Migrating from Currents API

Replace /v1/search and /v2/search with /v1/news/everything. The boolean query language carries over almost unchanged, and the category taxonomy maps one-to-one — the cleanest category mapping in this section. The blocker, stated up front, is Russian.

Why teams switch

	Currents API	APITube
Article text	description only — no full-text field	Full body and body_html
Enrichment	A category array	Entities with per-entity sentiment, sentiment on title/body/overall, IPTC plus topic and industry axes, readability, publisher bias
Results per page	300 max, but (page_number − 1) × page_size capped at 5 000	250 per request, no offset ceiling
Category taxonomy	16 canonical V2 categories	The same IPTC top level, plus thousands of child codes

Your first request

# Currents API
curl "https://api.currentsapi.services/v2/search?keywords=technology&language=en&country=US&category=science_technology&page_size=100" \
  -H "Authorization: Bearer YOUR_KEY"

# APITube
curl "https://api.apitube.io/v1/news/everything?title=technology&language.code=en&source.country.code=us&category.id=medtop:13000000&per_page=100" \
  -H "X-API-Key: YOUR_API_KEY"

The categories map one-to-one

Currents' V2 taxonomy is built on IPTC MediaTopics top-level codes, so the conversion is mechanical rather than judgement-based:

Currents V2	IPTC code
economy_business_finance	medtop:04000000
crime_law_justice	medtop:02000000
science_technology	medtop:13000000
sport	medtop:15000000
health	medtop:07000000

All 14 filterable categories translate this way. The full table — including the 47 legacy V1 categories — is in the migration kit.

Parameter mapping

Currents API	APITube	Note
keywords	title	Narrows to headlines — Currents matches the description too
query	query	Boolean syntax survives; every bare term needs a field prefix
language	language.code	See the language section below
country	source.country.code
category	category.id	One-to-one, max 3 values
domain	source.domain	Max 3
domain_not	ignore.source.domain
start_date / end_date	published_at.start / .end
page_size	per_page	250 max on APITube
page_number	page
type	—	No article-type filter

keywords takes priority over query when both are present on Currents. The shims in the kit reproduce that precedence.

Boolean queries

Both sides use AND / OR / NOT, quoted phrases and parentheses, and both bind AND tighter than OR. The only change is the field prefix:

# Currents
query=("AI" OR "Machine Learning") AND NOT "Ethics"

# APITube
query=(title:"AI" OR title:"Machine Learning") AND NOT title:"Ethics"

What does not carry over

Russian. language.code=ru returns 400 ER0237. Currents lists Russian among its 23 languages, and there is no substitute. If Russian coverage matters to you, this migration does not work.

Malay. Currents uses the non-standard code msa; ISO 639-1 says ms. APITube supports neither.

Wildcards. No wildcard support, and title= accepts one without error while returning the entire index with a 200. Expand into query=title:(term1 OR term2).

Article type. Currents' type filter has no APITube counterpart.

Description search. Currents matches keywords against the description as well as the title. APITube's title= is headlines-only, so expect fewer results — and compensate by filtering the returned body, or by using organization.name instead of a keyword.

What you gain

Full article body instead of a description, named entities with sentiment scored towards each one, sentiment on the headline and the body separately, IPTC MediaTopics with child codes rather than 16 top-level buckets, topic and industry axes, publisher political bias and Open Page Rank, readability scores, and export to eight formats.

Next steps

• currents-apitube-migration-kit — every /v1 and /v2 parameter mapped, all 16 V2 and 47 V1 categories with verified IPTC codes, all 23 language codes checked individually against the live API, a drop-in shim for Node.js and Python, and an AI prompt for bulk conversion
• Parameters reference
• Common migration issues