrence to keep, whether to ignore letter case, what to show, and the indent (0 minifies).
5. Copy the result. Everything runs locally in your browser — the JSON is never uploaded.

FAQ

Limits & edge cases

• The input must be valid JSON, and the target must be an array — a bare object is rejected with an error that points you at Path to the array.
• Up to 200,000 elements; larger arrays are rejected with a named error rather than freezing the tab. Very large inputs are still bounded by your browser's memory.
• Indent is clamped to 0–8 spaces; 0 minifies to a single line.
• Ignore letter case folds case in string values and in field names, so {"ID":1} and {"id":1} match while it is on. Leave it off for case-exact data such as IDs and hashes.
• An absent field and an explicit null are different: {} and {"id":null} are not duplicates when comparing by id. All rows missing that field do group together.
• Elements are compared, never rewritten: key order, spacing choices inside each surviving element, and every non-compared field come back exactly as pasted.
• Everything runs in-browser via WebAssembly — the JSON you paste is never uploaded.