Public publication API

Read OpenForex's published resource catalog and retrieve source Markdown over HTTPS. These read-only endpoints expose the same articles available in the website's resource library.

Resource catalog

GET https://openforex.io/?format=catalog

The response uses application/json; charset=utf-8. schemaVersion is 1. total is the number of resources, and resources is an array of catalog entries.

Each entry contains id, title, collection, summary, readingMinutes, url, and markdownUrl. Additive source fields are materialType, sourceVersion, sourceUpdated, contentSha256, and referenceUrls. sourceVersion and sourceUpdated are null when the source does not supply them. contentSha256 identifies the exact downloadable Markdown, including its bytes and line endings; it is not an approval or an archived edition. Reading time is an estimate based on article word count. The catalog contains the current published resources, not a historical snapshot or a standards-approval registry.

const response = await fetch('/?format=catalog');
if (!response.ok) throw new Error('Catalog request failed');
const catalog = await response.json();
if (catalog.schemaVersion !== 1) throw new Error('Unsupported schema');
console.log(catalog.resources.map(item => item.title));

This browser example runs on the OpenForex origin. Cross-origin browser access is not enabled by this endpoint; another website can retrieve the catalog through its own server if appropriate.

Article and Markdown routes

GET https://openforex.io/?page=seed-content/guides/value-comparison
GET https://openforex.io/?page=seed-content/guides/value-comparison&download=1

The first route returns the rendered article. The second returns a Markdown attachment. Treat a resource id as an opaque catalog identifier and use URL encoding when building a request.

Unknown page identifiers return HTTP 404. Unknown format values return HTTP 404 with a JSON error object. An unknown article with download=1 still returns the normal HTML 404 page, so inspect status and content type before saving a response.

Integration behavior

The catalog requires no account and has no write endpoint. No pagination or authentication token is part of version 1. Handle HTTP errors, timeouts, and malformed responses in your client. Cache a successful catalog response when repeated fetching is unnecessary.

Preserve source attribution, article version information where present, and the date you retrieved the content. A current URL can change as the publication is updated; retain an actual copy when an experiment depends on an exact text.

Choose the correct API

For numeric calculations, use the TVX SDK. For optional Wiki.js authoring integrations, use the separate authoring API reference. Wiki.js examples describe an optional instance; the live publication catalog is provided by this PHP website.

Search and caching

The website loads its full-text index on demand from ?format=search. Its schemaVersion is 1, with a documents array containing id, title, group, url, summary, minutes, and text. The index covers the current publication. HTML pages include only lightweight discovery metadata, so readers do not download every article before reading one.

Both JSON endpoints supply an ETag and a five-minute public cache lifetime. A matching If-None-Match request returns HTTP 304. Clients should tolerate new object fields. JSON and Markdown responses carry noindex directives; canonical HTML articles remain the primary search results.

For attribution and evidence boundaries, see the sourcing policy.