HSENI Publications Metadata Scraper
Pricing
from $2.00 / 1,000 recognized consultation/publication/alert metadata records
HSENI Publications Metadata Scraper
Northern Ireland official publication metadata, source dates, topics and document links. Bounded recent discovery; documents never downloaded.
Pricing
from $2.00 / 1,000 recognized consultation/publication/alert metadata records
Rating
0.0
(0)
Developer
Vixendata
Maintained by CommunityActor stats
0
Bookmarked
2
Total users
1
Monthly active users
10 days ago
Last modified
Categories
Share
Export official Northern Ireland safety publication metadata into an Apify dataset for regulatory monitoring, research inventories and document-link feeds. HSENI is the Health and Safety Executive for Northern Ireland, a separate regulator from Great Britain's HSE.
Scope
Reads only the first unfiltered https://www.hseni.gov.uk/publications catalog page and optional explicit publication detail URLs. Consultations are not supported. It does not paginate, filter via query strings, claim historical completeness, download documents, interpret legislation or score compliance. The catalog is a recent snapshot, not an exhaustive archive. Source metadata is observational, not legal advice.
At most 30 detail pages are inspected and 30 records emitted. Explicit URLs are prioritized, followed by first-page discovery. One canonical detail page yields one record, regardless of its document count. Filters do not expand discovery. Schedule repeated runs in Apify and compare canonical URLs downstream if monitoring is desired; this Actor does not persist change history.
Input
Defaults: discoverPublications true, maxItems 10. sourceUrls accepts up to 30 plain HTTPS www.hseni.gov.uk/publications/slug URLs. sinceDate is an inclusive YYYY-MM-DD publication-date filter; undated records are excluded when filtering. topicKeywords matches any keyword as a case-insensitive substring in official topics. No custom host, query, consultation or document-fetch options are supported.
Example:
{"discoverPublications":true,"maxItems":5,"sinceDate":"2026-01-01","topicKeywords":["Mining"]}
Dataset
Each record contains source, contentType (publication), url, title, publishedDate (ISO date or null), topics, publicationType (null when absent), summary, documentLinks (title/url/type/sizeText), observedAt (UTC retrieval time), attribution and diagnostics. Documents are links only and never fetched. Some source fields can be null; diagnostics identify unrecognized or missing dates.
RUN_REPORT in the default key-value store records status, error, requests, checked, emitted, filtered, scope and documentsFetched. Source denials, robots prohibitions, redirects and layout errors stop the run and mark it failed; previously written rows may remain. Inspect this report before treating output as a complete run.
Responsible access and reuse
The fixed HTTPS host and publication-only paths are enforced. robots.txt is checked before HTML fetches, including wildcard restrictions. Sequential requests have a minimum two-second delay and honor longer source crawl delays up to 60 seconds; larger delays stop. Request-rate directives stop conservatively for manual review. Invalid robots responses stop the run. Requests have 25-second timeouts and 2 MB limits. No retries, redirects, configurable scraper proxies, CAPTCHAs or alternative-host workarounds are used. Standard runtime network egress is used. Source robots prohibits document downloads and query/faceted URLs, so neither is fetched. Returned document URLs are references, not permission to crawl them.
Contains public sector information from the Health and Safety Executive for Northern Ireland licensed under the Open Government Licence v3.0, pursuant to the HSENI Crown copyright notice. Attribute HSENI and source URLs when reusing output. Third-party material, personal data, logos and maps are excluded from this reuse claim. No HSENI endorsement is implied. Assess intended downstream reuse independently.
Run and test
Python 3.12: install requirements.txt and run python main.py, or build the included Apify Dockerfile. Configure input on Apify and export the default dataset in JSON/CSV.
python -m unittest discover -s tests -v runs portable tests against bundled original official publication HTML, catalog and robots snapshots collected 2026-10-05. Covers dates, discovery, URL guards (including excluded consultation paths), robots, denied access and actual Apify SDK fixture-positive and denied-source runs. The SDK test emits one record into temporary local storage without cloud charges. Fixture tests alone are not evidence of current live availability.