@comunica/actor-http-fetch

Comunica Fetch HTTP Actor

An HTTP actor that uses fetch to perform HTTP requests.

This module is part of the Comunica framework, and should only be used by developers that want to build their own query engine.

When this actor is used, a custom fetch implementation may be provided via the context (fetch). If none is provided, the global fetch implementation from the runtime or a polyfill is used.

Click here if you just want to query with Comunica.

Install

$ yarn add @comunica/actor-http-fetch

Configure

After installing, this package can be added to your engine's configuration as follows:

{
  "@context": [
    "https://linkedsoftwaredependencies.org/bundles/npm/@comunica/actor-http-fetch/^4.0.0/components/context.jsonld"
  ],
  "actors": [
    {
      "@id": "urn:comunica:default:http/actors#fetch",
      "@type": "ActorHttpFetch"
    }
  ]
}

Config Parameters

• agentOptions: The agent JSON options for the HTTP agent in Node.js environments. (optional)

Changelog

v4.2.0 - 2025-04-29

Added

• Add HTTP rate limit actor
• Add workaround for DBPedia SPARQL HTTP status 405 with Retry-After
• Add VoID metadata extract actor and cardinality estimation for SPARQL sources
• Allow SPARQL source actor to skip COUNTs for large queries
• Allow sending COUNT queries to be disabled for SPARQL endpoints
• Allow forcing SPARQL source type from config
• Add join sorter based on join selectivities
• Report the url when a fetch fails
• Add distinctConstruct key to IQueryContextCommon

Changed

• Detect SPARQL endpoints with trailing slashes, Closes #1529
• Skip COUNT queries to singular SPARQL source
• Skip unnecessary SPARQL SD requests for a single source
• Update to rdf-streaming-store v2
• Try action.mediaType before doing heuristics for media type
• Avoid unnecessary parse attempts after dereferencing fails
• Avoid printing all metadata in join logging (#1509)
• Avoid creation of intermediary triple objects for in-memory stores

Fixed

• Fix variable names in ActorOptimizeQueryOperationPruneEmptySourceOperations
• Fix incorrect id of actor-query-source-hypermedia-identify-sparql
• Fix GraphQL queries being sent to SPARQL endpoints, Closes #1525
• Fix HTTP request count tracking, Closes #1526
• Fix bad plans sometimes being chosen due to requestTime in files
• Fix regressions of initialBindings with BIND clauses
• Fix queries with complex property paths not terminating
• Avoid flattening joins with input-specific metadata (#1522)
• Fix broken Dockerfile for query-sparql

<a name="v4.1.0"></a>