What export formats are available?#

The APITube News API can return your search results in eight formats besides JSON: CSV, TSV, XLSX, XML, RSS, JSONL, NDJSON and Parquet. You pick one by adding the export parameter to any article-returning request — the filters stay the same, only the output changes. Leave export off, or set export=json, and you get the default JSON response.

curl "https://api.apitube.io/v1/news/everything?title=tesla&export=csv&api_key=YOUR_API_KEY" -o results.csv

Which export formats does APITube support?#

Every format below is selected with the same export parameter. The API sets the matching Content-Type and, for downloadable files, a Content-Disposition: attachment header so a browser saves the file instead of displaying it.

export value	Output	Content-Type
json (default)	Standard paginated JSON response	application/json
csv	Comma-separated spreadsheet, results.csv	text/csv
tsv	Tab-separated spreadsheet, results.tsv	text/tab-separated-values
xlsx	Native Excel workbook, sheet named Articles	application/vnd.openxmlformats-officedocument.spreadsheetml.sheet
xml	Nested XML document	application/xml
rss	RSS 2.0 feed you can open in a reader	application/xml
jsonl	One JSON object per line, results.jsonl	application/x-ndjson
ndjson	Alias for jsonl — identical output	application/x-ndjson
parquet	Typed columnar file, results.parquet	application/octet-stream

How do I change the export format?#

Add export=<format> anywhere in your query string. The export step runs after your normal search, so the file holds exactly the articles your filters return — keywords, language, date range and any other filter still apply. Build the search you want first, then append the format:

curl "https://api.apitube.io/v1/news/everything?category.id=medtop:04000000&export=xlsx&api_key=YOUR_API_KEY" -o financial_news.xlsx

If you pass a value the API doesn’t recognise, it ignores it and falls back to the normal JSON response rather than erroring out. You can send the key as the api_key query parameter (shown above) or as an X-API-Key header.

Which formats are flat, and which keep the full structure?#

This is the main difference to know when you choose a format:

• Flat tabular formats — csv, tsv, xlsx, parquet. Each article becomes one row with the same 28 fixed columns (ID, Href, PublishedAT, Title, Description, Body, Language, Author, Image, Categories, Topics, Industries, Entities, Source, Sentiment, Summary, Keywords, Links, Media, Story, IsDuplicate, IsAccessibleForFree, IsBreaking, ReadTime, SentencesCount, ParagraphsCount, WordsCount, CharactersCount). Nested objects like source or entities are written as JSON text inside their cell. Parquet additionally keeps typed columns — booleans, integers and strings — which is why it loads cleanly into data tools.
• Structured formats — xml, jsonl, ndjson. These keep the full nested article shape. jsonl/ndjson write one complete JSON object per line, so you can stream the file line by line; xml mirrors the same nesting as the JSON response.

So if you need a spreadsheet, pick csv, tsv or xlsx. If you need the complete enrichment (entities, per-field sentiment, links, media), pick jsonl or xml. For analytics pipelines, parquet keeps the data typed — see how to export for data pipelines.

Which endpoints support export?#

The export parameter works on the article-returning news endpoints — not only everything. You can add it to /v1/news/everything, /v1/news/top-headlines, /v1/news/local, /v1/news/story/:id, /v1/news/article, /v1/news/entity, /v1/news/category, /v1/news/topic and /v1/news/industry. The same format list and column layout apply on each one. The full parameter reference lives in the News API parameters documentation.

---

Common Questions#

• How many articles does one export return?
• Does exporting cost extra requests?
• What’s the difference between JSONL and NDJSON?
• Which format should I choose?

How many articles does one export return?#

An export returns one page of results, just like a JSON search. Page size is controlled by per_page, which defaults to 100 and has a maximum of 250; asking for more returns a 400 error (ER0171, “Limit is out of range”). To export a larger result set, page through it with page and per_page and combine the files yourself:

curl "https://api.apitube.io/v1/news/everything?title=tesla&per_page=250&page=2&export=csv&api_key=YOUR_API_KEY" -o results_page2.csv

Does exporting cost extra requests?#

No. An export is the same endpoint call with one extra parameter, so it counts as a single API request — the same as the JSON search would. The format you choose does not change the cost.

What’s the difference between JSONL and NDJSON?#

There is none. export=ndjson is an alias for export=jsonl: both produce the same line-delimited JSON output with a Content-Type: application/x-ndjson header. Use whichever name your tooling expects.

Which format should I choose?#

• JSON — default, best for direct API consumption in code.
• CSV / XLSX — open in Excel, Google Sheets or Numbers without an import step.
• TSV — same as CSV but tab-separated, useful when article fields already contain commas.
• RSS — subscribe to a search in a feed reader; see how to turn a search into an RSS feed.
• XML — nested output for systems that consume XML.
• JSONL / NDJSON / Parquet — load into data pipelines and analytics tools.