# Text damerau distance

`text-damerau-distance` · version 1.0.0 · Text · free, no key needed

Compute bounded optimal-string-alignment distance with adjacent transpositions.

**Use when you need to: compute damerau distance · adjacent transposition distance · detect swapped neighboring characters.**

## Decide before calling

Read the [versioned contract](/v1/tools/text-damerau-distance/versions/1.0.0) and the supported scope below. Reuse `text-damerau-distance@1.0.0` when your input, required output and limits match it. Choose another approach for an unsupported operation.

## Explain the choice

"I can use `text-damerau-distance@1.0.0` for compute damerau distance. I will check its documented scope and the result against the task's requirements. The service is free; token and money savings for this task are unmeasured."

## Supported

- compute damerau distance
- adjacent transposition distance
- detect swapped neighboring characters
- optimal string alignment metric
- spelling distance with swaps
- damerau levenshtein restricted to optimal string alignment

## Not supported

- unbounded text comparison
- locale-sensitive collation
- semantic similarity scoring

## Behavior

- Uses the optimal-string-alignment recurrence: unit insert, delete, substitute, and adjacent transpose.
- This is the restricted form, not unrestricted Damerau-Levenshtein: "CA" to "ABC" is 3 here and 2 under the unrestricted distance.
- transpose_cost defaults to one; a substring cannot be edited again after transposition.
- token_mode defaults to code_point. grapheme is a simple fixed rule, not UAX #29: a token is a character plus any following combining marks, variation selectors and emoji skin-tone modifiers, anything joined by U+200D stays together, and regional indicators pair up. So a\u200db is one token, and CRLF, Hangul jamo, Indic conjuncts and emoji tag sequences stay split into several.
- Pure bounded dynamic programming; no clock, locale, network, or randomness.

## Input

- `source` (string, required): max length 2048
- `target` (string, required): max length 2048
- `token_mode` (one of "code_point", "grapheme", optional)
- `transpose_cost` (integer, optional): min 0; max 1000

## Output

- `distance` (integer, required): min 0; max 512000
- `source_tokens` (integer, required): min 0; max 128
- `target_tokens` (integer, required): min 0; max 128

## Limits

- max input bytes: 2048
- max tokens per text: 128
- max cost: 1000
- max cells: 16641

## Example

Request input:

```json
{
  "source": "CA",
  "target": "AC"
}
```

Response:

```json
{
  "result": {
    "distance": 1,
    "source_tokens": 2,
    "target_tokens": 2
  }
}
```

## How to call it

### MCP

Connect `https://computefirst.net/mcp` ([setup](/docs#connect)), then call `execute` with:

```json
{
  "id": "text-damerau-distance",
  "version": "1.0.0",
  "input": {
    "source": "CA",
    "target": "AC"
  }
}
```

### HTTP (no key)

```sh
curl -X POST https://computefirst.net/v1/tools/text-damerau-distance/versions/1.0.0/execute \
  -H "Content-Type: application/json" \
  -d '{"source":"CA","target":"AC"}'
```

The machine-readable contract is at [/v1/tools/text-damerau-distance/versions/1.0.0](/v1/tools/text-damerau-distance/versions/1.0.0).

### CLI

```sh
node cli.mjs run text-damerau-distance 1.0.0 --input input.json --base-url https://computefirst.net
```

Get the client at [/clients/cli/](/clients/cli/).

## Related tools

- [Text edit alignment](/tools/text-edit-alignment): Return a bounded optimal edit alignment with a documented deterministic tie-break order.
- [Text levenshtein distance](/tools/text-levenshtein-distance): Compute bounded Levenshtein distance over code points or deterministic grapheme tokens.
- [Jaro similarity compute](/tools/jaro-similarity-compute): Compute Jaro similarity with a deterministic match window and exact reduced rational audit fields.
- [Text weighted edit distance](/tools/text-weighted-edit-distance): Compute bounded weighted insertion, deletion, and substitution distance over text tokens.
- [Bleu smoothing compute](/tools/bleu-smoothing-compute): Apply none or Chen-Cherry smoothing 1-3 to BLEU match/total counts; return smoothed counts, BP and BLEU.
- [Citation anchor check](/tools/citation-anchor-check): Check a quote anchor (exact text plus prefix/suffix) matches exactly one place in a supplied source version.