Article Text with Paragraph Sources
Pricing
$1.50 / 1,000 complete articles
Article Text with Paragraph Sources
Extract permitted static article text with source URL and per-block location. Returns explicit reasons for restricted or ambiguous pages.
Pricing
$1.50 / 1,000 complete articles
Rating
0.0
(0)
Developer
Data Lantern
Maintained by CommunityActor stats
0
Bookmarked
2
Total users
1
Monthly active users
2 days ago
Last modified
Categories
Security
Limited permissionsShare
Supply 1–25 final public URLs to articles you own, or articles you have permission to extract and store. Set rightsConfirmed to true only if those permissions cover the submitted URLs. Output rows contain plain text and the source URL. Each block includes its HTML tag and its one-based ordinal among elements of that tag in the original parsed page. No article text comes from JavaScript or JSON-LD.
For example, a publisher can supply {"rightsConfirmed":true,"urls":["https://your-publication.example/story"]} after replacing the illustrative URL with an authorized one. A full extraction is marked extracted. Pages with no unique article, access restrictions, unverifiable robots rules, or unsupported formats are skipped with a reason. If an article reaches the output limit it is marked partial. A row's zero-based index corresponds to the input list even when its URL is rejected.
The static parser considers a single article or articleBody, or a main region with a heading and paragraphs. It drops common navigation, advertising and explicitly hidden elements. Sites without recognizable structure, dynamic content and unrecognized boilerplate can produce gaps. It does not render responsive CSS, execute scripts, follow redirects, log in or try alternate access paths. Explicit subscription/login restrictions and negative access metadata cause abstention. Customers are responsible for supplying URLs they have rights to process.
Limits: 128 KiB robots file and 512 KiB page, ten-second combined source-read deadline per URL, 256 blocks and 64,000 output characters per article. Processing is sequential, stops admitting new URLs after two minutes or three consecutive non-successes, and observes the configured event budget.
Candidate pricing: one article event per full extraction at $1.50/1,000. Partial and skipped rows incur no result event. No startup or automatic dataset-item fee is proposed. Local testing does not bill an Apify account.
The built-in fictional example works offline. Repeated URL fragments share one result and retain original input positions.
