> For the complete documentation index, see [llms.txt](https://docs.algenta.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.algenta.ai/guides/instant-map.md).

# Map a dataset instantly

Drop in CSV, JSON, or connected tables and get back a unified schema with every field mapped, join keys detected, and confidence for each match — no schema, no ETL.

`POST /v1/map` connects and understands any dataset with zero configuration. Hand it one or more sources — CSV text, JSON records, a URL, base64 Excel/Parquet, a connector config, or the id of a source you already registered — and the engine returns a unified schema: which fields mean the same thing across sources, their distribution shapes, polarity (revenue is positive, cost is negative), detected join keys, and a plain-English reason and confidence for every mapping.

Mapping runs a two-speed pipeline. Phase 1 (the unified schema plus column matches) returns immediately. Phase 2 (join-key detection and derived formulas like `A × B ≈ C`) is embedded in the same response when it finishes fast; otherwise the response sets `deep_analysis_pending: true` and you poll `GET /v1/map/{map_id}` for the rest.

## Before you start

You need:

* An API key sent as `Authorization: Bearer $ALGENTA_API_KEY`.
* At least one data source. The examples use inline `csv`, but every `DataSource` accepts exactly one payload field — pick whichever you have.

| Field                       | Use it for                                                              |
| --------------------------- | ----------------------------------------------------------------------- |
| `records`                   | Inline JSON array of objects (fastest inline path).                     |
| `csv`                       | Raw CSV text.                                                           |
| `json_str`                  | Raw JSON array/object text (nested JSON is auto-flattened).             |
| `url`                       | An HTTP(S) URL; format auto-detected from content type/extension.       |
| `excel_b64` / `parquet_b64` | A base64-encoded `.xlsx` or Parquet file.                               |
| `connection`                | A connector config dict (`sql`, `rest_api`, `s3`, `gcs`, `azure_blob`). |
| `source_id` / `dataset_id`  | A source you already registered — avoids re-uploading data.             |

## Map your sources

{% stepper %}
{% step %}

### Send the sources to `/v1/map`

Every source needs a `name`. `min_confidence` (default `0.5`) sets the threshold below which a mapping is dropped; `include_statistics` (default `true`) keeps distribution stats in the response.

{% tabs %}
{% tab title="cURL" %}

```bash
curl -sS -X POST "https://api.algenta.ai/v1/map" \
  -H "Authorization: Bearer $ALGENTA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "sources": [
      {"name": "sales",  "csv": "order_id,revenue,cost\n1,1200,800\n2,900,600"},
      {"name": "orders", "csv": "id,amount,qty\n1,1200,3\n2,900,2"}
    ],
    "min_confidence": 0.5
  }'
```

{% endtab %}

{% tab title="Python" %}

```python
import os
import requests

BASE = os.environ.get("ALGENTA_API_URL", "https://api.algenta.ai")

resp = requests.post(
    f"{BASE}/v1/map",
    headers={"Authorization": f"Bearer {os.environ['ALGENTA_API_KEY']}"},
    json={
        "sources": [
            {"name": "sales",  "csv": "order_id,revenue,cost\n1,1200,800\n2,900,600"},
            {"name": "orders", "csv": "id,amount,qty\n1,1200,3\n2,900,2"},
        ],
        "min_confidence": 0.5,
    },
)
resp.raise_for_status()
data = resp.json()

for m in data["mappings"]:
    print(f"{m['confidence']:.0%}  {m['source']}.{m['source_field']} -> {m['target']}.{m['target_field']}")
```

{% endtab %}

{% tab title="TypeScript" %}

```typescript
const base = process.env.ALGENTA_API_URL ?? "https://api.algenta.ai";

const res = await fetch(`${base}/v1/map`, {
  method: "POST",
  headers: {
    Authorization: `Bearer ${process.env.ALGENTA_API_KEY}`,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    sources: [
      { name: "sales",  csv: "order_id,revenue,cost\n1,1200,800\n2,900,600" },
      { name: "orders", csv: "id,amount,qty\n1,1200,3\n2,900,2" },
    ],
    min_confidence: 0.5,
  }),
});
if (!res.ok) throw new Error(`map failed: ${res.status}`);
const data = await res.json();
console.log(data.mappings);
```

{% endtab %}
{% endtabs %}

The `200` `MapResponse` carries the unified schema, the ranked field mappings, and the deep-analysis fields:

```json
{
  "map_id": "1f3c9a2e-4b7d-4a11-9c6e-0f2b8d5a7c31",
  "sources_analyzed": 2,
  "total_fields": 6,
  "mappings": [
    {
      "source_field": "revenue",
      "source": "sales",
      "target_field": "amount",
      "target": "orders",
      "confidence": 0.86,
      "reason": ["semantic match: revenue ~ amount", "overlapping value distribution"],
      "match_type": "semantic"
    }
  ],
  "unified_schema": [
    {
      "name": "revenue",
      "sources": ["sales", "orders"],
      "distribution": "lognormal",
      "unit": "USD",
      "polarity": 1,
      "sample_values": [1200, 900]
    }
  ],
  "join_keys": [],
  "derived_relationships": [],
  "simulation_ready": true,
  "simulation_payload": {},
  "deep_analysis_pending": false,
  "latency_ms": 42.3
}
```

Each mapping's `match_type` is one of `exact`, `semantic`, or `statistical`; `confidence` is `0`–`1`; `reason` is a short list of human-readable justifications. Keep `map_id` — you need it if deep analysis is still running.
{% endstep %}

{% step %}

### Poll for deep analysis if it is pending

When the response has `deep_analysis_pending: true`, join keys and derived relationships are still computing. Poll with the `map_id`:

`GET /v1/map/{map_id}`

```bash
curl -sS "https://api.algenta.ai/v1/map/$MAP_ID" \
  -H "Authorization: Bearer $ALGENTA_API_KEY"
```

A `200` returns the completed `join_keys`, `derived_relationships`, and `simulation_payload`. While the work is still in flight the endpoint returns `404` with `"map_id not found or still processing"` — retry after a short backoff. Results are scoped to your organization, so a `map_id` from another org is never readable.
{% endstep %}

{% step %}

### Promote a source to a queryable dataset (optional)

Mapping is stateless. To persist a source and query it later, connect it as a dataset with `POST /v1/data/connect`:

```bash
curl -sS -X POST "https://api.algenta.ai/v1/data/connect" \
  -H "Authorization: Bearer $ALGENTA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "dataset_name": "sales",
    "csv": "order_id,revenue,cost\n1,1200,800\n2,900,600",
    "visibility": "private"
  }'
```

The response carries a `dataset_id` you reuse across the `/v1/data/*` endpoints below and in [governed queries](/guides/governed-query.md).
{% endstep %}
{% endstepper %}

{% hint style="success" %}
**Expected result** — `/v1/map` returns `200` with `sources_analyzed` matching the number of sources you sent, a non-empty `unified_schema`, and a `mappings` list ranked by `confidence`. `simulation_ready: true` means the schema has enough numeric metric fields to run a decision on.
{% endhint %}

## Manage datasets

Once a source is connected, these routes list, inspect, refresh, and disconnect it. All are scoped to the caller's workspace.

| Method & path                        | What it does                                                                                   |
| ------------------------------------ | ---------------------------------------------------------------------------------------------- |
| `GET /v1/data`                       | List visible datasets. Supports `page`, `limit`, `search`, `status`, `source_name`, `compact`. |
| `GET /v1/data/{dataset_id}`          | Get one dataset plus its full schema.                                                          |
| `GET /v1/data/{dataset_id}/summary`  | Low-token discovery summary (row/column counts, query hints).                                  |
| `POST /v1/data/{dataset_id}/refresh` | Re-pull the dataset from its saved origin.                                                     |
| `DELETE /v1/data/{dataset_id}`       | Disconnect the dataset (and its connector when nothing else uses it).                          |

## Next steps

<table data-view="cards"><thead><tr><th></th><th></th><th data-hidden data-card-target data-type="content-ref"></th></tr></thead><tbody><tr><td><strong>Query your mapped data</strong></td><td>Ask questions over a connected dataset.</td><td><a href="/guides/governed-query.md">Query governed data</a></td></tr><tr><td><strong>Connect a data source</strong></td><td>Register databases and object storage as reusable connectors.</td><td><a href="/guides/connect-data.md">Connect a data source</a></td></tr><tr><td><strong>Recommend a decision</strong></td><td>Turn the mapped schema into a ranked recommendation.</td><td><a href="/guides/recommend-and-score.md">Recommend, score &amp; compare</a></td></tr></tbody></table>


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.algenta.ai/guides/instant-map.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.