dane.gov.pl Scraper - Poland Open Data Portal API

dataset-row

One dane.gov.pl dataset with its full 49-field record: Polish title and description in text and HTML, portal and API URLs, publishing institution with its type, city and website, EU DCAT themes with their codes, the legacy portal category, keywords, geographic regions, every file format and delivery type it offers, resource and showcase counts, licence name, code, plain-English description and every reuse condition, update frequency, openness score, the high value, EU high value, dynamic, research and promoted flags, view and download counts, the external catalogue it was harvested from, supplementary documents, and the created, modified, verified and resource-modified timestamps.

resource-row

One distribution: the downloadable file, API or service inside a dataset, with its format, media type, byte size, direct download URL, CSV and JSON-LD conversions, openness score, language, data date, the protected-data and high-value flags, the special-sign legend the publisher used for missing cells, and the dataset and institution it belongs to.

institution-row

One of the 7,476 publishing bodies with its REGON statistical number, institution type, postal address, city, email, telephone, fax, ePUAP inbox, electronic delivery address, website, dataset and resource counts, and the external catalogues it is harvested from.

showcase-row

One of the 119 public services, websites and apps built on dane.gov.pl data, with its author, external URL, licence type, platform, mobile and desktop store links, keywords, screenshots and how many datasets it reuses.

history-row

One entry from the portal audit log: the action, the timestamp, the editor account, the record it touched, which fields changed and the full before-and-after difference the portal recorded.

broken-link-row

One row from the portal's own dead-link report: the institution, the dataset, the resource and the URL that stopped resolving, with the date the report was last rebuilt.

news-row

One portal news article with its title, full text, keywords, author and publication date. News is only reachable through the cross-model search index, never through a listing endpoint.

knowledge-base-row

One knowledge base article: the portal guidance, reports and training material for data publishers, with its category, keywords and full text.

search-row

One hit from the cross-model search index, which is the only place the portal filters on licence, update frequency and date, and the only way to query datasets, resources, institutions, showcases, news and knowledge base in a single pass.

tabular-row

One real row of data out of a resource's own table, with the publisher's column headers restored instead of the API's col1, col2, col3, plus the row number and the date that row was last refreshed.

facet-row

One aggregation bucket: a format, EU theme, portal category, delivery type, openness score or publisher with its dataset count and its share of the filtered catalogue. This is the shape of the catalogue rather than its contents.

dataset-resources

Opt in. Every file inside one dataset attached to its row, each with its format, byte size, openness score, download URL, data date and visualisation types. Billed only when the dataset actually returned files.

dataset-showcases

Opt in. The public services and apps that reuse one dataset, attached to its row with their authors, URLs and licence types. Billed only when the dataset actually has one.

dataset-history

Opt in. The portal audit trail for one dataset attached to its row: every update with its timestamp, the editor account and the fields that changed, plus the total change count. Billed only when the audit log had entries.

institution-profile

Opt in. The publishing body's complete record attached to the row: REGON, postal address, email, telephone, fax, ePUAP inbox, electronic delivery address, website, dataset and resource counts and harvest sources. Billed only when the profile came back.

tabular-preview

Opt in. The column schema with the publisher's real header names and types, the total row count, and the first rows of a resource's table attached to its row. Only tabulated resources have any, and the rest are not billed.

file-probe

Opt in. One HEAD request against the actual download URL recording the HTTP status, content type and byte size, which is how you find files that stopped resolving before the portal notices. Billed only when the check completed.