Reference

REST API

Base URL: https://api.imibare.org
All endpoints are versioned under /v1/.
Responses are JSON unless a download endpoint is used.

GET /v1/datasets

Returns all datasets in the catalog.

Request

GET /v1/datasets

Query parameters

Parameter	Type	Description
`country`	string	Filter by ISO 3166-1 alpha-2 code (e.g., `RW`, `KE`). Case-insensitive.
`institution`	string	Filter by institution slug (e.g., `nisr`).
`topic`	string	Filter by topic slug (e.g., `prices`).
`frequency`	string	Filter by frequency slug (e.g., `monthly`).
`pipeline`	string	Filter by pipeline status: `automated`, `curated`, or `stale`.

Response 200 OK

{
  "datasets": [
    {
      "id": "rw.nisr.cpi.monthly",
      "country": "RW",
      "country_name": "Rwanda",
      "name": "Consumer Price Index (monthly)",
      "institution": "National Institute of Statistics of Rwanda",
      "institution_slug": "nisr",
      "topic": "prices",
      "frequency": "monthly",
      "coverage_start": "2009-01",
      "coverage_end": "present",
      "pipeline": "automated",
      "last_updated": "2026-05-15",
      "version": "2026-05-15",
      "formats": ["csv", "json", "parquet"],
      "license": "CC-BY-4.0",
      "license_basis": "publisher-terms",
      "license_url": "https://statistics.gov.rw/terms-use"
    }
  ],
  "total": 5
}

Examples

# All datasets
curl https://api.imibare.org/v1/datasets

# Rwanda only
curl "https://api.imibare.org/v1/datasets?country=RW"

# Monthly price datasets
curl "https://api.imibare.org/v1/datasets?topic=prices&frequency=monthly"

GET /v1/datasets/{id}

Returns full metadata for a single dataset, including column specifications.

Request

GET /v1/datasets/{id}

Path parameters

Parameter	Type	Description
`id`	string	Dataset ID (e.g., `rw.nisr.cpi.monthly`)

Response 200 OK

{
  "id": "rw.nisr.cpi.monthly",
  "country": "RW",
  "country_name": "Rwanda",
  "name": "Consumer Price Index (monthly)",
  "institution": "National Institute of Statistics of Rwanda",
  "institution_slug": "nisr",
  "topic": "prices",
  "frequency": "monthly",
  "coverage_start": "2009-01",
  "coverage_end": "present",
  "pipeline": "automated",
  "source_url": "https://statistics.gov.rw/statistical-publications/price-indices-cpi-ppi/",
  "source_format": "excel",
  "last_updated": "2026-05-15",
  "version": "2026-05-15",
  "formats": ["csv", "json", "parquet"],
  "license": "CC-BY-4.0",
  "license_basis": "publisher-terms",
  "license_url": "https://statistics.gov.rw/terms-use",
  "columns": [
    {
      "name": "date",
      "type": "date",
      "description": "First day of reference month",
      "unit": null
    },
    {
      "name": "category",
      "type": "string",
      "description": "CPI category (overall, food, housing, transport, …)",
      "unit": null
    },
    {
      "name": "value",
      "type": "float64",
      "description": "Index value (base year 2014=100)",
      "unit": "index"
    }
  ],
  "notes": "Base year changed from 2009 to 2014 in the January 2015 release."
}

Response 404 Not Found

{
  "error": "Dataset not found",
  "id": "rw.nisr.cpi.monthly"
}

Example

curl https://api.imibare.org/v1/datasets/rw.nisr.cpi.monthly

GET /v1/data/{id}/download

Returns the dataset file as a download.

Request

GET /v1/data/{id}/download

Path parameters

Parameter	Type	Description
`id`	string	Dataset ID

Query parameters

Parameter	Type	Default	Description
`format`	string	`parquet`	File format: `parquet`, `csv`, or `json`
`version`	string	latest	Version date `YYYY-MM-DD`. Returns the Parquet snapshot for that date.

Response 200 OK
Binary file stream with Content-Type set to the format’s MIME type:

application/octet-stream for Parquet
text/csv for CSV
application/json for JSON

Examples

# Download CPI as Parquet
curl -O "https://api.imibare.org/v1/data/rw.nisr.cpi.monthly/download"

# Download as CSV
curl -o cpi.csv "https://api.imibare.org/v1/data/rw.nisr.cpi.monthly/download?format=csv"

# Download a specific version
curl -O "https://api.imibare.org/v1/data/rw.nisr.cpi.monthly/download?version=2025-06-01"

GET /v1/search

Full-text search across dataset names and descriptions.

Request

GET /v1/search?q={query}

Query parameters

Parameter	Type	Required	Description
`q`	string	Yes	Search query

Response 200 OK

{
  "query": "exchange rate",
  "results": [
    {
      "id": "rw.bnr.fx.daily",
      "name": "Official exchange rates (daily)",
      "country": "RW",
      "country_name": "Rwanda",
      "institution": "National Bank of Rwanda",
      "institution_slug": "bnr",
      "topic": "exchange-rates",
      "frequency": "daily",
      "pipeline": "automated"
    }
  ],
  "total": 1
}

Search matches on id, name, country code and name, institution, topic, notes and tags, so q=kenya finds Kenyan datasets whose names never say Kenya.

Example

curl "https://api.imibare.org/v1/search?q=exchange+rate"

Error responses

All errors return JSON with an error field.

Status	Meaning
`400 Bad Request`	Invalid query parameter or format value
`404 Not Found`	Dataset ID not found
`500 Internal Server Error`	R2 read failure or upstream error

Rate limits

No authentication is required, and no formal rate limits are enforced. Treat the API as a public service and avoid automated bulk downloads. Use the Parquet files directly from R2 instead.